refactor(prometheus): serve the Prometheus query API from a /prometheus prefix

Moves the Prometheus HTTP query API out of the legacy query-service handler
into pkg/prometheus/promapi, served under /prometheus:

- GET|POST /prometheus/api/v1/query_range and /query: Prometheus parameter
  parsing (float-unix or RFC3339 times, float-seconds or duration-string
  durations), the Prometheus error envelope ({status, errorType, error}
  with 400/422/503), the 11,000-point cap, documented in openapi.yml.
- Removes the legacy GET /api/v1/query_range and /api/v1/query handlers
  and their now-dead plumbing (parseMetricsTime, parseMetricsDuration,
  parseInstantQueryMetricsRequest, parseQueryRangeRequest,
  GetInstantQueryMetricsResult, InstantQueryMetricsParams).
  GetQueryRangeResult stays - legacy dashboard queriers use it.

This is the breaking slice of the stack, deliberately last: everything
before it is invisible to API consumers.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

.github/workflows/integrationci.yaml

@@ -54,6 +54,7 @@ jobs:
           - querierscalar
           - queriercommon
           - rawexportdata
+          - promqlconformance
           - querierauthz
           - role
           - rootuser

docs/api/openapi.yml

@@ -24695,6 +24695,149 @@ paths:
       summary: Replace variables
       tags:
       - querier
+  /prometheus/api/v1/query:
+    get:
+      deprecated: false
+      description: Evaluate a PromQL expression at a single instant. Request and
+        response follow the Prometheus HTTP API (https://prometheus.io/docs/prometheus/latest/querying/api/);
+        the /prometheus prefix distinguishes these PromQL-only endpoints from the
+        SigNoz query APIs. Also accepts POST with form-encoded parameters.
+      operationId: PrometheusInstantQuery
+      parameters:
+      - description: PromQL expression to evaluate
+        in: query
+        name: query
+        required: true
+        schema:
+          type: string
+      - description: 'Evaluation timestamp: float unix seconds or RFC3339. Defaults
+          to the server''s current time.'
+        in: query
+        name: time
+        schema:
+          type: string
+      - description: 'Evaluation timeout: float seconds or a Prometheus duration
+          string (e.g. 30s).'
+        in: query
+        name: timeout
+        schema:
+          type: string
+      - description: Set to any value to include query statistics in the response.
+        in: query
+        name: stats
+        schema:
+          type: string
+      responses:
+        "200":
+          content:
+            application/json:
+              schema:
+                properties:
+                  data:
+                    properties:
+                      result: {}
+                      resultType:
+                        enum:
+                        - matrix
+                        - vector
+                        - scalar
+                        - string
+                        type: string
+                      stats: {}
+                    type: object
+                  status:
+                    enum:
+                    - success
+                    type: string
+                type: object
+          description: Query evaluated successfully
+        "400":
+          description: Unparsable expression or parameters (errorType bad_data)
+        "422":
+          description: Expression failed to evaluate (errorType execution)
+        "503":
+          description: Query timed out or was canceled
+      summary: Prometheus instant query
+      tags:
+      - prometheus
+  /prometheus/api/v1/query_range:
+    get:
+      deprecated: false
+      description: Evaluate a PromQL expression over a range of time on a fixed
+        step grid. Request and response follow the Prometheus HTTP API
+        (https://prometheus.io/docs/prometheus/latest/querying/api/); the
+        /prometheus prefix distinguishes these PromQL-only endpoints from the
+        SigNoz query APIs. Grids are capped at 11,000 points per series. Also
+        accepts POST with form-encoded parameters.
+      operationId: PrometheusRangeQuery
+      parameters:
+      - description: PromQL expression to evaluate
+        in: query
+        name: query
+        required: true
+        schema:
+          type: string
+      - description: 'Start timestamp: float unix seconds or RFC3339.'
+        in: query
+        name: start
+        required: true
+        schema:
+          type: string
+      - description: 'End timestamp: float unix seconds or RFC3339.'
+        in: query
+        name: end
+        required: true
+        schema:
+          type: string
+      - description: 'Grid step: float seconds or a Prometheus duration string
+          (e.g. 30s). Must be positive.'
+        in: query
+        name: step
+        required: true
+        schema:
+          type: string
+      - description: 'Evaluation timeout: float seconds or a Prometheus duration
+          string (e.g. 30s).'
+        in: query
+        name: timeout
+        schema:
+          type: string
+      - description: Set to any value to include query statistics in the response.
+        in: query
+        name: stats
+        schema:
+          type: string
+      responses:
+        "200":
+          content:
+            application/json:
+              schema:
+                properties:
+                  data:
+                    properties:
+                      result: {}
+                      resultType:
+                        enum:
+                        - matrix
+                        type: string
+                      stats: {}
+                    type: object
+                  status:
+                    enum:
+                    - success
+                    type: string
+                type: object
+          description: Query evaluated successfully
+        "400":
+          description: Unparsable expression or parameters, or a grid past the 11,000-point
+            cap (errorType bad_data)
+        "422":
+          description: Expression failed to evaluate (errorType execution)
+        "503":
+          description: Query timed out or was canceled
+      summary: Prometheus range query
+      tags:
+      - prometheus
 servers:
 - description: The fully qualified URL to the SigNoz APIServer.
   url: https://{host}:{port}{base_path}

frontend/src/api/ErrorResponseHandlerForGeneratedAPIs.ts

@@ -2,6 +2,29 @@ import { RenderErrorResponseDTO } from 'api/generated/services/sigNoz.schemas';
 import { AxiosError } from 'axios';
 import APIError from 'types/api/error';
 
+// The wire shape these handlers can actually rely on. The generated
+// RenderErrorResponseDTO marks code/message/url/errors as required, but the
+// server omits any of them even on valid errors (e.g. a 400 with just a
+// message), so a present `error` object is all the guard can guarantee.
+type ErrorEnvelope = {
+	error: {
+		code?: string;
+		message?: string;
+		url?: string;
+		errors?: { message?: string }[];
+	};
+};
+
+function isErrorEnvelope(data: unknown): data is ErrorEnvelope {
+	return (
+		typeof data === 'object' &&
+		data !== null &&
+		'error' in data &&
+		typeof (data as ErrorEnvelope).error === 'object' &&
+		(data as ErrorEnvelope).error !== null
+	);
+}
+
 // @deprecated Use convertToApiError instead
 export function ErrorResponseHandlerForGeneratedAPIs(
 	error: AxiosError<RenderErrorResponseDTO>,
@@ -10,15 +33,29 @@ export function ErrorResponseHandlerForGeneratedAPIs(
 	// The request was made and the server responded with a status code
 	// that falls out of the range of 2xx
 	if (response) {
+		// The body isn't guaranteed to be an error envelope — e.g. a gateway 5xx
+		// with an HTML/empty body during a deploy. Verify the shape before reading
+		// it; otherwise synthesize a consistent error from the status.
+		const data: unknown = response.data;
+		if (isErrorEnvelope(data)) {
+			const { code, message, url, errors } = data.error;
+			throw new APIError({
+				httpStatusCode: response.status || 500,
+				error: {
+					code: code ?? '',
+					message: message ?? '',
+					url: url ?? '',
+					errors: (errors ?? []).map((e) => ({ message: e.message ?? '' })),
+				},
+			});
+		}
 		throw new APIError({
 			httpStatusCode: response.status || 500,
 			error: {
-				code: response.data.error.code,
-				message: response.data.error.message,
-				url: response.data.error.url ?? '',
-				errors: (response.data.error.errors ?? []).map((e) => ({
-					message: e.message ?? '',
-				})),
+				code: 'UPSTREAM_UNAVAILABLE',
+				message: error.message || 'Something went wrong',
+				url: '',
+				errors: [],
 			},
 		});
 	}
@@ -62,9 +99,7 @@ export function convertToApiError(
 	return new APIError({
 		httpStatusCode: response?.status || error.status || 500,
 		error: {
-			code:
-				errorData?.code ||
-				String(response?.status || error.code || 'unknown_error'),
+			code: errorData?.code || 'UPSTREAM_UNAVAILABLE',
 			message:
 				errorData?.message ||
 				response?.statusText ||

frontend/src/api/ErrorResponseHandlerV2.ts

@@ -2,19 +2,38 @@ import { AxiosError } from 'axios';
 import { ErrorV2Resp } from 'types/api';
 import APIError from 'types/api/error';
 
+function isErrorV2Resp(data: unknown): data is ErrorV2Resp {
+	return (
+		typeof data === 'object' &&
+		data !== null &&
+		'error' in data &&
+		typeof (data as ErrorV2Resp).error?.code === 'string'
+	);
+}
+
 // reference - https://axios-http.com/docs/handling_errors
 export function ErrorResponseHandlerV2(error: AxiosError<ErrorV2Resp>): never {
 	const { response, request } = error;
 	// The request was made and the server responded with a status code
 	// that falls out of the range of 2xx
 	if (response) {
+		// response.data isn't guaranteed to be a V2 envelope (e.g. a gateway 5xx
+		// with an HTML/empty body during a deploy), so verify the shape first.
+		const data: unknown = response.data;
+		if (isErrorV2Resp(data)) {
+			const { code, message, url, errors } = data.error;
+			throw new APIError({
+				httpStatusCode: response.status || 500,
+				error: { code, message, url, errors },
+			});
+		}
 		throw new APIError({
 			httpStatusCode: response.status || 500,
 			error: {
-				code: response.data.error.code,
-				message: response.data.error.message,
-				url: response.data.error.url,
-				errors: response.data.error.errors,
+				code: 'UPSTREAM_UNAVAILABLE',
+				message: error.message || 'Something went wrong',
+				url: '',
+				errors: [],
 			},
 		});
 	}

frontend/src/api/__tests__/ErrorResponseHandlerV2.test.ts

@@ -0,0 +1,126 @@
+import { ErrorResponseHandlerV2 } from 'api/ErrorResponseHandlerV2';
+import { AxiosError } from 'axios';
+import { ErrorV2Resp } from 'types/api';
+import APIError from 'types/api/error';
+
+function asAxiosError(partial: Partial<AxiosError>): AxiosError<ErrorV2Resp> {
+	return partial as AxiosError<ErrorV2Resp>;
+}
+
+// ErrorResponseHandlerV2 always throws — capture the APIError so assertions stay
+// unconditional.
+function runHandler(error: AxiosError<ErrorV2Resp>): APIError {
+	try {
+		ErrorResponseHandlerV2(error);
+	} catch (thrown) {
+		return thrown as APIError;
+	}
+	throw new Error('expected ErrorResponseHandlerV2 to throw');
+}
+
+type ExpectedError = {
+	httpStatusCode: number;
+	code: string;
+	message: string;
+	errors: { message: string }[];
+};
+
+// One row per response shape the handler must normalize. New shapes (with
+// different bodies) can be added here without a new test block.
+const cases: {
+	name: string;
+	error: AxiosError<ErrorV2Resp>;
+	expected: ExpectedError;
+}[] = [
+	{
+		name: 'well-formed V2 error envelope',
+		error: asAxiosError({
+			message: 'Request failed with status code 400',
+			response: {
+				status: 400,
+				data: {
+					error: {
+						code: 'bad_request',
+						message: 'Invalid dashboard payload',
+						url: 'https://signoz.io/docs',
+						errors: [{ message: 'name is required' }, { message: 'name too long' }],
+					},
+				},
+			} as AxiosError['response'],
+		}),
+		expected: {
+			httpStatusCode: 400,
+			code: 'bad_request',
+			message: 'Invalid dashboard payload',
+			errors: [{ message: 'name is required' }, { message: 'name too long' }],
+		},
+	},
+	{
+		// Regression: during a deployment the gateway returns a 5xx with a
+		// non-envelope body. Reading response.data.error.code used to throw a
+		// TypeError from inside the handler itself. See engineering-pod#5760.
+		name: '5xx with a non-envelope HTML body',
+		error: asAxiosError({
+			message: 'Request failed with status code 503',
+			response: {
+				status: 503,
+				data: '<html><body>503 Service Temporarily Unavailable</body></html>',
+			} as AxiosError['response'],
+		}),
+		expected: {
+			httpStatusCode: 503,
+			code: 'UPSTREAM_UNAVAILABLE',
+			message: 'Request failed with status code 503',
+			errors: [],
+		},
+	},
+	{
+		name: '5xx with an empty body',
+		error: asAxiosError({
+			message: 'Request failed with status code 502',
+			response: {
+				status: 502,
+				data: undefined,
+			} as AxiosError['response'],
+		}),
+		expected: {
+			httpStatusCode: 502,
+			code: 'UPSTREAM_UNAVAILABLE',
+			message: 'Request failed with status code 502',
+			errors: [],
+		},
+	},
+	{
+		name: 'no response received (network error)',
+		error: asAxiosError({
+			message: 'Network Error',
+			code: 'ERR_NETWORK',
+			name: 'AxiosError',
+			request: {},
+		}),
+		expected: {
+			httpStatusCode: 500,
+			code: 'ERR_NETWORK',
+			message: 'Network Error',
+			errors: [],
+		},
+	},
+];
+
+describe('ErrorResponseHandlerV2', () => {
+	it.each(cases)(
+		'normalizes $name into a consistent APIError',
+		({ error, expected }) => {
+			const apiError = runHandler(error);
+
+			expect(apiError).toBeInstanceOf(APIError);
+			expect(apiError.getHttpStatusCode()).toBe(expected.httpStatusCode);
+			expect(apiError.getErrorCode()).toBe(expected.code);
+			expect(apiError.getErrorMessage()).toBe(expected.message);
+			// The sub-error messages feed several parts of the UI, so assert them.
+			expect(apiError.getErrorDetails().error.errors).toStrictEqual(
+				expected.errors,
+			);
+		},
+	);
+});

frontend/src/components/FieldsSelector/OtherFields.tsx

@@ -8,7 +8,6 @@ import { buildCompositeKey } from 'container/OptionsMenu/utils';
 import { useGetQueryKeySuggestions } from 'hooks/querySuggestions/useGetQueryKeySuggestions';
 import {
 	FieldContext,
-	FieldDataType,
 	SignalType,
 	TelemetryFieldKey,
 } from 'types/api/v5/queryRange';
@@ -55,7 +54,7 @@ function OtherFields({
 				key: buildCompositeKey(attr.name, attr.fieldContext as string),
 				signal: attr.signal as SignalType,
 				fieldContext: attr.fieldContext as FieldContext,
-				fieldDataType: attr.fieldDataType as FieldDataType,
+				fieldDataType: attr.fieldDataType,
 			}),
 		);
 		const addedIds = new Set(

frontend/src/components/QueryBuilderV2/QueryV2/QueryAggregation/QueryAggregationSelect.tsx

@@ -25,12 +25,12 @@ import CodeMirror, {
 } from '@uiw/react-codemirror';
 import { Button, Popover, Tooltip } from 'antd';
 import { getKeySuggestions } from 'api/querySuggestions/getKeySuggestions';
-import { QUERY_BUILDER_KEY_TYPES } from 'constants/antlrQueryConstants';
 import { QueryBuilderKeys } from 'constants/queryBuilder';
 import { tracesAggregateOperatorOptions } from 'constants/queryBuilderOperators';
 import { useIsDarkMode } from 'hooks/useDarkMode';
 import { Info, TriangleAlert } from '@signozhq/icons';
 import { IBuilderQuery } from 'types/api/queryBuilder/queryBuilderData';
+import { FieldDataType } from 'types/api/v5/queryRange';
 import { TracesAggregatorOperator } from 'types/common/queryBuilder';
 
 import { useQueryBuilderV2Context } from '../../QueryBuilderV2Context';
@@ -323,16 +323,16 @@ function QueryAggregationSelect({
 				TracesAggregatorOperator.RATE,
 			];
 
-			const fieldDataType =
+			const fieldDataType: FieldDataType | undefined =
 				functionContextForFetch &&
 				operatorsWithoutDataType.includes(functionContextForFetch)
 					? undefined
-					: QUERY_BUILDER_KEY_TYPES.NUMBER;
+					: 'number';
 
 			return getKeySuggestions({
 				signal: queryData.dataSource,
 				searchText: '',
-				fieldDataType: fieldDataType as QUERY_BUILDER_KEY_TYPES,
+				fieldDataType,
 			});
 		},
 		{

frontend/src/components/QuickFilters/FilterRenderers/Checkbox/Checkbox.styles.scss

@@ -92,52 +92,76 @@
 						color: var(--l3-foreground);
 					}
 
-					.only-btn {
-						cursor: not-allowed;
-						color: var(--l3-foreground);
-					}
-
+					.only-btn,
 					.toggle-btn {
 						cursor: not-allowed;
 						color: var(--l3-foreground);
 					}
 				}
 
-				.only-btn {
-					display: none;
+				// Stack "Only"/"All" and "Toggle" in a single cell so revealing
+				// one swaps in place instead of shifting the row layout.
+				.value-actions {
+					display: grid;
+					align-items: center;
+					justify-items: end;
+
+					> * {
+						grid-area: 1 / 1;
+					}
 				}
+
+				.only-btn,
 				.toggle-btn {
 					display: none;
-				}
-
-				.toggle-btn:hover {
-					background-color: unset;
-				}
-
-				.only-btn:hover {
-					background-color: unset;
-				}
-			}
-
-			.checkbox-value-section:hover {
-				.toggle-btn {
-					display: none;
-				}
-				.only-btn {
-					display: flex;
 					align-items: center;
 					justify-content: center;
 					height: 21px;
+					opacity: 0;
+					transform: translateX(4px);
+					// `display` is animated as a discrete property so the button
+					// stays mounted through its fade-out on exit.
+					transition:
+						opacity 0.16s ease,
+						transform 0.16s ease,
+						display 0.16s allow-discrete;
+
+					&:hover {
+						background-color: unset;
+					}
+				}
+			}
+
+			// Hovering the label/value area reveals the "Only"/"All" action.
+			.checkbox-value-section:hover .only-btn {
+				display: flex;
+				opacity: 1;
+				transform: translateX(0);
+
+				// Entry animation start state (fires on the display switch).
+				@starting-style {
+					opacity: 0;
+					transform: translateX(4px);
+				}
+			}
+
+			// Hovering the checkbox reveals the "Toggle" action.
+			.check-box:hover ~ .checkbox-value-section .toggle-btn {
+				display: flex;
+				opacity: 1;
+				transform: translateX(0);
+
+				@starting-style {
+					opacity: 0;
+					transform: translateX(4px);
 				}
 			}
 		}
 
-		.value:hover {
+		@media (prefers-reduced-motion: reduce) {
+			.only-btn,
 			.toggle-btn {
-				display: flex;
-				align-items: center;
-				justify-content: center;
-				height: 21px;
+				transition: none;
 			}
 		}
 	}

frontend/src/components/QuickFilters/FilterRenderers/Checkbox/CheckboxValueRow.tsx

@@ -53,12 +53,14 @@ function CheckboxValueRow({
 						</Typography.Text>
 					</TooltipSimple>
 				)}
-				<Button type="text" className="only-btn">
-					{onlyButtonLabel}
-				</Button>
-				<Button type="text" className="toggle-btn">
-					Toggle
-				</Button>
+				<div className="value-actions">
+					<Button type="text" className="only-btn">
+						{onlyButtonLabel}
+					</Button>
+					<Button type="text" className="toggle-btn">
+						Toggle
+					</Button>
+				</div>
 			</div>
 		</div>
 	);

frontend/src/components/WarningPopover/WarningPopover.styles.scss

@@ -1,6 +1,11 @@
+.warning-popover-overlay {
+	--antd-arrow-background-color: var(--l2-background);
+}
+
 .warning-content {
 	display: flex;
 	flex-direction: column;
+	background-color: var(--l2-background);
 
 	// === SECTION: Summary (Top)
 	&__summary-section {
@@ -10,16 +15,25 @@
 
 	&__summary {
 		display: flex;
+		align-items: flex-start;
 		justify-content: space-between;
+		gap: var(--spacing-8);
 		padding: 16px;
 	}
 
 	&__summary-left {
 		display: flex;
-		align-items: baseline;
+		align-items: flex-start;
 		gap: 8px;
 	}
 
+	&__icon-wrapper {
+		display: flex;
+		align-items: center;
+		flex-shrink: 0;
+		height: var(--spacing-12);
+	}
+
 	&__summary-text {
 		display: flex;
 		flex-direction: column;

frontend/src/components/WarningPopover/WarningPopover.tsx

@@ -2,6 +2,7 @@ import { ReactNode, useMemo } from 'react';
 import { Color } from '@signozhq/design-tokens';
 import { Button, Popover, PopoverProps } from 'antd';
 import ErrorIcon from 'assets/Error';
+import cx from 'classnames';
 import OverlayScrollbar from 'components/OverlayScrollbar/OverlayScrollbar';
 import { BookOpenText, ChevronsDown, TriangleAlert } from '@signozhq/icons';
 import KeyValueLabel from 'periscope/components/KeyValueLabel';
@@ -30,16 +31,18 @@ export function WarningContent({ warning }: WarningContentProps): JSX.Element {
 			{/* Summary Header */}
 			<section className="warning-content__summary-section">
 				<header className="warning-content__summary">
-					<div className="warning-content__summary-left">
-						<div className="warning-content__icon-wrapper">
-							<ErrorIcon />
-						</div>
+					{(warningCode || warningMessage) && (
+						<div className="warning-content__summary-left">
+							<div className="warning-content__icon-wrapper">
+								<ErrorIcon />
+							</div>
 
-						<div className="warning-content__summary-text">
-							<h2 className="warning-content__warning-code">{warningCode}</h2>
-							<p className="warning-content__warning-message">{warningMessage}</p>
+							<div className="warning-content__summary-text">
+								<h2 className="warning-content__warning-code">{warningCode}</h2>
+								<p className="warning-content__warning-message">{warningMessage}</p>
+							</div>
 						</div>
-					</div>
+					)}
 
 					{warningUrl && (
 						<div className="warning-content__summary-right">
@@ -154,6 +157,10 @@ function WarningPopover({
 			overlayInnerStyle={{ padding: 0 }}
 			autoAdjustOverflow
 			{...popoverProps}
+			overlayClassName={cx(
+				'warning-popover-overlay',
+				popoverProps.overlayClassName,
+			)}
 		>
 			{children || (
 				<TriangleAlert

frontend/src/constants/queryBuilder.ts

@@ -1,8 +1,5 @@
 // ** Helpers
-import {
-	MetrictypesTemporalityDTO,
-	MetrictypesTypeDTO,
-} from 'api/generated/services/sigNoz.schemas';
+import { MetrictypesTypeDTO } from 'api/generated/services/sigNoz.schemas';
 import { defaultTraceSelectedColumns } from 'container/OptionsMenu/constants';
 import { createIdFromObjectFields } from 'lib/createIdFromObjectFields';
 import { createNewBuilderItemName } from 'lib/newQueryBuilder/createNewBuilderItemName';
@@ -392,19 +389,11 @@ const METRIC_TYPE_TO_ATTRIBUTE_TYPE: Record<
 export function toAttributeType(
 	metricType: MetrictypesTypeDTO | undefined,
 	isMonotonic?: boolean,
-	temporality?: MetrictypesTemporalityDTO,
 ): ATTRIBUTE_TYPES | '' {
 	if (!metricType) {
 		return '';
 	}
-	// Monotonicity carries meaning only for cumulative sums: a non-monotonic
-	// cumulative sum is a gauge for all practical purposes. Delta sums are
-	// counters regardless of monotonicity.
-	if (
-		metricType === MetrictypesTypeDTO.sum &&
-		isMonotonic === false &&
-		temporality === MetrictypesTemporalityDTO.cumulative
-	) {
+	if (metricType === MetrictypesTypeDTO.sum && isMonotonic === false) {
 		return ATTRIBUTE_TYPES.GAUGE;
 	}
 	return METRIC_TYPE_TO_ATTRIBUTE_TYPE[metricType] || '';

frontend/src/container/MetricsExplorer/MetricDetails/AllAttributes.tsx

@@ -33,7 +33,6 @@ function AllAttributes({
 	metricName,
 	metricType,
 	isMonotonic,
-	temporality,
 	minTime,
 	maxTime,
 }: AllAttributesProps): JSX.Element {
@@ -72,7 +71,6 @@ function AllAttributes({
 				groupBy,
 				limit,
 				isMonotonic,
-				temporality,
 			);
 			handleExplorerTabChange(
 				PANEL_TYPES.TIME_SERIES,
@@ -91,7 +89,7 @@ function AllAttributes({
 				[MetricsExplorerEventKeys.AttributeKey]: groupBy,
 			});
 		},
-		[metricName, metricType, isMonotonic, temporality, handleExplorerTabChange],
+		[metricName, metricType, isMonotonic, handleExplorerTabChange],
 	);
 
 	const goToMetricsExploreWithAppliedAttribute = useCallback(
@@ -103,7 +101,6 @@ function AllAttributes({
 				undefined,
 				undefined,
 				isMonotonic,
-				temporality,
 			);
 			handleExplorerTabChange(
 				PANEL_TYPES.TIME_SERIES,
@@ -123,7 +120,7 @@ function AllAttributes({
 				[MetricsExplorerEventKeys.AttributeValue]: value,
 			});
 		},
-		[metricName, metricType, isMonotonic, temporality, handleExplorerTabChange],
+		[metricName, metricType, isMonotonic, handleExplorerTabChange],
 	);
 
 	const handleKeyMenuItemClick = useCallback(

frontend/src/container/MetricsExplorer/MetricDetails/DashboardsAndAlertsPopover.tsx

@@ -6,7 +6,7 @@ import { Skeleton } from 'antd';
 import { Typography } from '@signozhq/ui/typography';
 import {
 	useGetMetricAlerts,
-	useGetMetricDashboards,
+	useGetMetricDashboardsV2,
 } from 'api/generated/services/metrics';
 import { QueryParams } from 'constants/query';
 import ROUTES from 'constants/routes';
@@ -38,7 +38,7 @@ function DashboardsAndAlertsPopover({
 		data: dashboardsData,
 		isLoading: isLoadingDashboards,
 		isError: isErrorDashboards,
-	} = useGetMetricDashboards(
+	} = useGetMetricDashboardsV2(
 		{
 			metricName,
 		},
@@ -55,7 +55,8 @@ function DashboardsAndAlertsPopover({
 
 	const dashboards = useMemo(() => {
 		const currentDashboards = dashboardsData?.data.dashboards ?? [];
-		// Remove duplicate dashboards
+		// The API returns one entry per referencing panel, so a dashboard repeats
+		// once per panel that uses the metric.
 		return currentDashboards.filter(
 			(dashboard, index, self) =>
 				index === self.findIndex((t) => t.dashboardId === dashboard.dashboardId),

frontend/src/container/MetricsExplorer/MetricDetails/MetricDetails.tsx

@@ -86,7 +86,6 @@ function MetricDetails({
 				undefined,
 				undefined,
 				metadata?.isMonotonic,
-				metadata?.temporality,
 			);
 			handleExplorerTabChange(
 				PANEL_TYPES.TIME_SERIES,
@@ -109,7 +108,6 @@ function MetricDetails({
 		handleExplorerTabChange,
 		metadata?.type,
 		metadata?.isMonotonic,
-		metadata?.temporality,
 	]);
 
 	useEffect(() => {
@@ -198,7 +196,6 @@ function MetricDetails({
 					metricName={metricName}
 					metricType={metadata?.type}
 					isMonotonic={metadata?.isMonotonic}
-					temporality={metadata?.temporality}
 					minTime={minTime}
 					maxTime={maxTime}
 				/>

frontend/src/container/MetricsExplorer/MetricDetails/__tests__/DashboardsAndAlertsPopover.test.tsx

@@ -23,7 +23,7 @@ const useGetMetricAlertsMock = jest.spyOn(
 );
 const useGetMetricDashboardsMock = jest.spyOn(
 	metricsExplorerHooks,
-	'useGetMetricDashboards',
+	'useGetMetricDashboardsV2',
 );
 
 describe('DashboardsAndAlertsPopover', () => {
@@ -153,11 +153,15 @@ describe('DashboardsAndAlertsPopover', () => {
 		);
 	});
 
-	it('renders unique dashboards even when there are duplicates', async () => {
+	it('collapses multiple panels of the same dashboard into one entry', async () => {
 		useGetMetricDashboardsMock.mockReturnValue(
 			getMockDashboardsData({
 				data: {
-					dashboards: [MOCK_DASHBOARD_1, MOCK_DASHBOARD_2, MOCK_DASHBOARD_1],
+					dashboards: [
+						MOCK_DASHBOARD_1,
+						MOCK_DASHBOARD_2,
+						{ ...MOCK_DASHBOARD_1, panelId: '3', panelName: 'Panel 3' },
+					],
 				},
 			}),
 		);

frontend/src/container/MetricsExplorer/MetricDetails/__tests__/testUtlls.ts

@@ -2,7 +2,7 @@ import * as metricsExplorerHooks from 'api/generated/services/metrics';
 import {
 	GetMetricAlerts200,
 	GetMetricAttributes200,
-	GetMetricDashboards200,
+	GetMetricDashboardsV2200,
 	GetMetricHighlights200,
 	GetMetricMetadata200,
 	MetrictypesTemporalityDTO,
@@ -40,14 +40,14 @@ export function getMockMetricHighlightsData(
 export const MOCK_DASHBOARD_1 = {
 	dashboardName: 'Dashboard 1',
 	dashboardId: '1',
-	widgetId: '1',
-	widgetName: 'Widget 1',
+	panelId: '1',
+	panelName: 'Panel 1',
 };
 export const MOCK_DASHBOARD_2 = {
 	dashboardName: 'Dashboard 2',
 	dashboardId: '2',
-	widgetId: '2',
-	widgetName: 'Widget 2',
+	panelId: '2',
+	panelName: 'Panel 2',
 };
 export const MOCK_ALERT_1 = {
 	alertName: 'Alert 1',
@@ -59,7 +59,7 @@ export const MOCK_ALERT_2 = {
 };
 
 export function getMockDashboardsData(
-	overrides?: Partial<GetMetricDashboards200>,
+	overrides?: Partial<GetMetricDashboardsV2200>,
 	{
 		isLoading = false,
 		isError = false,
@@ -67,7 +67,7 @@ export function getMockDashboardsData(
 		isLoading?: boolean;
 		isError?: boolean;
 	} = {},
-): ReturnType<typeof metricsExplorerHooks.useGetMetricDashboards> {
+): ReturnType<typeof metricsExplorerHooks.useGetMetricDashboardsV2> {
 	return {
 		data: {
 			data: {
@@ -79,7 +79,7 @@ export function getMockDashboardsData(
 
 		isLoading,
 		isError,
-	} as ReturnType<typeof metricsExplorerHooks.useGetMetricDashboards>;
+	} as ReturnType<typeof metricsExplorerHooks.useGetMetricDashboardsV2>;
 }
 
 export function getMockAlertsData(

frontend/src/container/MetricsExplorer/MetricDetails/__tests__/utils.test.ts

@@ -147,44 +147,6 @@ describe('MetricDetails utils', () => {
 			expect(query.builder.queryData[0]?.spaceAggregation).toBe('sum');
 		});
 
-		it('should treat a non-monotonic cumulative SUM as a gauge', () => {
-			const query = getMetricDetailsQuery(
-				TEST_METRIC_NAME,
-				MetrictypesTypeDTO.sum,
-				undefined,
-				undefined,
-				undefined,
-				false,
-				MetrictypesTemporalityDTO.cumulative,
-			);
-
-			expect(query.builder.queryData[0]?.aggregateAttribute?.type).toBe(
-				ATTRIBUTE_TYPES.GAUGE,
-			);
-			expect(query.builder.queryData[0]?.aggregateOperator).toBe('avg');
-			expect(query.builder.queryData[0]?.timeAggregation).toBe('avg');
-			expect(query.builder.queryData[0]?.spaceAggregation).toBe('avg');
-		});
-
-		it('should treat a non-monotonic delta SUM as a counter', () => {
-			const query = getMetricDetailsQuery(
-				TEST_METRIC_NAME,
-				MetrictypesTypeDTO.sum,
-				undefined,
-				undefined,
-				undefined,
-				false,
-				MetrictypesTemporalityDTO.delta,
-			);
-
-			expect(query.builder.queryData[0]?.aggregateAttribute?.type).toBe(
-				ATTRIBUTE_TYPES.SUM,
-			);
-			expect(query.builder.queryData[0]?.aggregateOperator).toBe('rate');
-			expect(query.builder.queryData[0]?.timeAggregation).toBe('rate');
-			expect(query.builder.queryData[0]?.spaceAggregation).toBe('sum');
-		});
-
 		it('should create correct query for GAUGE metric type', () => {
 			const query = getMetricDetailsQuery(
 				TEST_METRIC_NAME,

frontend/src/container/MetricsExplorer/MetricDetails/types.ts

@@ -35,7 +35,6 @@ export interface AllAttributesProps {
 	metricName: string;
 	metricType: MetrictypesTypeDTO | undefined;
 	isMonotonic?: boolean;
-	temporality?: MetrictypesTemporalityDTO;
 	minTime?: number;
 	maxTime?: number;
 }

frontend/src/container/MetricsExplorer/MetricDetails/utils.tsx

@@ -89,21 +89,16 @@ export function getMetricDetailsQuery(
 	groupBy?: string,
 	limit?: number,
 	isMonotonic?: boolean,
-	temporality?: MetrictypesTemporalityDTO,
 ): Query {
 	let timeAggregation;
 	let spaceAggregation;
 	let aggregateOperator;
-	// Monotonicity carries meaning only for cumulative sums; delta sums are
-	// counters regardless of monotonicity.
-	const isNonMonotonicCumulativeSum =
-		metricType === MetrictypesTypeDTO.sum &&
-		isMonotonic === false &&
-		temporality === MetrictypesTemporalityDTO.cumulative;
+	const isNonMonotonicSum =
+		metricType === MetrictypesTypeDTO.sum && isMonotonic === false;
 
 	switch (metricType) {
 		case MetrictypesTypeDTO.sum:
-			if (isNonMonotonicCumulativeSum) {
+			if (isNonMonotonicSum) {
 				timeAggregation = 'avg';
 				spaceAggregation = 'avg';
 				aggregateOperator = 'avg';
@@ -136,7 +131,7 @@ export function getMetricDetailsQuery(
 			break;
 	}
 
-	const attributeType = toAttributeType(metricType, isMonotonic, temporality);
+	const attributeType = toAttributeType(metricType, isMonotonic);
 
 	return {
 		...initialQueriesMap[DataSource.METRICS],

frontend/src/container/OptionsMenu/useOptionsMenu.ts

@@ -113,7 +113,7 @@ const useOptionsMenu = ({
 					(suggestion) => ({
 						name: suggestion.name,
 						signal: suggestion.signal as SignalType,
-						fieldDataType: suggestion.fieldDataType as FieldDataType,
+						fieldDataType: suggestion.fieldDataType,
 						fieldContext: suggestion.fieldContext as FieldContext,
 					}),
 				);
@@ -192,7 +192,7 @@ const useOptionsMenu = ({
 				name: e.name,
 				signal: e.signal as SignalType,
 				fieldContext: e.fieldContext as FieldContext,
-				fieldDataType: e.fieldDataType as FieldDataType,
+				fieldDataType: e.fieldDataType,
 			}));
 		}
 		if (dataSource === DataSource.TRACES) {

frontend/src/container/QueryBuilder/filters/MetricNameSelector/MetricNameSelector.test.tsx

@@ -8,7 +8,6 @@ import {
 } from '@testing-library/react';
 import {
 	MetricsexplorertypesListMetricDTO,
-	MetrictypesTemporalityDTO,
 	MetrictypesTypeDTO,
 } from 'api/generated/services/sigNoz.schemas';
 import { ENTITY_VERSION_V5 } from 'constants/app';
@@ -71,7 +70,7 @@ function makeMetric(
 		type: MetrictypesTypeDTO.sum,
 		isMonotonic: true,
 		description: '',
-		temporality: MetrictypesTemporalityDTO.cumulative,
+		temporality: 'cumulative' as never,
 		unit: '',
 		...overrides,
 	};
@@ -394,13 +393,12 @@ describe('selecting a metric type updates the aggregation options', () => {
 		]);
 	});
 
-	it('non-monotonic cumulative Sum metric is treated as Gauge', () => {
+	it('non-monotonic Sum metric is treated as Gauge', () => {
 		returnMetrics([
 			makeMetric({
 				metricName: 'active_connections',
 				type: MetrictypesTypeDTO.sum,
 				isMonotonic: false,
-				temporality: MetrictypesTemporalityDTO.cumulative,
 			}),
 		]);
 
@@ -429,36 +427,6 @@ describe('selecting a metric type updates the aggregation options', () => {
 		]);
 	});
 
-	it('non-monotonic delta Sum metric remains a counter with Rate/Increase options', () => {
-		returnMetrics([
-			makeMetric({
-				metricName: 'gcp_reject_connections_count',
-				type: MetrictypesTypeDTO.sum,
-				isMonotonic: false,
-				temporality: MetrictypesTemporalityDTO.delta,
-			}),
-		]);
-
-		render(<MetricQueryHarness query={makeQuery()} />);
-
-		const input = screen.getByRole('combobox');
-		fireEvent.change(input, {
-			target: { value: 'gcp_reject_connections_count' },
-		});
-		fireEvent.blur(input);
-
-		expect(getOptionLabels('time-agg-options')).toStrictEqual([
-			'Rate',
-			'Increase',
-		]);
-		expect(getOptionLabels('space-agg-options')).toStrictEqual([
-			'Sum',
-			'Avg',
-			'Min',
-			'Max',
-		]);
-	});
-
 	it('Histogram metric shows no time options and P50–P99 space options', () => {
 		returnMetrics([
 			makeMetric({

frontend/src/container/QueryBuilder/filters/MetricNameSelector/MetricNameSelector.tsx

@@ -34,7 +34,7 @@ export type MetricNameSelectorProps = {
 function getAttributeType(
 	metric: MetricsexplorertypesListMetricDTO,
 ): ATTRIBUTE_TYPES | '' {
-	return toAttributeType(metric.type, metric.isMonotonic, metric.temporality);
+	return toAttributeType(metric.type, metric.isMonotonic);
 }
 
 function createAutocompleteData(

frontend/src/container/QueryTable/Drilldown/BreakoutOptions.tsx

@@ -4,11 +4,11 @@ import { Input } from '@signozhq/ui/input';
 import { Skeleton } from 'antd';
 import { getKeySuggestions } from 'api/querySuggestions/getKeySuggestions';
 import OverlayScrollbar from 'components/OverlayScrollbar/OverlayScrollbar';
-import { QUERY_BUILDER_KEY_TYPES } from 'constants/antlrQueryConstants';
 import useDebounce from 'hooks/useDebounce';
 import { ContextMenu } from 'periscope/components/ContextMenu';
 import { BaseAutocompleteData } from 'types/api/queryBuilder/queryAutocompleteResponse';
 import { MetricAggregation } from 'types/api/v5/queryRange';
+import { fieldDataTypeToDataType } from 'utils/fieldDataType';
 
 import { BreakoutOptionsProps } from './contextConfig';
 import { BreakoutAttributeType } from './types';
@@ -80,7 +80,7 @@ function BreakoutOptions({
 			keyArray.forEach((keyData) => {
 				transformedOptions.push({
 					key: keyData.name,
-					dataType: keyData.fieldDataType as QUERY_BUILDER_KEY_TYPES,
+					dataType: fieldDataTypeToDataType(keyData.fieldDataType),
 					type: '',
 				});
 			});

frontend/src/container/QueryTable/Drilldown/__tests__/Breakout.test.tsx

@@ -238,6 +238,10 @@ describe('TableDrilldown Breakout Functionality', () => {
 		expect(aggregateQueryData.groupBy).toHaveLength(1);
 		expect(aggregateQueryData.groupBy[0].key).toBe('deployment.environment');
 
+		// The picked field's type travels with it — dropping it leaves the breakout query
+		// untyped, so a drilldown on its result can't tell a number from a string.
+		expect(aggregateQueryData.groupBy[0].dataType).toBe('string');
+
 		// Verify that orderBy has been cleared (as per getBreakoutQuery logic)
 		expect(aggregateQueryData.orderBy).toStrictEqual([]);
 

frontend/src/container/QueryTable/Drilldown/tableDrilldownUtils.tsx

@@ -1,6 +1,5 @@
 import { OPERATORS, PANEL_TYPES } from 'constants/queryBuilder';
 import cloneDeep from 'lodash-es/cloneDeep';
-import { BaseAutocompleteData } from 'types/api/queryBuilder/queryAutocompleteResponse';
 import { IBuilderQuery, Query } from 'types/api/queryBuilder/queryBuilderData';
 
 import { addFilterToSelectedQuery, FilterData } from './drilldownUtils';
@@ -89,7 +88,11 @@ export const getBreakoutQuery = (
 		.filter((item: IBuilderQuery) => item.queryName === aggregateData.queryName)
 		.map((item: IBuilderQuery) => ({
 			...item,
-			groupBy: [{ key: groupBy.key, type: groupBy.type } as BaseAutocompleteData],
+			// The picked field's type travels with it: dropped, the breakout query goes out
+			// untyped and a drilldown on its result can't tell a number from a string.
+			groupBy: [
+				{ key: groupBy.key, dataType: groupBy.dataType, type: groupBy.type },
+			],
 			orderBy: [],
 			legend: item.legend && groupBy.key ? `{{${groupBy.key}}}` : '',
 		}));

frontend/src/container/QueryTable/Drilldown/types.ts

@@ -1,6 +1,8 @@
 import { ReactNode } from 'react';
-import { QUERY_BUILDER_KEY_TYPES } from 'constants/antlrQueryConstants';
-import { BaseAutocompleteData } from 'types/api/queryBuilder/queryAutocompleteResponse';
+import {
+	BaseAutocompleteData,
+	DataTypes,
+} from 'types/api/queryBuilder/queryAutocompleteResponse';
 import { IBuilderQuery } from 'types/api/queryBuilder/queryBuilderData';
 
 export type ContextMenuItem = ReactNode;
@@ -31,7 +33,8 @@ export interface AggregateContextMenuConfig {
 
 export interface BreakoutAttributeType {
 	key: string;
-	dataType: QUERY_BUILDER_KEY_TYPES;
+	/** The picked field's type, in the vocabulary the group-by it becomes is read with. */
+	dataType: DataTypes;
 	type: string;
 }
 

frontend/src/pages/DashboardPageV2/DashboardContainer/DashboardPageToolbar/DashboardActions/DashboardActions.module.scss

@@ -11,3 +11,9 @@
 	border: 1px solid var(--l3-border) !important;
 	box-shadow: none !important;
 }
+
+/* Highlights the dashboard name inside the delete-confirm title. */
+.deleteName {
+	color: var(--danger-background);
+	font-weight: var(--font-weight-medium);
+}

frontend/src/pages/DashboardPageV2/DashboardContainer/DashboardPageToolbar/DashboardActions/DashboardActions.tsx

@@ -27,16 +27,13 @@ import logEvent from 'api/common/logEvent';
 import { cloneDashboardV2 } from 'api/generated/services/dashboard';
 import type { DashboardtypesGettableDashboardV2DTO } from 'api/generated/services/sigNoz.schemas';
 import ROUTES from 'constants/routes';
-import { useDeleteDashboard } from 'hooks/dashboard/useDeleteDashboard';
 import { useSafeNavigate } from 'hooks/useSafeNavigate';
-import history from 'lib/history';
 import { DashboardDetailEvents } from 'pages/DashboardPageV2/constants/events';
 import { useAppContext } from 'providers/App/App';
 import { useErrorModal } from 'providers/ErrorModalProvider';
 import APIError from 'types/api/error';
 import { USER_ROLES } from 'types/roles';
 
-import ConfirmDeleteDialog from '../../components/ConfirmDeleteDialog/ConfirmDeleteDialog';
 import DisabledControlTooltip from '../../components/DisabledControlTooltip/DisabledControlTooltip';
 import DisabledMenuItemLabel from '../../components/DisabledMenuItemLabel/DisabledMenuItemLabel';
 import DashboardSettings from '../../DashboardSettings';
@@ -45,6 +42,7 @@ import SectionTitleModal from '../../PanelsAndSectionsLayout/Section/SectionTitl
 import JsonEditorDrawer from '../JsonEditorDrawer/JsonEditorDrawer';
 import SettingsDrawer from '../SettingsDrawer';
 import styles from './DashboardActions.module.scss';
+import { useDeleteDashboardAction } from './useDeleteDashboardAction';
 import { DASHBOARD_LOCKED_REASON } from '../../hooks/useDashboardEditGuard';
 import { useDashboardStore } from '../../store/useDashboardStore';
 
@@ -84,8 +82,12 @@ function DashboardActions({
 	const [isCloning, setIsCloning] = useState<boolean>(false);
 	const [isNewSectionOpen, setIsNewSectionOpen] = useState<boolean>(false);
 
-	const [isDeleteOpen, setIsDeleteOpen] = useState<boolean>(false);
-	const deleteDashboardMutation = useDeleteDashboard(dashboard.id);
+	const { contextHolder: deleteConfirmHolder, confirmDeleteDashboard } =
+		useDeleteDashboardAction({
+			dashboardId: dashboard.id,
+			dashboardName: title,
+			panelCount: Object.keys(dashboard.spec.panels).length,
+		});
 
 	// Open the settings drawer when something in the tree requests it (e.g. the
 	// variables bar's "Add variable" button).
@@ -132,19 +134,6 @@ function DashboardActions({
 		}
 	}, [dashboard.id, title, safeNavigate, showErrorModal]);
 
-	const handleConfirmDelete = useCallback((): void => {
-		deleteDashboardMutation.mutate(undefined, {
-			onSuccess: () => {
-				void logEvent(DashboardDetailEvents.Deleted, {
-					dashboardId: dashboard.id,
-					panelCount: Object.keys(dashboard.spec.panels).length,
-				});
-				setIsDeleteOpen(false);
-				history.replace(ROUTES.ALL_DASHBOARD);
-			},
-		});
-	}, [deleteDashboardMutation, dashboard.id, dashboard.spec.panels]);
-
 	const handleOpenSettings = useCallback((): void => {
 		void logEvent(DashboardDetailEvents.SettingsOpened, {
 			dashboardId: dashboard.id,
@@ -248,7 +237,7 @@ function DashboardActions({
 					icon: <Trash2 size={14} />,
 					danger: true,
 					disabled: isLocked,
-					onClick: (): void => setIsDeleteOpen(true),
+					onClick: confirmDeleteDashboard,
 				},
 			);
 		}
@@ -266,6 +255,7 @@ function DashboardActions({
 		handleClone,
 		onLockToggle,
 		handleEnterFullScreen,
+		confirmDeleteDashboard,
 	]);
 
 	return (
@@ -348,14 +338,7 @@ function DashboardActions({
 				isOpen={isJsonEditorOpen}
 				onClose={(): void => setIsJsonEditorOpen(false)}
 			/>
-			<ConfirmDeleteDialog
-				open={isDeleteOpen}
-				title={`Delete dashboard"?`}
-				description={`Are you sure you want to delete this dashboard - "${title}"? This action cannot be undone.`}
-				isLoading={deleteDashboardMutation.isLoading}
-				onConfirm={handleConfirmDelete}
-				onClose={(): void => setIsDeleteOpen(false)}
-			/>
+			{deleteConfirmHolder}
 			<SectionTitleModal
 				open={isNewSectionOpen}
 				heading="New section"

frontend/src/pages/DashboardPageV2/DashboardContainer/DashboardPageToolbar/DashboardActions/useDeleteDashboardAction.tsx

@@ -0,0 +1,89 @@
+import { type ReactNode, useCallback } from 'react';
+import { useQueryClient } from 'react-query';
+import { toast } from '@signozhq/ui/sonner';
+import { Typography } from '@signozhq/ui/typography';
+import logEvent from 'api/common/logEvent';
+import {
+	invalidateListDashboardsForUserV2,
+	useDeleteDashboardV2,
+} from 'api/generated/services/dashboard';
+import { useDeleteConfirm } from 'components/DeleteConfirmModal/useDeleteConfirm';
+import ROUTES from 'constants/routes';
+import { useDashboardPreferencesStore } from 'hooks/dashboard/useDashboardPreference';
+import history from 'lib/history';
+import { DashboardDetailEvents } from 'pages/DashboardPageV2/constants/events';
+import { useErrorModal } from 'providers/ErrorModalProvider';
+import APIError from 'types/api/error';
+
+import styles from './DashboardActions.module.scss';
+
+interface UseDeleteDashboardActionArgs {
+	dashboardId: string;
+	dashboardName: string;
+	panelCount: number;
+}
+
+interface UseDeleteDashboardAction {
+	/** Must be rendered in the calling component for the modal to appear. */
+	contextHolder: ReactNode;
+	confirmDeleteDashboard: () => void;
+}
+
+/**
+ * Deletes the open dashboard through the v2 endpoint behind the shared
+ * destructive confirmation, then drops its local preferences, refreshes the
+ * dashboard list cache and sends the user back to the list.
+ */
+export function useDeleteDashboardAction({
+	dashboardId,
+	dashboardName,
+	panelCount,
+}: UseDeleteDashboardActionArgs): UseDeleteDashboardAction {
+	const queryClient = useQueryClient();
+	const { showErrorModal } = useErrorModal();
+	const { contextHolder, confirmDelete } = useDeleteConfirm();
+	const removePreferences = useDashboardPreferencesStore(
+		(state) => state.removePreferences,
+	);
+
+	const { mutate: deleteDashboard } = useDeleteDashboardV2({
+		mutation: {
+			onSuccess: async (): Promise<void> => {
+				void logEvent(DashboardDetailEvents.Deleted, { dashboardId, panelCount });
+				removePreferences(dashboardId);
+				await invalidateListDashboardsForUserV2(queryClient);
+				toast.success('Dashboard deleted successfully');
+				history.replace(ROUTES.ALL_DASHBOARD);
+			},
+			onError: (error: unknown): void => {
+				showErrorModal(error as APIError);
+			},
+		},
+	});
+
+	const confirmDeleteDashboard = useCallback((): void => {
+		confirmDelete({
+			title: (
+				<Typography.Title level={5}>
+					Are you sure you want to delete the
+					<Typography.Text className={styles.deleteName}>
+						{' '}
+						{dashboardName}{' '}
+					</Typography.Text>
+					dashboard?
+				</Typography.Title>
+			),
+			content: 'This action cannot be undone.',
+			// Keeps the Delete button loading until the mutation settles, then closes.
+			onConfirm: () =>
+				new Promise<void>((resolve) => {
+					deleteDashboard(
+						{ pathParams: { id: dashboardId } },
+						{ onSettled: () => resolve() },
+					);
+				}),
+		});
+	}, [confirmDelete, dashboardName, deleteDashboard, dashboardId]);
+
+	return { contextHolder, confirmDeleteDashboard };
+}

frontend/src/pages/DashboardPageV2/DashboardContainer/PanelEditor/Header/Header.tsx

@@ -1,5 +1,6 @@
 import { useCallback } from 'react';
 import { SolidAlertTriangle, X } from '@signozhq/icons';
+import { Badge } from '@signozhq/ui/badge';
 import { Button } from '@signozhq/ui/button';
 import { DialogWrapper } from '@signozhq/ui/dialog';
 import { Divider } from '@signozhq/ui/divider';
@@ -13,6 +14,7 @@ import DisabledControlTooltip from '../../components/DisabledControlTooltip/Disa
 import styles from './Header.module.scss';
 
 interface HeaderProps {
+	/** Unsaved edits exist — shows the "Unsaved Changes" badge and gates the discard confirmation on close (not the Save button). */
 	isDirty: boolean;
 	isSaving: boolean;
 	showSwitchToView?: boolean;
@@ -66,6 +68,11 @@ function Header({
 				/>
 				<Divider type="vertical" />
 				<Typography.Text>Configure panel</Typography.Text>
+				{isDirty && (
+					<Badge color="warning" data-testid="panel-editor-v2-unsaved-badge">
+						Unsaved Changes
+					</Badge>
+				)}
 			</div>
 			<div className={styles.actions}>
 				<HeaderRightSection
@@ -88,7 +95,7 @@ function Header({
 						variant="solid"
 						color="primary"
 						data-testid="panel-editor-v2-save"
-						disabled={readOnly || !isDirty || isSaving}
+						disabled={readOnly || isSaving}
 						loading={!readOnly && isSaving}
 						onClick={readOnly ? undefined : onSave}
 					>

frontend/src/pages/DashboardPageV2/DashboardContainer/PanelEditor/Header/__tests__/Header.test.tsx

@@ -1,3 +1,4 @@
+import type { ComponentProps } from 'react';
 import { MemoryRouter } from 'react-router-dom';
 import { render, screen } from '@testing-library/react';
 import { TooltipProvider } from '@signozhq/ui/tooltip';
@@ -23,7 +24,9 @@ jest.mock('api/common/logEvent', () => ({
 
 const mockUseIsAIAssistantEnabled = useIsAIAssistantEnabled as jest.Mock;
 
-function renderHeader(): void {
+function renderHeader(
+	props: Partial<ComponentProps<typeof Header>> = {},
+): void {
 	// AppLayout supplies the TooltipProvider in the app; the header is rendered bare here.
 	render(
 		<MemoryRouter>
@@ -33,6 +36,7 @@ function renderHeader(): void {
 					isSaving={false}
 					onSave={jest.fn()}
 					onClose={jest.fn()}
+					{...props}
 				/>
 			</TooltipProvider>
 		</MemoryRouter>,
@@ -66,4 +70,34 @@ describe('PanelEditor Header', () => {
 		).not.toBeInTheDocument();
 		expect(screen.getByTestId('panel-editor-v2-save')).toBeInTheDocument();
 	});
+
+	it('keeps Save enabled even when there are no unsaved edits', () => {
+		mockUseIsAIAssistantEnabled.mockReturnValue(false);
+
+		renderHeader({ isDirty: false });
+
+		expect(screen.getByTestId('panel-editor-v2-save')).toBeEnabled();
+	});
+
+	it('disables Save only while read-only or saving', () => {
+		mockUseIsAIAssistantEnabled.mockReturnValue(false);
+
+		renderHeader({ isDirty: true, readOnly: true, readOnlyReason: 'Locked' });
+
+		expect(screen.getByTestId('panel-editor-v2-save')).toBeDisabled();
+	});
+
+	it('shows the Unsaved Changes badge only when there are unsaved edits', () => {
+		mockUseIsAIAssistantEnabled.mockReturnValue(false);
+
+		renderHeader({ isDirty: false });
+		expect(
+			screen.queryByTestId('panel-editor-v2-unsaved-badge'),
+		).not.toBeInTheDocument();
+
+		renderHeader({ isDirty: true });
+		expect(
+			screen.getByTestId('panel-editor-v2-unsaved-badge'),
+		).toBeInTheDocument();
+	});
 });

frontend/src/pages/DashboardPageV2/DashboardContainer/PanelEditor/__tests__/PanelEditorContainer.test.tsx

@@ -1,5 +1,6 @@
 import { render, screen, waitFor } from '@testing-library/react';
 import userEvent from '@testing-library/user-event';
+import { toast } from '@signozhq/ui/sonner';
 import type { DashboardtypesPanelDTO } from 'api/generated/services/sigNoz.schemas';
 import { PANEL_TYPES } from 'constants/queryBuilder';
 import { getPanelDefinition } from 'pages/DashboardPageV2/DashboardContainer/Panels/registry';
@@ -100,6 +101,12 @@ jest.mock('@signozhq/ui/resizable', () => ({
 jest.mock('@signozhq/ui/sonner', () => ({
 	toast: { success: jest.fn(), error: jest.fn() },
 }));
+const mockShowErrorModal = jest.fn();
+jest.mock('providers/ErrorModalProvider', () => ({
+	useErrorModal: (): { showErrorModal: jest.Mock } => ({
+		showErrorModal: mockShowErrorModal,
+	}),
+}));
 
 // Children mocked to capture props (and expose a Save trigger / footer slot).
 const mockHeaderProps = jest.fn();
@@ -261,13 +268,13 @@ describe('PanelEditorContainer composition', () => {
 		);
 	});
 
-	it('keeps a query-less new panel unsaveable but still serializes its seed query', () => {
+	it('keeps a query-less new panel clean but still serializes its seed query', () => {
 		setup(makePanel('signoz/TimeSeriesPanel'), { isNew: true });
 
 		expect(mockUseQuerySync).toHaveBeenCalledWith(
 			expect.objectContaining({ alwaysSerializeQuery: true }),
 		);
-		// No query and no edits yet → nothing to save, so Save stays disabled.
+		// No query and no edits yet → not dirty, so closing won't prompt to discard.
 		expect(mockHeaderProps).toHaveBeenCalledWith(
 			expect.objectContaining({ isDirty: false }),
 		);
@@ -290,7 +297,7 @@ describe('PanelEditorContainer composition', () => {
 		);
 	});
 
-	it('marks a new panel that already has a query saveable (e.g. list auto-runs one)', () => {
+	it('marks a new panel that already has a query dirty (e.g. list auto-runs one)', () => {
 		const seededQuery = {
 			spec: { plugin: { kind: 'signoz/BuilderQuery', spec: { signal: 'logs' } } },
 		};
@@ -327,6 +334,24 @@ describe('PanelEditorContainer composition', () => {
 		expect(mockSave).toHaveBeenCalledWith(panel.spec);
 	});
 
+	it('surfaces a save failure through the error modal', async () => {
+		// The raw thrown value flows straight to the modal, which normalizes it to an
+		// APIError itself (that normalization is covered by ErrorModalProvider's tests).
+		const failure = {
+			response: {
+				status: 400,
+				data: { error: { code: 'INVALID', message: 'Panel name already exists' } },
+			},
+		};
+		mockSave.mockRejectedValueOnce(failure);
+		setup(makePanel('signoz/TimeSeriesPanel'));
+
+		await userEvent.click(screen.getByTestId('editor-save'));
+
+		await waitFor(() => expect(mockShowErrorModal).toHaveBeenCalledWith(failure));
+		expect(toast.error).not.toHaveBeenCalled();
+	});
+
 	it('marks the saved panel to be scrolled into view on the dashboard', async () => {
 		setup(makePanel('signoz/TimeSeriesPanel'));
 

frontend/src/pages/DashboardPageV2/DashboardContainer/PanelEditor/hooks/__tests__/usePanelEditorQuerySync.integration.test.tsx

@@ -117,6 +117,49 @@ describe('usePanelEditorQuerySync (real query builder)', () => {
 		},
 	);
 
+	it('a NEW panel stays query-dirty after Stage & Run commits the edited query into the draft', async () => {
+		// Repro of the P0: new panel → edit → Run commits the query into the draft; the
+		// dirty baseline must not drift onto it, or Save re-disables.
+		const editedInUrl: Query = {
+			...initialQueriesMap[DataSource.METRICS],
+			id: 'edited-new-panel',
+			builder: {
+				...initialQueriesMap[DataSource.METRICS].builder,
+				queryData: [
+					{
+						...initialQueriesMap[DataSource.METRICS].builder.queryData[0],
+						legend: 'edited-legend',
+					},
+				],
+			},
+		};
+		const committed = toPerses(editedInUrl, panelType);
+
+		const { result, rerender } = renderHook(
+			({ draftQueries }: { draftQueries: DashboardtypesQueryDTO[] }) =>
+				usePanelEditorQuerySync({
+					draft: makePanel(draftQueries),
+					panelType,
+					setSpec: jest.fn(),
+					refetch: jest.fn(),
+					alwaysSerializeQuery: true,
+					signal: DataSource.METRICS as unknown as TelemetrytypesSignalDTO,
+					// savedQueries omitted — new panel.
+				}),
+			{
+				wrapper: makeUrlWrapper(editedInUrl),
+				initialProps: { draftQueries: [] as DashboardtypesQueryDTO[] },
+			},
+		);
+
+		// The edited query (from the URL) diverges from the seeded default → dirty.
+		await waitFor(() => expect(result.current.isQueryDirty).toBe(true));
+
+		// Stage & Run commits the edited query into the draft → must stay dirty.
+		rerender({ draftQueries: committed });
+		expect(result.current.isQueryDirty).toBe(true);
+	});
+
 	it('an untouched panel with a minimal/older stored query is NOT dirty (drift fix)', async () => {
 		// An older saved query carries only a few fields; the builder re-emits many more
 		// (source, stepInterval, filter, spaceAggregation, …). Comparing raw would read

frontend/src/pages/DashboardPageV2/DashboardContainer/PanelEditor/hooks/usePanelEditorQuerySync.ts

@@ -79,6 +79,11 @@ export function usePanelEditorQuerySync({
 	// in-editor edit in the URL survives a refresh / browser Back-Forward.
 	useShareBuilderUrl({ defaultValue: seedQuery });
 
+	// Frozen seed for the new-panel dirty baseline: Stage & Run commits the live query
+	// into `draft.spec.queries`, so re-reading `seedQuery` would drift the baseline onto
+	// the run query and Save would re-disable right after a run.
+	const initialSeedRef = useRef(seedQuery);
+
 	// Commit the live query into the draft (what the preview fetches).
 	const commitQuery = useCallback(
 		(query: Query): boolean => {
@@ -152,16 +157,19 @@ export function usePanelEditorQuerySync({
 	// the V5 envelope level. Anchoring to `savedQueries` (not the builder-seed) keeps a
 	// handed-off / URL-restored edit reading as dirty; routing both sides through the same
 	// `fromPerses → toPerses` round-trip stops builder-added defaults (absent from an older
-	// stored query) reading an untouched panel as modified. New panel: fall back to seed.
+	// stored query) reading an untouched panel as modified. New panel: fall back to the
+	// frozen initial seed (see `initialSeedRef`).
 	const baselineEnvelopes = useMemo(
 		() =>
 			toQueryEnvelopes(
 				toPerses(
-					savedQueries ? fromPerses(savedQueries, panelType) : seedQuery,
+					savedQueries
+						? fromPerses(savedQueries, panelType)
+						: initialSeedRef.current,
 					panelType,
 				),
 			),
-		[savedQueries, seedQuery, panelType],
+		[savedQueries, panelType],
 	);
 	const isQueryDirty = useMemo(
 		() =>

frontend/src/pages/DashboardPageV2/DashboardContainer/PanelEditor/index.tsx

@@ -19,6 +19,7 @@ import {
 	SectionKind,
 } from 'pages/DashboardPageV2/DashboardContainer/Panels/types/sections';
 import { getBuilderQueries } from 'pages/DashboardPageV2/DashboardContainer/Panels/utils/getBuilderQueries';
+import { useErrorModal } from 'providers/ErrorModalProvider';
 
 import { getExecStats } from '../queryV5/v5ResponseData';
 import { usePanelInteractions } from '../PanelsAndSectionsLayout/Panel/hooks/usePanelInteractions';
@@ -159,9 +160,10 @@ function PanelEditorContainer({
 		return section?.controls;
 	}, [panelDefinition]);
 
-	// New panels are savable once seeded with a query (List auto-seeds one). Read the
-	// seed `panel`, not the live `draft` — the staged-query sync commits the seed into
-	// the draft on open, which would falsely dirty an untouched query-less new panel.
+	// Unsaved-edits flag driving the discard confirmation on close (Save is always
+	// enabled). Read the seed `panel`, not the live `draft` — the staged-query sync
+	// commits the seed into the draft on open, which would falsely dirty an untouched
+	// query-less new panel.
 	const isDirty = useMemo(
 		() => isSpecDirty || isQueryDirty || (isNew && panel.spec.queries.length > 0),
 		[isSpecDirty, isQueryDirty, isNew, panel.spec.queries.length],
@@ -226,6 +228,7 @@ function PanelEditorContainer({
 	});
 
 	const setScrollTargetId = useScrollIntoViewStore((s) => s.setScrollTargetId);
+	const { showErrorModal } = useErrorModal();
 
 	const onSave = useCallback(async (): Promise<void> => {
 		if (!isEditable) {
@@ -236,12 +239,22 @@ function PanelEditorContainer({
 			const savedPanelId = await save(buildSaveSpec(draft.spec));
 			// Reveal the saved panel once the dashboard re-renders.
 			setScrollTargetId(savedPanelId);
-			toast.success('Panel saved');
+			toast.success('Panel saved', {
+				position: 'top-center',
+			});
 			onSaved();
-		} catch {
-			toast.error('Failed to save panel');
+		} catch (err) {
+			showErrorModal(err);
 		}
-	}, [isEditable, save, buildSaveSpec, draft.spec, setScrollTargetId, onSaved]);
+	}, [
+		isEditable,
+		save,
+		buildSaveSpec,
+		draft.spec,
+		setScrollTargetId,
+		onSaved,
+		showErrorModal,
+	]);
 
 	// Leaving an existing panel's editor (without saving) still returns to it, so
 	// the dashboard lands on that panel rather than scrolled to the top. A new,

frontend/src/pages/DashboardPageV2/DashboardContainer/PanelsAndSectionsLayout/Panel/PanelStatus/__tests__/utils.test.ts

@@ -51,7 +51,7 @@ describe('panelStatusFromError', () => {
 
 	it('falls back to the error message when there is no structured body', () => {
 		expect(panelStatusFromError(new Error('boom'))).toStrictEqual({
-			code: 'unknown_error',
+			code: 'UPSTREAM_UNAVAILABLE',
 			message: 'boom',
 			docsUrl: undefined,
 			messages: [],

frontend/src/pages/DashboardsListPageV2/components/ActionsPopover/DeleteActionItem.tsx

@@ -12,13 +12,13 @@ import {
 	deleteDashboardV2,
 	invalidateListDashboardsForUserV2,
 } from 'api/generated/services/dashboard';
+import { useDeleteConfirm } from 'components/DeleteConfirmModal/useDeleteConfirm';
 import { DashboardListEvents } from 'pages/DashboardsListPageV2/constants/events';
 import { useAppContext } from 'providers/App/App';
 import { useErrorModal } from 'providers/ErrorModalProvider';
 import APIError from 'types/api/error';
 import { USER_ROLES } from 'types/roles';
 
-import { useDeleteConfirm } from '../DeleteConfirmModal/useDeleteConfirm';
 import styles from './ActionsPopover.module.scss';
 
 interface Props {

frontend/src/pages/DashboardsListPageV2/components/ViewsRail/ViewsRail.tsx

@@ -6,11 +6,11 @@ import { Typography } from '@signozhq/ui/typography';
 import { Bookmark, PenLine, Plus, Search, Trash2 } from '@signozhq/icons';
 import cx from 'classnames';
 
+import { useDeleteConfirm } from 'components/DeleteConfirmModal/useDeleteConfirm';
 import { DashboardListEvents } from 'pages/DashboardsListPageV2/constants/events';
 
 import type { SavedView } from '../../types';
 import { type BuiltinView } from '../../utils/views';
-import { useDeleteConfirm } from '../DeleteConfirmModal/useDeleteConfirm';
 import ViewNamePopover from './ViewNamePopover';
 
 import styles from './ViewsRail.module.scss';

frontend/src/types/api/querySuggestions/types.ts

@@ -1,4 +1,4 @@
-import { QUERY_BUILDER_KEY_TYPES } from 'constants/antlrQueryConstants';
+import { FieldDataType } from 'types/api/v5/queryRange';
 
 export interface QueryKeyDataSuggestionsProps {
 	label: string;
@@ -7,7 +7,12 @@ export interface QueryKeyDataSuggestionsProps {
 	apply?: string;
 	detail?: string;
 	fieldContext?: 'resource' | 'scope' | 'attribute' | 'span';
-	fieldDataType?: QUERY_BUILDER_KEY_TYPES;
+	/**
+	 * The field's type as the API reports it. Was declared as the antlr key-type enum
+	 * (`string | number | boolean`), which the endpoint never returns — every consumer cast it
+	 * back to `FieldDataType`.
+	 */
+	fieldDataType?: FieldDataType;
 	name: string;
 	signal: 'traces' | 'logs' | 'metrics';
 }
@@ -26,7 +31,7 @@ export interface QueryKeyRequestProps {
 	signal: 'traces' | 'logs' | 'metrics';
 	searchText: string;
 	fieldContext?: 'resource' | 'scope' | 'attribute' | 'span';
-	fieldDataType?: QUERY_BUILDER_KEY_TYPES;
+	fieldDataType?: FieldDataType;
 	metricName?: string;
 	metricNamespace?: string;
 	signalSource?: 'meter' | '';

frontend/src/utils/__tests__/fieldDataType.test.ts

@@ -0,0 +1,34 @@
+import { DataTypes } from 'types/api/queryBuilder/queryAutocompleteResponse';
+import type { FieldDataType } from 'types/api/v5/queryRange';
+
+import { fieldDataTypeToDataType } from '../fieldDataType';
+
+describe('fieldDataTypeToDataType', () => {
+	it('maps the reported numeric spellings onto the query-builder numerics', () => {
+		expect(fieldDataTypeToDataType('number')).toBe(DataTypes.Float64);
+		expect(fieldDataTypeToDataType('int64')).toBe(DataTypes.Int64);
+		expect(fieldDataTypeToDataType('float64')).toBe(DataTypes.Float64);
+	});
+
+	it('maps the list spellings onto the array types', () => {
+		expect(fieldDataTypeToDataType('[]string')).toBe(DataTypes.ArrayString);
+		expect(fieldDataTypeToDataType('[]bool')).toBe(DataTypes.ArrayBool);
+		expect(fieldDataTypeToDataType('[]int64')).toBe(DataTypes.ArrayInt64);
+		expect(fieldDataTypeToDataType('[]float64')).toBe(DataTypes.ArrayFloat64);
+		expect(fieldDataTypeToDataType('[]number')).toBe(DataTypes.ArrayFloat64);
+	});
+
+	it('passes through the spellings the two vocabularies share', () => {
+		expect(fieldDataTypeToDataType('string')).toBe(DataTypes.String);
+		expect(fieldDataTypeToDataType('bool')).toBe(DataTypes.bool);
+	});
+
+	it('returns EMPTY for empty, missing and not-yet-known types', () => {
+		expect(fieldDataTypeToDataType('')).toBe(DataTypes.EMPTY);
+		expect(fieldDataTypeToDataType(undefined)).toBe(DataTypes.EMPTY);
+		// The backend can store `[]json` / `[]dynamic`, which the API union doesn't list.
+		expect(fieldDataTypeToDataType('[]json' as FieldDataType)).toBe(
+			DataTypes.EMPTY,
+		);
+	});
+});

frontend/src/utils/errorUtils.ts

@@ -42,7 +42,12 @@ export function toAPIError(
 	try {
 		ErrorResponseHandlerForGeneratedAPIs(error);
 	} catch (apiError) {
-		if (apiError instanceof APIError) {
+		// UPSTREAM_UNAVAILABLE means the handler couldn't find a backend error code
+		// (non-envelope body); prefer the caller's context-specific defaultMessage.
+		if (
+			apiError instanceof APIError &&
+			apiError.getErrorCode() !== 'UPSTREAM_UNAVAILABLE'
+		) {
 			return apiError;
 		}
 	}

frontend/src/utils/fieldDataType.ts

@@ -0,0 +1,31 @@
+import { DataTypes } from 'types/api/queryBuilder/queryAutocompleteResponse';
+import type { FieldDataType } from 'types/api/v5/queryRange';
+
+/**
+ * The field metadata APIs and the query builder spell field types differently: the metadata
+ * collapses the numerics to `number` and writes lists as `[]string`, while `DataTypes` — which
+ * keys the operator and attribute-value lookups — uses `float64` and `array(string)`. Convert
+ * where a reported type becomes a query-builder one, so those lookups can't miss.
+ */
+const TO_DATA_TYPE: Record<FieldDataType, DataTypes> = {
+	'': DataTypes.EMPTY,
+	string: DataTypes.String,
+	bool: DataTypes.bool,
+	int64: DataTypes.Int64,
+	float64: DataTypes.Float64,
+	number: DataTypes.Float64,
+	'[]string': DataTypes.ArrayString,
+	'[]bool': DataTypes.ArrayBool,
+	'[]int64': DataTypes.ArrayInt64,
+	'[]float64': DataTypes.ArrayFloat64,
+	'[]number': DataTypes.ArrayFloat64,
+};
+
+/**
+ * Exhaustive over `FieldDataType`, so a type added to the API union fails to compile here
+ * rather than going missing at runtime. The fallback is for a spelling the union doesn't know
+ * yet — the backend can store `[]json` and `[]dynamic` — not for any of the cases above.
+ */
+export const fieldDataTypeToDataType = (
+	fieldDataType?: FieldDataType,
+): DataTypes => TO_DATA_TYPE[fieldDataType ?? ''] ?? DataTypes.EMPTY;

go.mod

@@ -4,7 +4,7 @@ go 1.25.7
 
 require (
 	dario.cat/mergo v1.0.2
-	github.com/AfterShip/clickhouse-sql-parser v0.4.16
+	github.com/AfterShip/clickhouse-sql-parser v0.5.2
 	github.com/ClickHouse/clickhouse-go/v2 v2.44.0
 	github.com/DATA-DOG/go-sqlmock v1.5.2
 	github.com/SigNoz/clickhouse-go-mock v0.14.0

go.sum

@@ -66,8 +66,8 @@ dario.cat/mergo v1.0.2/go.mod h1:E/hbnu0NxMFBjpMIE34DRGLWqDy0g5FuKDhCb31ngxA=
 dmitri.shuralyov.com/gpu/mtl v0.0.0-20190408044501-666a987793e9/go.mod h1:H6x//7gZCb22OMCxBHrMx7a5I7Hp++hsVxbQ4BYO7hU=
 filippo.io/edwards25519 v1.2.0 h1:crnVqOiS4jqYleHd9vaKZ+HKtHfllngJIiOpNpoJsjo=
 filippo.io/edwards25519 v1.2.0/go.mod h1:xzAOLCNug/yB62zG1bQ8uziwrIqIuxhctzJT18Q77mc=
-github.com/AfterShip/clickhouse-sql-parser v0.4.16 h1:gpl+wXclYUKT0p4+gBq22XeRYWwEoZ9f35vogqMvkLQ=
-github.com/AfterShip/clickhouse-sql-parser v0.4.16/go.mod h1:W0Z82wJWkJxz2RVun/RMwxue3g7ut47Xxl+SFqdJGus=
+github.com/AfterShip/clickhouse-sql-parser v0.5.2 h1:KCxdmvuVawVF4yl0ZS33q7olgmLZwvZG5BjWHp6o414=
+github.com/AfterShip/clickhouse-sql-parser v0.5.2/go.mod h1:Qi3qvPTfZb/aFwI5V4WFOahgjsLJa4MzVijIAfwOhDw=
 github.com/Azure/azure-sdk-for-go v68.0.0+incompatible h1:fcYLmCpyNYRnvJbPerq7U0hS+6+I79yEDJBqVNcqUzU=
 github.com/Azure/azure-sdk-for-go/sdk/azcore v1.21.0 h1:fou+2+WFTib47nS+nz/ozhEBnvU96bKHy6LjRsY4E28=
 github.com/Azure/azure-sdk-for-go/sdk/azcore v1.21.0/go.mod h1:t76Ruy8AHvUAC8GfMWJMa0ElSbuIcO03NLpynfbgsPA=

pkg/flagger/registry.go

@@ -14,6 +14,8 @@ var (
 	FeatureEnableAIObservability  = featuretypes.MustNewName("enable_ai_observability")
 	FeatureEnableMetricsReduction = featuretypes.MustNewName("enable_metrics_reduction")
 	FeatureUseInfraMonitoringV2   = featuretypes.MustNewName("use_infra_monitoring_v2")
+
+	FeatureUsePrometheusClickhouseV2 = featuretypes.MustNewName("use_prometheus_clickhouse_v2")
 )
 
 func MustNewRegistry() featuretypes.Registry {
@@ -106,6 +108,14 @@ func MustNewRegistry() featuretypes.Registry {
 			DefaultVariant: featuretypes.MustNewName("disabled"),
 			Variants:       featuretypes.NewBooleanVariants(),
 		},
+		&featuretypes.Feature{
+			Name:           FeatureUsePrometheusClickhouseV2,
+			Kind:           featuretypes.KindBoolean,
+			Stage:          featuretypes.StageExperimental,
+			Description:    "Runs PromQL queries on the clickhousev2 provider alongside the served engine result and logs any difference; serving is unaffected.",
+			DefaultVariant: featuretypes.MustNewName("disabled"),
+			Variants:       featuretypes.NewBooleanVariants(),
+		},
 	)
 	if err != nil {
 		panic(err)

pkg/prometheus/clickhouseprometheusv2/capture.go

@@ -0,0 +1,85 @@
+package clickhouseprometheusv2
+
+import (
+	"context"
+	"sync"
+
+	"github.com/SigNoz/signoz/pkg/prometheus"
+	"github.com/prometheus/prometheus/model/labels"
+	"github.com/prometheus/prometheus/storage"
+	"github.com/prometheus/prometheus/util/annotations"
+)
+
+// statementRecorder collects the statements a PromQL evaluation would run.
+// Safe for concurrent use: the engine may Select selectors concurrently.
+type statementRecorder struct {
+	mu         sync.Mutex
+	statements []prometheus.CapturedStatement
+}
+
+func (r *statementRecorder) record(query string, args []any) {
+	r.mu.Lock()
+	defer r.mu.Unlock()
+	r.statements = append(r.statements, prometheus.CapturedStatement{Query: query, Args: args})
+}
+
+func (r *statementRecorder) Statements() []prometheus.CapturedStatement {
+	r.mu.Lock()
+	defer r.mu.Unlock()
+	out := make([]prometheus.CapturedStatement, len(r.statements))
+	copy(out, r.statements)
+	return out
+}
+
+type captureQueryable struct {
+	client   *client
+	recorder *statementRecorder
+}
+
+func (c *captureQueryable) Querier(mint, maxt int64) (storage.Querier, error) {
+	return &captureQuerier{
+		querier:  querier{mint: mint, maxt: maxt, client: c.client},
+		recorder: c.recorder,
+	}, nil
+}
+
+// captureQuerier builds the same SQL as the live querier but records it and
+// returns no data. The fingerprint filter always takes the subquery form:
+// without executing the series lookup, the inline literal set is unknown.
+type captureQuerier struct {
+	querier
+	recorder *statementRecorder
+}
+
+func (c *captureQuerier) Select(ctx context.Context, _ bool, hints *storage.SelectHints, matchers ...*labels.Matcher) storage.SeriesSet {
+
+	start, end := c.window(hints)
+
+	samplesQuery, args, err := buildSamplesQuery(start, end, metricNamesFromMatchers(matchers), matchers, c.lastSamplePerStepFor(ctx, hints))
+	if err != nil {
+		return storage.ErrSeriesSet(err)
+	}
+	c.recorder.record(samplesQuery, args)
+
+	return storage.EmptySeriesSet()
+}
+
+func (c *captureQuerier) LabelValues(context.Context, string, *storage.LabelHints, ...*labels.Matcher) ([]string, annotations.Annotations, error) {
+	return nil, nil, nil
+}
+
+func (c *captureQuerier) LabelNames(context.Context, *storage.LabelHints, ...*labels.Matcher) ([]string, annotations.Annotations, error) {
+	return nil, nil, nil
+}
+
+// metricNamesFromMatchers extracts the statically known metric name, if any.
+// The live path derives names from the matched series; the capture path has
+// no execution results, so only a __name__ equality contributes.
+func metricNamesFromMatchers(matchers []*labels.Matcher) []string {
+	for _, m := range matchers {
+		if m.Name == metricNameLabel && m.Type == labels.MatchEqual && m.Value != "" {
+			return []string{m.Value}
+		}
+	}
+	return nil
+}

pkg/prometheus/clickhouseprometheusv2/classification_golden_test.go

@@ -0,0 +1,119 @@
+package clickhouseprometheusv2
+
+import (
+	"encoding/json"
+	"flag"
+	"os"
+	"path/filepath"
+	"sort"
+	"testing"
+
+	"github.com/prometheus/prometheus/promql/parser"
+	"github.com/stretchr/testify/require"
+)
+
+var updateGolden = flag.Bool("update", false, "rewrite the classification golden file")
+
+const goldenFile = "testdata/classification_golden.json"
+
+// corpusFile is the conformance corpus the integration suite replays; the
+// golden freezes how the classifier routes every one of its expressions.
+const corpusFile = "../../../tests/integration/testdata/promqltestcorpus/corpus.json"
+
+type goldenEntry struct {
+	Expr    string `json:"expr"`
+	StartMs int64  `json:"start_ms"`
+	EndMs   int64  `json:"end_ms"`
+	StepMs  int64  `json:"step_ms"`
+	// Plan is the routing decision: "full" (whole query in ClickHouse),
+	// "hybrid" (units substituted, engine on top), "fallback" (engine over
+	// the native querier).
+	Plan string `json:"plan"`
+	// Units is the substituted-unit count for hybrid plans.
+	Units int `json:"units,omitempty"`
+	// Reason is the coarse fallback bucket (fallbackShape).
+	Reason string `json:"reason,omitempty"`
+}
+
+// TestClassificationGolden freezes the classifier's routing decision for
+// every (expression, grid) of the conformance corpus. Routing is a
+// correctness surface of its own: a change that silently sends rate() to the
+// engine path costs the pushdown, and one that silently starts transpiling a
+// shape never proven equivalent risks wrong numbers — both must show up in
+// review as a diff of this file, with the corpus suite's clickhousev2 leg
+// judging whether the new routing still returns the reference answers.
+//
+// Regenerate after intentional classifier changes:
+//
+//	go test ./pkg/prometheus/clickhouseprometheusv2 -run TestClassificationGolden -update
+func TestClassificationGolden(t *testing.T) {
+	raw, err := os.ReadFile(corpusFile)
+	require.NoError(t, err)
+
+	var corpus struct {
+		Cases []struct {
+			Expr    string `json:"expr"`
+			StartMs int64  `json:"start_ms"`
+			EndMs   int64  `json:"end_ms"`
+			StepMs  int64  `json:"step_ms"`
+		} `json:"cases"`
+	}
+	require.NoError(t, json.Unmarshal(raw, &corpus))
+	require.NotEmpty(t, corpus.Cases)
+
+	promParser := parser.NewParser(parser.Options{})
+	seen := map[goldenEntry]bool{}
+	var entries []goldenEntry
+	for _, c := range corpus.Cases {
+		key := goldenEntry{Expr: c.Expr, StartMs: c.StartMs, EndMs: c.EndMs, StepMs: c.StepMs}
+		if seen[key] {
+			continue
+		}
+		seen[key] = true
+
+		expr, err := promParser.ParseExpr(c.Expr)
+		require.NoError(t, err, "corpus expression must parse: %q", c.Expr)
+
+		entry := key
+		plan, ok := classify(expr, gridContext{startMs: c.StartMs, endMs: c.EndMs, stepMs: c.StepMs})
+		switch {
+		case ok && plan.full:
+			entry.Plan = "full"
+		case ok:
+			entry.Plan = "hybrid"
+			entry.Units = len(plan.units)
+		default:
+			entry.Plan = "fallback"
+			entry.Reason = fallbackShape(expr)
+		}
+		entries = append(entries, entry)
+	}
+	sort.Slice(entries, func(i, j int) bool {
+		a, b := entries[i], entries[j]
+		if a.Expr != b.Expr {
+			return a.Expr < b.Expr
+		}
+		if a.StartMs != b.StartMs {
+			return a.StartMs < b.StartMs
+		}
+		if a.EndMs != b.EndMs {
+			return a.EndMs < b.EndMs
+		}
+		return a.StepMs < b.StepMs
+	})
+
+	got, err := json.MarshalIndent(entries, "", "  ")
+	require.NoError(t, err)
+	got = append(got, '\n')
+
+	if *updateGolden {
+		require.NoError(t, os.MkdirAll(filepath.Dir(goldenFile), 0o755))
+		require.NoError(t, os.WriteFile(goldenFile, got, 0o644))
+		return
+	}
+
+	want, err := os.ReadFile(goldenFile)
+	require.NoError(t, err, "golden missing — generate it with -update")
+	require.Equal(t, string(want), string(got),
+		"classification routing changed; if intentional, regenerate with -update and justify the diff in review")
+}

pkg/prometheus/clickhouseprometheusv2/client.go

@@ -0,0 +1,180 @@
+package clickhouseprometheusv2
+
+import (
+	"context"
+	"encoding/json"
+	"log/slog"
+	"math"
+	"slices"
+
+	"github.com/SigNoz/signoz/pkg/factory"
+	"github.com/SigNoz/signoz/pkg/prometheus"
+	"github.com/SigNoz/signoz/pkg/telemetrystore"
+	"github.com/SigNoz/signoz/pkg/types/ctxtypes"
+	"github.com/SigNoz/signoz/pkg/types/instrumentationtypes"
+	"github.com/SigNoz/signoz/pkg/types/telemetrytypes"
+	"github.com/prometheus/prometheus/model/labels"
+	promValue "github.com/prometheus/prometheus/model/value"
+)
+
+// seriesLookup holds a series lookup's result: matched fingerprints with
+// their labels, and the distinct metric names seen on them.
+type seriesLookup struct {
+	fingerprints map[uint64]labels.Labels
+	metricNames  []string
+}
+
+// client executes the series, samples and raw queries against ClickHouse.
+type client struct {
+	settings       factory.ScopedProviderSettings
+	telemetryStore telemetrystore.TelemetryStore
+	lookbackMs     int64
+}
+
+func newClient(settings factory.ScopedProviderSettings, telemetryStore telemetrystore.TelemetryStore, cfg prometheus.Config) *client {
+	lookback := cfg.LookbackDelta
+	if lookback <= 0 {
+		// Mirror the engine: promql defaults an unset lookback to 5m.
+		lookback = defaultLookbackDelta
+	}
+	return &client{
+		settings:       settings,
+		telemetryStore: telemetryStore,
+		lookbackMs:     lookback.Milliseconds(),
+	}
+}
+
+func (c *client) withContext(ctx context.Context, functionName string) context.Context {
+	return ctxtypes.NewContextWithCommentVals(ctx, map[string]string{
+		instrumentationtypes.TelemetrySignal:  telemetrytypes.SignalMetrics.StringValue(),
+		instrumentationtypes.CodeNamespace:    "clickhouse-prometheus-v2",
+		instrumentationtypes.CodeFunctionName: functionName,
+	})
+}
+
+func (c *client) selectSeries(ctx context.Context, query string, args []any) (*seriesLookup, error) {
+	ctx = c.withContext(ctx, "selectSeries")
+	rows, err := c.telemetryStore.ClickhouseDB().Query(ctx, query, args...)
+	if err != nil {
+		return nil, err
+	}
+	defer rows.Close()
+
+	lookup := &seriesLookup{fingerprints: make(map[uint64]labels.Labels)}
+	names := make(map[string]struct{})
+
+	var fingerprint uint64
+	var labelsJSON string
+	for rows.Next() {
+		if err := rows.Scan(&fingerprint, &labelsJSON); err != nil {
+			return nil, err
+		}
+		lset, err := unmarshalLabels(labelsJSON)
+		if err != nil {
+			return nil, err
+		}
+		lookup.fingerprints[fingerprint] = lset
+		if name := lset.Get(metricNameLabel); name != "" {
+			names[name] = struct{}{}
+		}
+	}
+	if err := rows.Err(); err != nil {
+		return nil, err
+	}
+
+	for name := range names {
+		lookup.metricNames = append(lookup.metricNames, name)
+	}
+	slices.Sort(lookup.metricNames)
+
+	return lookup, nil
+}
+
+// unmarshalLabels parses the labels JSON column, dropping empty-valued
+// labels: empty means "absent" in Prometheus, but stored attribute JSON can
+// carry them.
+func unmarshalLabels(s string) (labels.Labels, error) {
+	m := make(map[string]string)
+	if err := json.Unmarshal([]byte(s), &m); err != nil {
+		return labels.EmptyLabels(), err
+	}
+	builder := labels.NewScratchBuilder(len(m))
+	for k, v := range m {
+		if v == "" {
+			continue
+		}
+		builder.Add(k, v)
+	}
+	builder.Sort()
+	return builder.Labels(), nil
+}
+
+// selectSamples assembles per-series sample slices from a samples query (raw
+// or last-sample-per-step; same column shape), whose rows arrive ordered by
+// (fingerprint, unix_milli). Fingerprints missing from the lookup are skipped:
+// the samples query's semi-join re-runs the series predicates and can match
+// series registered after the lookup ran — the lookup is the read snapshot.
+// Stale flags map to the engine's StaleNaN. Duplicate
+// timestamps pass through: uniqueness is ingest's job, and v1 feeds them to
+// the engine as-is over the same dirty data.
+func (c *client) selectSamples(ctx context.Context, query string, args []any, lookup *seriesLookup) ([]*series, error) {
+	ctx = c.withContext(ctx, "selectSamples")
+	rows, err := c.telemetryStore.ClickhouseDB().Query(ctx, query, args...)
+	if err != nil {
+		return nil, err
+	}
+	defer rows.Close()
+
+	var (
+		result       []*series
+		current      *series
+		fingerprint  uint64
+		prevFp       uint64
+		timestampMs  int64
+		val          float64
+		flags        uint32
+		first        = true
+		haveCurrent  bool
+		staleMarker  = math.Float64frombits(promValue.StaleNaN)
+		unknownCount int
+	)
+
+	for rows.Next() {
+		if err := rows.Scan(&fingerprint, &timestampMs, &val, &flags); err != nil {
+			return nil, err
+		}
+
+		if first || fingerprint != prevFp {
+			first = false
+			prevFp = fingerprint
+			lset, ok := lookup.fingerprints[fingerprint]
+			if !ok {
+				unknownCount++
+				haveCurrent = false
+				continue
+			}
+			current = &series{lset: lset}
+			result = append(result, current)
+			haveCurrent = true
+		}
+		if !haveCurrent {
+			continue
+		}
+
+		if flags&1 == 1 {
+			val = staleMarker
+		}
+		current.ts = append(current.ts, timestampMs)
+		current.vs = append(current.vs, val)
+	}
+	if err := rows.Err(); err != nil {
+		return nil, err
+	}
+
+	if unknownCount > 0 {
+		c.settings.Logger().DebugContext(ctx, "skipped samples of fingerprints missing from series lookup",
+			slog.Int("unknown_fingerprints", unknownCount))
+	}
+
+	return result, nil
+}

pkg/prometheus/clickhouseprometheusv2/doc.go

@@ -0,0 +1,337 @@
+// Package clickhouseprometheusv2 is the second-generation ClickHouse-backed
+// Prometheus provider. It exists because the v1 provider fetches every raw
+// sample of a query's union window through the remote-read protobuf layer
+// and hands it to the engine — the cost is a function of ingested data, not
+// of the question asked, which is how a dashboard of PromQL panels takes an
+// instance down.
+//
+// Every query runs in one of two ways, decided per query:
+//
+//   - Transpiled: the query is evaluated entirely inside ClickHouse and only
+//     final (or near-final) per-group grid arrays come back, built on the
+//     timeSeries*ToGrid aggregate functions (the supported ClickHouse floor
+//     is >= 25.6, so they are assumed available).
+//   - Engine: the stock promql engine evaluates over this package's native
+//     storage.Querier. This is the path for everything not transpilable.
+//
+// Correctness is the constraint that shaped both paths: a PromQL result that
+// differs from upstream Prometheus is a lost user, so anything that cannot
+// reproduce engine semantics exactly falls back rather than approximate.
+// The rest of this comment is the PromQL -> SQL story, because that mapping
+// is where correctness is won or lost.
+//
+// # The evaluation model the SQL must reproduce
+//
+// A PromQL range query is an instant query evaluated at every grid point
+// t_i = start + i*step, i = 0..(end-start)/step. At each t_i:
+//
+//   - an instant selector resolves to the latest sample in the left-open
+//     lookback window (t_i - lookback, t_i], and to nothing when that latest
+//     sample is a stale marker — even if older real samples sit inside the
+//     window;
+//   - a range selector [r] collects every sample in (t_i - r, t_i], stale
+//     markers excluded;
+//   - offset d shifts both windows to (t_i - d - w, t_i - d].
+//
+// The transpilation invariant follows from this: every transpiled construct
+// produces, per output series, one array with exactly one slot per grid
+// point — slot i holds the value at t_i, NULL means absent. This is what
+// makes composition correct, not just convenient: the engine evaluates
+// these operators independently per t_i, so any representation that gets
+// every slot right gets the whole query right, and spatial aggregation over
+// arrays is sound because it combines values that belong to the same t_i by
+// construction. Slot index i maps back to t_i = start + i*step at scan time
+// (toMatrix). Everything below is about filling those slots with exactly
+// the numbers the engine would compute — and each equivalence was validated
+// against the vendored engine on live data before its shape entered the
+// allowlist; anything unproven stays on the engine path.
+//
+// # Classification: finding what a statement can answer
+//
+// classify walks the parsed AST looking for "core units" — maximal subtrees
+// of the shape
+//
+//	[agg by/without (...)] [fn(] selector[range] [offset d] [)] [op scalar]...
+//
+// classifyCore peels that chain from the outside in: an optional
+// sum/min/max/avg/count aggregation, then one of the allowlisted functions
+// or a bare instant selector, then the selector with its offset; on the way
+// out it accumulates number-literal arithmetic, comparisons (including
+// bool) and unary minus into a scalar-op pipeline. A node qualifies only if
+// its type, arguments and children are in the proven set — an allowlist, so
+// an overlooked construct becomes a fallback instead of a wrong number.
+//
+// Three unit kinds come out of this, each with its own SQL form:
+// unitRange (rate, irate, increase, delta, idelta over a range selector),
+// unitInstant (instant vector selection, bare or comparison-filtered) and
+// unitOverTime (avg/min/max/sum/count/last _over_time).
+//
+// If the entire tree is one unit, the plan is "full": the statement's rows
+// are the query result. Otherwise every maximal unit is cut out and replaced
+// in the expression with a synthetic selector __signoz_transpiled_N__, and
+// the rewritten expression runs in the engine over the units' materialized
+// results ("hybrid") — histogram_quantile, topk, or/and/unless and vector
+// matching keep exact engine semantics while their expensive inputs were
+// aggregated server-side.
+//
+// Classification refuses when exact semantics cannot be guaranteed
+// server-side: the @ modifier anywhere and default-resolution subqueries
+// (their resolution is a server runtime setting the transpiler cannot see);
+// duration expressions (offset step(), [range()], ...) anywhere — they are
+// resolved into the selector's static fields only at evaluation time, so at
+// classification time those fields still hold their zero values and
+// transpiling would silently use the wrong offset or range;
+// steps or ranges that are not whole seconds (the grid functions take
+// whole-second parameters); grouping by or matching on __name__ in hybrid
+// plans (the synthetic name would leak into results); name-keeping units —
+// bare/comparison instant selectors and last_over_time keep their real
+// __name__ (keepsName), which substitution would replace, so they transpile
+// only as full plans; and every function outside the allowlist (changes,
+// resets, quantile_over_time, absent, native-histogram functions, ...).
+//
+// Units inside a fixed-resolution subquery evaluate on the subquery's own
+// grid instead of the query grid: epoch-aligned multiples of the resolution
+// strictly after outerStart - offset - range, ending at outer end - offset —
+// the exact derivation the engine uses, because a grid shifted by one step
+// changes which samples every window sees.
+//
+// # From one unit to one statement
+//
+// buildUnitSQL renders each unit as a single statement. For
+// sum by (pod) (rate(m{job="api"}[5m])) the skeleton is:
+//
+//	SELECT gkey, sumForEach(grid) AS grid FROM (
+//	    SELECT series.gkey AS gkey,
+//	           timeSeriesRateToGrid(<start>, <end>, <step>, <range>)(fromUnixTimestamp64Milli(unix_milli), value) AS grid
+//	    FROM signoz_metrics.distributed_samples_v4 AS points
+//	    INNER JOIN (
+//	        SELECT fingerprint, <group key expr> AS gkey
+//	        FROM signoz_metrics.time_series_v4
+//	        WHERE <series predicates>
+//	        GROUP BY fingerprint, gkey
+//	    ) AS series ON points.fingerprint = series.fingerprint
+//	    WHERE metric_name = ? AND temporality IN ['Cumulative', 'Unspecified']
+//	      AND points.fingerprint IN (<matched fingerprints>)
+//	      AND unix_milli > <start - range> AND unix_milli <= <end>
+//	      AND bitAnd(flags, 1) = 0
+//	    GROUP BY points.fingerprint, series.gkey
+//	) GROUP BY gkey
+//	SETTINGS allow_experimental_ts_to_grid_aggregate_function = 1
+//
+// Reading it inside out:
+//
+// The time window is the selector's semantics verbatim: strict > on the
+// lower bound and <= on the upper is the left-open (t - w, t] rule, with the
+// whole window shifted by the offset. bitAnd(flags, 1) = 0 drops stale
+// markers, which PromQL excludes from range vectors.
+//
+// The inner GROUP BY computes one grid array per series.
+// timeSeriesRateToGrid(start, end, step, range) is a parametric aggregate:
+// fed (timestamp, value) pairs it produces Array(Nullable(Float64)) with one
+// slot per grid point. Correct because it implements the engine's
+// extrapolatedRate decision for decision — counter resets, the zero-point
+// clamp, the extrapolation thresholds, the >= 2 samples rule, the left-open
+// window — verified by feeding identical samples to both and comparing
+// slot for slot: the only difference ever observed is the last bit
+// (ClickHouse's C++ and Go round the same formula differently), which is
+// the floating-point floor, not a semantic gap. irate/delta/idelta map to
+// their own timeSeries*ToGrid functions with the same verification;
+// increase has no function of its own and is emitted as
+// arrayMap(x -> x * <range seconds>, <rate expr>), exact by definition —
+// extrapolatedRate computes the same extrapolated delta for both and
+// divides by the range only when isRate, so multiplying it back is the
+// identity, not an approximation. The grid parameters are rendered as
+// literals, not bound args — they are aggregate-function parameters — and
+// the experimental gate rides as a SETTINGS clause on the statement itself
+// so telemetrystore hooks cannot clobber it.
+//
+// The join annotates each series with its group key, in one of two forms.
+// by (...) extracts each listed label as a plain column
+// (JSONExtractString(labels, 'pod') AS g0) and groups on the columns
+// directly: the projection is a known short list and the label names live
+// in Go, so building, sorting and stringifying every label pair per row
+// would be waste. Correct because column-tuple equality is label-set
+// equality on the projection, and an extracted '' is the label being
+// absent — Prometheus semantics for by() over missing labels, and empties
+// are skipped when the columns turn back into labels. without and
+// no-aggregation project a label SET that varies per series, so they get
+// the canonical key: toJSONString of the sorted [label, value] pairs the
+// unit projects (without excludes the listed labels plus __name__; no
+// aggregation keeps everything, the name coming off in Go per the engine's
+// name-dropping rules). There the sort is load-bearing — stored JSON key
+// order is not canonical across fingerprints, and two orderings of the same
+// labels must land in one group — empty values are filtered for the same
+// absent-label reason, and the same string parses back into the output
+// label set (labelsFromGroupKey).
+//
+// The outer GROUP BY is the spatial aggregation: sum/min/max/avg/count
+// by/without become the -ForEach combinators. Element-wise aggregation over
+// grid arrays is the engine's per-t_i aggregation, because slot i of every
+// input array refers to the same t_i; the combinators skip NULLs, which is
+// the engine aggregating only the series present at t_i, and an index where
+// every series is absent stays NULL. Two edges need explicit handling:
+// countForEach wraps in a mapping of 0 back to NULL, because a count over
+// an all-absent index is an absent point, not 0; and a unit without
+// aggregation still passes through maxForEach — the identity for the common
+// one-fingerprint group, and a deterministic NULL-skipping merge when a
+// regex __name__ selector collapses distinct metrics onto one projected
+// label set. One caveat is inherent: summation order over series differs
+// from the engine's, so spatial aggregates can differ in the last ULP —
+// float addition is not associative; no ordering reproduces the engine's
+// bit-exactly from inside a GROUP BY.
+//
+// # Instant selectors: staleness needs two aggregates
+//
+// unitInstant uses window = lookback and must reproduce the shadowing rule:
+// the point is absent when the latest in-window sample is a stale marker.
+// timeSeriesLastToGrid alone cannot express that — skipping stale rows in
+// WHERE would resurrect the older real sample the marker was written to
+// bury. So stale rows stay in the scan for this kind only, and the grid
+// expression compares three aggregates per slot:
+//
+//	arrayMap((tall, tok, vok) -> if(tall IS NULL OR tok IS NULL OR tall != tok, NULL, vok),
+//	         timeSeriesLastToGrid(...)(ts, toFloat64(unix_milli)),                           -- last sample overall
+//	         timeSeriesLastToGridIf(...)(ts, toFloat64(unix_milli), bitAnd(flags, 1) = 0),   -- last non-stale, its timestamp
+//	         timeSeriesLastToGridIf(...)(ts, value, bitAnd(flags, 1) = 0))                   -- last non-stale, its value
+//
+// Correct by cases on a slot's window. No samples at all: both timestamp
+// aggregates are NULL, the slot is NULL — absent, as the engine says. Latest
+// sample non-stale: it is the latest overall and the latest non-stale, the
+// timestamps agree, the slot takes its value — the engine's pick. Latest
+// sample stale: the last-overall timestamp is the marker's, the
+// last-non-stale timestamp is older (or NULL when only markers are in
+// window), they disagree, the slot is NULL — the marker shadows, exactly
+// the engine's rule. Timestamps are unique per series (ingest dedups), so
+// timestamp equality identifies "the same sample" without ambiguity. The
+// -If combinator's applicability to these experimental aggregates was
+// probed before being trusted, not assumed.
+//
+// # Windowed *_over_time: whole buckets instead of a grid function
+//
+// avg/min/max/sum/count _over_time aggregate every raw sample in the window,
+// and no timeSeries*ToGrid function computes them. (last_over_time is the
+// exception: the last sample of a range vector — stale markers excluded from
+// range vectors by PromQL, excluded here in WHERE — is exactly
+// timeSeriesLastToGrid.) These transpile only when the range is a whole
+// multiple of the step, and then the window needs no per-sample fan-out at
+// all: with W = range/step, the window (t_k - range, t_k] is exactly the
+// union of W step buckets — both are left-open on the same boundaries — so
+// bucket membership fully determines window membership. Each sample lands
+// in exactly one bucket by a plain GROUP BY:
+//
+//	intDiv(unix_milli - <start> + <range> - 1, <step>) AS jj
+//
+// (ceil((ts - start)/step) shifted by W-1 so the earliest in-window sample
+// sits at 0; slot k's window is buckets jj in [k, k+W-1]). The alternative —
+// fanning each sample into all W windows that cover it — multiplies rows by
+// W, which for a long range over a short step is a row explosion measured
+// in billions; the bucketed form's row count is series x buckets, the size
+// of the output, regardless of W.
+//
+// The shard level aggregates per (series, group key, bucket): a bucket
+// count plus the function's value aggregate (sum for sum/avg, min, max).
+// The assembly level places the partials into dense arrays
+// (groupArrayInsertAt — positions are unique, one row per bucket; counts
+// and sums default to 0, which contributes nothing) and slides: slot k
+// combines its at-most-W bucket partials by direct aggregation, so window
+// sums are added the way the engine adds them — no prefix-sum differencing,
+// whose large-minus-large cancellation would drift past the shadow
+// tolerance on counter-sized values. Correct per slot because the bucket
+// union is the exact window multiset and avg/min/max/sum/count are
+// order-insensitive on a multiset (sum/avg up to summation order, the float
+// caveat above). A slot with zero window count is absent; min/max filter
+// their slices on the bucket counts, so an empty bucket's default can never
+// be mistaken for a value — a real sample can legitimately be +Inf.
+//
+// Ranges that don't divide the step, and windows wider than
+// maxWindowBuckets buckets (the slide costs W combines per slot), fall back
+// to the engine path, which is exact.
+//
+// # Scalar ops, full plans, hybrid plans
+//
+// The scalar-op pipeline applies in Go to the returned arrays
+// (applyScalarOps), slot by slot: arithmetic operators compute, comparisons
+// filter (the slot keeps the vector-side value or becomes NULL) or return
+// 0/1 under bool. Correct trivially: it is the same float64 operation the
+// engine would apply to the same slot value, in the same operator order the
+// AST dictates — running it in Go instead of another SQL layer changes
+// where, not what.
+//
+// A full plan's arrays map straight to the result matrix. A hybrid plan
+// materializes each unit's arrays as synthetic series under its
+// __signoz_transpiled_N__ name and evaluates the rewritten expression over
+// a storage that serves synthetic names from memory and everything else
+// live. Substitution is sound because a unit's output is a plain instant
+// vector to the engine — same values at same timestamps under a different
+// name, and the name cannot matter: plans that group by or match on
+// __name__ were refused at classification, and name-keeping units are never
+// substituted. One subtlety makes it exact: stale markers are written at
+// absent grid points, because the engine's lookback would otherwise
+// resurrect a point from up to lookback earlier — the marker encodes
+// "absent here" the way the engine itself encodes it. Units evaluate
+// concurrently; each is one series lookup plus one grid statement. A step
+// of 0 is an instant query: a single evaluation at end.
+//
+// # Series lookup
+//
+// Both paths resolve matchers the same way, once per selector
+// (selectSeries), against the series tables holding one row per
+// (fingerprint, bucket) at 1h/6h/1d/1w granularities; timeSeriesTableFor
+// picks the table whose bucket fits the window and rounds the window start
+// down to the bucket boundary. How matchers become SQL, and why regexes are
+// anchored, is documented at applySeriesConditions. Empty-valued labels come
+// off at this boundary: an empty value means "label absent" in Prometheus,
+// but stored attribute JSON can carry them.
+//
+// # The engine path
+//
+// Queries that do not transpile run in the stock engine over this package's
+// storage.Querier, which is still not the v1 path. Samples are fetched per
+// selector using the engine's per-selector hints, not the query-wide union
+// window, so foo / foo offset 1d reads two narrow windows instead of the
+// widest one twice. Instant selectors of subquery-free queries fetch only
+// the last sample per step bucket (lastSamplePerStep): buckets anchor at the
+// selector's first evaluation timestamp — recovered from the hints as
+// hints.Start + lookback - 1ms, the inverse of how the engine derives
+// hints.Start — so bucket boundaries coincide with evaluation timestamps and
+// a non-final sample of a bucket can never be the latest sample in
+// (t - lookback, t] for any grid t. Real timestamps are preserved, so the
+// engine's own lookback and staleness handling stay exact. Range selectors
+// always fetch raw — every sample feeds the range function — and the
+// subquery-free proof travels in the context as prometheus.QueryTraits,
+// because subquery selectors evaluate at the subquery's step while the
+// hints carry the top-level step. Row assembly maps stale flags to the
+// engine's StaleNaN and merges series with identical label sets
+// (sortAndMerge) — the engine assumes storages never emit duplicates.
+//
+// # Sharding
+//
+// samples_v4 and time_series_v4 (and all their rollups) shard on the same
+// key — cityHash64(env, temporality, metric_name, fingerprint) — so a
+// series' samples and catalog rows live on the same shard. The transpiled
+// statement above exploits that: the distributed samples table at the
+// top-level FROM makes ClickHouse rewrite the whole inner query per shard,
+// where the join against the shard-local series table and the per-series
+// grid aggregation run next to the data; the initiator only merges
+// aggregate states and applies the spatial -ForEach step. Same layout as
+// the telemetrymetrics statement builder. The group-key join alone
+// restricts the transpiled scan to the matched series; the engine path's
+// samples fetch restricts by the same predicates as a shard-local
+// semi-join, not a GLOBAL broadcast of the matched set. The temporality
+// filter on every
+// samples statement is a semantic no-op — the matched fingerprints already
+// come from those temporalities — that engages the leading samples
+// primary-key column. Delta-temporality series stay invisible to PromQL
+// here exactly as they are in v1: the rollout gate is parity with v1, and
+// making Delta visible is its own change with its own semantics to design —
+// a Delta stream fed to rate() as-if-cumulative would be wrong, not just
+// new.
+//
+// # Observability
+//
+// Every statement carries a log_comment with
+// code.namespace=clickhouse-prometheus-v2 and code.function.name naming the
+// call site (selectSeries, selectSamples, transpiledUnit, LabelValues,
+// LabelNames), so this provider's work is attributable in system.query_log
+// without guessing from query text.
+package clickhouseprometheusv2

pkg/prometheus/clickhouseprometheusv2/provider.go

@@ -0,0 +1,84 @@
+package clickhouseprometheusv2
+
+import (
+	"context"
+	"time"
+
+	"github.com/SigNoz/signoz/pkg/factory"
+	"github.com/SigNoz/signoz/pkg/prometheus"
+	"github.com/SigNoz/signoz/pkg/telemetrystore"
+	"github.com/prometheus/prometheus/promql"
+	"github.com/prometheus/prometheus/storage"
+)
+
+// Provider ties the package together: its own engine and parser, the
+// ClickHouse client behind the native storage.Querier, and the transpiler
+// executor. See the package documentation for what runs where and why. It is
+// exported as a concrete type — pkg/querier holds it directly for shadow
+// comparison and pinned serving, and an interface with a single
+// implementation would only hide that dependency.
+type Provider struct {
+	settings factory.ScopedProviderSettings
+	engine   *prometheus.Engine
+	parser   prometheus.Parser
+	client   *client
+	executor *executor
+}
+
+var (
+	_ prometheus.Prometheus        = (*Provider)(nil)
+	_ prometheus.StatementCapturer = (*Provider)(nil)
+)
+
+func NewFactory(telemetryStore telemetrystore.TelemetryStore) factory.ProviderFactory[prometheus.Prometheus, prometheus.Config] {
+	return factory.NewProviderFactory(factory.MustNewName("clickhousev2"), func(ctx context.Context, providerSettings factory.ProviderSettings, config prometheus.Config) (prometheus.Prometheus, error) {
+		return New(ctx, providerSettings, config, telemetryStore)
+	})
+}
+
+func New(_ context.Context, providerSettings factory.ProviderSettings, config prometheus.Config, telemetryStore telemetrystore.TelemetryStore) (*Provider, error) {
+	settings := factory.NewScopedProviderSettings(providerSettings, "github.com/SigNoz/signoz/pkg/prometheus/clickhouseprometheusv2")
+
+	engine := prometheus.NewEngine(settings.Logger(), config)
+	parser := prometheus.NewParser()
+	client := newClient(settings, telemetryStore, config)
+
+	return &Provider{
+		settings: settings,
+		engine:   engine,
+		parser:   parser,
+		client:   client,
+		executor: &executor{client: client, engine: engine, parser: parser},
+	}, nil
+}
+
+// TryExecuteRange evaluates transpilable query shapes directly in ClickHouse
+// (see transpiler.go). ok=false means the shape is not transpilable and the
+// caller should evaluate through Engine over Storage instead.
+func (p *Provider) TryExecuteRange(ctx context.Context, query string, start, end time.Time, step time.Duration) (promql.Matrix, bool, error) {
+	return p.executor.TryExecuteRange(ctx, query, start, end, step)
+}
+
+func (p *Provider) Engine() *prometheus.Engine {
+	return p.engine
+}
+
+func (p *Provider) Parser() prometheus.Parser {
+	return p.parser
+}
+
+func (p *Provider) Storage() storage.Queryable {
+	return p
+}
+
+func (p *Provider) Querier(mint, maxt int64) (storage.Querier, error) {
+	return &querier{mint: mint, maxt: maxt, client: p.client}, nil
+}
+
+// CapturingStorage implements prometheus.StatementCapturer: a storage that
+// records each selector's SQL without executing it, for the preview path.
+// A fresh recorder per call keeps concurrent dry-runs isolated.
+func (p *Provider) CapturingStorage() (storage.Queryable, prometheus.StatementRecorder) {
+	recorder := &statementRecorder{}
+	return &captureQueryable{client: p.client, recorder: recorder}, recorder
+}

pkg/prometheus/clickhouseprometheusv2/querier.go

@@ -0,0 +1,170 @@
+package clickhouseprometheusv2
+
+import (
+	"context"
+	"fmt"
+	"slices"
+	"time"
+
+	"github.com/SigNoz/signoz/pkg/prometheus"
+	"github.com/huandu/go-sqlbuilder"
+	"github.com/prometheus/prometheus/model/labels"
+	"github.com/prometheus/prometheus/storage"
+	"github.com/prometheus/prometheus/util/annotations"
+)
+
+// defaultLookbackDelta mirrors promql's default when the config leaves the
+// lookback unset; the engine and the storage must agree on it for
+// last-sample-per-step bucket anchoring.
+const defaultLookbackDelta = 5 * time.Minute
+
+// querier is a native storage.Querier over ClickHouse: Select builds SQL
+// directly from the matchers and hints, with no remote-read protobuf layer.
+type querier struct {
+	mint, maxt int64
+	client     *client
+}
+
+var _ storage.Querier = (*querier)(nil)
+
+func (q *querier) Select(ctx context.Context, sortSeries bool, hints *storage.SelectHints, matchers ...*labels.Matcher) storage.SeriesSet {
+	start, end := q.window(hints)
+
+	seriesQuery, seriesArgs, err := buildSeriesQuery(start, end, matchers)
+	if err != nil {
+		return storage.ErrSeriesSet(err)
+	}
+	lookup, err := q.client.selectSeries(ctx, seriesQuery, seriesArgs)
+	if err != nil {
+		return storage.ErrSeriesSet(err)
+	}
+	if len(lookup.fingerprints) == 0 {
+		return storage.EmptySeriesSet()
+	}
+
+	list, err := q.fetchSamples(ctx, start, end, matchers, lookup, q.lastSamplePerStepFor(ctx, hints))
+	if err != nil {
+		return storage.ErrSeriesSet(err)
+	}
+
+	// The engine assumes storages never emit duplicate label sets.
+	list = sortAndMerge(list)
+	return newSeriesSet(list)
+}
+
+func (q *querier) LabelValues(ctx context.Context, name string, hints *storage.LabelHints, matchers ...*labels.Matcher) ([]string, annotations.Annotations, error) {
+	sb := sqlbuilder.NewSelectBuilder()
+	if name == metricNameLabel {
+		sb.Select("DISTINCT metric_name AS value")
+	} else {
+		sb.Select(fmt.Sprintf("DISTINCT JSONExtractString(labels, %s) AS value", sb.Var(name)))
+	}
+	adjustedStart, table := timeSeriesTableFor(q.mint, q.maxt)
+	sb.From(fmt.Sprintf("%s.%s", databaseName, table))
+	if err := applySeriesConditions(sb, adjustedStart, q.maxt, matchers); err != nil {
+		return nil, nil, err
+	}
+	sb.Where("value != ''")
+	if hints != nil && hints.Limit > 0 {
+		sb.Limit(hints.Limit)
+	}
+
+	query, args := sb.BuildWithFlavor(sqlbuilder.ClickHouse)
+	values, err := q.selectStrings(ctx, "LabelValues", query, args)
+	if err != nil {
+		return nil, nil, err
+	}
+	slices.Sort(values)
+	return values, nil, nil
+}
+
+func (q *querier) LabelNames(ctx context.Context, hints *storage.LabelHints, matchers ...*labels.Matcher) ([]string, annotations.Annotations, error) {
+	sb := sqlbuilder.NewSelectBuilder()
+	sb.Select("DISTINCT arrayJoin(JSONExtractKeys(labels)) AS name")
+	adjustedStart, table := timeSeriesTableFor(q.mint, q.maxt)
+	sb.From(fmt.Sprintf("%s.%s", databaseName, table))
+	if err := applySeriesConditions(sb, adjustedStart, q.maxt, matchers); err != nil {
+		return nil, nil, err
+	}
+	if hints != nil && hints.Limit > 0 {
+		sb.Limit(hints.Limit)
+	}
+
+	query, args := sb.BuildWithFlavor(sqlbuilder.ClickHouse)
+	names, err := q.selectStrings(ctx, "LabelNames", query, args)
+	if err != nil {
+		return nil, nil, err
+	}
+	slices.Sort(names)
+	return names, nil, nil
+}
+
+func (q *querier) Close() error {
+	return nil
+}
+
+// window returns the per-selector fetch window from the hints (already
+// adjusted for offset, @, range and lookback); mint/maxt span the union of
+// all selectors and are the fallback.
+func (q *querier) window(hints *storage.SelectHints) (int64, int64) {
+	if hints != nil && hints.Start != 0 && hints.End != 0 && hints.Start <= hints.End {
+		return hints.Start, hints.End
+	}
+	return q.mint, q.maxt
+}
+
+// lastSamplePerStepFor decides whether the fetch can keep only the last
+// sample per step bucket (see lastSamplePerStep). Only instant selectors
+// (hints.Range == 0) of subquery-free queries qualify: range selectors need
+// every raw sample, and subquery selectors evaluate at the subquery's own
+// step while hints carry the top-level step — the subquery-free proof
+// arrives as QueryTraits in the context. The first evaluation timestamp is
+// recovered as hints.Start + lookback - 1ms, inverting how the engine
+// derives hints.Start.
+func (q *querier) lastSamplePerStepFor(ctx context.Context, hints *storage.SelectHints) *lastSamplePerStep {
+	if hints == nil || hints.Range != 0 || hints.Start <= 0 {
+		return nil
+	}
+	traits, ok := prometheus.QueryTraitsFromContext(ctx)
+	if !ok || !traits.SubqueryFree {
+		return nil
+	}
+	firstEval := hints.Start + q.client.lookbackMs - 1
+	if firstEval > hints.End {
+		// Defensive: never anchor a bucket past the window.
+		firstEval = hints.End
+	}
+	return &lastSamplePerStep{firstEvalMs: firstEval, stepMs: hints.Step}
+}
+
+// fetchSamples runs the samples query for the matched series (see
+// buildSamplesQuery).
+func (q *querier) fetchSamples(ctx context.Context, start, end int64, matchers []*labels.Matcher, lookup *seriesLookup, lastPerStep *lastSamplePerStep) ([]*series, error) {
+	query, args, err := buildSamplesQuery(start, end, lookup.metricNames, matchers, lastPerStep)
+	if err != nil {
+		return nil, err
+	}
+	return q.client.selectSamples(ctx, query, args, lookup)
+}
+
+func (q *querier) selectStrings(ctx context.Context, fn, query string, args []any) ([]string, error) {
+	ctx = q.client.withContext(ctx, fn)
+	rows, err := q.client.telemetryStore.ClickhouseDB().Query(ctx, query, args...)
+	if err != nil {
+		return nil, err
+	}
+	defer rows.Close()
+
+	var out []string
+	var v string
+	for rows.Next() {
+		if err := rows.Scan(&v); err != nil {
+			return nil, err
+		}
+		out = append(out, v)
+	}
+	if err := rows.Err(); err != nil {
+		return nil, err
+	}
+	return out, nil
+}

pkg/prometheus/clickhouseprometheusv2/seriesset.go

@@ -0,0 +1,184 @@
+package clickhouseprometheusv2
+
+import (
+	"sort"
+
+	"github.com/prometheus/prometheus/model/histogram"
+	"github.com/prometheus/prometheus/model/labels"
+	"github.com/prometheus/prometheus/storage"
+	"github.com/prometheus/prometheus/tsdb/chunkenc"
+	"github.com/prometheus/prometheus/util/annotations"
+)
+
+// series is one time series with samples as parallel slices ordered by
+// timestamp. Deliberately not storage.NewListSeries: that boxes every sample
+// as an interface value, a per-sample allocation this fetch path exists to
+// avoid.
+type series struct {
+	lset labels.Labels
+	ts   []int64
+	vs   []float64
+}
+
+var _ storage.Series = (*series)(nil)
+
+func (s *series) Labels() labels.Labels {
+	return s.lset
+}
+
+func (s *series) Iterator(it chunkenc.Iterator) chunkenc.Iterator {
+	if fit, ok := it.(*floatIterator); ok {
+		fit.reset(s)
+		return fit
+	}
+	fit := &floatIterator{}
+	fit.reset(s)
+	return fit
+}
+
+// floatIterator implements chunkenc.Iterator over a series' sample slices.
+type floatIterator struct {
+	s *series
+	i int
+}
+
+var _ chunkenc.Iterator = (*floatIterator)(nil)
+
+func (it *floatIterator) reset(s *series) {
+	it.s = s
+	it.i = -1
+}
+
+func (it *floatIterator) Next() chunkenc.ValueType {
+	it.i++
+	if it.i >= len(it.s.ts) {
+		return chunkenc.ValNone
+	}
+	return chunkenc.ValFloat
+}
+
+func (it *floatIterator) Seek(t int64) chunkenc.ValueType { //nolint:govet // stdmethods flags io.Seeker; this is chunkenc.Iterator's Seek
+	if it.i < 0 {
+		it.i = 0
+	}
+	if it.i >= len(it.s.ts) {
+		return chunkenc.ValNone
+	}
+	// The current position, once valid, must not move backwards.
+	if it.s.ts[it.i] >= t {
+		return chunkenc.ValFloat
+	}
+	it.i += sort.Search(len(it.s.ts)-it.i, func(j int) bool {
+		return it.s.ts[it.i+j] >= t
+	})
+	if it.i >= len(it.s.ts) {
+		return chunkenc.ValNone
+	}
+	return chunkenc.ValFloat
+}
+
+func (it *floatIterator) At() (int64, float64) {
+	return it.s.ts[it.i], it.s.vs[it.i]
+}
+
+func (it *floatIterator) AtHistogram(*histogram.Histogram) (int64, *histogram.Histogram) {
+	return 0, nil
+}
+
+func (it *floatIterator) AtFloatHistogram(*histogram.FloatHistogram) (int64, *histogram.FloatHistogram) {
+	return 0, nil
+}
+
+func (it *floatIterator) AtT() int64 {
+	return it.s.ts[it.i]
+}
+
+// AtST returns the current start timestamp; not tracked by this storage.
+func (it *floatIterator) AtST() int64 {
+	return 0
+}
+
+func (it *floatIterator) Err() error {
+	return nil
+}
+
+// seriesSet iterates a fully materialized, label-sorted list of series.
+type seriesSet struct {
+	series []*series
+	i      int
+}
+
+var _ storage.SeriesSet = (*seriesSet)(nil)
+
+func newSeriesSet(list []*series) *seriesSet {
+	return &seriesSet{series: list, i: -1}
+}
+
+func (s *seriesSet) Next() bool {
+	s.i++
+	return s.i < len(s.series)
+}
+
+func (s *seriesSet) At() storage.Series {
+	return s.series[s.i]
+}
+
+func (s *seriesSet) Err() error {
+	return nil
+}
+
+func (s *seriesSet) Warnings() annotations.Annotations {
+	return nil
+}
+
+// sortAndMerge orders series by label set and merges identical label sets
+// by timestamp (first sample wins ties): distinct fingerprints can carry
+// identical label sets, and the engine assumes storages never emit
+// duplicates.
+func sortAndMerge(list []*series) []*series {
+	if len(list) < 2 {
+		return list
+	}
+	sort.Slice(list, func(i, j int) bool {
+		return labels.Compare(list[i].lset, list[j].lset) < 0
+	})
+	out := list[:1]
+	for _, s := range list[1:] {
+		last := out[len(out)-1]
+		if labels.Compare(last.lset, s.lset) != 0 {
+			out = append(out, s)
+			continue
+		}
+		merged := mergeSamples(last, s)
+		out[len(out)-1] = merged
+	}
+	return out
+}
+
+func mergeSamples(a, b *series) *series {
+	ts := make([]int64, 0, len(a.ts)+len(b.ts))
+	vs := make([]float64, 0, len(a.ts)+len(b.ts))
+	i, j := 0, 0
+	for i < len(a.ts) && j < len(b.ts) {
+		switch {
+		case a.ts[i] < b.ts[j]:
+			ts = append(ts, a.ts[i])
+			vs = append(vs, a.vs[i])
+			i++
+		case a.ts[i] > b.ts[j]:
+			ts = append(ts, b.ts[j])
+			vs = append(vs, b.vs[j])
+			j++
+		default:
+			ts = append(ts, a.ts[i])
+			vs = append(vs, a.vs[i])
+			i++
+			j++
+		}
+	}
+	ts = append(ts, a.ts[i:]...)
+	vs = append(vs, a.vs[i:]...)
+	ts = append(ts, b.ts[j:]...)
+	vs = append(vs, b.vs[j:]...)
+	return &series{lset: a.lset, ts: ts, vs: vs}
+}

pkg/prometheus/clickhouseprometheusv2/sql.go

@@ -0,0 +1,164 @@
+package clickhouseprometheusv2
+
+import (
+	"fmt"
+
+	"github.com/SigNoz/signoz/pkg/errors"
+	"github.com/huandu/go-sqlbuilder"
+	"github.com/prometheus/prometheus/model/labels"
+)
+
+// buildSeriesQuery renders the series lookup: one row per matched fingerprint
+// with its labels.
+func buildSeriesQuery(start, end int64, matchers []*labels.Matcher) (string, []any, error) {
+	adjustedStart, table := timeSeriesTableFor(start, end)
+	sb := sqlbuilder.NewSelectBuilder()
+	sb.Select("fingerprint", "any(labels)")
+	sb.From(fmt.Sprintf("%s.%s", databaseName, table))
+	if err := applySeriesConditions(sb, adjustedStart, end, matchers); err != nil {
+		return "", nil, err
+	}
+	sb.GroupBy("fingerprint")
+	query, args := sb.BuildWithFlavor(sqlbuilder.ClickHouse)
+	return query, args, nil
+}
+
+// buildSamplesQuery renders the samples fetch for the matched series: a
+// semi-join re-runs the series predicates against the shard-local series
+// table (complete by fingerprint co-locality, see localTimeSeriesTable — a
+// GLOBAL broadcast would ship the matched set to every shard, and ClickHouse
+// materializes the subquery's set per shard before the scan, so it still
+// engages the fingerprint primary-key column). metricNames (observed on the
+// matched series when the selector had no __name__ equality) narrows the
+// primary-key scan. A non-nil lastPerStep groups to the last sample per
+// step bucket.
+func buildSamplesQuery(start, end int64, metricNames []string, matchers []*labels.Matcher, lastPerStep *lastSamplePerStep) (string, []any, error) {
+	sb := sqlbuilder.NewSelectBuilder()
+	if lastPerStep != nil {
+		// Aliases must not shadow source columns: ClickHouse resolves aliases
+		// in WHERE too, and "max(unix_milli) AS unix_milli" would put an
+		// aggregate into the WHERE clause (error 184).
+		sb.Select("fingerprint", "max(unix_milli) AS ts", "argMax(value, unix_milli) AS val", "argMax(flags, unix_milli) AS fl")
+	} else {
+		sb.Select("fingerprint", "unix_milli", "value", "flags")
+	}
+	sb.From(fmt.Sprintf("%s.%s", databaseName, distributedSamplesV4))
+
+	switch len(metricNames) {
+	case 0:
+		// No name constraint derivable; the primary-key prefix goes unused.
+	case 1:
+		sb.Where(sb.EQ("metric_name", metricNames[0]))
+	default:
+		sb.Where(sb.In("metric_name", sqlbuilder.List(metricNames)))
+	}
+	// Semantically redundant (the fingerprints already come from these
+	// temporalities) but engages the leading primary-key column.
+	sb.Where("temporality IN ['Cumulative', 'Unspecified']")
+	sub := sqlbuilder.NewSelectBuilder()
+	sub.Select("fingerprint")
+	adjustedStart, table := timeSeriesTableFor(start, end)
+	sub.From(fmt.Sprintf("%s.%s", databaseName, localTimeSeriesTable(table)))
+	if err := applySeriesConditions(sub, adjustedStart, end, matchers); err != nil {
+		return "", nil, err
+	}
+	sb.Where(sb.In("fingerprint", sub))
+	sb.Where(sb.GTE("unix_milli", start), sb.LTE("unix_milli", end))
+
+	if lastPerStep != nil {
+		sb.GroupBy("fingerprint")
+		if expr := lastPerStep.bucketExpr(); expr != "" {
+			sb.GroupBy(expr)
+		}
+		sb.OrderBy("fingerprint", "ts")
+	} else {
+		sb.OrderBy("fingerprint", "unix_milli")
+	}
+
+	query, args := sb.BuildWithFlavor(sqlbuilder.ClickHouse)
+	return query, args, nil
+}
+
+// applySeriesConditions adds the WHERE conditions of a series table scan:
+// __name__ matchers (all four types) translate to the metric_name column,
+// every other matcher to a JSONExtractString condition on the labels column.
+// An equality matcher against "" matches series without the label, mirroring
+// PromQL, because JSONExtractString returns "" for missing keys. Regexes are
+// anchored: PromQL matchers match the whole value, ClickHouse match()
+// searches for a substring.
+func applySeriesConditions(sb *sqlbuilder.SelectBuilder, start, end int64, matchers []*labels.Matcher) error {
+	for _, m := range matchers {
+		if m.Name != metricNameLabel {
+			continue
+		}
+		switch m.Type {
+		case labels.MatchEqual:
+			sb.Where(sb.EQ("metric_name", m.Value))
+		case labels.MatchNotEqual:
+			sb.Where(sb.NE("metric_name", m.Value))
+		case labels.MatchRegexp:
+			sb.Where(fmt.Sprintf("match(metric_name, %s)", sb.Var(anchorRegex(m.Value))))
+		case labels.MatchNotRegexp:
+			sb.Where(fmt.Sprintf("NOT match(metric_name, %s)", sb.Var(anchorRegex(m.Value))))
+		default:
+			return errors.NewInvalidInputf(errors.CodeInvalidInput, "unsupported matcher type %q for __name__", m.Type)
+		}
+	}
+
+	sb.Where("temporality IN ['Cumulative', 'Unspecified']")
+	// Inclusive upper bound: registration rows are hour-floored (and 6h/1d/1w
+	// for the rollup tables) by the exporter, so a series first registered in
+	// the bucket starting exactly at `end` would otherwise be invisible while
+	// its samples (<= end) are in range.
+	sb.Where(sb.GTE("unix_milli", start), sb.LTE("unix_milli", end))
+
+	for _, m := range matchers {
+		if m.Name == metricNameLabel {
+			continue
+		}
+		switch m.Type {
+		case labels.MatchEqual:
+			sb.Where(fmt.Sprintf("JSONExtractString(labels, %s) = %s", sb.Var(m.Name), sb.Var(m.Value)))
+		case labels.MatchNotEqual:
+			sb.Where(fmt.Sprintf("JSONExtractString(labels, %s) != %s", sb.Var(m.Name), sb.Var(m.Value)))
+		case labels.MatchRegexp:
+			sb.Where(fmt.Sprintf("match(JSONExtractString(labels, %s), %s)", sb.Var(m.Name), sb.Var(anchorRegex(m.Value))))
+		case labels.MatchNotRegexp:
+			sb.Where(fmt.Sprintf("NOT match(JSONExtractString(labels, %s), %s)", sb.Var(m.Name), sb.Var(anchorRegex(m.Value))))
+		default:
+			return errors.NewInvalidInputf(errors.CodeInvalidInput, "unsupported matcher type %q", m.Type)
+		}
+	}
+
+	return nil
+}
+
+// anchorRegex turns a PromQL regex into its fully-anchored form (see
+// applySeriesConditions).
+func anchorRegex(v string) string {
+	return "^(?:" + v + ")$"
+}
+
+// lastSamplePerStep reduces an instant-selector fetch to the last sample of
+// each step bucket: bucket 0 is (start, firstEval], bucket i is
+// (firstEval+(i-1)·step, firstEval+i·step]. Anchoring buckets at the first
+// evaluation timestamp makes every bucket boundary an evaluation timestamp,
+// so a non-final sample of a bucket can never be the latest sample in any
+// (t-lookback, t] the engine resolves — the reduction is lossless. Real
+// timestamps are preserved, so the engine's own lookback and staleness
+// handling remain exact.
+type lastSamplePerStep struct {
+	firstEvalMs int64
+	stepMs      int64
+}
+
+func (t *lastSamplePerStep) bucketExpr() string {
+	if t.stepMs <= 0 {
+		// Instant query: a single evaluation at firstEval; one bucket.
+		return ""
+	}
+	return fmt.Sprintf(
+		"if(unix_milli <= %d, 0, intDiv(unix_milli - %d - 1, %d) + 1)",
+		t.firstEvalMs, t.firstEvalMs, t.stepMs,
+	)
+}

pkg/prometheus/clickhouseprometheusv2/sql_test.go

@@ -0,0 +1,139 @@
+package clickhouseprometheusv2
+
+import (
+	"testing"
+	"time"
+
+	"github.com/prometheus/prometheus/model/labels"
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+)
+
+func mustMatcher(t *testing.T, mt labels.MatchType, name, value string) *labels.Matcher {
+	t.Helper()
+	m, err := labels.NewMatcher(mt, name, value)
+	require.NoError(t, err)
+	return m
+}
+
+func TestTimeSeriesTableFor(t *testing.T) {
+	base := time.Date(2026, 7, 10, 3, 27, 0, 0, time.UTC).UnixMilli()
+
+	tests := []struct {
+		name      string
+		span      time.Duration
+		wantTable string
+		roundTo   time.Duration
+	}{
+		{"under 6h uses hourly table", 2 * time.Hour, distributedTimeSeriesV4, time.Hour},
+		{"under 1d uses 6h table", 12 * time.Hour, distributedTimeSeriesV46hrs, 6 * time.Hour},
+		{"under 1w uses 1d table", 3 * 24 * time.Hour, distributedTimeSeriesV41day, 24 * time.Hour},
+		{"over 1w uses 1w table", 10 * 24 * time.Hour, distributedTimeSeriesV41week, 7 * 24 * time.Hour},
+	}
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			start, table := timeSeriesTableFor(base, base+tt.span.Milliseconds())
+			assert.Equal(t, tt.wantTable, table)
+			assert.Zero(t, start%tt.roundTo.Milliseconds())
+			assert.LessOrEqual(t, start, base)
+		})
+	}
+}
+
+func TestBuildSeriesQuery(t *testing.T) {
+	start := int64(1_700_000_000_000)
+	end := start + time.Hour.Milliseconds()
+	// The series table window rounds down to the table's bucket boundary.
+	adjustedStart := start - (start % time.Hour.Milliseconds())
+
+	t.Run("equality name and label matchers", func(t *testing.T) {
+		query, args, err := buildSeriesQuery(start, end, []*labels.Matcher{
+			mustMatcher(t, labels.MatchEqual, "__name__", "http_requests_total"),
+			mustMatcher(t, labels.MatchEqual, "job", "api"),
+		})
+		require.NoError(t, err)
+		assert.Equal(t,
+			"SELECT fingerprint, any(labels) FROM signoz_metrics.distributed_time_series_v4 WHERE metric_name = ? AND temporality IN ['Cumulative', 'Unspecified'] AND unix_milli >= ? AND unix_milli <= ? AND JSONExtractString(labels, ?) = ? GROUP BY fingerprint",
+			query,
+		)
+		assert.Equal(t, []any{"http_requests_total", adjustedStart, end, "job", "api"}, args)
+	})
+
+	t.Run("regex matchers are anchored", func(t *testing.T) {
+		_, args, err := buildSeriesQuery(start, end, []*labels.Matcher{
+			mustMatcher(t, labels.MatchEqual, "__name__", "up"),
+			mustMatcher(t, labels.MatchRegexp, "instance", "prod.*"),
+			mustMatcher(t, labels.MatchNotRegexp, "env", "dev|test"),
+		})
+		require.NoError(t, err)
+		assert.Equal(t, []any{"up", adjustedStart, end, "instance", "^(?:prod.*)$", "env", "^(?:dev|test)$"}, args)
+	})
+
+	t.Run("regex name matcher uses metric_name column", func(t *testing.T) {
+		query, args, err := buildSeriesQuery(start, end, []*labels.Matcher{
+			mustMatcher(t, labels.MatchRegexp, "__name__", "node_cpu.*|node_memory.*"),
+		})
+		require.NoError(t, err)
+		assert.Contains(t, query, "match(metric_name, ?)")
+		assert.NotContains(t, query, "JSONExtractString")
+		assert.Equal(t, []any{"^(?:node_cpu.*|node_memory.*)$", adjustedStart, end}, args)
+	})
+
+	t.Run("no name matcher omits metric_name condition", func(t *testing.T) {
+		query, _, err := buildSeriesQuery(start, end, []*labels.Matcher{
+			mustMatcher(t, labels.MatchEqual, "job", "api"),
+		})
+		require.NoError(t, err)
+		assert.NotContains(t, query, "metric_name")
+	})
+}
+
+func TestBuildSamplesQuery(t *testing.T) {
+	start := int64(1_700_000_000_000)
+	end := start + time.Hour.Milliseconds()
+	adjustedStart := start - (start % time.Hour.Milliseconds())
+	matchers := []*labels.Matcher{
+		mustMatcher(t, labels.MatchEqual, "__name__", "up"),
+		mustMatcher(t, labels.MatchEqual, "job", "api"),
+	}
+
+	t.Run("raw fetch filters by a shard-local semi-join", func(t *testing.T) {
+		query, args, err := buildSamplesQuery(start, end, []string{"up"}, matchers, nil)
+		require.NoError(t, err)
+		assert.Contains(t, query, "fingerprint IN (SELECT fingerprint FROM signoz_metrics.time_series_v4 WHERE ")
+		assert.NotContains(t, query, "GLOBAL IN")
+		assert.Contains(t, query, "ORDER BY fingerprint, unix_milli")
+		// Args follow placeholder order: samples metric name, the semi-join's
+		// series predicates, then the samples window bounds.
+		assert.Equal(t, []any{"up", "up", adjustedStart, end, "job", "api", start, end}, args)
+	})
+
+	t.Run("last-sample-per-step groups by step bucket anchored at first eval", func(t *testing.T) {
+		lastPerStep := &lastSamplePerStep{firstEvalMs: start + 299_999, stepMs: 60_000}
+		query, _, err := buildSamplesQuery(start, end, []string{"up"}, matchers, lastPerStep)
+		require.NoError(t, err)
+		assert.Contains(t, query, "argMax(value, unix_milli) AS val")
+		assert.Contains(t, query, "argMax(flags, unix_milli) AS fl")
+		assert.Contains(t, query, "GROUP BY fingerprint, if(unix_milli <= 1700000299999, 0, intDiv(unix_milli - 1700000299999 - 1, 60000) + 1)")
+		assert.Contains(t, query, "ORDER BY fingerprint, ts")
+		// Aliases must not shadow the source columns referenced in WHERE.
+		assert.NotContains(t, query, "AS unix_milli")
+		assert.NotContains(t, query, "AS value")
+		assert.NotContains(t, query, "AS flags")
+	})
+
+	t.Run("instant query keeps one bucket", func(t *testing.T) {
+		lastPerStep := &lastSamplePerStep{firstEvalMs: end, stepMs: 0}
+		query, _, err := buildSamplesQuery(start, end, []string{"up"}, matchers, lastPerStep)
+		require.NoError(t, err)
+		assert.Contains(t, query, "GROUP BY fingerprint ORDER BY fingerprint, ts")
+		assert.NotContains(t, query, "intDiv")
+	})
+
+	t.Run("multiple metric names from regex selector", func(t *testing.T) {
+		query, args, err := buildSamplesQuery(start, end, []string{"node_cpu", "node_memory"}, matchers, nil)
+		require.NoError(t, err)
+		assert.Contains(t, query, "metric_name IN (?, ?)")
+		assert.Equal(t, []any{"node_cpu", "node_memory", "up", adjustedStart, end, "job", "api", start, end}, args)
+	})
+}

pkg/prometheus/clickhouseprometheusv2/tables.go

@@ -0,0 +1,70 @@
+package clickhouseprometheusv2
+
+import "time"
+
+// TODO(srikanthccv): consolidate the metrics table names and the
+// window-to-table selection across the query builders (telemetrymetrics)
+// and the prometheus providers instead of each package carrying its own
+// copy.
+
+const (
+	// metricNameLabel is the reserved PromQL label holding the metric name.
+	metricNameLabel string = "__name__"
+
+	databaseName                 string = "signoz_metrics"
+	distributedTimeSeriesV4      string = "distributed_time_series_v4"
+	distributedTimeSeriesV46hrs  string = "distributed_time_series_v4_6hrs"
+	distributedTimeSeriesV41day  string = "distributed_time_series_v4_1day"
+	distributedTimeSeriesV41week string = "distributed_time_series_v4_1week"
+	distributedSamplesV4         string = "distributed_samples_v4"
+
+	localTimeSeriesV4      string = "time_series_v4"
+	localTimeSeriesV46hrs  string = "time_series_v4_6hrs"
+	localTimeSeriesV41day  string = "time_series_v4_1day"
+	localTimeSeriesV41week string = "time_series_v4_1week"
+)
+
+// localTimeSeriesTable maps a distributed time series table to its shard-local
+// table. Samples and time series shard on the same key
+// (cityHash64(env, temporality, metric_name, fingerprint)), so a query whose
+// top-level FROM is the distributed samples table can join or semi-join the
+// local time series table inside each shard: the shard rewrite runs the
+// subquery against the shard's own series rows, which are exactly the series
+// of the shard's samples. No broadcast, no initiator-side join.
+func localTimeSeriesTable(distributed string) string {
+	switch distributed {
+	case distributedTimeSeriesV46hrs:
+		return localTimeSeriesV46hrs
+	case distributedTimeSeriesV41day:
+		return localTimeSeriesV41day
+	case distributedTimeSeriesV41week:
+		return localTimeSeriesV41week
+	default:
+		return localTimeSeriesV4
+	}
+}
+
+var (
+	oneHourInMilliseconds  = time.Hour.Milliseconds()
+	sixHoursInMilliseconds = time.Hour.Milliseconds() * 6
+	oneDayInMilliseconds   = time.Hour.Milliseconds() * 24
+	oneWeekInMilliseconds  = time.Hour.Milliseconds() * 24 * 7
+)
+
+// timeSeriesTableFor returns the adjusted start and the time series table for
+// the window. Time series tables hold one row per (fingerprint, bucket), with
+// bucket granularities of 1h, 6h, 1d and 1w; the start is rounded down to the
+// bucket boundary so a window beginning mid-bucket still matches the bucket's
+// row.
+func timeSeriesTableFor(start, end int64) (int64, string) {
+	switch {
+	case end-start < sixHoursInMilliseconds:
+		return start - (start % oneHourInMilliseconds), distributedTimeSeriesV4
+	case end-start < oneDayInMilliseconds:
+		return start - (start % sixHoursInMilliseconds), distributedTimeSeriesV46hrs
+	case end-start < oneWeekInMilliseconds:
+		return start - (start % oneDayInMilliseconds), distributedTimeSeriesV41day
+	default:
+		return start - (start % oneWeekInMilliseconds), distributedTimeSeriesV41week
+	}
+}

pkg/prometheus/clickhouseprometheusv2/transpiler.go

@@ -0,0 +1,492 @@
+package clickhouseprometheusv2
+
+import (
+	"fmt"
+	"strings"
+
+	"github.com/prometheus/prometheus/model/labels"
+	"github.com/prometheus/prometheus/promql/parser"
+)
+
+// The compiler turns PromQL subtrees into single ClickHouse queries built on
+// the timeSeries*ToGrid aggregate functions (CH >= 25.6), whose semantics
+// were verified against this repo's vendored engine: exact extrapolatedRate
+// behavior including counter resets, the counter zero-point clamp, the
+// 1.1x-average extrapolation threshold, left-open windows, the >= 2 samples
+// rule, stale-marker shadowing, and millisecond grid starts. Sample rows
+// never leave ClickHouse: one row per output series comes back, holding the
+// whole grid as an array.
+//
+// Scope (the allowlist): an optional sum/min/max/avg/count by/without
+// aggregation over a core unit — a rate/increase/delta/irate/idelta range
+// selection, an instant vector selection, or an avg/min/max/sum/count/last
+// _over_time window — plus number-literal arithmetic/comparisons and unary
+// minus on top. Units inside fixed-resolution subqueries evaluate on the
+// subquery's own grid. Everything else either falls back to the engine over
+// this package's querier, or — when a transpilable subtree sits under a
+// non-transpilable node — runs hybrid: the subtree's grids are computed in
+// ClickHouse and substituted into the engine as synthetic series (see
+// compiler_exec.go). See doc.go for the fallback list and the reasons behind
+// each entry.
+
+// rangeFn is a transpilable range-vector function.
+type rangeFn string
+
+const (
+	fnRate     rangeFn = "rate"
+	fnIncrease rangeFn = "increase"
+	fnDelta    rangeFn = "delta"
+	fnIRate    rangeFn = "irate"
+	fnIDelta   rangeFn = "idelta"
+)
+
+var gridFunction = map[rangeFn]string{
+	fnRate:     "timeSeriesRateToGrid",
+	fnIncrease: "timeSeriesRateToGrid", // increase == rate * range seconds, exactly (same factor algebra)
+	fnDelta:    "timeSeriesDeltaToGrid",
+	fnIRate:    "timeSeriesInstantRateToGrid",
+	fnIDelta:   "timeSeriesInstantDeltaToGrid",
+}
+
+// scalarOp is one number-literal arithmetic or comparison applied to a
+// compiled vector, evaluated in Go during assembly with the same float64
+// operations the engine uses.
+type scalarOp struct {
+	op           parser.ItemType
+	scalar       float64
+	scalarOnLeft bool
+	returnBool   bool
+}
+
+// isComparison reports whether the op is a filtering/bool comparison, which
+// preserves the metric name (arithmetic drops it).
+func (o scalarOp) isComparison() bool {
+	return o.op.IsComparisonOperator()
+}
+
+// unitKind is the selector shape at the bottom of a core unit.
+type unitKind int
+
+const (
+	// unitRange: rate/increase/delta/irate/idelta over a matrix selector.
+	unitRange unitKind = iota
+	// unitInstant: a plain vector selector resolved per grid point with
+	// lookback and stale-marker shadowing.
+	unitInstant
+	// unitOverTime: avg/min/max/sum/count/last_over_time over a matrix
+	// selector (aggregation over the window's samples, stale rows excluded).
+	unitOverTime
+)
+
+// coreUnit is one transpilable subtree: selector [-> range function] ->
+// optional aggregation -> scalar op pipeline.
+type coreUnit struct {
+	kind     unitKind
+	matchers []*labels.Matcher
+	offsetMs int64
+	fn       rangeFn // unitRange
+	overFn   string  // unitOverTime: avg|min|max|sum|count|last
+	rangeMs  int64   // unitRange/unitOverTime window
+
+	hasAgg   bool
+	aggOp    parser.ItemType // SUM MIN MAX AVG COUNT
+	by       bool
+	grouping []string
+
+	ops []scalarOp
+}
+
+// keepsName reports whether the unit's output series keep their real
+// __name__: bare/comparison-filtered instant selectors and last_over_time do
+// (it returns the raw sample, name included); range functions, the other
+// *_over_time functions, aggregations, arithmetic and bool comparisons all
+// drop it — a bool comparison returns 0/1, not the sample, so the engine
+// drops the name there too. Units that keep the name cannot be substituted
+// as synthetic series in hybrid plans — the synthetic name would replace
+// the real one — but transpile fine as full plans, where assembly emits the
+// real names.
+func (u *coreUnit) keepsName() bool {
+	nameKeepingSelector := u.kind == unitInstant || (u.kind == unitOverTime && u.overFn == "last")
+	if !nameKeepingSelector || u.hasAgg {
+		return false
+	}
+	for _, op := range u.ops {
+		if !op.isComparison() || op.returnBool {
+			return false
+		}
+	}
+	return true
+}
+
+// gridContext is the evaluation grid a unit computes on. The query grid for
+// top-level units; for units inside subqueries, the subquery's own grid:
+// epoch-aligned multiples of its resolution covering the subquery window,
+// exactly as the engine derives it (engine.go, *parser.SubqueryExpr case).
+type gridContext struct {
+	startMs int64
+	endMs   int64
+	stepMs  int64
+}
+
+// subqueryGrid derives the inner grid for a subquery evaluated on outer:
+// interval S, end = outer end − offset, start = first multiple of S strictly
+// greater than outer start − offset − range.
+func subqueryGrid(outer gridContext, rangeMs, stepMs, offsetMs int64) gridContext {
+	lower := outer.startMs - offsetMs - rangeMs
+	start := stepMs * (lower / stepMs)
+	if start <= lower {
+		start += stepMs
+	}
+	return gridContext{startMs: start, endMs: outer.endMs - offsetMs, stepMs: stepMs}
+}
+
+// transpiledUnit is a coreUnit scheduled for execution, named for hybrid
+// substitution, carrying the grid it evaluates on.
+type transpiledUnit struct {
+	core coreUnit
+	name string // __signoz_transpiled_<n>__
+	grid gridContext
+}
+
+// transpilePlan is the outcome of classifying a query.
+type transpilePlan struct {
+	units []*transpiledUnit
+	grid  gridContext // the query's top-level grid
+	// full is set when the entire query is units[0]; otherwise rewritten
+	// holds the query with each unit replaced by a synthetic selector, to be
+	// evaluated by the engine over a hybrid storage.
+	full      bool
+	rewritten string
+}
+
+const syntheticNamePrefix = "__signoz_transpiled_"
+
+func syntheticName(i int) string {
+	return fmt.Sprintf("%s%d__", syntheticNamePrefix, i)
+}
+
+// classifyCore matches a subtree against the transpilable core shape.
+// stepMs gates second-granularity: the grid functions take whole-second step
+// and window parameters (grid *starts* are millisecond-precise).
+func classifyCore(node parser.Expr, stepMs int64) (*coreUnit, bool) {
+	unit := &coreUnit{}
+
+	expr := node
+	// Peel scalar ops and parens off the top, outermost first; ops apply in
+	// evaluation order, so prepend while peeling.
+	for {
+		switch n := expr.(type) {
+		case *parser.ParenExpr:
+			expr = n.Expr
+			continue
+		case *parser.UnaryExpr:
+			if n.Op != parser.SUB {
+				expr = n.Expr // unary '+' is a no-op
+				continue
+			}
+			// -x == -1 * x for every float64 (incl. NaN and signed zero).
+			unit.ops = append([]scalarOp{{op: parser.MUL, scalar: -1}}, unit.ops...)
+			expr = n.Expr
+			continue
+		case *parser.StepInvariantExpr:
+			// @-pinned expressions evaluate on a different grid.
+			return nil, false
+		case *parser.BinaryExpr:
+			lit, litOnLeft, ok := numberLiteralSide(n)
+			if !ok {
+				return nil, false
+			}
+			if !n.Op.IsOperator() && !n.Op.IsComparisonOperator() {
+				return nil, false
+			}
+			if n.Op == parser.ATAN2 {
+				// atan2 is arithmetic in PromQL but rarely used; keep the
+				// allowlist tight.
+				return nil, false
+			}
+			returnBool := n.ReturnBool
+			unit.ops = append([]scalarOp{{op: n.Op, scalar: lit, scalarOnLeft: litOnLeft, returnBool: returnBool}}, unit.ops...)
+			if litOnLeft {
+				expr = n.RHS
+			} else {
+				expr = n.LHS
+			}
+			continue
+		}
+		break
+	}
+
+	// Optional aggregation.
+	if agg, ok := expr.(*parser.AggregateExpr); ok {
+		switch agg.Op {
+		case parser.SUM, parser.MIN, parser.MAX, parser.AVG, parser.COUNT:
+		default:
+			return nil, false
+		}
+		for _, g := range agg.Grouping {
+			if g == metricNameLabel {
+				// by(__name__)/without(__name__) over synthetic or compiled
+				// output needs name bookkeeping the compiler doesn't do.
+				return nil, false
+			}
+		}
+		unit.hasAgg = true
+		unit.aggOp = agg.Op
+		unit.by = !agg.Without
+		unit.grouping = agg.Grouping
+		expr = agg.Expr
+		for {
+			if p, ok := expr.(*parser.ParenExpr); ok {
+				expr = p.Expr
+				continue
+			}
+			break
+		}
+	}
+
+	// The grid functions take whole-second steps; stepMs == 0 is an instant
+	// query (single-point grid).
+	if stepMs < 0 || stepMs%1000 != 0 {
+		return nil, false
+	}
+
+	// Bare instant selector: resolved per grid point with lookback and
+	// stale-marker shadowing (see compiler_sql.go).
+	if vs, ok := expr.(*parser.VectorSelector); ok {
+		// A duration expression (offset step(), offset range()*2, ...) is
+		// resolved into OriginalOffset only at evaluation time; at
+		// classification time the field still holds its zero value, so
+		// transpiling would silently use the wrong offset.
+		if vs.Timestamp != nil || vs.StartOrEnd != 0 || vs.Anchored || vs.Smoothed || vs.OriginalOffsetExpr != nil {
+			return nil, false
+		}
+		offsetMs := vs.OriginalOffset.Milliseconds()
+		if offsetMs < 0 {
+			return nil, false
+		}
+		unit.kind = unitInstant
+		unit.offsetMs = offsetMs
+		unit.matchers = vs.LabelMatchers
+		return unit, true
+	}
+
+	// Range or *_over_time function over a plain matrix selector.
+	call, ok := expr.(*parser.Call)
+	if !ok {
+		return nil, false
+	}
+	var fn rangeFn
+	var overFn string
+	switch call.Func.Name {
+	case "rate":
+		fn = fnRate
+	case "increase":
+		fn = fnIncrease
+	case "delta":
+		fn = fnDelta
+	case "irate":
+		fn = fnIRate
+	case "idelta":
+		fn = fnIDelta
+	case "avg_over_time", "min_over_time", "max_over_time", "sum_over_time", "count_over_time", "last_over_time":
+		overFn = strings.TrimSuffix(call.Func.Name, "_over_time")
+	default:
+		return nil, false
+	}
+	if len(call.Args) != 1 {
+		return nil, false
+	}
+	ms, ok := call.Args[0].(*parser.MatrixSelector)
+	if !ok {
+		return nil, false
+	}
+	vs, ok := ms.VectorSelector.(*parser.VectorSelector)
+	if !ok {
+		return nil, false
+	}
+	// Duration expressions resolve at evaluation time (see the instant
+	// selector case above); Range/OriginalOffset would be read as zero here.
+	if vs.Timestamp != nil || vs.StartOrEnd != 0 || vs.Anchored || vs.Smoothed || vs.OriginalOffsetExpr != nil || ms.RangeExpr != nil {
+		return nil, false
+	}
+
+	rangeMs := ms.Range.Milliseconds()
+	offsetMs := vs.OriginalOffset.Milliseconds()
+	if rangeMs <= 0 || rangeMs%1000 != 0 || offsetMs < 0 {
+		return nil, false
+	}
+
+	if overFn != "" {
+		unit.kind = unitOverTime
+		unit.overFn = overFn
+	} else {
+		unit.kind = unitRange
+		unit.fn = fn
+	}
+	unit.rangeMs = rangeMs
+	unit.offsetMs = offsetMs
+	unit.matchers = vs.LabelMatchers
+	return unit, true
+}
+
+// numberLiteralSide returns the number literal on one side of a binary
+// expression (peeling parens and unary minus), and which side it is on.
+func numberLiteralSide(b *parser.BinaryExpr) (float64, bool, bool) {
+	if v, ok := literalValue(b.LHS); ok {
+		return v, true, true
+	}
+	if v, ok := literalValue(b.RHS); ok {
+		return v, false, true
+	}
+	return 0, false, false
+}
+
+func literalValue(e parser.Expr) (float64, bool) {
+	neg := false
+	for {
+		switch n := e.(type) {
+		case *parser.ParenExpr:
+			e = n.Expr
+			continue
+		case *parser.StepInvariantExpr:
+			e = n.Expr
+			continue
+		case *parser.UnaryExpr:
+			if n.Op == parser.SUB {
+				neg = !neg
+			}
+			e = n.Expr
+			continue
+		case *parser.NumberLiteral:
+			if neg {
+				return -n.Val, true
+			}
+			return n.Val, true
+		default:
+			return 0, false
+		}
+	}
+}
+
+// classify builds the compile plan for a query: full when the root is a core
+// unit, hybrid when core units sit strictly below the root (including inside
+// fixed-resolution subqueries, computed on the subquery grid), none
+// otherwise.
+func classify(root parser.Expr, grid gridContext) (*transpilePlan, bool) {
+	if unit, ok := classifyCore(root, grid.stepMs); ok {
+		return &transpilePlan{
+			units: []*transpiledUnit{{core: *unit, name: syntheticName(0), grid: grid}},
+			grid:  grid,
+			full:  true,
+		}, true
+	}
+
+	plan := &transpilePlan{grid: grid}
+	rewritten := rewrite(root, grid, plan, false)
+	if len(plan.units) == 0 {
+		return nil, false
+	}
+	plan.rewritten = rewritten.String()
+	return plan, true
+}
+
+// rewrite walks top-down replacing maximal transpilable subtrees with synthetic
+// vector selectors. nameSensitive marks scopes where an ancestor's semantics
+// depend on __name__ (grouping or vector matching on it): synthetic series
+// carry a synthetic __name__, so substitution there would change results.
+// Fixed-resolution subqueries recurse with the subquery's own grid; scopes
+// whose evaluation grid is unknowable (@-pinned, default-resolution
+// subqueries) are not entered.
+func rewrite(node parser.Expr, grid gridContext, plan *transpilePlan, nameSensitive bool) parser.Expr {
+	if node == nil {
+		return nil
+	}
+
+	if !nameSensitive {
+		// Units whose output keeps the real __name__ (bare instant selectors)
+		// cannot be substituted: the synthetic name would replace it in the
+		// engine's output. They still compile as full plans.
+		if unit, ok := classifyCore(node, grid.stepMs); ok && !unit.keepsName() {
+			cu := &transpiledUnit{core: *unit, name: syntheticName(len(plan.units)), grid: grid}
+			plan.units = append(plan.units, cu)
+			return &parser.VectorSelector{
+				Name: cu.name,
+				LabelMatchers: []*labels.Matcher{
+					labels.MustNewMatcher(labels.MatchEqual, metricNameLabel, cu.name),
+				},
+				PosRange: node.PositionRange(),
+			}
+		}
+	}
+
+	switch n := node.(type) {
+	case *parser.ParenExpr:
+		n.Expr = rewrite(n.Expr, grid, plan, nameSensitive)
+	case *parser.UnaryExpr:
+		n.Expr = rewrite(n.Expr, grid, plan, nameSensitive)
+	case *parser.AggregateExpr:
+		sensitive := nameSensitive || groupingUsesName(n.Grouping)
+		n.Expr = rewrite(n.Expr, grid, plan, sensitive)
+		// n.Param is a scalar/string; nothing transpilable inside for our core.
+	case *parser.Call:
+		for i, arg := range n.Args {
+			n.Args[i] = rewrite(arg, grid, plan, nameSensitive)
+		}
+	case *parser.BinaryExpr:
+		sensitive := nameSensitive || vectorMatchingUsesName(n.VectorMatching)
+		n.LHS = rewrite(n.LHS, grid, plan, sensitive)
+		n.RHS = rewrite(n.RHS, grid, plan, sensitive)
+	case *parser.SubqueryExpr:
+		// The alert-smoothing idiom fn_over_time((expr)[R:S]) dominates real
+		// rule fleets; inner units evaluate on the subquery grid, and the
+		// engine does the smoothing over the synthetic series. Requires an
+		// explicit whole-second resolution (S == 0 needs the engine's
+		// default-interval function) and no @ pinning.
+		stepMs := n.Step.Milliseconds()
+		rangeMs := n.Range.Milliseconds()
+		offsetMs := n.OriginalOffset.Milliseconds()
+		if n.Timestamp == nil && n.StartOrEnd == 0 &&
+			n.RangeExpr == nil && n.StepExpr == nil && n.OriginalOffsetExpr == nil &&
+			stepMs > 0 && stepMs%1000 == 0 && rangeMs%1000 == 0 && offsetMs >= 0 {
+			inner := subqueryGrid(grid, rangeMs, stepMs, offsetMs)
+			n.Expr = rewrite(n.Expr, inner, plan, nameSensitive)
+		}
+	case *parser.StepInvariantExpr, *parser.MatrixSelector,
+		*parser.VectorSelector, *parser.NumberLiteral, *parser.StringLiteral:
+		// Leaves, or scopes substitution must not enter.
+	}
+	return node
+}
+
+func groupingUsesName(grouping []string) bool {
+	for _, g := range grouping {
+		if g == metricNameLabel {
+			return true
+		}
+	}
+	return false
+}
+
+func vectorMatchingUsesName(vm *parser.VectorMatching) bool {
+	if vm == nil {
+		return false
+	}
+	for _, l := range append(append([]string{}, vm.MatchingLabels...), vm.Include...) {
+		if l == metricNameLabel {
+			return true
+		}
+	}
+	// Default (all-labels) matching ignores __name__, and by()/ignoring()
+	// lists were checked above.
+	return false
+}
+
+// isSyntheticSelector reports whether matchers target a compiled unit.
+func isSyntheticSelector(matchers []*labels.Matcher) (string, bool) {
+	for _, m := range matchers {
+		if m.Name == metricNameLabel && m.Type == labels.MatchEqual && strings.HasPrefix(m.Value, syntheticNamePrefix) {
+			return m.Value, true
+		}
+	}
+	return "", false
+}

pkg/prometheus/clickhouseprometheusv2/transpiler_corpus_test.go

@@ -0,0 +1,161 @@
+package clickhouseprometheusv2
+
+import (
+	"bufio"
+	"encoding/json"
+	"fmt"
+	"os"
+	"regexp"
+	"sort"
+	"strings"
+	"testing"
+
+	"github.com/prometheus/prometheus/promql/parser"
+	"github.com/stretchr/testify/require"
+)
+
+// TestClassifyCorpus measures real-workload compiler coverage: it classifies
+// every query of a JSON-lines corpus (one JSON-encoded PromQL string per
+// line) with the live classifier and reports full / hybrid / fallback
+// shares. Skipped unless PROMQL_CORPUS points to one or more files
+// (comma-separated). Dashboard template variables are substituted with
+// placeholder values before parsing, mirroring the production render step.
+//
+//	PROMQL_CORPUS=corpus-a.jsonl,corpus-b.jsonl go test -run TestClassifyCorpus -v
+func TestClassifyCorpus(t *testing.T) {
+	corpus := os.Getenv("PROMQL_CORPUS")
+	if corpus == "" {
+		t.Skip("PROMQL_CORPUS not set")
+	}
+
+	varRe := regexp.MustCompile(`\{\{\s*\.?[\w.]+\s*\}\}|\[\[\s*[\w.]+\s*\]\]|\$[\w.]+`)
+	promParser := parser.NewParser(parser.Options{})
+
+	for _, path := range strings.Split(corpus, ",") {
+		f, err := os.Open(path)
+		require.NoError(t, err)
+
+		var full, hybrid, fallbackInstant, fallbackOther, parseErrs int
+		fallbackReasons := map[string]int{}
+
+		scanner := bufio.NewScanner(f)
+		scanner.Buffer(make([]byte, 1024*1024), 1024*1024)
+		for scanner.Scan() {
+			var query string
+			require.NoError(t, json.Unmarshal(scanner.Bytes(), &query))
+			query = varRe.ReplaceAllString(query, "placeholder")
+
+			expr, err := promParser.ParseExpr(query)
+			if err != nil {
+				parseErrs++
+				continue
+			}
+
+			plan, ok := classify(expr, gridContext{startMs: 1_700_000_000_000, endMs: 1_700_007_200_000, stepMs: 60_000})
+			switch {
+			case ok && plan.full:
+				full++
+			case ok:
+				hybrid++
+			default:
+				reason := fallbackShape(expr)
+				fallbackReasons[reason]++
+				if reason == "instant-selector shape (last-sample-per-step engine path)" {
+					fallbackInstant++
+				} else {
+					fallbackOther++
+				}
+			}
+		}
+		require.NoError(t, scanner.Err())
+		_ = f.Close()
+
+		total := full + hybrid + fallbackInstant + fallbackOther
+		if total == 0 {
+			t.Logf("%s: no parseable queries (%d parse errors)", path, parseErrs)
+			continue
+		}
+		t.Logf("%s: %d queries — full=%d (%.0f%%) hybrid=%d (%.0f%%) fallback=%d (%.0f%%; instant-shape=%d) parse_errors=%d",
+			path, total,
+			full, 100*float64(full)/float64(total),
+			hybrid, 100*float64(hybrid)/float64(total),
+			fallbackInstant+fallbackOther, 100*float64(fallbackInstant+fallbackOther)/float64(total),
+			fallbackInstant, parseErrs)
+
+		reasons := make([]string, 0, len(fallbackReasons))
+		for r := range fallbackReasons {
+			reasons = append(reasons, r)
+		}
+		sort.Slice(reasons, func(i, j int) bool { return fallbackReasons[reasons[i]] > fallbackReasons[reasons[j]] })
+		for _, r := range reasons {
+			t.Logf("  fallback %4d  %s", fallbackReasons[r], r)
+		}
+	}
+}
+
+// fallbackShape buckets a non-transpilable query by why it stays on the engine
+// path, to separate "already served well" (instant selectors on the last-sample-per-step
+// path) from genuine compiler gaps.
+func fallbackShape(expr parser.Expr) string {
+	var hasMatrix, hasSubquery, hasAt, hasDurationExpr, overTime bool
+	rangeFns := map[string]bool{"rate": true, "increase": true, "delta": true, "irate": true, "idelta": true}
+	var unsupportedFns []string
+	parser.Inspect(expr, func(node parser.Node, _ []parser.Node) error {
+		switch n := node.(type) {
+		case *parser.MatrixSelector:
+			hasMatrix = true
+			if n.RangeExpr != nil {
+				hasDurationExpr = true
+			}
+		case *parser.SubqueryExpr:
+			hasSubquery = true
+			if n.RangeExpr != nil || n.StepExpr != nil || n.OriginalOffsetExpr != nil {
+				hasDurationExpr = true
+			}
+		case *parser.VectorSelector:
+			if n.Timestamp != nil || n.StartOrEnd != 0 {
+				hasAt = true
+			}
+			if n.OriginalOffsetExpr != nil {
+				hasDurationExpr = true
+			}
+		case *parser.Call:
+			if strings.HasSuffix(n.Func.Name, "_over_time") {
+				overTime = true
+			} else if !rangeFns[n.Func.Name] {
+				unsupportedFns = append(unsupportedFns, n.Func.Name)
+			}
+		}
+		return nil
+	})
+
+	switch {
+	case hasDurationExpr:
+		return "duration expression (resolved only at evaluation time)"
+	case hasSubquery:
+		return "subquery"
+	case hasAt:
+		return "@ modifier"
+	case overTime:
+		return "*_over_time range function"
+	case !hasMatrix:
+		return "instant-selector shape (last-sample-per-step engine path)"
+	case len(unsupportedFns) > 0:
+		return fmt.Sprintf("range shape with unsupported function(s): %s", strings.Join(dedupe(unsupportedFns), ",")) //nolint:makezero
+	default:
+		return "other range shape"
+	}
+}
+
+func dedupe(in []string) []string {
+	seen := map[string]bool{}
+	var out []string
+	for _, s := range in {
+		if !seen[s] {
+			seen[s] = true
+			out = append(out, s)
+		}
+	}
+	sort.Strings(out)
+	return out
+}

pkg/prometheus/clickhouseprometheusv2/transpiler_exec.go

@@ -0,0 +1,550 @@
+package clickhouseprometheusv2
+
+import (
+	"context"
+	"encoding/json"
+	"math"
+	"sort"
+	"time"
+
+	"github.com/SigNoz/signoz/pkg/errors"
+	"github.com/SigNoz/signoz/pkg/prometheus"
+	"github.com/prometheus/prometheus/model/labels"
+	promValue "github.com/prometheus/prometheus/model/value"
+	"github.com/prometheus/prometheus/promql"
+	"github.com/prometheus/prometheus/promql/parser"
+	"github.com/prometheus/prometheus/storage"
+	"golang.org/x/sync/errgroup"
+)
+
+// executor evaluates transpilable PromQL directly in ClickHouse, falling
+// back (ok=false) whenever the query shape or the step doesn't qualify. The
+// timeSeries*ToGrid functions it builds on are assumed available: the
+// supported ClickHouse floor is >= 25.6.
+type executor struct {
+	client *client
+	engine *prometheus.Engine
+	parser prometheus.Parser
+}
+
+// maxWindowBuckets caps range/step for the windowed *_over_time form: every
+// grid slot combines that many bucket partials, and the fleet's windows sit
+// well under it ([1m]..[17m] at 30-60s steps) — anything larger is a
+// long-range query whose step a dashboard scales up anyway, and the engine
+// path serves the rest.
+const maxWindowBuckets = 64
+
+// TryExecuteRange transpiles and runs the query in ClickHouse when its shape
+// is in the allowlist. ok=false means "not transpilable" and carries no
+// error; the caller runs the engine path.
+func (e *executor) TryExecuteRange(ctx context.Context, qs string, start, end time.Time, step time.Duration) (promql.Matrix, bool, error) {
+	expr, err := e.parser.ParseExpr(qs)
+	if err != nil {
+		// Let the engine path produce the (enhanced) parse error.
+		return nil, false, nil
+	}
+
+	plan, ok := classify(expr, queryGrid(start, end, step))
+	if !ok {
+		return nil, false, nil
+	}
+
+	// timeSeriesLastToGrid widens its window to max(window, step) — probed: a
+	// sample aged (window, step] still fills the slot — while the rate/delta
+	// family enforces the window strictly. The Last-style kinds used to fall
+	// back when window < step because of that widening; the window-sliver
+	// filter (see samplesConditions) makes the widening harmless there:
+	// samples exist only inside (t_k - window, t_k] slivers, so the widened
+	// window intersected with the data IS the lookback window — and if a
+	// future ClickHouse stops widening, the unwidened window is the sliver
+	// too. Correct either way. A non-positive window still falls back: the
+	// sliver argument needs a real window to filter to.
+	//
+	// The windowed *_over_time form gates only the range >= step regime: it
+	// decomposes the window into whole step buckets (see windowedInner),
+	// which is exact only when the range is a multiple of the step, and its
+	// per-slot slide costs range/step bucket combines — bounded by
+	// maxWindowBuckets so a long-range short-step query cannot turn the
+	// slide into the bottleneck. range < step needs neither gate: the
+	// windows are disjoint slivers, aggregated one slot each with no slide.
+	// Every miss falls back to the engine path, which is exact.
+	for _, unit := range plan.units {
+		stepMs := unit.grid.stepMs
+		if stepMs == 0 {
+			stepMs = 1000
+		}
+		switch {
+		case unit.core.kind == unitInstant || (unit.core.kind == unitOverTime && unit.core.overFn == "last"):
+			windowMs := unit.core.rangeMs
+			if unit.core.kind == unitInstant {
+				windowMs = e.client.lookbackMs
+			}
+			if windowMs <= 0 {
+				return nil, false, nil
+			}
+		case unit.core.kind == unitOverTime:
+			if unit.core.rangeMs < unit.grid.stepMs {
+				// Disjoint slivers: no divisibility or width requirement.
+				continue
+			}
+			if unit.core.rangeMs%stepMs != 0 || unit.core.rangeMs/stepMs > maxWindowBuckets {
+				return nil, false, nil
+			}
+		}
+	}
+
+	// Evaluate every unit concurrently on its own grid (the query grid, or a
+	// subquery grid); each is one series lookup plus one grid query.
+	results := make([][]transpiledSeries, len(plan.units))
+	eg, egCtx := errgroup.WithContext(ctx)
+	for i, unit := range plan.units {
+		eg.Go(func() error {
+			res, err := e.executeUnit(egCtx, &unit.core, unit.grid)
+			if err != nil {
+				return err
+			}
+			results[i] = res
+			return nil
+		})
+	}
+	if err := eg.Wait(); err != nil {
+		return nil, true, err
+	}
+
+	if plan.full {
+		g := plan.units[0].grid
+		return toMatrix(results[0], g.startMs, g.stepMs), true, nil
+	}
+
+	matrix, err := e.executeHybrid(ctx, plan, results)
+	if err != nil {
+		return nil, true, err
+	}
+	return matrix, true, nil
+}
+
+// queryGrid derives the top-level evaluation grid; step 0 is an instant
+// query: a single evaluation at end, whatever start was.
+func queryGrid(start, end time.Time, step time.Duration) gridContext {
+	startMs, endMs, stepMs := start.UnixMilli(), end.UnixMilli(), step.Milliseconds()
+	if stepMs == 0 {
+		startMs = endMs
+	}
+	return gridContext{startMs: startMs, endMs: endMs, stepMs: stepMs}
+}
+
+// transpiledSeries is one output series of a unit: projected labels and one
+// value pointer per grid point (nil = absent).
+type transpiledSeries struct {
+	lset   labels.Labels
+	values []*float64
+}
+
+// executeUnit runs one core unit on its grid: series lookup (budgets,
+// fingerprints, metric names), then the single grid query, then the
+// scalar-op pipeline.
+func (e *executor) executeUnit(ctx context.Context, unit *coreUnit, grid gridContext) ([]transpiledSeries, error) {
+	startMs, endMs, stepMs := grid.startMs, grid.endMs, grid.stepMs
+	windowMs := unit.rangeMs
+	if unit.kind == unitInstant {
+		windowMs = e.client.lookbackMs
+	}
+	dataStart := startMs - unit.offsetMs - windowMs
+	dataEnd := endMs - unit.offsetMs
+
+	seriesQuery, seriesArgs, err := buildSeriesQuery(dataStart, dataEnd, unit.matchers)
+	if err != nil {
+		return nil, err
+	}
+	lookup, err := e.client.selectSeries(ctx, seriesQuery, seriesArgs)
+	if err != nil {
+		return nil, err
+	}
+	if len(lookup.fingerprints) == 0 {
+		return nil, nil
+	}
+
+	query, args, err := buildUnitSQL(unit, lookup.metricNames, dataStart, dataEnd, startMs, endMs, stepMs, e.client.lookbackMs)
+	if err != nil {
+		return nil, err
+	}
+
+	rows, err := e.client.telemetryStore.ClickhouseDB().Query(e.client.withContext(ctx, "transpiledUnit"), query, args...)
+	if err != nil {
+		return nil, err
+	}
+	defer rows.Close()
+
+	// Name-dropping units keep __name__ in the SQL group key so distinct
+	// metrics never merge server-side; the name comes off here. Two metrics
+	// can then share a labelset — the engine merges their samples into one
+	// series when they never overlap in time (a selector spanning metrics
+	// whose series alternate across lookback windows) and raises the
+	// duplicate-labelset error only when two samples land on the same
+	// evaluation timestamp. mergeSameLabelsetSeries reproduces exactly that.
+	stripName := !unit.hasAgg && !unit.keepsName()
+
+	// by (...) units return one plain column per grouped label; everything
+	// else returns the single canonical JSON key (see groupKeyColumns).
+	keyNames := groupKeyColumns(unit)
+	keyVals := make([]string, max(len(keyNames), 1))
+	targets := make([]any, 0, len(keyVals)+1)
+	for i := range keyVals {
+		targets = append(targets, &keyVals[i])
+	}
+	var gridValues []*float64
+	targets = append(targets, &gridValues)
+
+	var out []transpiledSeries
+	for rows.Next() {
+		if err := rows.Scan(targets...); err != nil {
+			return nil, err
+		}
+		var lset labels.Labels
+		if keyNames != nil {
+			builder := labels.NewScratchBuilder(len(keyNames))
+			for i, name := range keyNames {
+				// An empty extracted value is the label being absent.
+				if keyVals[i] != "" {
+					builder.Add(name, keyVals[i])
+				}
+			}
+			builder.Sort()
+			lset = builder.Labels()
+		} else {
+			lset, err = labelsFromGroupKey(keyVals[0])
+			if err != nil {
+				return nil, err
+			}
+		}
+		if stripName {
+			lset = labels.NewBuilder(lset).Del(metricNameLabel).Labels()
+		}
+		values := make([]*float64, len(gridValues))
+		copy(values, gridValues)
+		applyScalarOps(unit.ops, values)
+		out = append(out, transpiledSeries{lset: lset, values: values})
+	}
+	if err := rows.Err(); err != nil {
+		return nil, err
+	}
+
+	if stripName {
+		if out, err = mergeSameLabelsetSeries(out); err != nil {
+			return nil, err
+		}
+	}
+	sort.Slice(out, func(i, j int) bool { return labels.Compare(out[i].lset, out[j].lset) < 0 })
+	return out, nil
+}
+
+// mergeSameLabelsetSeries combines series left with identical labelsets by a
+// name strip, slot by slot: the engine assembles its result matrix by
+// labelset, so post-strip twins whose points interleave in time are one
+// series to it, and two values on the same evaluation timestamp are its
+// duplicate-labelset error — v1 would have errored there too, so silently
+// picking one value would be a divergence.
+func mergeSameLabelsetSeries(in []transpiledSeries) ([]transpiledSeries, error) {
+	index := make(map[uint64]int, len(in))
+	out := in[:0]
+	for _, s := range in {
+		hash := s.lset.Hash()
+		idx, ok := index[hash]
+		if ok && labels.Equal(out[idx].lset, s.lset) {
+			dst := out[idx].values
+			for k, v := range s.values {
+				if v == nil {
+					continue
+				}
+				if dst[k] != nil {
+					return nil, errors.NewInvalidInputf(errors.CodeInvalidInput, "vector cannot contain metrics with the same labelset")
+				}
+				dst[k] = v
+			}
+			continue
+		}
+		index[hash] = len(out)
+		out = append(out, s)
+	}
+	return out, nil
+}
+
+// labelsFromGroupKey parses the toJSONString'd sorted [key, value] pairs.
+func labelsFromGroupKey(gkey string) (labels.Labels, error) {
+	var pairs [][]string
+	if err := json.Unmarshal([]byte(gkey), &pairs); err != nil {
+		return labels.EmptyLabels(), errors.WrapInternalf(err, errors.CodeInternal, "malformed compiled group key %q", gkey)
+	}
+	builder := labels.NewScratchBuilder(len(pairs))
+	for _, p := range pairs {
+		if len(p) != 2 {
+			return labels.EmptyLabels(), errors.NewInternalf(errors.CodeInternal, "malformed compiled group key pair %q", gkey)
+		}
+		builder.Add(p[0], p[1])
+	}
+	builder.Sort()
+	return builder.Labels(), nil
+}
+
+// applyScalarOps applies the number-literal op pipeline in place, with the
+// same float64 arithmetic and comparison-filter semantics as the engine.
+func applyScalarOps(ops []scalarOp, values []*float64) {
+	for _, op := range ops {
+		for i, v := range values {
+			if v == nil {
+				continue
+			}
+			lhs, rhs := *v, op.scalar
+			if op.scalarOnLeft {
+				lhs, rhs = op.scalar, *v
+			}
+			switch op.op {
+			case parser.ADD:
+				res := lhs + rhs
+				values[i] = &res
+			case parser.SUB:
+				res := lhs - rhs
+				values[i] = &res
+			case parser.MUL:
+				res := lhs * rhs
+				values[i] = &res
+			case parser.DIV:
+				res := lhs / rhs
+				values[i] = &res
+			case parser.MOD:
+				res := math.Mod(lhs, rhs)
+				values[i] = &res
+			case parser.POW:
+				res := math.Pow(lhs, rhs)
+				values[i] = &res
+			default:
+				keep := compare(op.op, lhs, rhs)
+				switch {
+				case op.returnBool:
+					res := 0.0
+					if keep {
+						res = 1.0
+					}
+					values[i] = &res
+				case keep:
+					// Filter comparisons keep the vector-side value.
+					vec := *v
+					values[i] = &vec
+				default:
+					values[i] = nil
+				}
+			}
+		}
+	}
+}
+
+func compare(op parser.ItemType, lhs, rhs float64) bool {
+	switch op {
+	case parser.EQLC:
+		return lhs == rhs
+	case parser.NEQ:
+		return lhs != rhs
+	case parser.GTR:
+		return lhs > rhs
+	case parser.LSS:
+		return lhs < rhs
+	case parser.GTE:
+		return lhs >= rhs
+	case parser.LTE:
+		return lhs <= rhs
+	}
+	return false
+}
+
+// toMatrix converts a unit result to a promql matrix on the query grid.
+func toMatrix(series []transpiledSeries, startMs, stepMs int64) promql.Matrix {
+	matrix := make(promql.Matrix, 0, len(series))
+	for _, s := range series {
+		var floats []promql.FPoint
+		for i, v := range s.values {
+			if v == nil {
+				continue
+			}
+			floats = append(floats, promql.FPoint{T: startMs + int64(i)*stepMs, F: *v})
+		}
+		if len(floats) == 0 {
+			continue
+		}
+		matrix = append(matrix, promql.Series{Metric: s.lset, Floats: floats})
+	}
+	return matrix
+}
+
+// executeHybrid substitutes each unit's grids into the engine as synthetic
+// series and evaluates the rewritten query over a storage that serves
+// synthetic selectors from memory and everything else from the live querier.
+// Absent grid points become stale markers so the engine's lookback cannot
+// resurrect the previous grid point. Each unit's synthetic samples sit on its
+// own grid (query grid, or subquery grid for units inside subqueries).
+func (e *executor) executeHybrid(ctx context.Context, plan *transpilePlan, results [][]transpiledSeries) (promql.Matrix, error) {
+	synthetic := make(map[string][]*series, len(plan.units))
+	staleMarker := math.Float64frombits(promValue.StaleNaN)
+
+	queryGrid := plan.grid
+
+	for i, unit := range plan.units {
+		g := unit.grid
+		gridLen := 1
+		if g.stepMs > 0 {
+			gridLen = int((g.endMs-g.startMs)/g.stepMs) + 1
+		}
+		list := make([]*series, 0, len(results[i]))
+		for _, cs := range results[i] {
+			builder := labels.NewBuilder(cs.lset)
+			builder.Set(metricNameLabel, unit.name)
+			s := &series{lset: builder.Labels()}
+			s.ts = make([]int64, 0, gridLen)
+			s.vs = make([]float64, 0, gridLen)
+			for idx := 0; idx < gridLen; idx++ {
+				t := g.startMs + int64(idx)*g.stepMs
+				var v float64
+				if idx < len(cs.values) && cs.values[idx] != nil {
+					v = *cs.values[idx]
+				} else {
+					v = staleMarker
+				}
+				s.ts = append(s.ts, t)
+				s.vs = append(s.vs, v)
+			}
+			list = append(list, s)
+		}
+		synthetic[unit.name] = list
+	}
+
+	hybrid := &hybridQueryable{client: e.client, synthetic: synthetic}
+
+	var qry promql.Query
+	var err error
+	if queryGrid.stepMs == 0 {
+		qry, err = e.engine.NewInstantQuery(ctx, hybrid, nil, plan.rewritten, time.UnixMilli(queryGrid.endMs))
+	} else {
+		qry, err = e.engine.NewRangeQuery(ctx, hybrid, nil, plan.rewritten, time.UnixMilli(queryGrid.startMs), time.UnixMilli(queryGrid.endMs), time.Duration(queryGrid.stepMs)*time.Millisecond)
+	}
+	if err != nil {
+		return nil, err
+	}
+	defer qry.Close()
+
+	res := qry.Exec(ctx)
+	if res.Err != nil {
+		return nil, res.Err
+	}
+
+	matrix, err := resultToMatrix(res)
+	if err != nil {
+		return nil, err
+	}
+
+	// Deep-copy before Close returns the result's slices to the engine pool,
+	// and drop the synthetic __name__ that filter comparisons preserve.
+	out := make(promql.Matrix, 0, len(matrix))
+	for _, s := range matrix {
+		lset := s.Metric
+		if name := lset.Get(metricNameLabel); len(name) >= len(syntheticNamePrefix) && name[:len(syntheticNamePrefix)] == syntheticNamePrefix {
+			builder := labels.NewBuilder(lset)
+			builder.Del(metricNameLabel)
+			lset = builder.Labels()
+		}
+		floats := make([]promql.FPoint, len(s.Floats))
+		copy(floats, s.Floats)
+		out = append(out, promql.Series{Metric: lset.Copy(), Floats: floats})
+	}
+	// The strip can leave twins: two units' outputs distinguishable only by
+	// their synthetic names (e.g. -metric_a or -metric_b, both {} to the
+	// engine's real evaluation once names dropped). The engine assembles its
+	// matrix by labelset, merging such temporally-disjoint elements into one
+	// series; reproduce that, with its duplicate error on same-timestamp
+	// overlap.
+	out, err = mergeMatrixByLabelset(out)
+	if err != nil {
+		return nil, err
+	}
+	sort.Slice(out, func(i, j int) bool { return labels.Compare(out[i].Metric, out[j].Metric) < 0 })
+	return out, nil
+}
+
+// mergeMatrixByLabelset merges series sharing a labelset by interleaving
+// their points in timestamp order; a timestamp present in both is the
+// engine's duplicate-labelset error.
+func mergeMatrixByLabelset(matrix promql.Matrix) (promql.Matrix, error) {
+	index := make(map[uint64]int, len(matrix))
+	out := matrix[:0]
+	for _, s := range matrix {
+		hash := s.Metric.Hash()
+		idx, ok := index[hash]
+		if ok && labels.Equal(out[idx].Metric, s.Metric) {
+			merged := make([]promql.FPoint, 0, len(out[idx].Floats)+len(s.Floats))
+			a, b := out[idx].Floats, s.Floats
+			for len(a) > 0 && len(b) > 0 {
+				switch {
+				case a[0].T < b[0].T:
+					merged, a = append(merged, a[0]), a[1:]
+				case b[0].T < a[0].T:
+					merged, b = append(merged, b[0]), b[1:]
+				default:
+					return nil, errors.NewInvalidInputf(errors.CodeInvalidInput, "vector cannot contain metrics with the same labelset")
+				}
+			}
+			out[idx].Floats = append(append(merged, a...), b...)
+			continue
+		}
+		index[hash] = len(out)
+		out = append(out, s)
+	}
+	return out, nil
+}
+
+func resultToMatrix(res *promql.Result) (promql.Matrix, error) {
+	switch v := res.Value.(type) {
+	case promql.Matrix:
+		return v, nil
+	case promql.Vector:
+		matrix := make(promql.Matrix, 0, len(v))
+		for _, s := range v {
+			matrix = append(matrix, promql.Series{Metric: s.Metric, Floats: []promql.FPoint{{T: s.T, F: s.F}}})
+		}
+		return matrix, nil
+	case promql.Scalar:
+		return promql.Matrix{{Metric: labels.EmptyLabels(), Floats: []promql.FPoint{{T: v.T, F: v.V}}}}, nil
+	default:
+		return nil, errors.NewInternalf(errors.CodeInternal, "unexpected hybrid result type %T", res.Value)
+	}
+}
+
+// hybridQueryable serves synthetic (compiled) selectors from memory and
+// everything else from the live storage.
+type hybridQueryable struct {
+	client    *client
+	synthetic map[string][]*series
+}
+
+func (h *hybridQueryable) Querier(mint, maxt int64) (storage.Querier, error) {
+	return &hybridQuerier{
+		querier:   querier{mint: mint, maxt: maxt, client: h.client},
+		synthetic: h.synthetic,
+	}, nil
+}
+
+type hybridQuerier struct {
+	querier
+	synthetic map[string][]*series
+}
+
+func (h *hybridQuerier) Select(ctx context.Context, sortSeries bool, hints *storage.SelectHints, matchers ...*labels.Matcher) storage.SeriesSet {
+	if name, ok := isSyntheticSelector(matchers); ok {
+		list := h.synthetic[name]
+		if sortSeries {
+			sorted := make([]*series, len(list))
+			copy(sorted, list)
+			sort.Slice(sorted, func(i, j int) bool { return labels.Compare(sorted[i].lset, sorted[j].lset) < 0 })
+			list = sorted
+		}
+		return newSeriesSet(list)
+	}
+	return h.querier.Select(ctx, sortSeries, hints, matchers...)
+}

pkg/prometheus/clickhouseprometheusv2/transpiler_sql.go

@@ -0,0 +1,412 @@
+package clickhouseprometheusv2
+
+import (
+	"fmt"
+	"strings"
+
+	"github.com/huandu/go-sqlbuilder"
+)
+
+// experimental gate for the timeSeries*ToGrid aggregate functions; attached
+// as a SETTINGS clause so telemetrystore hooks cannot clobber it.
+const gridFunctionsSetting = "SETTINGS allow_experimental_ts_to_grid_aggregate_function = 1"
+
+var aggForEach = map[string]string{
+	"sum":   "sumForEach",
+	"min":   "minForEach",
+	"max":   "maxForEach",
+	"avg":   "avgForEach",
+	"count": "countForEach",
+}
+
+// buildUnitSQL renders the single ClickHouse query evaluating a core unit
+// over the [startMs, endMs] / stepMs evaluation grid: per-series grids via a
+// timeSeries*ToGrid aggregate (or a windowed aggregation for *_over_time),
+// then spatial aggregation with -ForEach combinators grouped by a canonical
+// JSON key of the projected label pairs.
+//
+// The heavy level is shaped to run on the shards: the top-level FROM is the
+// distributed samples table and the group-key join partner is a subquery on
+// the shard-local time series table, so the shard rewrite executes the join
+// and the per-(fingerprint, group key) aggregation next to the data —
+// complete by fingerprint co-locality (see localTimeSeriesTable) — and the
+// initiator only merges the per-series states and applies the spatial
+// -ForEach step. Same layout as the telemetrymetrics statement builder.
+// The windowed *_over_time form shares the frame but aggregates per
+// (series, group key, step bucket) instead of straight to grids
+// (see windowedInner).
+//
+// The selector's data window is offset-shifted; the resulting grid indices
+// map 1:1 onto the query grid (output ts = startMs + i*stepMs). Grid
+// parameters are rendered as literals — they are aggregate-function
+// parameters, not bindable values.
+//
+// Statements nest builder-rendered SQL as text, so the returned args must be
+// ordered by where each fragment lands in the final statement: ClickHouse
+// binds ? placeholders by position. A JOIN renders before WHERE, so a joined
+// subquery's args precede the outer query's own condition args.
+//
+// Row shape: the group-key columns (see groupKeyColumns) followed by
+// grid Array(Nullable(Float64)); NULL grid points are absent points (the
+// engine's "no value here"), which the -ForEach combinators preserve: an
+// index where every series is NULL aggregates to NULL, and countForEach's 0
+// is mapped back to NULL.
+func buildUnitSQL(unit *coreUnit, metricNames []string, dataStart, dataEnd int64, startMs, endMs, stepMs, lookbackMs int64) (string, []any, error) {
+	selStart := startMs - unit.offsetMs
+	selEnd := endMs - unit.offsetMs
+	stepSec := stepMs / 1000
+	if stepSec == 0 {
+		// Instant query: start == end, so the grid has one point for any
+		// positive step.
+		stepSec = 1
+	}
+	windowMs := unit.rangeMs
+	if unit.kind == unitInstant {
+		windowMs = lookbackMs
+	}
+	windowSec := windowMs / 1000
+
+	adjustedTsStart, tsTable := timeSeriesTableFor(dataStart, dataEnd)
+	keyNames := groupKeyColumns(unit)
+
+	// seriesSub computes fingerprint -> group key columns. It reads the
+	// local series table when it rides inside the shard-rewritten samples
+	// query, and the distributed one when it joins at the initiator
+	// (windowed form).
+	seriesSub := func(table string) (string, []any, error) {
+		sub := sqlbuilder.NewSelectBuilder()
+		selects := []string{"fingerprint"}
+		if keyNames == nil {
+			selects = append(selects, groupKeyExpr(unit)+" AS gkey")
+		} else {
+			// by (...) grouping extracts exactly the listed labels as plain
+			// columns: no reason to build, sort and stringify every label
+			// pair per row when the projection is a known short list and
+			// the label names live in Go anyway.
+			for i, name := range keyNames {
+				selects = append(selects, fmt.Sprintf("JSONExtractString(labels, %s) AS g%d", sub.Var(name), i))
+			}
+		}
+		sub.Select(selects...)
+		sub.From(fmt.Sprintf("%s.%s", databaseName, table))
+		if err := applySeriesConditions(sub, adjustedTsStart, dataEnd, unit.matchers); err != nil {
+			return "", nil, err
+		}
+		sub.GroupBy(append([]string{"fingerprint"}, keyColumnAliases(keyNames)...)...)
+		q, args := sub.BuildWithFlavor(sqlbuilder.ClickHouse)
+		return q, args, nil
+	}
+
+	// samplesConditions adds the samples-side WHERE. The group-key join
+	// restricts to the matched series; no fingerprint condition is added
+	// here.
+	samplesConditions := func(sb *sqlbuilder.SelectBuilder, excludeStale bool) {
+		switch len(metricNames) {
+		case 0:
+			// No name constraint derivable; correct but unable to use the
+			// metric_name primary-key prefix.
+		case 1:
+			sb.Where(sb.EQ("metric_name", metricNames[0]))
+		default:
+			sb.Where(sb.In("metric_name", sqlbuilder.List(metricNames)))
+		}
+		// temporality precedes metric_name in the samples primary key; the
+		// fingerprints already come from these temporalities, so this only
+		// helps granule pruning.
+		sb.Where("temporality IN ['Cumulative', 'Unspecified']")
+		// When the window is narrower than the step, the grid windows
+		// (t_k − window, t_k] cover only window/step of the selector's
+		// timeline; a sample in a gap belongs to no window and cannot move
+		// any grid point, but the grid aggregate buffers every row it is
+		// fed. Keeping only in-window rows cut a 36k-series 1w rate from
+		// 74s/28GiB to 16s/4.3GiB on fleet data — the read stays the same,
+		// the aggregate input shrinks by the coverage ratio. The lattice
+		// anchors at selStart (end may sit off-lattice on unaligned grids),
+		// positiveModulo because samples above selStart make the dividend
+		// negative, and the upper bound tightens to the last grid point —
+		// rows past it are equally windowless. window >= step tiles the
+		// timeline and keeps today's plain bounds.
+		sliver := stepMs > 0 && windowMs > 0 && windowMs < stepMs
+		upper := selEnd
+		if sliver {
+			upper = selStart + (selEnd-selStart)/stepMs*stepMs
+		}
+		// Left-open window: a sample exactly at the window's lower boundary
+		// is never used (range selectors and lookback are both left-open).
+		sb.Where(sb.GT("unix_milli", selStart-windowMs), sb.LTE("unix_milli", upper))
+		if sliver {
+			sb.Where(fmt.Sprintf("positiveModulo(%s - unix_milli, %s) < %s",
+				sb.Var(selStart), sb.Var(stepMs), sb.Var(windowMs)))
+		}
+		if excludeStale {
+			// PromQL excludes stale markers from range vectors. Instant
+			// selectors need the stale rows for shadowing instead.
+			sb.Where("bitAnd(flags, 1) = 0")
+		}
+	}
+
+	keyCols := keyColumnAliases(keyNames)
+
+	// joinedInner builds the shard-side SELECT for the single-pass kinds:
+	// grid expression per (fingerprint, group key), group-key join against
+	// the local series table.
+	joinedInner := func(gridExpr string, excludeStale bool) (string, []any, error) {
+		seriesSQL, seriesArgs, err := seriesSub(localTimeSeriesTable(tsTable))
+		if err != nil {
+			return "", nil, err
+		}
+		sb := sqlbuilder.NewSelectBuilder()
+		selects := make([]string, 0, len(keyCols)+1)
+		// A fingerprint is the hash of one labelset, so every group-key
+		// column is functionally dependent on it: any() is exact, and
+		// grouping by the fingerprint alone spares hashing the joined
+		// string per sample row — measured -10-13% on a 1.9B-row rate.
+		for _, col := range keyCols {
+			selects = append(selects, fmt.Sprintf("any(series.%s) AS %s", col, col))
+		}
+		sb.Select(append(selects, gridExpr+" AS grid")...)
+		sb.From(fmt.Sprintf("%s.%s AS points", databaseName, distributedSamplesV4))
+		sb.JoinWithOption(sqlbuilder.InnerJoin, fmt.Sprintf("(%s) AS series", seriesSQL), "points.fingerprint = series.fingerprint")
+		samplesConditions(sb, excludeStale)
+		sb.GroupBy("points.fingerprint")
+		q, args := sb.BuildWithFlavor(sqlbuilder.ClickHouse)
+		// The join text renders before WHERE: its args come first.
+		return q, append(seriesArgs, args...), nil
+	}
+
+	var inner string
+	var innerArgs []any
+	var err error
+	switch unit.kind {
+	case unitInstant:
+		// Instant selection with stale shadowing: the grid value is the last
+		// non-stale sample in (t-lookback, t], absent when the overall last
+		// sample in that window is a stale marker (verified semantics: the
+		// -If combinator applies to the grid aggregates, and NULL comparisons
+		// make a stale-latest point absent).
+		gridParams := fmt.Sprintf("(fromUnixTimestamp64Milli(%d), fromUnixTimestamp64Milli(%d), %d, %d)", selStart, selEnd, stepSec, windowSec)
+		gridExpr := fmt.Sprintf(
+			"arrayMap((tall, tok, vok) -> if(tall IS NULL OR tok IS NULL OR tall != tok, NULL, vok), timeSeriesLastToGrid%s(fromUnixTimestamp64Milli(unix_milli), toFloat64(unix_milli)), timeSeriesLastToGridIf%s(fromUnixTimestamp64Milli(unix_milli), toFloat64(unix_milli), bitAnd(flags, 1) = 0), timeSeriesLastToGridIf%s(fromUnixTimestamp64Milli(unix_milli), value, bitAnd(flags, 1) = 0))",
+			gridParams, gridParams, gridParams,
+		)
+		inner, innerArgs, err = joinedInner(gridExpr, false)
+	case unitOverTime:
+		if unit.overFn == "last" {
+			// last_over_time == last non-stale sample in the window: the
+			// stale rows are already excluded in WHERE.
+			gridExpr := fmt.Sprintf(
+				"timeSeriesLastToGrid(fromUnixTimestamp64Milli(%d), fromUnixTimestamp64Milli(%d), %d, %d)(fromUnixTimestamp64Milli(unix_milli), value)",
+				selStart, selEnd, stepSec, windowSec,
+			)
+			inner, innerArgs, err = joinedInner(gridExpr, true)
+			break
+		}
+		inner, innerArgs, err = windowedInner(unit, samplesConditions, seriesSub, keyCols, localTimeSeriesTable(tsTable), selStart, selEnd, stepMs, windowMs)
+	default: // unitRange
+		gridExpr := fmt.Sprintf(
+			"%s(fromUnixTimestamp64Milli(%d), fromUnixTimestamp64Milli(%d), %d, %d)(fromUnixTimestamp64Milli(unix_milli), value)",
+			gridFunction[unit.fn], selStart, selEnd, stepSec, windowSec,
+		)
+		if unit.fn == fnIncrease {
+			// increase == rate * range-seconds, exactly: extrapolatedRate
+			// divides by the range only when isRate.
+			gridExpr = fmt.Sprintf("arrayMap(x -> x * %d, %s)", windowSec, gridExpr)
+		}
+		inner, innerArgs, err = joinedInner(gridExpr, true)
+	}
+	if err != nil {
+		return "", nil, err
+	}
+
+	spatial := "maxForEach(grid)"
+	switch {
+	case !unit.hasAgg:
+		// Per-series output: one row per (labels-minus-__name__) group.
+		// Distinct fingerprints can collapse onto the same projected label
+		// set only via a regex __name__ selector over metrics with identical
+		// other labels; maxForEach is a deterministic NULL-skipping merge and
+		// the identity for the overwhelmingly common one-fingerprint group.
+	case unit.aggOp.String() == "count":
+		// count over an all-absent index is an absent point, not 0.
+		spatial = "arrayMap(c -> if(c = 0, NULL, toFloat64(c)), countForEach(grid))"
+	default:
+		spatial = fmt.Sprintf("%s(grid)", aggForEach[unit.aggOp.String()])
+	}
+
+	keyList := strings.Join(keyCols, ", ")
+	query := fmt.Sprintf("SELECT %s, %s AS grid FROM (%s) GROUP BY %s %s", keyList, spatial, inner, keyList, gridFunctionsSetting)
+	return query, innerArgs, nil
+}
+
+// groupKeyColumns returns the label names to extract as plain group-key
+// columns, or nil when the unit needs the canonical JSON key instead. Only
+// by (...) grouping qualifies: its projection is a known short list, so
+// extracting each label directly beats building, sorting and stringifying
+// every label pair per row. without and no-aggregation project a label SET
+// that varies per series — there the sorted-JSON key is load-bearing: the
+// sort is what makes two fingerprints with different stored JSON key order
+// land in one group, and the string carries the labels back out.
+func groupKeyColumns(unit *coreUnit) []string {
+	if unit.hasAgg && unit.by && len(unit.grouping) > 0 {
+		return unit.grouping
+	}
+	return nil
+}
+
+// keyColumnAliases names the group-key columns in every SELECT level: g0..gN
+// for direct extraction, the single canonical gkey otherwise.
+func keyColumnAliases(keyNames []string) []string {
+	if keyNames == nil {
+		return []string{"gkey"}
+	}
+	cols := make([]string, len(keyNames))
+	for i := range keyNames {
+		cols[i] = fmt.Sprintf("g%d", i)
+	}
+	return cols
+}
+
+// windowedInner builds the avg/min/max/sum/count _over_time form without
+// fanning samples out. It runs only when the range is a whole multiple of
+// the step (see the transpile gate), because then the window
+// (t_k - range, t_k] is exactly the union of W = range/step step buckets —
+// both are left-open on the same boundaries — so bucket membership fully
+// determines window membership. Fanning each sample into all W windows it
+// covers (ARRAY JOIN) multiplies rows by W, which at long ranges over short
+// steps is a row explosion measured in billions.
+//
+// The bucketing itself is the -Resample combinator: one group per (series,
+// group key) whose state is a fixed array of per-bucket aggregates, updated
+// in place per sample. Grouping by (series, bucket) instead — measured on a
+// 100k-series x 371-bucket workload — creates a 37M-entry hash aggregation
+// whose per-thread partial tables scale memory WITH max_threads (12 -> 48
+// GiB from 2 to 8 threads, dead at 16) and ships one row per group to the
+// initiator; the Resample form carries the same numbers in 100k compact
+// array states, like every other unit kind.
+//
+// The wrapper level slides the window: slot k combines buckets k..k+W-1 by
+// direct aggregation over at most W partials — no prefix-sum tricks, so no
+// large-minus-large cancellation against the engine's directly-summed
+// windows. A slot with zero window count is absent, which also keeps
+// min/max honest: their slices filter on the bucket counts, so an empty
+// bucket's zero-fill can never be mistaken for a value (a real sample can
+// legitimately be 0 or +Inf).
+func windowedInner(unit *coreUnit, samplesConditions func(*sqlbuilder.SelectBuilder, bool), seriesSub func(string) (string, []any, error), keyCols []string, localSeriesTable string, selStart, selEnd, stepMs, windowMs int64) (string, []any, error) {
+	effStepMs := stepMs
+	if effStepMs == 0 {
+		effStepMs = 1000
+	}
+	lastIdx := (selEnd - selStart) / effStepMs
+	gridLen := lastIdx + 1
+	w := windowMs / effStepMs
+	bucketLen := gridLen + w
+
+	// A window narrower than the step makes the windows (t_k - range, t_k]
+	// pairwise disjoint: there is nothing to slide, each slot reads exactly
+	// its own window's aggregate. This is exact ONLY over sliver-filtered
+	// rows (samplesConditions adds the window<step predicate): the index
+	// below assigns every gap sample to the window above it, and the filter
+	// is what removes them. Requires a real step — instant queries carry no
+	// sliver filter, so they keep the tiled form and its gates.
+	disjoint := stepMs > 0 && windowMs < stepMs
+	if disjoint {
+		w = 1
+		bucketLen = gridLen
+	}
+
+	seriesSQL, seriesArgs, err := seriesSub(localSeriesTable)
+	if err != nil {
+		return "", nil, err
+	}
+
+	// Bucket index, shifted so the earliest in-window sample lands at 0:
+	// jj = ceil((ts - selStart)/step) + W - 1, folded into one intDiv. Slot
+	// k's window is then buckets jj in [k, k+W-1]. In the disjoint form the
+	// same ceil lands each in-window sample directly on its slot (W = 1),
+	// and the numerator stays positive: the fetch floor is
+	// selStart - range > selStart - step.
+	jjShift := windowMs
+	if disjoint {
+		jjShift = effStepMs
+	}
+	jj := fmt.Sprintf("intDiv(unix_milli - %d + %d - 1, %d)", selStart, jjShift, effStepMs)
+	buckets := sqlbuilder.NewSelectBuilder()
+	selects := make([]string, 0, len(keyCols)+2)
+	// any() over the group key: exact because the key is functionally
+	// dependent on the fingerprint (see joinedInner).
+	for _, col := range keyCols {
+		selects = append(selects, fmt.Sprintf("any(series.%s) AS %s", col, col))
+	}
+	selects = append(selects, fmt.Sprintf("countResample(0, %d, 1)(value, %s) AS cnts", bucketLen, jj))
+	if unit.overFn != "count" {
+		selects = append(selects, fmt.Sprintf("%sResample(0, %d, 1)(value, %s) AS vals", map[string]string{
+			"avg": "sum",
+			"sum": "sum",
+			"min": "min",
+			"max": "max",
+		}[unit.overFn], bucketLen, jj))
+	}
+	buckets.Select(selects...)
+	buckets.From(fmt.Sprintf("%s.%s AS points", databaseName, distributedSamplesV4))
+	buckets.JoinWithOption(sqlbuilder.InnerJoin, fmt.Sprintf("(%s) AS series", seriesSQL), "points.fingerprint = series.fingerprint")
+	samplesConditions(buckets, true)
+	buckets.GroupBy("points.fingerprint")
+	bucketsSQL, bucketsArgs := buckets.BuildWithFlavor(sqlbuilder.ClickHouse)
+
+	windowCnt := fmt.Sprintf("arraySum(arraySlice(cnts, k + 1, %d))", w)
+	var slot string
+	switch unit.overFn {
+	case "count":
+		slot = fmt.Sprintf("if(%s = 0, NULL, toFloat64(%s))", windowCnt, windowCnt)
+	case "sum":
+		slot = fmt.Sprintf("if(%s = 0, NULL, arraySum(arraySlice(vals, k + 1, %d)))", windowCnt, w)
+	case "avg":
+		slot = fmt.Sprintf("if(%s = 0, NULL, arraySum(arraySlice(vals, k + 1, %d)) / %s)", windowCnt, w, windowCnt)
+	case "min":
+		slot = fmt.Sprintf("if(%s = 0, NULL, arrayMin(arrayFilter((v, c) -> c > 0, arraySlice(vals, k + 1, %d), arraySlice(cnts, k + 1, %d))))", windowCnt, w, w)
+	case "max":
+		slot = fmt.Sprintf("if(%s = 0, NULL, arrayMax(arrayFilter((v, c) -> c > 0, arraySlice(vals, k + 1, %d), arraySlice(cnts, k + 1, %d))))", windowCnt, w, w)
+	}
+
+	keyList := strings.Join(keyCols, ", ")
+	inner := fmt.Sprintf(
+		"SELECT %s, arrayMap(k -> %s, range(toUInt64(%d))) AS grid FROM (%s)",
+		keyList, slot, gridLen, bucketsSQL,
+	)
+	return inner, append(seriesArgs, bucketsArgs...), nil
+}
+
+// groupKeyExpr renders the canonical JSON group key for the units whose
+// projected label SET varies per series (see groupKeyColumns): the sorted
+// [key, value] pairs of the projected labels, JSON-encoded.
+//   - by () with no labels: one constant group;
+//   - without (a, b): keep everything except the listed labels and __name__;
+//   - no aggregation: keep everything including __name__ — even when the
+//     unit drops the name from its OUTPUT, the key must keep it so distinct
+//     metrics never merge in SQL; executeUnit strips the name afterwards and
+//     turns a post-strip collision into the engine's duplicate-labelset
+//     error instead of a silently invented merge.
+func groupKeyExpr(unit *coreUnit) string {
+	// An empty label value means "label absent" in Prometheus; the stored
+	// labels JSON can carry empty attribute values, which must not become
+	// output labels or group keys.
+	pairs := "arraySort(JSONExtractKeysAndValues(labels, 'String'))"
+	if !unit.hasAgg {
+		return fmt.Sprintf("toJSONString(arrayFilter(p -> p.2 != '', %s))", pairs)
+	}
+	if unit.by {
+		// Non-empty by (...) never reaches here; groupKeyColumns extracts
+		// those labels as plain columns instead.
+		return "'[]'"
+	}
+	excluded := append([]string{metricNameLabel}, unit.grouping...)
+	return fmt.Sprintf("toJSONString(arrayFilter(p -> p.2 != '' AND p.1 NOT IN (%s), %s))", quotedList(excluded), pairs)
+}
+
+func quotedList(items []string) string {
+	quoted := make([]string, len(items))
+	for i, s := range items {
+		quoted[i] = "'" + strings.ReplaceAll(s, "'", "\\'") + "'"
+	}
+	return strings.Join(quoted, ", ")
+}

pkg/prometheus/clickhouseprometheusv2/transpiler_test.go

@@ -0,0 +1,751 @@
+package clickhouseprometheusv2
+
+import (
+	"context"
+	"testing"
+	"time"
+
+	"github.com/DATA-DOG/go-sqlmock"
+	cmock "github.com/SigNoz/clickhouse-go-mock"
+	"github.com/SigNoz/signoz/pkg/errors"
+	"github.com/SigNoz/signoz/pkg/factory"
+	"github.com/SigNoz/signoz/pkg/instrumentation/instrumentationtest"
+	"github.com/SigNoz/signoz/pkg/prometheus"
+	"github.com/SigNoz/signoz/pkg/telemetrystore"
+	"github.com/SigNoz/signoz/pkg/telemetrystore/telemetrystoretest"
+	"github.com/prometheus/prometheus/model/labels"
+	"github.com/prometheus/prometheus/promql"
+	"github.com/prometheus/prometheus/promql/parser"
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+)
+
+func newTestClient(t *testing.T) (*client, *telemetrystoretest.Provider) {
+	t.Helper()
+	store := telemetrystoretest.New(telemetrystore.Config{Provider: "clickhouse"}, sqlmock.QueryMatcherRegexp)
+	settings := factory.NewScopedProviderSettings(instrumentationtest.New().ToProviderSettings(), "clickhouseprometheusv2_test")
+	return newClient(settings, store, prometheus.Config{}), store
+}
+
+var seriesCols = []cmock.ColumnType{
+	{Name: "fingerprint", Type: "UInt64"},
+	{Name: "labels", Type: "String"},
+}
+
+func parse(t *testing.T, q string) parser.Expr {
+	t.Helper()
+	expr, err := parser.NewParser(parser.Options{}).ParseExpr(q)
+	require.NoError(t, err)
+	return expr
+}
+
+func TestClassifyFullShapes(t *testing.T) {
+	tests := []struct {
+		name  string
+		query string
+		check func(t *testing.T, u *coreUnit)
+	}{
+		{
+			name:  "sum by rate",
+			query: `sum by (pod) (rate(http_requests_total{job="api"}[5m]))`,
+			check: func(t *testing.T, u *coreUnit) {
+				assert.Equal(t, fnRate, u.fn)
+				assert.Equal(t, int64(300_000), u.rangeMs)
+				assert.True(t, u.hasAgg)
+				assert.True(t, u.by)
+				assert.Equal(t, []string{"pod"}, u.grouping)
+			},
+		},
+		{
+			name:  "bare increase with offset",
+			query: `increase(errors_total[10m] offset 30m)`,
+			check: func(t *testing.T, u *coreUnit) {
+				assert.Equal(t, fnIncrease, u.fn)
+				assert.Equal(t, int64(1_800_000), u.offsetMs)
+				assert.False(t, u.hasAgg)
+			},
+		},
+		{
+			name:  "avg without over delta",
+			query: `avg without (instance) (delta(gauge_metric[15m]))`,
+			check: func(t *testing.T, u *coreUnit) {
+				assert.Equal(t, fnDelta, u.fn)
+				assert.True(t, u.hasAgg)
+				assert.False(t, u.by)
+			},
+		},
+		{
+			name:  "scalar pipeline with comparison",
+			query: `sum(rate(x[5m])) * 100 > 5`,
+			check: func(t *testing.T, u *coreUnit) {
+				require.Len(t, u.ops, 2)
+				assert.Equal(t, parser.ItemType(parser.MUL), u.ops[0].op)
+				assert.Equal(t, 100.0, u.ops[0].scalar)
+				assert.Equal(t, parser.ItemType(parser.GTR), u.ops[1].op)
+			},
+		},
+		{
+			name:  "scalar on left with unary minus",
+			query: `-1 * sum(rate(x[5m]))`,
+			check: func(t *testing.T, u *coreUnit) {
+				require.Len(t, u.ops, 1)
+				assert.True(t, u.ops[0].scalarOnLeft)
+				assert.Equal(t, -1.0, u.ops[0].scalar)
+			},
+		},
+		{
+			name:  "bool comparison",
+			query: `sum(rate(x[5m])) >= bool 0.5`,
+			check: func(t *testing.T, u *coreUnit) {
+				require.Len(t, u.ops, 1)
+				assert.True(t, u.ops[0].returnBool)
+			},
+		},
+		{
+			name:  "irate utf8 name",
+			query: `sum by ("k8s.pod.name") (irate({"k8s.container.cpu.time"}[2m]))`,
+			check: func(t *testing.T, u *coreUnit) {
+				assert.Equal(t, fnIRate, u.fn)
+				assert.Equal(t, []string{"k8s.pod.name"}, u.grouping)
+			},
+		},
+		{
+			name:  "bare instant selector keeps name",
+			query: `up{job="api"}`,
+			check: func(t *testing.T, u *coreUnit) {
+				assert.Equal(t, unitInstant, u.kind)
+				assert.True(t, u.keepsName())
+			},
+		},
+		{
+			name:  "gauge aggregation",
+			query: `sum by (pod) (container_memory offset 5m)`,
+			check: func(t *testing.T, u *coreUnit) {
+				assert.Equal(t, unitInstant, u.kind)
+				assert.Equal(t, int64(300_000), u.offsetMs)
+				assert.True(t, u.hasAgg)
+				assert.False(t, u.keepsName())
+			},
+		},
+		{
+			name:  "gauge comparison keeps name",
+			query: `container_memory > 100`,
+			check: func(t *testing.T, u *coreUnit) {
+				assert.Equal(t, unitInstant, u.kind)
+				assert.True(t, u.keepsName())
+			},
+		},
+		{
+			name:  "gauge arithmetic drops name",
+			query: `container_memory / 1024`,
+			check: func(t *testing.T, u *coreUnit) {
+				assert.Equal(t, unitInstant, u.kind)
+				assert.False(t, u.keepsName())
+			},
+		},
+		{
+			name:  "avg_over_time",
+			query: `max by (node) (avg_over_time(load1[10m]))`,
+			check: func(t *testing.T, u *coreUnit) {
+				assert.Equal(t, unitOverTime, u.kind)
+				assert.Equal(t, "avg", u.overFn)
+				assert.Equal(t, int64(600_000), u.rangeMs)
+			},
+		},
+		{
+			name:  "last_over_time keeps name",
+			query: `last_over_time(load1[10m])`,
+			check: func(t *testing.T, u *coreUnit) {
+				assert.Equal(t, unitOverTime, u.kind)
+				assert.Equal(t, "last", u.overFn)
+				assert.True(t, u.keepsName())
+			},
+		},
+	}
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			plan, ok := classify(parse(t, tt.query), testGrid(60_000))
+			require.True(t, ok, "expected transpilable")
+			require.True(t, plan.full, "expected full compilation")
+			require.Len(t, plan.units, 1)
+			tt.check(t, &plan.units[0].core)
+		})
+	}
+}
+
+func TestClassifyFallbackShapes(t *testing.T) {
+	queries := []struct {
+		name  string
+		query string
+		step  int64
+	}{
+		{"default-resolution subquery", `max_over_time(rate(x[5m])[30m:])`, 60_000},
+		{"at modifier", `sum(rate(x[5m] @ 1609746000))`, 60_000},
+		{"at modifier on gauge", `sum(container_memory @ 1609746000)`, 60_000},
+		{"sub-second step", `sum(rate(x[5m]))`, 500},
+		{"sub-second range", `sum(rate(x[1500ms]))`, 60_000},
+		{"by __name__ full", `sum by (__name__) (rate({__name__=~"a|b"}[5m]))`, 60_000},
+		{"quantile_over_time unsupported", `quantile_over_time(0.9, load1[10m])`, 60_000},
+		// Duration expressions resolve into the selectors' static fields only
+		// at evaluation time; classification reads those fields as zero, so
+		// transpiling would silently use the wrong offset (caught by the
+		// conformance corpus' duration_expression.test cases). Offset
+		// expressions parse without the experimental-parser flag, so they do
+		// reach the transpiler; range-position expressions are rejected at
+		// parse (the RangeExpr/StepExpr guards are defense-in-depth).
+		{"duration expression offset on instant", `x offset step()`, 60_000},
+		{"duration expression offset arithmetic", `x offset -step()*2`, 60_000},
+		{"duration expression offset on range", `sum(rate(x[5m] offset max(3s, step())))`, 60_000},
+		{"duration expression subquery step", `max_over_time(rate(x[5m])[30m:step()])`, 60_000},
+	}
+	for _, tt := range queries {
+		t.Run(tt.name, func(t *testing.T) {
+			_, ok := classify(parse(t, tt.query), testGrid(tt.step))
+			assert.False(t, ok, "expected fallback for %s", tt.query)
+		})
+	}
+}
+
+func TestClassifyHybridShapes(t *testing.T) {
+	tests := []struct {
+		name          string
+		query         string
+		wantUnits     int
+		wantRewritten string
+	}{
+		{
+			name:          "histogram quantile",
+			query:         `histogram_quantile(0.95, sum by (le) (rate(http_bucket[5m])))`,
+			wantUnits:     1,
+			wantRewritten: `histogram_quantile(0.95, __signoz_transpiled_0__)`,
+		},
+		{
+			name:          "topk over compiled",
+			query:         `topk(5, sum by (pod) (rate(x[5m])))`,
+			wantUnits:     1,
+			wantRewritten: `topk(5, __signoz_transpiled_0__)`,
+		},
+		{
+			name:          "ratio of compiled units",
+			query:         `sum(rate(a[5m])) / sum(rate(b[5m]))`,
+			wantUnits:     2,
+			wantRewritten: `__signoz_transpiled_0__ / __signoz_transpiled_1__`,
+		},
+		{
+			name:          "or vector zero",
+			query:         `sum(rate(a[5m])) or vector(0)`,
+			wantUnits:     1,
+			wantRewritten: `__signoz_transpiled_0__ or vector(0)`,
+		},
+		{
+			name:          "quantile agg over compiled rate",
+			query:         `quantile(0.9, rate(x[5m]))`,
+			wantUnits:     1,
+			wantRewritten: `quantile(0.9, __signoz_transpiled_0__)`,
+		},
+		{
+			name:          "non-literal scalar side stays engine-side",
+			query:         `sum(rate(x[5m])) * scalar(y)`,
+			wantUnits:     1,
+			wantRewritten: `__signoz_transpiled_0__ * scalar(y)`,
+		},
+		{
+			name:          "compiled mixed with raw selector",
+			query:         `sum by (pod) (rate(a[5m])) / on (pod) group_left () b`,
+			wantUnits:     1,
+			wantRewritten: `__signoz_transpiled_0__ / on (pod) group_left () b`,
+		},
+	}
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			plan, ok := classify(parse(t, tt.query), testGrid(60_000))
+			require.True(t, ok)
+			assert.False(t, plan.full)
+			assert.Len(t, plan.units, tt.wantUnits)
+			assert.Equal(t, tt.wantRewritten, plan.rewritten)
+		})
+	}
+}
+
+func TestClassifyHybridGuards(t *testing.T) {
+	t.Run("no substitution under on(__name__)", func(t *testing.T) {
+		plan, ok := classify(parse(t, `sum(rate(a[5m])) * on (__name__) b`), testGrid(60_000))
+		_ = plan
+		assert.False(t, ok, "matching on __name__ must not see synthetic names")
+	})
+	t.Run("no substitution inside @-pinned subquery", func(t *testing.T) {
+		_, ok := classify(parse(t, `max_over_time(rate(x[5m])[30m:1m] @ 1609746000)`), testGrid(60_000))
+		assert.False(t, ok)
+	})
+}
+
+// The alert-smoothing idiom: units inside a fixed-resolution subquery
+// evaluate on the subquery grid — epoch-aligned multiples of the resolution,
+// starting strictly after (outer start - range), exactly as the engine
+// derives it.
+func TestClassifySubqueryUnits(t *testing.T) {
+	grid := gridContext{startMs: 1_700_000_030_000, endMs: 1_700_007_200_000, stepMs: 60_000}
+
+	plan, ok := classify(parse(t, `min_over_time((sum by (ns) (increase(x[5m])))[10m:5m]) > 0`), grid)
+	require.True(t, ok)
+	require.False(t, plan.full)
+	require.Len(t, plan.units, 1)
+	assert.Equal(t, `min_over_time(__signoz_transpiled_0__[10m:5m]) > 0`, plan.rewritten)
+
+	unit := plan.units[0]
+	// lower bound = outer start - range = 1_699_999_430_000; first multiple
+	// of 300_000 strictly greater is 1_699_999_500_000.
+	assert.Equal(t, int64(1_699_999_500_000), unit.grid.startMs)
+	assert.Equal(t, grid.endMs, unit.grid.endMs)
+	assert.Equal(t, int64(300_000), unit.grid.stepMs)
+	assert.Equal(t, fnIncrease, unit.core.fn)
+
+	t.Run("subquery offset shifts the grid", func(t *testing.T) {
+		plan, ok := classify(parse(t, `max_over_time((sum(rate(x[5m])))[10m:5m] offset 30m)`), grid)
+		require.True(t, ok)
+		require.Len(t, plan.units, 1)
+		// lower = start - offset - range = 1_699_997_630_000 -> first
+		// multiple of 300_000 above = 1_699_997_700_000; end shifts too.
+		assert.Equal(t, int64(1_699_997_700_000), plan.units[0].grid.startMs)
+		assert.Equal(t, grid.endMs-1_800_000, plan.units[0].grid.endMs)
+	})
+
+	t.Run("mollusk ratio-inside-subquery idiom", func(t *testing.T) {
+		q := `min_over_time(((sum by (a) (rate(m1[5m]))) / (avg by (a) (m2)))[5m:1m])`
+		plan, ok := classify(parse(t, q), grid)
+		require.True(t, ok)
+		// Both sides compile on the subquery grid: the rate side and the
+		// gauge aggregation side; the engine joins them and smooths.
+		require.Len(t, plan.units, 2)
+		assert.Equal(t, int64(60_000), plan.units[0].grid.stepMs)
+		assert.Equal(t, unitInstant, plan.units[1].core.kind)
+		assert.Contains(t, plan.rewritten, `__signoz_transpiled_0__ / __signoz_transpiled_1__`)
+	})
+}
+
+func TestBuildUnitSQL(t *testing.T) {
+	unit := &coreUnit{
+		fn:       fnRate,
+		rangeMs:  300_000,
+		hasAgg:   true,
+		aggOp:    parser.SUM,
+		by:       true,
+		grouping: []string{"pod"},
+		matchers: []*labels.Matcher{mustMatcher(t, labels.MatchEqual, "__name__", "http_requests_total")},
+	}
+	sql, args, err := buildUnitSQL(unit, []string{"http_requests_total"}, 1_699_999_700_000, 1_700_003_600_000, 1_700_000_000_000, 1_700_003_600_000, 60_000, 300_000)
+	require.NoError(t, err)
+
+	assert.Contains(t, sql, "timeSeriesRateToGrid(fromUnixTimestamp64Milli(1700000000000), fromUnixTimestamp64Milli(1700003600000), 60, 300)(fromUnixTimestamp64Milli(unix_milli), value)")
+	assert.Contains(t, sql, "unix_milli > ? AND unix_milli <= ?")
+	assert.Contains(t, sql, "bitAnd(flags, 1) = 0")
+	assert.Contains(t, sql, "sumForEach(grid)")
+	// The group-key join rides inside the shard query: distributed samples
+	// at the top level, the local series table in the join subquery, the
+	// grid aggregation grouped per (fingerprint, group key) shard-side.
+	assert.Contains(t, sql, "FROM signoz_metrics.distributed_samples_v4 AS points INNER JOIN (SELECT fingerprint,")
+	assert.Contains(t, sql, "FROM signoz_metrics.time_series_v4 WHERE")
+	// The group key is functionally dependent on the fingerprint (one
+	// labelset per fingerprint): any() is exact and the per-row hash key
+	// shrinks to the fingerprint alone.
+	assert.Contains(t, sql, "any(series.g0) AS g0")
+	assert.Contains(t, sql, "GROUP BY points.fingerprint)")
+	// No samples-side fingerprint condition: the group-key join restricts.
+	assert.NotContains(t, sql, "points.fingerprint IN (")
+	// by (pod) extracts the grouped label directly — no per-row JSON
+	// build/sort/stringify for a known projection.
+	assert.Contains(t, sql, "JSONExtractString(labels, ?) AS g0")
+	assert.NotContains(t, sql, "toJSONString")
+	assert.Contains(t, sql, "SETTINGS allow_experimental_ts_to_grid_aggregate_function = 1")
+	// Args follow placeholder order: the joined series subquery renders
+	// before the samples WHERE, and its select list ('pod') renders before
+	// its own conditions.
+	assert.Equal(t, []any{"pod", "http_requests_total", int64(1_699_999_200_000), int64(1_700_003_600_000), "http_requests_total", int64(1_699_999_700_000), int64(1_700_003_600_000)}, args)
+}
+
+func TestBuildUnitSQLIncreaseAndOffset(t *testing.T) {
+	unit := &coreUnit{
+		fn:       fnIncrease,
+		rangeMs:  600_000,
+		offsetMs: 1_800_000,
+		matchers: []*labels.Matcher{mustMatcher(t, labels.MatchEqual, "__name__", "errors_total")},
+	}
+	sql, _, err := buildUnitSQL(unit, nil, 1_699_997_600_000, 1_700_001_800_000, 1_700_000_000_000, 1_700_003_600_000, 60_000, 300_000)
+	require.NoError(t, err)
+
+	// Grid and window shift by the offset; increase multiplies rate by the
+	// range in seconds.
+	assert.Contains(t, sql, "fromUnixTimestamp64Milli(1699998200000), fromUnixTimestamp64Milli(1700001800000)")
+	assert.Contains(t, sql, "arrayMap(x -> x * 600, timeSeriesRateToGrid")
+	assert.Contains(t, sql, "maxForEach(grid)")
+}
+
+func TestBuildUnitSQLOverLimitJoinOnly(t *testing.T) {
+	// Past the inline limit no fingerprint filter is rendered: the series
+	// join restricts to the matched fingerprints on its own.
+	unit := &coreUnit{
+		fn:       fnRate,
+		rangeMs:  300_000,
+		hasAgg:   true,
+		aggOp:    parser.SUM,
+		by:       true,
+		matchers: []*labels.Matcher{mustMatcher(t, labels.MatchEqual, "__name__", "http_requests_total")},
+	}
+	sql, _, err := buildUnitSQL(unit, []string{"http_requests_total"}, 1_699_999_700_000, 1_700_003_600_000, 1_700_000_000_000, 1_700_003_600_000, 60_000, 300_000)
+	require.NoError(t, err)
+
+	assert.NotContains(t, sql, "points.fingerprint IN")
+	assert.Contains(t, sql, "INNER JOIN (SELECT fingerprint,")
+	assert.Contains(t, sql, "FROM signoz_metrics.time_series_v4 WHERE")
+}
+
+func TestBuildUnitSQLWindowSliver(t *testing.T) {
+	// rate[5m] on a 30m grid evaluates only a 5m sliver before each grid
+	// point — samples in the gaps belong to no window and would only be
+	// buffered by the grid aggregate. The WHERE must keep exactly the
+	// in-window rows: positiveModulo anchored at the selector start (end
+	// can sit off-lattice on unaligned grids, and samples above the start
+	// make the plain modulo dividend negative), and the scan capped at the
+	// last grid point — rows past it are equally windowless.
+	unit := &coreUnit{
+		fn:       fnRate,
+		rangeMs:  300_000,
+		hasAgg:   true,
+		aggOp:    parser.SUM,
+		by:       true,
+		grouping: []string{"pod"},
+		matchers: []*labels.Matcher{mustMatcher(t, labels.MatchEqual, "__name__", "http_requests_total")},
+	}
+	sql, args, err := buildUnitSQL(unit, []string{"http_requests_total"}, 1_699_999_700_000, 1_700_003_600_000, 1_700_000_000_000, 1_700_003_600_000, 1_800_000, 300_000)
+	require.NoError(t, err)
+
+	assert.Contains(t, sql, "positiveModulo(? - unix_milli, ?) < ?")
+	assert.Equal(t, []any{"pod", "http_requests_total", int64(1_699_999_200_000), int64(1_700_003_600_000), "http_requests_total", int64(1_699_999_700_000), int64(1_700_003_600_000), int64(1_700_000_000_000), int64(1_800_000), int64(300_000)}, args)
+
+	t.Run("off-lattice end caps the scan at the last grid point", func(t *testing.T) {
+		// end - start = 50m at a 30m step: the only grid points are start
+		// and start+30m; samples in the trailing 20m serve no window.
+		_, args, err := buildUnitSQL(unit, []string{"http_requests_total"}, 1_699_999_700_000, 1_700_003_000_000, 1_700_000_000_000, 1_700_003_000_000, 1_800_000, 300_000)
+		require.NoError(t, err)
+		assert.Contains(t, args, int64(1_700_001_800_000))
+	})
+
+	t.Run("window covering the step keeps plain bounds", func(t *testing.T) {
+		sql, _, err := buildUnitSQL(unit, []string{"http_requests_total"}, 1_699_999_700_000, 1_700_003_600_000, 1_700_000_000_000, 1_700_003_600_000, 60_000, 300_000)
+		require.NoError(t, err)
+		assert.NotContains(t, sql, "positiveModulo")
+	})
+}
+
+func TestBuildUnitSQLWindowedBucketsWithoutFanOut(t *testing.T) {
+	// The window is W = range/step whole buckets, so each sample lands in
+	// exactly one bucket via GROUP BY and the window slides over bucket
+	// partials — fanning samples into every covered window (ARRAY JOIN)
+	// multiplies rows by W, a row explosion at long ranges.
+	unit := &coreUnit{
+		kind:     unitOverTime,
+		overFn:   "avg",
+		rangeMs:  600_000,
+		matchers: []*labels.Matcher{mustMatcher(t, labels.MatchEqual, "__name__", "node_load1")},
+	}
+	sql, _, err := buildUnitSQL(unit, []string{"node_load1"}, 1_699_999_400_000, 1_700_003_600_000, 1_700_000_000_000, 1_700_003_600_000, 60_000, 600_000)
+	require.NoError(t, err)
+
+	assert.NotContains(t, sql, "ARRAY JOIN")
+	// One group per series with fixed per-bucket arrays (-Resample); the
+	// bucket index jj = ceil((ts - start)/step) + W - 1 folded into a single
+	// intDiv. Grouping by (series, bucket) instead measured 37M hash groups
+	// whose per-thread partials scale memory with max_threads.
+	assert.Contains(t, sql, "countResample(0, 71, 1)(value, intDiv(unix_milli - 1700000000000 + 600000 - 1, 60000)) AS cnts")
+	assert.Contains(t, sql, "sumResample(0, 71, 1)(value, intDiv(unix_milli - 1700000000000 + 600000 - 1, 60000)) AS vals")
+	assert.Contains(t, sql, "any(series.gkey) AS gkey")
+	assert.Contains(t, sql, "GROUP BY points.fingerprint)")
+	assert.NotContains(t, sql, "jj) AS jj")
+	assert.Contains(t, sql, "INNER JOIN (SELECT fingerprint,")
+	assert.Contains(t, sql, "FROM signoz_metrics.time_series_v4 WHERE")
+	// Slide: W = 10 buckets per slot, absent when the window count is 0.
+	assert.Contains(t, sql, "arraySum(arraySlice(cnts, k + 1, 10))")
+	assert.Contains(t, sql, "arraySum(arraySlice(vals, k + 1, 10))")
+}
+
+func TestBuildUnitSQLDisjointOverTime(t *testing.T) {
+	// avg_over_time[5m] on a 30m grid: the windows are pairwise disjoint,
+	// so there is no slide — one Resample bucket per grid slot, read
+	// directly. Exact only together with the window-sliver predicate, which
+	// removes the gap samples the ceil index would otherwise assign to the
+	// window above them.
+	unit := &coreUnit{
+		kind:     unitOverTime,
+		overFn:   "avg",
+		rangeMs:  300_000,
+		matchers: []*labels.Matcher{mustMatcher(t, labels.MatchEqual, "__name__", "node_load1")},
+	}
+	sql, _, err := buildUnitSQL(unit, []string{"node_load1"}, 1_699_999_700_000, 1_700_003_600_000, 1_700_000_000_000, 1_700_003_600_000, 1_800_000, 300_000)
+	require.NoError(t, err)
+
+	assert.NotContains(t, sql, "ARRAY JOIN")
+	// gridLen = 3 slots, bucket array the same length — no W tail.
+	assert.Contains(t, sql, "countResample(0, 3, 1)(value, intDiv(unix_milli - 1700000000000 + 1800000 - 1, 1800000)) AS cnts")
+	assert.Contains(t, sql, "sumResample(0, 3, 1)(value, intDiv(unix_milli - 1700000000000 + 1800000 - 1, 1800000)) AS vals")
+	// Single-bucket window: the slide degenerates to reading one slot.
+	assert.Contains(t, sql, "arraySum(arraySlice(cnts, k + 1, 1))")
+	// The sliver predicate is the correctness precondition of this form.
+	assert.Contains(t, sql, "positiveModulo(? - unix_milli, ?) < ?")
+}
+
+// TestDisjointWindowLattice brute-forces the disjoint-form arithmetic: a
+// sample survives the sliver predicate exactly when some grid window
+// contains it, and the ceil bucket index then lands it on that window's
+// slot. This is the pure-Go mirror of the SQL expressions — the predicate
+// in samplesConditions and jj in windowedInner — over random lattices,
+// including off-lattice ends and samples beyond the last grid point.
+func TestDisjointWindowLattice(t *testing.T) {
+	rng := func(seed *uint64) int64 {
+		*seed = *seed*6364136223846793005 + 1442695040888963407
+		return int64(*seed >> 33)
+	}
+	seed := uint64(42)
+	for trial := 0; trial < 2000; trial++ {
+		stepMs := 1_000 * (1 + rng(&seed)%3600)
+		windowMs := 1 + rng(&seed)%(stepMs-1) // strictly below the step
+		selStart := 1_700_000_000_000 + rng(&seed)%1_000_000
+		selEnd := selStart + rng(&seed)%(50*stepMs) // end may sit off-lattice
+		lastIdx := (selEnd - selStart) / stepMs
+		upper := selStart + lastIdx*stepMs
+
+		for i := 0; i < 50; i++ {
+			u := selStart - windowMs - stepMs + rng(&seed)%(selEnd-selStart+3*stepMs)
+
+			// Oracle: is u inside any window (t_k - window, t_k]?
+			inWindow := false
+			var slot int64 = -1
+			for k := int64(0); k <= lastIdx; k++ {
+				tk := selStart + k*stepMs
+				if u > tk-windowMs && u <= tk {
+					inWindow = true
+					slot = k
+					break
+				}
+			}
+
+			// The SQL: fetch bounds, then the sliver predicate
+			// positiveModulo(selStart - u, step) < window.
+			kept := u > selStart-windowMs && u <= upper
+			if kept {
+				pmod := (selStart - u) % stepMs
+				if pmod < 0 {
+					pmod += stepMs
+				}
+				kept = pmod < windowMs
+			}
+
+			require.Equal(t, inWindow, kept,
+				"sliver keep mismatch: u=%d selStart=%d step=%d window=%d", u, selStart, stepMs, windowMs)
+			if !kept {
+				continue
+			}
+			// jj = ceil((u - selStart)/step) via one intDiv; numerator is
+			// positive because u > selStart - window > selStart - step.
+			jj := (u - selStart + stepMs - 1) / stepMs
+			require.Equal(t, slot, jj,
+				"slot mismatch: u=%d selStart=%d step=%d window=%d", u, selStart, stepMs, windowMs)
+		}
+	}
+}
+
+func TestTryExecuteRange_WindowedGateFallsBack(t *testing.T) {
+	c, store := newTestClient(t)
+	e := &executor{client: c, parser: prometheus.NewParser()}
+
+	start := time.UnixMilli(1_700_000_000_000)
+	end := time.UnixMilli(1_700_003_600_000)
+
+	// 10m range at 90s step: the window is not a whole number of buckets.
+	_, ok, err := e.TryExecuteRange(context.Background(), `avg_over_time(up[10m])`, start, end, 90*time.Second)
+	require.NoError(t, err)
+	assert.False(t, ok, "range not divisible by step must not transpile")
+
+	// 1d range at 60s step: 1440 bucket combines per slot, over the cap.
+	_, ok, err = e.TryExecuteRange(context.Background(), `avg_over_time(up[1d])`, start, end, time.Minute)
+	require.NoError(t, err)
+	assert.False(t, ok, "range/step above maxWindowBuckets must not transpile")
+
+	// 1m range at 5m step: the windows are disjoint slivers — no
+	// divisibility or width requirement, so this transpiles.
+	store.Mock().ExpectQuery("SELECT fingerprint, any\\(labels\\)").WithArgs("up", int64(1_699_999_200_000), int64(1_700_003_600_000)).WillReturnRows(cmock.NewRows(seriesCols, [][]any{}))
+	_, ok, err = e.TryExecuteRange(context.Background(), `avg_over_time(up[1m])`, start, end, 5*time.Minute)
+	require.NoError(t, err)
+	assert.True(t, ok, "range below step is the disjoint form and must transpile")
+}
+
+func TestApplyScalarOps(t *testing.T) {
+	f := func(v float64) *float64 { return &v }
+
+	t.Run("arithmetic chain", func(t *testing.T) {
+		values := []*float64{f(2), nil, f(4)}
+		applyScalarOps([]scalarOp{{op: parser.MUL, scalar: 100}, {op: parser.ADD, scalar: 1}}, values)
+		require.NotNil(t, values[0])
+		assert.Equal(t, 201.0, *values[0])
+		assert.Nil(t, values[1])
+		assert.Equal(t, 401.0, *values[2])
+	})
+
+	t.Run("comparison filters points", func(t *testing.T) {
+		values := []*float64{f(1), f(10)}
+		applyScalarOps([]scalarOp{{op: parser.GTR, scalar: 5}}, values)
+		assert.Nil(t, values[0])
+		require.NotNil(t, values[1])
+		assert.Equal(t, 10.0, *values[1], "filter comparisons keep the original value")
+	})
+
+	t.Run("bool comparison emits 0/1", func(t *testing.T) {
+		values := []*float64{f(1), f(10)}
+		applyScalarOps([]scalarOp{{op: parser.GTR, scalar: 5, returnBool: true}}, values)
+		assert.Equal(t, 0.0, *values[0])
+		assert.Equal(t, 1.0, *values[1])
+	})
+
+	t.Run("scalar on left division", func(t *testing.T) {
+		values := []*float64{f(4)}
+		applyScalarOps([]scalarOp{{op: parser.DIV, scalar: 100, scalarOnLeft: true}}, values)
+		assert.Equal(t, 25.0, *values[0])
+	})
+}
+
+func TestLabelsFromGroupKey(t *testing.T) {
+	lset, err := labelsFromGroupKey(`[["pod","api-0"],["ns","prod"]]`)
+	require.NoError(t, err)
+	assert.Equal(t, "api-0", lset.Get("pod"))
+	assert.Equal(t, "prod", lset.Get("ns"))
+
+	empty, err := labelsFromGroupKey(`[]`)
+	require.NoError(t, err)
+	assert.True(t, empty.IsEmpty())
+}
+
+// testGrid is a 2h query grid ending on a round timestamp.
+func testGrid(stepMs int64) gridContext {
+	return gridContext{startMs: 1_700_000_000_000, endMs: 1_700_007_200_000, stepMs: stepMs}
+}
+
+// A bool comparison returns 0/1, not the sample, so the engine drops
+// __name__; keeping it would change downstream vector matching.
+func TestKeepsName_BoolComparisonDropsName(t *testing.T) {
+	plan, ok := classify(parse(t, `up > bool 0`), testGrid(60_000))
+	require.True(t, ok)
+	assert.False(t, plan.units[0].core.keepsName())
+
+	plan, ok = classify(parse(t, `up > 0`), testGrid(60_000))
+	require.True(t, ok)
+	assert.True(t, plan.units[0].core.keepsName())
+}
+
+// timeSeriesLastToGrid widens its window to max(window, step) — probed on
+// 25.12 — so Last-style units at window < step must fall back or they would
+// resurrect samples the engine's lookback already dropped.
+func TestTryExecuteRange_LastStyleWindowBelowStepTranspiles(t *testing.T) {
+	// These used to fall back because timeSeriesLastToGrid widens its window
+	// to max(window, step). Over sliver-filtered rows the widening is
+	// harmless — the widened window intersected with the data IS the
+	// lookback window — so the gate is gone and both shapes transpile. The
+	// mock returns no series: the point here is the routing, the value
+	// semantics are the parity suite's job.
+	c, store := newTestClient(t)
+	e := &executor{client: c, parser: prometheus.NewParser()}
+
+	start := time.UnixMilli(1_700_000_000_000)
+	end := time.UnixMilli(1_700_003_600_000)
+
+	store.Mock().ExpectQuery("SELECT fingerprint, any\\(labels\\)").WithArgs("up", int64(1_699_999_200_000), int64(1_700_003_600_000)).WillReturnRows(cmock.NewRows(seriesCols, [][]any{}))
+	_, ok, err := e.TryExecuteRange(context.Background(), `sum by (pod) (up)`, start, end, time.Hour)
+	require.NoError(t, err)
+	assert.True(t, ok, "instant selection at step > lookback must transpile")
+
+	store.Mock().ExpectQuery("SELECT fingerprint, any\\(labels\\)").WithArgs("up", int64(1_699_999_200_000), int64(1_700_003_600_000)).WillReturnRows(cmock.NewRows(seriesCols, [][]any{}))
+	_, ok, err = e.TryExecuteRange(context.Background(), `last_over_time(up[10m])`, start, end, time.Hour)
+	require.NoError(t, err)
+	assert.True(t, ok, "last_over_time at range < step must transpile")
+}
+
+// Two metrics collapsing onto one labelset after the name drop, with values
+// on the same grid slot, is the engine's duplicate-labelset error; merging
+// them would invent a series no engine would produce. (Temporally disjoint
+// twins merge instead — see TestMergeSameLabelsetSeries.)
+func TestExecuteUnit_NameCollisionErrors(t *testing.T) {
+	c, store := newTestClient(t)
+	e := &executor{client: c, parser: prometheus.NewParser()}
+
+	store.Mock().ExpectQuery("SELECT fingerprint, any\\(labels\\)").WithArgs("^(?:a|b)$", int64(1_699_999_200_000), int64(1_700_003_600_000)).WillReturnRows(cmock.NewRows(seriesCols, [][]any{
+		{uint64(1), `{"__name__":"a","job":"x"}`},
+		{uint64(2), `{"__name__":"b","job":"x"}`},
+	}))
+	store.Mock().ExpectQuery("SELECT gkey").
+		WithArgs("^(?:a|b)$", int64(1_699_999_200_000), int64(1_700_003_600_000), "a", "b", int64(1_699_999_700_000), int64(1_700_003_600_000)).
+		WillReturnRows(cmock.NewRows(gkeyCols, [][]any{
+			{`[["__name__","a"],["job","x"]]`, []*float64{f64(1)}},
+			{`[["__name__","b"],["job","x"]]`, []*float64{f64(2)}},
+		}))
+
+	plan, ok := classify(parse(t, `rate({__name__=~"a|b"}[5m])`), gridContext{startMs: 1_700_000_000_000, endMs: 1_700_003_600_000, stepMs: 60_000})
+	require.True(t, ok)
+
+	_, err := e.executeUnit(context.Background(), &plan.units[0].core, plan.units[0].grid)
+	require.Error(t, err)
+	assert.Contains(t, err.Error(), "vector cannot contain metrics with the same labelset")
+}
+
+var gkeyCols = []cmock.ColumnType{
+	{Name: "gkey", Type: "String"},
+	{Name: "grid", Type: "Array(Nullable(Float64))"},
+}
+
+func f64(v float64) *float64 { return &v }
+
+// A nameless selector can span metrics whose series alternate in time (one
+// dies inside the lookback before the other appears); after the name drop
+// the engine merges them into ONE series and errors only when two samples
+// share an evaluation timestamp. Pinned by conformance cases
+// operators.test:994/997 (-{job="api"} over http_requests/http_errors).
+func TestMergeSameLabelsetSeries(t *testing.T) {
+	f := func(v float64) *float64 { return &v }
+	api := labels.FromStrings("job", "api")
+
+	out, err := mergeSameLabelsetSeries([]transpiledSeries{
+		{lset: api, values: []*float64{f(-2), nil}},
+		{lset: api, values: []*float64{nil, f(-4)}},
+		{lset: labels.FromStrings("job", "web"), values: []*float64{f(7), nil}},
+	})
+	require.NoError(t, err)
+	require.Len(t, out, 2)
+	assert.Equal(t, []*float64{f(-2), f(-4)}, out[0].values, "temporally disjoint twins must merge into one series")
+
+	_, err = mergeSameLabelsetSeries([]transpiledSeries{
+		{lset: api, values: []*float64{f(1), nil}},
+		{lset: api, values: []*float64{f(2), nil}},
+	})
+	require.Error(t, err, "two values on one evaluation timestamp is the engine's duplicate error")
+	assert.True(t, errors.Ast(err, errors.TypeInvalidInput))
+}
+
+// Hybrid twin case: stripping the synthetic __name__ can leave two engine
+// output series distinguishable only by those names (-metric_a or -metric_b:
+// both {} once real names are dropped). Pinned by conformance cases
+// name_label_dropping.test:137 and operators.test:1016.
+func TestMergeMatrixByLabelset(t *testing.T) {
+	empty := labels.EmptyLabels()
+
+	out, err := mergeMatrixByLabelset(promql.Matrix{
+		{Metric: empty, Floats: []promql.FPoint{{T: 0, F: -1}}},
+		{Metric: empty, Floats: []promql.FPoint{{T: 600_000, F: -4}}},
+	})
+	require.NoError(t, err)
+	require.Len(t, out, 1)
+	assert.Equal(t, []promql.FPoint{{T: 0, F: -1}, {T: 600_000, F: -4}}, out[0].Floats)
+
+	_, err = mergeMatrixByLabelset(promql.Matrix{
+		{Metric: empty, Floats: []promql.FPoint{{T: 0, F: -1}}},
+		{Metric: empty, Floats: []promql.FPoint{{T: 0, F: -3}}},
+	})
+	require.Error(t, err)
+	assert.True(t, errors.Ast(err, errors.TypeInvalidInput))
+}

pkg/prometheus/config.go

@@ -24,6 +24,10 @@ type Config struct {
 
 	// Timeout is the maximum time a query is allowed to run before being aborted.
 	Timeout time.Duration `mapstructure:"timeout"`
+
+	// ProviderName selects the storage provider: "clickhouse" (default) or
+	// "clickhousev2".
+	ProviderName string `mapstructure:"provider"`
 }
 
 func NewConfigFactory() factory.ConfigFactory {
@@ -37,7 +41,8 @@ func newConfig() factory.Config {
 			Path:          "",
 			MaxConcurrent: 20,
 		},
-		Timeout: 2 * time.Minute,
+		Timeout:      2 * time.Minute,
+		ProviderName: "clickhouse",
 	}
 }
 
@@ -45,9 +50,15 @@ func (c Config) Validate() error {
 	if c.Timeout <= 0 {
 		return errors.Newf(errors.TypeInvalidInput, errors.CodeInvalidInput, "prometheus::timeout must be greater than 0")
 	}
+	if c.ProviderName != "" && c.ProviderName != "clickhouse" && c.ProviderName != "clickhousev2" {
+		return errors.Newf(errors.TypeInvalidInput, errors.CodeInvalidInput, "prometheus::provider must be one of [clickhouse, clickhousev2], got %q", c.ProviderName)
+	}
 	return nil
 }
 
 func (c Config) Provider() string {
-	return "clickhouse"
+	if c.ProviderName == "" {
+		return "clickhouse"
+	}
+	return c.ProviderName
 }

pkg/prometheus/promapi/handler.go

@@ -0,0 +1,233 @@
+// Package promapi serves the Prometheus HTTP query API over a
+// prometheus.Prometheus provider: /query and /query_range in the shape of
+// Prometheus' /api/v1 endpoints (https://prometheus.io/docs/prometheus/latest/querying/api/),
+// intended to be mounted under a distinguishing prefix (/prometheus/api/v1)
+// so PromQL-only endpoints are separate from the SigNoz query APIs. The
+// request and response contracts follow Prometheus: form-encoded GET/POST
+// params, {"status":"success","data":{resultType,result}} on success and
+// {"status":"error","errorType","error"} with Prometheus' status codes on
+// failure — so Prometheus-compatible clients can point at the prefix.
+package promapi
+
+import (
+	"context"
+	"encoding/json"
+	"log/slog"
+	"math"
+	"net/http"
+	"strconv"
+	"time"
+
+	promModel "github.com/prometheus/common/model"
+	"github.com/prometheus/prometheus/promql"
+	"github.com/prometheus/prometheus/promql/parser"
+	"github.com/prometheus/prometheus/util/stats"
+
+	"github.com/SigNoz/signoz/pkg/errors"
+	"github.com/SigNoz/signoz/pkg/prometheus"
+)
+
+// Handler serves the Prometheus query API over the configured provider.
+type Handler struct {
+	logger *slog.Logger
+	prom   prometheus.Prometheus
+}
+
+func NewHandler(logger *slog.Logger, prom prometheus.Prometheus) *Handler {
+	return &Handler{logger: logger, prom: prom}
+}
+
+type errorType string
+
+const (
+	errBadData  errorType = "bad_data"
+	errExec     errorType = "execution"
+	errCanceled errorType = "canceled"
+	errTimeout  errorType = "timeout"
+	errInternal errorType = "internal"
+)
+
+type queryData struct {
+	ResultType parser.ValueType  `json:"resultType"`
+	Result     parser.Value      `json:"result"`
+	Stats      stats.QueryStats  `json:"stats,omitempty"`
+}
+
+type response struct {
+	Status    string     `json:"status"`
+	Data      *queryData `json:"data,omitempty"`
+	ErrorType errorType  `json:"errorType,omitempty"`
+	Error     string     `json:"error,omitempty"`
+}
+
+// QueryRange evaluates an expression over a grid: query, start, end, step,
+// and optional timeout/stats params, all in Prometheus' formats.
+func (h *Handler) QueryRange(w http.ResponseWriter, r *http.Request) {
+	start, err := parseTime(r.FormValue("start"))
+	if err != nil {
+		h.respondError(r.Context(), w, errBadData, err)
+		return
+	}
+	end, err := parseTime(r.FormValue("end"))
+	if err != nil {
+		h.respondError(r.Context(), w, errBadData, err)
+		return
+	}
+	if end.Before(start) {
+		h.respondError(r.Context(), w, errBadData, errors.NewInvalidInputf(errors.CodeInvalidInput, "end timestamp must not be before start time"))
+		return
+	}
+	step, err := parseDuration(r.FormValue("step"))
+	if err != nil {
+		h.respondError(r.Context(), w, errBadData, err)
+		return
+	}
+	if step <= 0 {
+		h.respondError(r.Context(), w, errBadData, errors.NewInvalidInputf(errors.CodeInvalidInput, "zero or negative query resolution step widths are not accepted. Try a positive integer"))
+		return
+	}
+	// The engine materializes every point of every series; an unbounded
+	// grid is an unbounded allocation. 11,000 points covers 60s resolution
+	// for a week or 1h resolution for a year.
+	if end.Sub(start)/step > 11000 {
+		h.respondError(r.Context(), w, errBadData, errors.NewInvalidInputf(errors.CodeInvalidInput, "exceeded maximum resolution of 11,000 points per timeseries. Try decreasing the query resolution (?step=XX)"))
+		return
+	}
+
+	ctx, cancel, err := h.contextWithTimeout(r)
+	if err != nil {
+		h.respondError(r.Context(), w, errBadData, err)
+		return
+	}
+	defer cancel()
+
+	qry, err := h.prom.Engine().NewRangeQuery(ctx, h.prom.Storage(), nil, r.FormValue("query"), start, end, step)
+	if err != nil {
+		h.respondError(r.Context(), w, errBadData, err)
+		return
+	}
+	h.exec(ctx, w, r, qry)
+}
+
+// Query evaluates an expression at a single instant: query and optional
+// time/timeout/stats params. A missing time evaluates at the server's now,
+// as in Prometheus.
+func (h *Handler) Query(w http.ResponseWriter, r *http.Request) {
+	ts := time.Now()
+	if t := r.FormValue("time"); t != "" {
+		var err error
+		ts, err = parseTime(t)
+		if err != nil {
+			h.respondError(r.Context(), w, errBadData, err)
+			return
+		}
+	}
+
+	ctx, cancel, err := h.contextWithTimeout(r)
+	if err != nil {
+		h.respondError(r.Context(), w, errBadData, err)
+		return
+	}
+	defer cancel()
+
+	qry, err := h.prom.Engine().NewInstantQuery(ctx, h.prom.Storage(), nil, r.FormValue("query"), ts)
+	if err != nil {
+		h.respondError(r.Context(), w, errBadData, err)
+		return
+	}
+	h.exec(ctx, w, r, qry)
+}
+
+func (h *Handler) exec(ctx context.Context, w http.ResponseWriter, r *http.Request, qry promql.Query) {
+	defer qry.Close()
+	res := qry.Exec(ctx)
+	if res.Err != nil {
+		h.logger.ErrorContext(ctx, "error evaluating promql query", errors.Attr(res.Err))
+		switch res.Err.(type) {
+		case promql.ErrQueryCanceled:
+			h.respondError(ctx, w, errCanceled, res.Err)
+		case promql.ErrQueryTimeout:
+			h.respondError(ctx, w, errTimeout, res.Err)
+		case promql.ErrStorage:
+			h.respondError(ctx, w, errInternal, res.Err)
+		default:
+			h.respondError(ctx, w, errExec, res.Err)
+		}
+		return
+	}
+
+	data := &queryData{ResultType: res.Value.Type(), Result: res.Value}
+	if r.FormValue("stats") != "" {
+		data.Stats = stats.NewQueryStats(qry.Stats())
+	}
+	h.respond(ctx, w, data)
+}
+
+func (h *Handler) contextWithTimeout(r *http.Request) (context.Context, context.CancelFunc, error) {
+	ctx := r.Context()
+	if to := r.FormValue("timeout"); to != "" {
+		timeout, err := parseDuration(to)
+		if err != nil {
+			return nil, nil, err
+		}
+		ctx, cancel := context.WithTimeout(ctx, timeout)
+		return ctx, cancel, nil
+	}
+	ctx, cancel := context.WithCancel(ctx)
+	return ctx, cancel, nil
+}
+
+func (h *Handler) respond(ctx context.Context, w http.ResponseWriter, data *queryData) {
+	w.Header().Set("Content-Type", "application/json")
+	w.WriteHeader(http.StatusOK)
+	if err := json.NewEncoder(w).Encode(&response{Status: "success", Data: data}); err != nil {
+		h.logger.ErrorContext(ctx, "error writing prometheus api response", errors.Attr(err))
+	}
+}
+
+// respondError follows Prometheus' status-code mapping: bad_data 400,
+// execution 422, canceled/timeout 503, internal 500.
+func (h *Handler) respondError(ctx context.Context, w http.ResponseWriter, typ errorType, err error) {
+	code := http.StatusInternalServerError
+	switch typ {
+	case errBadData:
+		code = http.StatusBadRequest
+	case errExec:
+		code = http.StatusUnprocessableEntity
+	case errCanceled, errTimeout:
+		code = http.StatusServiceUnavailable
+	}
+	w.Header().Set("Content-Type", "application/json")
+	w.WriteHeader(code)
+	if encErr := json.NewEncoder(w).Encode(&response{Status: "error", ErrorType: typ, Error: err.Error()}); encErr != nil {
+		h.logger.ErrorContext(ctx, "error writing prometheus api error response", errors.Attr(encErr))
+	}
+}
+
+// parseTime accepts Prometheus' time formats: float unix seconds or RFC3339.
+func parseTime(s string) (time.Time, error) {
+	if t, err := strconv.ParseFloat(s, 64); err == nil {
+		sec, ns := math.Modf(t)
+		return time.Unix(int64(sec), int64(ns*float64(time.Second))), nil
+	}
+	if t, err := time.Parse(time.RFC3339Nano, s); err == nil {
+		return t, nil
+	}
+	return time.Time{}, errors.NewInvalidInputf(errors.CodeInvalidInput, "cannot parse %q to a valid timestamp", s)
+}
+
+// parseDuration accepts Prometheus' duration formats: float seconds or a
+// duration string like 5m.
+func parseDuration(s string) (time.Duration, error) {
+	if d, err := strconv.ParseFloat(s, 64); err == nil {
+		ts := d * float64(time.Second)
+		if ts > float64(math.MaxInt64) || ts < float64(math.MinInt64) {
+			return 0, errors.NewInvalidInputf(errors.CodeInvalidInput, "cannot parse %q to a valid duration. It overflows int64", s)
+		}
+		return time.Duration(ts), nil
+	}
+	if d, err := promModel.ParseDuration(s); err == nil {
+		return time.Duration(d), nil
+	}
+	return 0, errors.NewInvalidInputf(errors.CodeInvalidInput, "cannot parse %q to a valid duration", s)
+}

pkg/prometheus/prometheus.go

@@ -35,3 +35,9 @@ type StatementRecorder interface {
 type StatementCapturer interface {
 	CapturingStorage() (storage.Queryable, StatementRecorder)
 }
+
+// ProviderClickhouseV2 is the clickhousev2 provider name: the factory
+// registration, the prometheus::provider config value and the
+// X-SigNoz-PromQL-Provider request header all use it, so they cannot drift
+// apart.
+const ProviderClickhouseV2 = "clickhousev2"

pkg/prometheus/traits.go

@@ -0,0 +1,49 @@
+package prometheus
+
+import (
+	"context"
+
+	"github.com/prometheus/prometheus/promql/parser"
+)
+
+type queryTraitsKey struct{}
+
+// QueryTraits carries per-query facts a storage implementation cannot derive
+// from SelectHints alone. Call sites that parse the PromQL expression attach
+// traits to the context before handing it to the engine; storages treat a
+// missing traits value as "unknown" and stay conservative.
+type QueryTraits struct {
+	// SubqueryFree is true when the query contains no subquery expression.
+	// Subquery selectors are evaluated at the subquery's own step, but
+	// SelectHints.Step always carries the top-level step, so step-aligned
+	// storage optimizations (e.g. keeping only the last sample per step
+	// bucket) are safe only when this is true.
+	SubqueryFree bool
+}
+
+// DetectQueryTraits derives QueryTraits from a parsed PromQL expression.
+func DetectQueryTraits(expr parser.Expr) QueryTraits {
+	subqueryFree := true
+	parser.Inspect(expr, func(node parser.Node, _ []parser.Node) error {
+		if _, ok := node.(*parser.SubqueryExpr); ok {
+			subqueryFree = false
+		}
+		return nil
+	})
+	return QueryTraits{SubqueryFree: subqueryFree}
+}
+
+// NewContextWithQueryTraits returns a context carrying the given traits.
+func NewContextWithQueryTraits(ctx context.Context, traits QueryTraits) context.Context {
+	return context.WithValue(ctx, queryTraitsKey{}, traits)
+}
+
+// QueryTraitsFromContext returns the traits attached to ctx, if any.
+//
+// Context is used here, unlike for backend selection, because traits must
+// cross the promql engine to reach storage.Querier.Select, and the engine's
+// interfaces offer no other channel; the alternative is a Prometheus fork.
+func QueryTraitsFromContext(ctx context.Context) (QueryTraits, bool) {
+	traits, ok := ctx.Value(queryTraitsKey{}).(QueryTraits)
+	return traits, ok
+}

pkg/querier/api.go

@@ -50,6 +50,7 @@ func (handler *handler) QueryRange(rw http.ResponseWriter, req *http.Request) {
 		render.Error(rw, err)
 		return
 	}
+	queryRangeRequest.PromQLProvider = req.Header.Get("X-SigNoz-PromQL-Provider")
 
 	// Validate the query request
 	if err := queryRangeRequest.Validate(); err != nil {

pkg/querier/clickhouse_query.go

@@ -100,9 +100,20 @@ func (q *chSQLQuery) renderVars(query string, vars map[string]qbtypes.VariableIt
 	return newQuery.String(), nil
 }
 
-// Statement renders the SQL without executing it, for the preview path.
-func (q *chSQLQuery) Statement(_ context.Context) (*qbtypes.Statement, error) {
+func (q *chSQLQuery) render(ctx context.Context) (string, error) {
 	rendered, err := q.renderVars(q.query.Query, q.vars, q.fromMS, q.toMS)
+	if err != nil {
+		return "", err
+	}
+
+	querybuilder.LogIfStatementIsNotValid(ctx, q.logger, rendered)
+
+	return rendered, nil
+}
+
+// Statement renders the SQL without executing it, for the preview path.
+func (q *chSQLQuery) Statement(ctx context.Context) (*qbtypes.Statement, error) {
+	rendered, err := q.render(ctx)
 	if err != nil {
 		return nil, err
 	}
@@ -124,7 +135,7 @@ func (q *chSQLQuery) Execute(ctx context.Context) (*qbtypes.Result, error) {
 		elapsed += p.Elapsed
 	}))
 
-	query, err := q.renderVars(q.query.Query, q.vars, q.fromMS, q.toMS)
+	query, err := q.render(ctx)
 	if err != nil {
 		return nil, err
 	}
@@ -135,11 +146,11 @@ func (q *chSQLQuery) Execute(ctx context.Context) (*qbtypes.Result, error) {
 	}
 	defer rows.Close()
 
-	// TODO: map the errors from ClickHouse to our error types
 	payload, err := consume(rows, q.kind, nil, qbtypes.Step{}, q.query.Name)
 	if err != nil {
 		return nil, err
 	}
+
 	return &qbtypes.Result{
 		Type:  q.kind,
 		Value: payload,

pkg/querier/preview.go

@@ -231,7 +231,7 @@ func (q *querier) buildPreviewProviders(
 			sub.CompositeQuery = qbtypes.CompositeQuery{Queries: []qbtypes.QueryEnvelope{query}}
 		}
 
-		built, _, bErr := q.buildQueries(orgID, &sub, deps, missingMetricQuerySet, event)
+		built, _, bErr := q.buildQueries(orgID, &sub, deps, missingMetricQuerySet, event, promqlOptions{})
 		if bErr != nil {
 			errs[name] = bErr
 			continue

pkg/querier/promql_query.go

@@ -8,15 +8,19 @@ import (
 	"regexp"
 	"sort"
 	"strings"
+	"sync"
 	"text/template"
 	"time"
 
+	"github.com/ClickHouse/clickhouse-go/v2"
+
 	"github.com/prometheus/prometheus/model/labels"
 	"github.com/prometheus/prometheus/promql"
 	"github.com/prometheus/prometheus/promql/parser"
 
 	"github.com/SigNoz/signoz/pkg/errors"
 	"github.com/SigNoz/signoz/pkg/prometheus"
+	"github.com/SigNoz/signoz/pkg/prometheus/clickhouseprometheusv2"
 	"github.com/SigNoz/signoz/pkg/querybuilder"
 	"github.com/SigNoz/signoz/pkg/types/ctxtypes"
 	"github.com/SigNoz/signoz/pkg/types/instrumentationtypes"
@@ -98,6 +102,24 @@ type promqlQuery struct {
 	tr          qbv5.TimeRange
 	requestType qbv5.RequestType
 	vars        map[string]qbv5.VariableItem
+	opts        promqlOptions
+}
+
+// promqlOptions is how a PromQL query relates to the clickhousev2 provider
+// (see querier.promqlOptions for where the fields come from and why they are
+// flag-gated). Both providers are nil for a plain request, so a plain
+// request costs nothing extra.
+type promqlOptions struct {
+	// shadow, when set, runs the query on this provider after serving and
+	// logs any result difference; the response is never affected.
+	shadow *clickhouseprometheusv2.Provider
+	// shadowSlots is the querier-wide admission for shadow runs, shared by
+	// every query so the bound holds per process.
+	shadowSlots chan struct{}
+	// serve, when set, serves the response from this provider instead of the
+	// default path. Comparison callers fetch the default and the pinned
+	// result as two API calls and diff them.
+	serve *clickhouseprometheusv2.Provider
 }
 
 var _ qbv5.Query = (*promqlQuery)(nil)
@@ -110,6 +132,7 @@ func newPromqlQuery(
 	tr qbv5.TimeRange,
 	requestType qbv5.RequestType,
 	variables map[string]qbv5.VariableItem,
+	opts promqlOptions,
 ) *promqlQuery {
 	return &promqlQuery{
 		logger:      logger,
@@ -119,10 +142,19 @@ func newPromqlQuery(
 		tr:          tr,
 		requestType: requestType,
 		vars:        variables,
+		opts:        opts,
 	}
 }
 
 func (q *promqlQuery) Fingerprint() string {
+	// A pinned request must not share cache entries with default serving: a
+	// cached default result would satisfy the pin without running the pinned
+	// provider, and a pinned result would poison normal serving. No
+	// fingerprint means no caching at all — the pin exists to observe a
+	// provider, so a cache in front of it defeats the point.
+	if q.opts.serve != nil {
+		return ""
+	}
 	if q.requestType != qbv5.RequestTypeTimeSeries {
 		return ""
 	}
@@ -252,7 +284,16 @@ func (q *promqlQuery) PreviewStatements(ctx context.Context) ([]prometheus.Captu
 	start := int64(querybuilder.ToNanoSecs(q.tr.From))
 	end := int64(querybuilder.ToNanoSecs(q.tr.To))
 
+	// Attach the same query traits as Execute so the captured statements
+	// match what the live path would run.
+	if expr, parseErr := q.parser.ParseExpr(rendered); parseErr == nil {
+		ctx = prometheus.NewContextWithQueryTraits(ctx, prometheus.DetectQueryTraits(expr))
+	}
+
 	capStorage, recorder := storer.CapturingStorage()
+	if capStorage == nil {
+		return nil, nil
+	}
 	qry, err := q.promEngine.Engine().NewRangeQuery(
 		ctx,
 		capStorage,
@@ -296,6 +337,58 @@ func (q *promqlQuery) Execute(ctx context.Context) (*qbv5.Result, error) {
 		return nil, err
 	}
 
+	// Attach query traits so the storage can prove step-aligned optimizations
+	// safe (see prometheus.QueryTraits). A parse failure surfaces below via
+	// the engine with the enhanced error message.
+	if expr, parseErr := q.parser.ParseExpr(query); parseErr == nil {
+		ctx = prometheus.NewContextWithQueryTraits(ctx, prometheus.DetectQueryTraits(expr))
+	}
+
+	// Accumulate ClickHouse-side scan stats across every storage query this
+	// evaluation issues (engine selectors or the compiled executor): progress
+	// options propagate to each ClickHouse query through the context.
+	var statsMu sync.Mutex
+	var rowsScanned, bytesScanned uint64
+	ctx = clickhouse.Context(ctx, clickhouse.WithProgress(func(p *clickhouse.Progress) {
+		statsMu.Lock()
+		rowsScanned += p.Rows
+		bytesScanned += p.Bytes
+		statsMu.Unlock()
+	}))
+
+	began := time.Now()
+
+	// A pinned provider serves directly from it: comparison callers fetch
+	// the default result and the pinned result as two API calls and diff
+	// them.
+	if q.opts.serve != nil {
+		matrix, err := q.serveFromProvider(ctx, query, start, end)
+		if err != nil {
+			if enhanced := tryEnhancePromQLExecError(err); enhanced != nil {
+				return nil, enhanced
+			}
+			return nil, err
+		}
+		return q.toResult(matrix, nil, began, &statsMu, &rowsScanned, &bytesScanned), nil
+	}
+
+	// When the serving provider itself is clickhousev2
+	// (prometheus::provider: clickhousev2), serve the way the provider is
+	// designed to serve: transpiled when the shape allows. Without this the
+	// override would silently run the engine path only.
+	if prov, ok := q.promEngine.(*clickhouseprometheusv2.Provider); ok {
+		matrix, served, err := prov.TryExecuteRange(ctx, query, time.Unix(0, start), time.Unix(0, end), q.query.Step.Duration)
+		if err != nil {
+			if enhanced := tryEnhancePromQLExecError(err); enhanced != nil {
+				return nil, enhanced
+			}
+			return nil, err
+		}
+		if served {
+			return q.toResult(matrix, nil, began, &statsMu, &rowsScanned, &bytesScanned), nil
+		}
+	}
+
 	qry, err := q.promEngine.Engine().NewRangeQuery(
 		ctx,
 		q.promEngine.Storage(),
@@ -331,11 +424,42 @@ func (q *promqlQuery) Execute(ctx context.Context) (*qbv5.Result, error) {
 		return nil, errors.WrapInternalf(promErr, errors.CodeInternal, "error getting matrix from promql query %q", query)
 	}
 
-	excludeLabel := func(labelName string) bool {
-		if labelName == "__name__" {
-			return false
+	if q.opts.shadow != nil {
+		// Shadows detach from the request, so without admission a dashboard
+		// burst would stack unbounded ClickHouse work for up to the shadow
+		// timeout — the concurrency pattern behind the original outages.
+		// Non-blocking: at the cap the comparison is skipped, not queued;
+		// a sampled shadow stream is exactly as useful for rollout evidence.
+		select {
+		case q.opts.shadowSlots <- struct{}{}:
+			// The engine pools the result's sample slices on Close; the
+			// shadow comparison needs a stable copy of what was served.
+			served := copyMatrix(matrix)
+			servedIn := time.Since(began)
+			go func() {
+				defer func() { <-q.opts.shadowSlots }()
+				q.runShadowCompare(context.WithoutCancel(ctx), query, start, end, served, servedIn)
+			}()
+		default:
+			q.logger.DebugContext(ctx, "promql shadow skipped: at concurrency cap", slog.String("query", query))
 		}
-		return strings.HasPrefix(labelName, "__") || labelName == "fingerprint"
+	}
+
+	warnings, _ := res.Warnings.AsStrings(query, 10, 0)
+	return q.toResult(matrix, warnings, began, &statsMu, &rowsScanned, &bytesScanned), nil
+}
+
+// toResult converts an evaluated matrix into the v5 result shape, attaching
+// the ClickHouse scan stats accumulated during evaluation.
+func (q *promqlQuery) toResult(matrix promql.Matrix, warnings []string, began time.Time, statsMu *sync.Mutex, rowsScanned, bytesScanned *uint64) *qbv5.Result {
+	// Hide only known SigNoz storage keys: label names are user data and may
+	// legitimately start with "__" (e.g. __address__), so a blanket dunder
+	// strip mangles user labelsets. The __scope./__resource. prefixes cover
+	// every exporter version's keys.
+	excludeLabel := func(labelName string) bool {
+		return labelName == "__temporality__" ||
+			strings.HasPrefix(labelName, "__scope.") ||
+			strings.HasPrefix(labelName, "__resource.")
 	}
 
 	var series []*qbv5.TimeSeries
@@ -363,7 +487,13 @@ func (q *promqlQuery) Execute(ctx context.Context) (*qbv5.Result, error) {
 		series = append(series, &s)
 	}
 
-	warnings, _ := res.Warnings.AsStrings(query, 10, 0)
+	statsMu.Lock()
+	stats := qbv5.ExecStats{
+		RowsScanned:  *rowsScanned,
+		BytesScanned: *bytesScanned,
+		DurationMS:   uint64(time.Since(began).Milliseconds()),
+	}
+	statsMu.Unlock()
 
 	tsData := &qbv5.TimeSeriesData{
 		QueryName: q.query.Name,
@@ -397,6 +527,6 @@ func (q *promqlQuery) Execute(ctx context.Context) (*qbv5.Result, error) {
 		Type:     q.requestType,
 		Value:    payload,
 		Warnings: warnings,
-		// TODO: map promql stats?
-	}, nil
+		Stats:    stats,
+	}
 }

pkg/querier/promql_query_test.go

@@ -7,6 +7,7 @@ import (
 
 	"github.com/SigNoz/signoz/pkg/errors"
 	"github.com/SigNoz/signoz/pkg/prometheus"
+	"github.com/SigNoz/signoz/pkg/prometheus/clickhouseprometheusv2"
 	qbv5 "github.com/SigNoz/signoz/pkg/types/querybuildertypes/querybuildertypesv5"
 	"github.com/stretchr/testify/assert"
 )
@@ -440,3 +441,15 @@ func TestQuotedMetricOutsideBracesPattern(t *testing.T) {
 		})
 	}
 }
+
+// A pinned request must not share cache entries with default serving: a
+// cached default result would satisfy the pin without running the pinned
+// provider.
+func TestFingerprint_PinnedProviderBypassesCache(t *testing.T) {
+	q := &promqlQuery{
+		logger: slog.Default(),
+		query:  qbv5.PromQuery{Query: "up"},
+		opts:   promqlOptions{serve: &clickhouseprometheusv2.Provider{}},
+	}
+	assert.Empty(t, q.Fingerprint())
+}

pkg/querier/promql_shadow.go

@@ -0,0 +1,186 @@
+package querier
+
+import (
+	"context"
+	"fmt"
+	"log/slog"
+	"math"
+	"sort"
+	"time"
+
+	"github.com/ClickHouse/clickhouse-go/v2"
+	"github.com/SigNoz/signoz/pkg/prometheus"
+	"github.com/SigNoz/signoz/pkg/prometheus/clickhouseprometheusv2"
+	"github.com/prometheus/prometheus/model/labels"
+	"github.com/prometheus/prometheus/promql"
+)
+
+// shadowTimeout bounds a shadow evaluation; a shadow run must never outlive
+// the request by much or pile up.
+const shadowTimeout = 2 * time.Minute
+
+// runShadowCompare executes the query on the clickhousev2 provider exactly
+// as it would serve (transpiled when the shape allows, engine over the v2
+// querier otherwise), compares against the served result and logs the
+// outcome. Serving is never affected: this runs after the response, off the
+// request context, and only logs. The mismatch and failure logs are the
+// rollout evidence — serving cuts over to v2 only after they stay clean.
+func (q *promqlQuery) runShadowCompare(ctx context.Context, query string, startNs, endNs int64, served promql.Matrix, servedIn time.Duration) {
+	defer func() {
+		if r := recover(); r != nil {
+			q.logger.ErrorContext(ctx, "promql shadow comparison panicked", slog.Any("panic", r), slog.String("query", query))
+		}
+	}()
+
+	ctx, cancel := context.WithTimeout(ctx, shadowTimeout)
+	defer cancel()
+
+	// The request context carries the served response's scan-stats progress
+	// callback; without replacing it the shadow's ClickHouse progress would
+	// race into the served stats. The response itself was already sent.
+	ctx = clickhouse.Context(ctx, clickhouse.WithProgress(func(*clickhouse.Progress) {}))
+
+	if expr, parseErr := q.parser.ParseExpr(query); parseErr == nil {
+		ctx = prometheus.NewContextWithQueryTraits(ctx, prometheus.DetectQueryTraits(expr))
+	}
+
+	start, end := time.Unix(0, startNs), time.Unix(0, endNs)
+	began := time.Now()
+	shadow, transpiled, err := executeOnProvider(ctx, q.opts.shadow, query, start, end, q.query.Step.Duration)
+	shadowIn := time.Since(began)
+
+	logAttrs := []any{
+		slog.String("query", query),
+		slog.Int64("start_ms", startNs/int64(time.Millisecond)),
+		slog.Int64("end_ms", endNs/int64(time.Millisecond)),
+		slog.Duration("step", q.query.Step.Duration),
+		slog.Bool("transpiled", transpiled),
+		slog.Duration("served_in", servedIn),
+		slog.Duration("shadow_in", shadowIn),
+	}
+
+	if err != nil {
+		// A shadow failure would be a serving failure after rollout; surface
+		// it at the same level as a result mismatch.
+		q.logger.WarnContext(ctx, "promql shadow execution failed", append(logAttrs, slog.Any("error", err))...)
+		return
+	}
+
+	servedNorm := normalizeShadowMatrix(served)
+	shadowNorm := normalizeShadowMatrix(shadow)
+	if diff := diffShadowMatrices(servedNorm, shadowNorm); diff != "" {
+		q.logger.WarnContext(ctx, "promql shadow comparison mismatch", append(logAttrs,
+			slog.String("diff", diff),
+			slog.Int("served_series", len(servedNorm)),
+			slog.Int("shadow_series", len(shadowNorm)),
+		)...)
+		return
+	}
+	// Matches log the timings: served_in vs shadow_in across the fleet is
+	// the perf evidence for the cutover, gathered for free.
+	q.logger.DebugContext(ctx, "promql shadow comparison matched", logAttrs...)
+}
+
+// serveFromProvider evaluates the query the way the pinned provider would
+// serve it.
+func (q *promqlQuery) serveFromProvider(ctx context.Context, query string, startNs, endNs int64) (promql.Matrix, error) {
+	matrix, _, err := executeOnProvider(ctx, q.opts.serve, query, time.Unix(0, startNs), time.Unix(0, endNs), q.query.Step.Duration)
+	return matrix, err
+}
+
+// executeOnProvider evaluates the query the way the provider would serve it:
+// transpiled in ClickHouse when the shape allows, the engine over the
+// provider's storage otherwise. The returned matrix is an owned copy.
+func executeOnProvider(ctx context.Context, prov *clickhouseprometheusv2.Provider, query string, start, end time.Time, step time.Duration) (promql.Matrix, bool, error) {
+	matrix, ok, err := prov.TryExecuteRange(ctx, query, start, end, step)
+	if err != nil {
+		return nil, true, err
+	}
+	if ok {
+		return matrix, true, nil
+	}
+
+	qry, err := prov.Engine().NewRangeQuery(ctx, prov.Storage(), nil, query, start, end, step)
+	if err != nil {
+		return nil, false, err
+	}
+	defer qry.Close()
+
+	res := qry.Exec(ctx)
+	if res.Err != nil {
+		return nil, false, res.Err
+	}
+	matrix, err = res.Matrix()
+	if err != nil {
+		return nil, false, err
+	}
+	// Close returns the result's sample slices to the engine pool.
+	return copyMatrix(matrix), false, nil
+}
+
+func copyMatrix(matrix promql.Matrix) promql.Matrix {
+	out := make(promql.Matrix, 0, len(matrix))
+	for _, s := range matrix {
+		floats := make([]promql.FPoint, len(s.Floats))
+		copy(floats, s.Floats)
+		out = append(out, promql.Series{Metric: s.Metric.Copy(), Floats: floats})
+	}
+	return out
+}
+
+// normalizeShadowMatrix sorts by label set for order-independent
+// comparison. Both providers now resolve series identity the same way
+// (empty-valued labels dropped at read, no synthetic fingerprint label
+// since the v1 series-identity fix), so labels need no normalization.
+func normalizeShadowMatrix(matrix promql.Matrix) promql.Matrix {
+	out := make(promql.Matrix, 0, len(matrix))
+	out = append(out, matrix...)
+	sort.Slice(out, func(i, j int) bool { return labels.Compare(out[i].Metric, out[j].Metric) < 0 })
+	return out
+}
+
+// diffShadowMatrices returns a description of the first difference, or "".
+// Values compare with relative tolerance: spatial aggregations accumulate
+// floats in storage order, which differs between the providers in the last
+// ULP.
+func diffShadowMatrices(served, shadow promql.Matrix) string {
+	const relTol = 1e-9
+	if len(served) != len(shadow) {
+		return fmt.Sprintf("series count: served=%d shadow=%d", len(served), len(shadow))
+	}
+	for i := range served {
+		if labels.Compare(served[i].Metric, shadow[i].Metric) != 0 {
+			return fmt.Sprintf("series %d labels: served=%s shadow=%s", i, served[i].Metric, shadow[i].Metric)
+		}
+		if len(served[i].Floats) != len(shadow[i].Floats) {
+			return fmt.Sprintf("series %s points: served=%d shadow=%d", served[i].Metric, len(served[i].Floats), len(shadow[i].Floats))
+		}
+		for j := range served[i].Floats {
+			a, b := served[i].Floats[j], shadow[i].Floats[j]
+			if a.T != b.T {
+				return fmt.Sprintf("series %s point %d ts: served=%d shadow=%d", served[i].Metric, j, a.T, b.T)
+			}
+			// NaN and infinities first: NaN != NaN and Inf-Inf arithmetic
+			// would otherwise make one-sided NaN and Inf-vs-finite compare
+			// as equal (NaN > x and Inf > Inf are both false).
+			if math.IsNaN(a.F) || math.IsNaN(b.F) {
+				if math.IsNaN(a.F) != math.IsNaN(b.F) {
+					return fmt.Sprintf("series %s @%d value: served=%v shadow=%v", served[i].Metric, a.T, a.F, b.F)
+				}
+				continue
+			}
+			if math.IsInf(a.F, 0) || math.IsInf(b.F, 0) {
+				if a.F != b.F {
+					return fmt.Sprintf("series %s @%d value: served=%v shadow=%v", served[i].Metric, a.T, a.F, b.F)
+				}
+				continue
+			}
+			diff := math.Abs(a.F - b.F)
+			scale := math.Max(math.Abs(a.F), math.Abs(b.F))
+			if diff > relTol*math.Max(scale, 1e-300) && diff > 1e-12 {
+				return fmt.Sprintf("series %s @%d value: served=%v shadow=%v", served[i].Metric, a.T, a.F, b.F)
+			}
+		}
+	}
+	return ""
+}

pkg/querier/promql_shadow_test.go

@@ -0,0 +1,67 @@
+package querier
+
+import (
+	"math"
+	"testing"
+
+	"github.com/prometheus/prometheus/model/labels"
+	"github.com/prometheus/prometheus/promql"
+	"github.com/stretchr/testify/assert"
+)
+
+func TestNormalizeShadowMatrix(t *testing.T) {
+	matrix := promql.Matrix{
+		{
+			Metric: labels.FromStrings("__name__", "up", "job", "api"),
+			Floats: []promql.FPoint{{T: 1000, F: 1}},
+		},
+		{
+			Metric: labels.FromStrings("a", "1"),
+			Floats: []promql.FPoint{{T: 1000, F: 2}},
+		},
+	}
+	norm := normalizeShadowMatrix(matrix)
+	// sorted by label set; labels pass through untouched — both providers
+	// resolve series identity identically since the v1 series-identity fix
+	assert.Equal(t, labels.FromStrings("__name__", "up", "job", "api"), norm[0].Metric)
+	assert.Equal(t, labels.FromStrings("a", "1"), norm[1].Metric)
+}
+
+func TestDiffShadowMatrices(t *testing.T) {
+	series := func(v float64) promql.Matrix {
+		return promql.Matrix{{Metric: labels.FromStrings("a", "1"), Floats: []promql.FPoint{{T: 1000, F: v}}}}
+	}
+
+	assert.Empty(t, diffShadowMatrices(series(1.5), series(1.5)))
+	// last-ULP differences from storage-order float accumulation are expected
+	assert.Empty(t, diffShadowMatrices(series(0.08888888888888889), series(0.08888888888888888)))
+	assert.Empty(t, diffShadowMatrices(series(math.NaN()), series(math.NaN())))
+
+	assert.Contains(t, diffShadowMatrices(series(1.5), series(1.6)), "value")
+	assert.Contains(t, diffShadowMatrices(series(1.5), promql.Matrix{}), "series count")
+	assert.Contains(t, diffShadowMatrices(
+		series(1.5),
+		promql.Matrix{{Metric: labels.FromStrings("a", "2"), Floats: []promql.FPoint{{T: 1000, F: 1.5}}}},
+	), "labels")
+	assert.Contains(t, diffShadowMatrices(
+		series(1.5),
+		promql.Matrix{{Metric: labels.FromStrings("a", "1"), Floats: []promql.FPoint{{T: 2000, F: 1.5}}}},
+	), "ts")
+}
+
+// One-sided NaN makes every float comparison false, and Inf-Inf arithmetic
+// yields Inf > Inf == false; without explicit handling both divergences log
+// as matched — a shadow comparator that cannot see them would green-light a
+// broken rollout.
+func TestDiffShadowMatrices_SpecialFloats(t *testing.T) {
+	point := func(v float64) promql.Matrix {
+		return promql.Matrix{{Metric: labels.FromStrings("a", "1"), Floats: []promql.FPoint{{T: 1000, F: v}}}}
+	}
+
+	assert.NotEmpty(t, diffShadowMatrices(point(math.NaN()), point(1.5)), "one-sided NaN must diff")
+	assert.NotEmpty(t, diffShadowMatrices(point(1.5), point(math.NaN())), "one-sided NaN must diff either way")
+	assert.NotEmpty(t, diffShadowMatrices(point(math.Inf(1)), point(1.5)), "Inf vs finite must diff")
+	assert.NotEmpty(t, diffShadowMatrices(point(math.Inf(1)), point(math.Inf(-1))), "opposite infinities must diff")
+	assert.Empty(t, diffShadowMatrices(point(math.Inf(1)), point(math.Inf(1))), "equal infinities match")
+	assert.Empty(t, diffShadowMatrices(point(math.NaN()), point(math.NaN())), "both NaN match")
+}

pkg/querier/querier.go

@@ -19,10 +19,12 @@ import (
 	"github.com/SigNoz/signoz/pkg/factory"
 	"github.com/SigNoz/signoz/pkg/flagger"
 	"github.com/SigNoz/signoz/pkg/prometheus"
+	"github.com/SigNoz/signoz/pkg/prometheus/clickhouseprometheusv2"
 	"github.com/SigNoz/signoz/pkg/query-service/utils"
 	"github.com/SigNoz/signoz/pkg/querybuilder"
 	"github.com/SigNoz/signoz/pkg/telemetrystore"
 	"github.com/SigNoz/signoz/pkg/types/ctxtypes"
+	"github.com/SigNoz/signoz/pkg/types/featuretypes"
 	"github.com/SigNoz/signoz/pkg/types/instrumentationtypes"
 	"github.com/SigNoz/signoz/pkg/types/metrictypes"
 	"github.com/SigNoz/signoz/pkg/types/telemetrytypes"
@@ -36,11 +38,19 @@ var (
 )
 
 type querier struct {
-	logger                   *slog.Logger
-	fl                       flagger.Flagger
-	telemetryStore           telemetrystore.TelemetryStore
-	metadataStore            telemetrytypes.MetadataStore
-	promEngine               prometheus.Prometheus
+	logger         *slog.Logger
+	fl             flagger.Flagger
+	telemetryStore telemetrystore.TelemetryStore
+	metadataStore  telemetrytypes.MetadataStore
+	promEngine     prometheus.Prometheus
+	// promV2 is the clickhousev2 prometheus provider, wired only when the
+	// serving provider is the default one (nil otherwise). It reads the same
+	// ClickHouse data through a different implementation; PromQL queries
+	// shadow-compare against it behind the use_prometheus_clickhouse_v2 flag
+	// and can be pinned to it for a response (see promqlOptions). It never
+	// serves by default — that cutover happens only after the shadow logs
+	// stay clean.
+	promV2                   *clickhouseprometheusv2.Provider
 	traceStmtBuilder         qbtypes.StatementBuilder[qbtypes.TraceAggregation]
 	logStmtBuilder           qbtypes.StatementBuilder[qbtypes.LogAggregation]
 	auditStmtBuilder         qbtypes.StatementBuilder[qbtypes.LogAggregation]
@@ -51,8 +61,16 @@ type querier struct {
 	liveDataRefresh          time.Duration
 	builderConfig            builderConfig
 	maxConcurrentQueries     int
+	// shadowSlots bounds concurrent shadow comparisons per process; shadows
+	// detach from their requests, so nothing else limits how many pile up.
+	shadowSlots chan struct{}
 }
 
+// maxConcurrentShadows is deliberately small: a shadow is a full extra
+// ClickHouse evaluation, and a sampled stream of comparisons is exactly as
+// useful for rollout evidence as an exhaustive one under load.
+const maxConcurrentShadows = 8
+
 var _ Querier = (*querier)(nil)
 
 func New(
@@ -60,6 +78,7 @@ func New(
 	telemetryStore telemetrystore.TelemetryStore,
 	metadataStore telemetrytypes.MetadataStore,
 	promEngine prometheus.Prometheus,
+	promV2 *clickhouseprometheusv2.Provider,
 	traceStmtBuilder qbtypes.StatementBuilder[qbtypes.TraceAggregation],
 	logStmtBuilder qbtypes.StatementBuilder[qbtypes.LogAggregation],
 	auditStmtBuilder qbtypes.StatementBuilder[qbtypes.LogAggregation],
@@ -81,6 +100,7 @@ func New(
 		telemetryStore:           telemetryStore,
 		metadataStore:            metadataStore,
 		promEngine:               promEngine,
+		promV2:                   promV2,
 		traceStmtBuilder:         traceStmtBuilder,
 		logStmtBuilder:           logStmtBuilder,
 		auditStmtBuilder:         auditStmtBuilder,
@@ -93,6 +113,7 @@ func New(
 			logTraceIDWindowPaddingMS: uint64(logTraceIDWindowPadding.Milliseconds()),
 		},
 		maxConcurrentQueries: maxConcurrentQueries,
+		shadowSlots:          make(chan struct{}, maxConcurrentShadows),
 	}
 }
 
@@ -132,7 +153,11 @@ func (q *querier) QueryRange(ctx context.Context, orgID valuer.UUID, req *qbtype
 		missingMetricQuerySet[name] = true
 	}
 
-	queries, steps, err := q.buildQueries(orgID, req, dependencyQueries, missingMetricQuerySet, event)
+	promqlOpts, err := q.promqlOptions(ctx, orgID, req)
+	if err != nil {
+		return nil, err
+	}
+	queries, steps, err := q.buildQueries(orgID, req, dependencyQueries, missingMetricQuerySet, event, promqlOpts)
 	if err != nil {
 		return nil, err
 	}
@@ -175,12 +200,41 @@ func (q *querier) QueryRange(ctx context.Context, orgID valuer.UUID, req *qbtype
 	return qbResp, qbErr
 }
 
+// promqlOptions derives the PromQL execution options for a request. With the
+// org's use_prometheus_clickhouse_v2 flag on, queries are shadow-compared
+// against the clickhousev2 provider (serving unaffected, diffs logged; see
+// promql_shadow.go). The X-SigNoz-PromQL-Provider header may instead pin the
+// response to that provider — integration tests and support fetch both
+// results for comparison — so it is deliberately flag-gated too: without the
+// gate the header would be an unaudited switch onto a provider still under
+// validation.
+func (q *querier) promqlOptions(ctx context.Context, orgID valuer.UUID, req *qbtypes.QueryRangeRequest) (promqlOptions, error) {
+	enabled := q.fl.BooleanOrEmpty(ctx, flagger.FeatureUsePrometheusClickhouseV2, featuretypes.NewFlaggerEvaluationContext(orgID))
+	if req.PromQLProvider == "" {
+		if enabled && q.promV2 != nil {
+			return promqlOptions{shadow: q.promV2, shadowSlots: q.shadowSlots}, nil
+		}
+		return promqlOptions{}, nil
+	}
+	if req.PromQLProvider != prometheus.ProviderClickhouseV2 {
+		return promqlOptions{}, errors.NewInvalidInputf(errors.CodeInvalidInput, "unknown promql provider %q", req.PromQLProvider)
+	}
+	if !enabled {
+		return promqlOptions{}, errors.NewInvalidInputf(errors.CodeInvalidInput, "promql provider %q requires the use_prometheus_clickhouse_v2 flag", req.PromQLProvider)
+	}
+	if q.promV2 == nil {
+		return promqlOptions{}, errors.NewInvalidInputf(errors.CodeInvalidInput, "promql provider %q is not available", req.PromQLProvider)
+	}
+	return promqlOptions{serve: q.promV2}, nil
+}
+
 func (q *querier) buildQueries(
 	orgID valuer.UUID,
 	req *qbtypes.QueryRangeRequest,
 	dependencyQueries map[string]bool,
 	missingMetricQuerySet map[string]bool,
 	event *qbtypes.QBEvent,
+	promqlOpts promqlOptions,
 ) (map[string]qbtypes.Query, map[string]qbtypes.Step, error) {
 
 	tmplVars := req.Variables
@@ -205,7 +259,7 @@ func (q *querier) buildQueries(
 			if !ok {
 				return nil, nil, errors.NewInvalidInputf(errors.CodeInvalidInput, "invalid promql query spec %T", query.Spec)
 			}
-			promqlQuery := newPromqlQuery(q.logger, q.promEngine, promQuery, qbtypes.TimeRange{From: req.Start, To: req.End}, req.RequestType, tmplVars)
+			promqlQuery := newPromqlQuery(q.logger, q.promEngine, promQuery, qbtypes.TimeRange{From: req.Start, To: req.End}, req.RequestType, tmplVars, promqlOpts)
 			queries[promQuery.Name] = promqlQuery
 			steps[promQuery.Name] = promQuery.Step
 		case qbtypes.QueryTypeClickHouseSQL:
@@ -848,7 +902,7 @@ func (q *querier) createRangedQuery(_ valuer.UUID, originalQuery qbtypes.Query,
 	switch qt := originalQuery.(type) {
 	case *promqlQuery:
 		queryCopy := qt.query.Copy()
-		return newPromqlQuery(q.logger, q.promEngine, queryCopy, timeRange, qt.requestType, qt.vars)
+		return newPromqlQuery(q.logger, qt.promEngine, queryCopy, timeRange, qt.requestType, qt.vars, qt.opts)
 
 	case *chSQLQuery:
 		queryCopy := qt.query.Copy()

pkg/querier/querier_test.go

@@ -48,6 +48,7 @@ func TestQueryRange_MetricTypeMissing(t *testing.T) {
 		nil, // telemetryStore
 		metadataStore,
 		nil,                // prometheus
+		nil,                // promV2
 		nil,                // traceStmtBuilder
 		nil,                // logStmtBuilder
 		nil,                // auditStmtBuilder
@@ -120,6 +121,7 @@ func TestQueryRange_MetricTypeFromStore(t *testing.T) {
 		telemetryStore,
 		metadataStore,
 		nil,                      // prometheus
+		nil,                      // promV2
 		nil,                      // traceStmtBuilder
 		nil,                      // logStmtBuilder
 		nil,                      // auditStmtBuilder

pkg/querier/signozquerier/provider.go

@@ -7,6 +7,7 @@ import (
 	"github.com/SigNoz/signoz/pkg/factory"
 	"github.com/SigNoz/signoz/pkg/flagger"
 	"github.com/SigNoz/signoz/pkg/prometheus"
+	"github.com/SigNoz/signoz/pkg/prometheus/clickhouseprometheusv2"
 	"github.com/SigNoz/signoz/pkg/querier"
 	"github.com/SigNoz/signoz/pkg/querybuilder"
 	"github.com/SigNoz/signoz/pkg/telemetryaudit"
@@ -22,6 +23,7 @@ import (
 func NewFactory(
 	telemetryStore telemetrystore.TelemetryStore,
 	prometheus prometheus.Prometheus,
+	promV2 *clickhouseprometheusv2.Provider,
 	cache cache.Cache,
 	flagger flagger.Flagger,
 ) factory.ProviderFactory[querier.Querier, querier.Config] {
@@ -32,7 +34,7 @@ func NewFactory(
 			settings factory.ProviderSettings,
 			cfg querier.Config,
 		) (querier.Querier, error) {
-			return newProvider(ctx, settings, cfg, telemetryStore, prometheus, cache, flagger)
+			return newProvider(ctx, settings, cfg, telemetryStore, prometheus, promV2, cache, flagger)
 		},
 	)
 }
@@ -43,6 +45,7 @@ func newProvider(
 	cfg querier.Config,
 	telemetryStore telemetrystore.TelemetryStore,
 	prometheus prometheus.Prometheus,
+	promV2 *clickhouseprometheusv2.Provider,
 	cache cache.Cache,
 	flagger flagger.Flagger,
 ) (querier.Querier, error) {
@@ -180,6 +183,7 @@ func newProvider(
 		telemetryStore,
 		telemetryMetadataStore,
 		prometheus,
+		promV2,
 		traceStmtBuilder,
 		logStmtBuilder,
 		auditStmtBuilder,

pkg/query-service/app/clickhouseReader/reader.go

@@ -232,25 +232,6 @@ func NewReader(
 	}
 }
 
-func (r *ClickHouseReader) GetInstantQueryMetricsResult(ctx context.Context, queryParams *model.InstantQueryMetricsParams) (*promql.Result, *stats.QueryStats, *model.ApiError) {
-	qry, err := r.prometheus.Engine().NewInstantQuery(ctx, r.prometheus.Storage(), nil, queryParams.Query, queryParams.Time)
-	if err != nil {
-		return nil, nil, &model.ApiError{Typ: model.ErrorBadData, Err: err}
-	}
-
-	res := qry.Exec(ctx)
-
-	// Optional stats field in response if parameter "stats" is not empty.
-	var qs stats.QueryStats
-	if queryParams.Stats != "" {
-		qs = stats.NewQueryStats(qry.Stats())
-	}
-
-	qry.Close()
-	return res, &qs, nil
-
-}
-
 func (r *ClickHouseReader) GetQueryRangeResult(ctx context.Context, query *model.QueryRangeParams) (*promql.Result, *stats.QueryStats, *model.ApiError) {
 	qry, err := r.prometheus.Engine().NewRangeQuery(ctx, r.prometheus.Storage(), nil, query.Query, query.Start, query.End, query.Step)
 

pkg/query-service/app/http_handler.go

@@ -11,6 +11,7 @@ import (
 	"github.com/SigNoz/signoz/pkg/errors"
 	"github.com/SigNoz/signoz/pkg/flagger"
 	"github.com/SigNoz/signoz/pkg/modules/thirdpartyapi"
+	"github.com/SigNoz/signoz/pkg/querybuilder"
 	"github.com/SigNoz/signoz/pkg/queryparser"
 
 	"log/slog"
@@ -25,11 +26,10 @@ import (
 	"text/template"
 	"time"
 
-	"github.com/prometheus/prometheus/promql"
-
 	"github.com/SigNoz/signoz/pkg/http/middleware"
 	"github.com/SigNoz/signoz/pkg/http/render"
 	"github.com/SigNoz/signoz/pkg/licensing"
+	"github.com/SigNoz/signoz/pkg/prometheus/promapi"
 	"github.com/SigNoz/signoz/pkg/query-service/app/integrations"
 	"github.com/SigNoz/signoz/pkg/signoz"
 	"github.com/SigNoz/signoz/pkg/types/retentiontypes"
@@ -482,8 +482,12 @@ func (aH *APIHandler) Respond(w http.ResponseWriter, data interface{}) {
 
 // RegisterRoutes registers routes for this handler on the given router
 func (aH *APIHandler) RegisterRoutes(router *mux.Router, am *middleware.AuthZ) {
-	router.HandleFunc("/api/v1/query_range", am.ViewAccess(aH.queryRangeMetrics)).Methods(http.MethodGet)
-	router.HandleFunc("/api/v1/query", am.ViewAccess(aH.queryMetrics)).Methods(http.MethodGet)
+	// PromQL-only endpoints, in Prometheus' own API shape, live under a
+	// /prometheus prefix so they are distinguishable from the SigNoz query
+	// APIs; Prometheus-compatible clients can be pointed at the prefix.
+	promAPI := promapi.NewHandler(aH.logger, aH.Signoz.Prometheus)
+	router.HandleFunc("/prometheus/api/v1/query_range", am.ViewAccess(promAPI.QueryRange)).Methods(http.MethodGet, http.MethodPost)
+	router.HandleFunc("/prometheus/api/v1/query", am.ViewAccess(promAPI.Query)).Methods(http.MethodGet, http.MethodPost)
 	router.HandleFunc("/api/v1/rules", am.ViewAccess(aH.listRules)).Methods(http.MethodGet)
 	router.HandleFunc("/api/v1/rules/{id}", am.ViewAccess(aH.getRule)).Methods(http.MethodGet)
 	router.HandleFunc("/api/v1/rules", am.EditAccess(aH.createRule)).Methods(http.MethodPost)
@@ -1092,6 +1096,8 @@ func (aH *APIHandler) queryDashboardVarsV2(w http.ResponseWriter, r *http.Reques
 		return
 	}
 
+	querybuilder.LogIfStatementIsNotValid(r.Context(), aH.logger, query)
+
 	dashboardVars, err := aH.reader.QueryDashboardVars(r.Context(), query)
 	if err != nil {
 		RespondError(w, &model.ApiError{Typ: model.ErrorBadData, Err: err}, nil)
@@ -1100,115 +1106,6 @@ func (aH *APIHandler) queryDashboardVarsV2(w http.ResponseWriter, r *http.Reques
 	aH.Respond(w, dashboardVars)
 }
 
-func (aH *APIHandler) queryRangeMetrics(w http.ResponseWriter, r *http.Request) {
-
-	query, apiErrorObj := parseQueryRangeRequest(r)
-
-	if apiErrorObj != nil {
-		RespondError(w, apiErrorObj, nil)
-		return
-	}
-
-	// TODO: add structured logging for query and apiError if needed
-
-	ctx := r.Context()
-	if to := r.FormValue("timeout"); to != "" {
-		var cancel context.CancelFunc
-		timeout, err := parseMetricsDuration(to)
-		if aH.HandleError(w, err, http.StatusBadRequest) {
-			return
-		}
-
-		ctx, cancel = context.WithTimeout(ctx, timeout)
-		defer cancel()
-	}
-
-	res, qs, apiError := aH.reader.GetQueryRangeResult(ctx, query)
-
-	if apiError != nil {
-		RespondError(w, apiError, nil)
-		return
-	}
-
-	if res.Err != nil {
-		aH.logger.ErrorContext(r.Context(), "error in query range metrics", errors.Attr(res.Err))
-	}
-
-	if res.Err != nil {
-		switch res.Err.(type) {
-		case promql.ErrQueryCanceled:
-			RespondError(w, &model.ApiError{Typ: model.ErrorCanceled, Err: res.Err}, nil)
-		case promql.ErrQueryTimeout:
-			RespondError(w, &model.ApiError{Typ: model.ErrorTimeout, Err: res.Err}, nil)
-		}
-		RespondError(w, &model.ApiError{Typ: model.ErrorExec, Err: res.Err}, nil)
-		return
-	}
-
-	response_data := &model.QueryData{
-		ResultType: res.Value.Type(),
-		Result:     res.Value,
-		Stats:      qs,
-	}
-
-	aH.Respond(w, response_data)
-
-}
-
-func (aH *APIHandler) queryMetrics(w http.ResponseWriter, r *http.Request) {
-
-	queryParams, apiErrorObj := parseInstantQueryMetricsRequest(r)
-
-	if apiErrorObj != nil {
-		RespondError(w, apiErrorObj, nil)
-		return
-	}
-
-	// TODO: add structured logging for query and apiError if needed
-
-	ctx := r.Context()
-	if to := r.FormValue("timeout"); to != "" {
-		var cancel context.CancelFunc
-		timeout, err := parseMetricsDuration(to)
-		if aH.HandleError(w, err, http.StatusBadRequest) {
-			return
-		}
-
-		ctx, cancel = context.WithTimeout(ctx, timeout)
-		defer cancel()
-	}
-
-	res, qs, apiError := aH.reader.GetInstantQueryMetricsResult(ctx, queryParams)
-
-	if apiError != nil {
-		RespondError(w, apiError, nil)
-		return
-	}
-
-	if res.Err != nil {
-		aH.logger.ErrorContext(r.Context(), "error in query range metrics", errors.Attr(res.Err))
-	}
-
-	if res.Err != nil {
-		switch res.Err.(type) {
-		case promql.ErrQueryCanceled:
-			RespondError(w, &model.ApiError{Typ: model.ErrorCanceled, Err: res.Err}, nil)
-		case promql.ErrQueryTimeout:
-			RespondError(w, &model.ApiError{Typ: model.ErrorTimeout, Err: res.Err}, nil)
-		}
-		RespondError(w, &model.ApiError{Typ: model.ErrorExec, Err: res.Err}, nil)
-	}
-
-	responseData := &model.QueryData{
-		ResultType: res.Value.Type(),
-		Result:     res.Value,
-		Stats:      qs,
-	}
-
-	aH.Respond(w, responseData)
-
-}
-
 func (aH *APIHandler) registerEvent(w http.ResponseWriter, r *http.Request) {
 	request, err := parseRegisterEventRequest(r)
 	if aH.HandleError(w, err, http.StatusBadRequest) {

pkg/query-service/app/parser.go

@@ -27,7 +27,6 @@ import (
 	queues2 "github.com/SigNoz/signoz/pkg/query-service/app/integrations/messagingQueues/queues"
 
 	"github.com/gorilla/mux"
-	promModel "github.com/prometheus/common/model"
 	"go.uber.org/multierr"
 
 	errorsV2 "github.com/SigNoz/signoz/pkg/errors"
@@ -88,95 +87,6 @@ func parseRegisterEventRequest(r *http.Request) (*model.RegisterEventParams, err
 	return postData, nil
 }
 
-func parseMetricsTime(s string) (time.Time, error) {
-	if t, err := strconv.ParseFloat(s, 64); err == nil {
-		s, ns := math.Modf(t)
-		return time.Unix(int64(s), int64(ns*float64(time.Second))), nil
-		// return time.Unix(0, t), nil
-	}
-	if t, err := time.Parse(time.RFC3339Nano, s); err == nil {
-		return t, nil
-	}
-	return time.Time{}, fmt.Errorf("cannot parse %q to a valid timestamp", s)
-}
-
-func parseMetricsDuration(s string) (time.Duration, error) {
-	if d, err := strconv.ParseFloat(s, 64); err == nil {
-		ts := d * float64(time.Second)
-		if ts > float64(math.MaxInt64) || ts < float64(math.MinInt64) {
-			return 0, fmt.Errorf("cannot parse %q to a valid duration. It overflows int64", s)
-		}
-		return time.Duration(ts), nil
-	}
-	if d, err := promModel.ParseDuration(s); err == nil {
-		return time.Duration(d), nil
-	}
-	return 0, fmt.Errorf("cannot parse %q to a valid duration", s)
-}
-
-func parseInstantQueryMetricsRequest(r *http.Request) (*model.InstantQueryMetricsParams, *model.ApiError) {
-	var ts time.Time
-	if t := r.FormValue("time"); t != "" {
-		var err error
-		ts, err = parseMetricsTime(t)
-		if err != nil {
-			return nil, &model.ApiError{Typ: model.ErrorBadData, Err: err}
-		}
-	} else {
-		ts = time.Now()
-	}
-
-	return &model.InstantQueryMetricsParams{
-		Time:  ts,
-		Query: r.FormValue("query"),
-		Stats: r.FormValue("stats"),
-	}, nil
-
-}
-
-func parseQueryRangeRequest(r *http.Request) (*model.QueryRangeParams, *model.ApiError) {
-
-	start, err := parseMetricsTime(r.FormValue("start"))
-	if err != nil {
-		return nil, &model.ApiError{Typ: model.ErrorBadData, Err: err}
-	}
-	end, err := parseMetricsTime(r.FormValue("end"))
-	if err != nil {
-		return nil, &model.ApiError{Typ: model.ErrorBadData, Err: err}
-	}
-	if end.Before(start) {
-		err := errors.New("end timestamp must not be before start time")
-		return nil, &model.ApiError{Typ: model.ErrorBadData, Err: err}
-	}
-
-	step, err := parseMetricsDuration(r.FormValue("step"))
-	if err != nil {
-		return nil, &model.ApiError{Typ: model.ErrorBadData, Err: err}
-	}
-
-	if step <= 0 {
-		err := errors.New("zero or negative query resolution step widths are not accepted. Try a positive integer")
-		return nil, &model.ApiError{Typ: model.ErrorBadData, Err: err}
-	}
-
-	// For safety, limit the number of returned points per timeseries.
-	// This is sufficient for 60s resolution for a week or 1h resolution for a year.
-	if end.Sub(start)/step > 11000 {
-		err := errors.New("exceeded maximum resolution of 11,000 points per timeseries. Try decreasing the query resolution (?step=XX)")
-		return nil, &model.ApiError{Typ: model.ErrorBadData, Err: err}
-	}
-
-	queryRangeParams := model.QueryRangeParams{
-		Start: start,
-		End:   end,
-		Step:  step,
-		Query: r.FormValue("query"),
-		Stats: r.FormValue("stats"),
-	}
-
-	return &queryRangeParams, nil
-}
-
 func parseGetUsageRequest(r *http.Request) (*model.GetUsageParams, error) {
 	startTime, err := parseTime("start", r)
 	if err != nil {

pkg/query-service/interfaces/interface.go

@@ -14,7 +14,6 @@ import (
 )
 
 type Reader interface {
-	GetInstantQueryMetricsResult(ctx context.Context, query *model.InstantQueryMetricsParams) (*promql.Result, *stats.QueryStats, *model.ApiError)
 	GetQueryRangeResult(ctx context.Context, query *model.QueryRangeParams) (*promql.Result, *stats.QueryStats, *model.ApiError)
 	GetTopLevelOperations(ctx context.Context, start, end time.Time, services []string) (*map[string][]string, *model.ApiError)
 	GetEntryPointOperations(ctx context.Context, query *model.GetTopOperationsParams) (*[]model.TopOperationsItem, error)

pkg/query-service/model/queryParams.go

@@ -4,12 +4,6 @@ import (
 	"time"
 )
 
-type InstantQueryMetricsParams struct {
-	Time  time.Time
-	Query string
-	Stats string
-}
-
 type QueryRangeParams struct {
 	Start time.Time
 	End   time.Time

pkg/query-service/rules/managertestfactory.go

@@ -105,7 +105,7 @@ func NewTestManager(t *testing.T, testOpts *TestManagerOptions) *Manager {
 	}
 
 	// Create querier with test values
-	providerFactory := signozquerier.NewFactory(telemetryStore, prometheus, cache, flagger)
+	providerFactory := signozquerier.NewFactory(telemetryStore, prometheus, nil, cache, flagger)
 	mockQuerier, err := providerFactory.New(context.Background(), providerSettings, querier.Config{})
 	require.NoError(t, err)
 

pkg/query-service/rules/setups_test.go

@@ -47,6 +47,7 @@ func prepareQuerierForMetrics(t *testing.T, telemetryStore telemetrystore.Teleme
 		telemetryStore,
 		metadataStore,
 		nil, // prometheus
+		nil, // promV2
 		nil, // traceStmtBuilder
 		nil, // logStmtBuilder
 		nil, // auditStmtBuilder
@@ -100,6 +101,7 @@ func prepareQuerierForLogs(t *testing.T, telemetryStore telemetrystore.Telemetry
 		telemetryStore,
 		metadataStore,
 		nil,            // prometheus
+		nil,            // promV2
 		nil,            // traceStmtBuilder
 		logStmtBuilder, // logStmtBuilder
 		nil,            // auditStmtBuilder
@@ -149,6 +151,7 @@ func prepareQuerierForTraces(t *testing.T, telemetryStore telemetrystore.Telemet
 		telemetryStore,
 		metadataStore,
 		nil,              // prometheus
+		nil,              // promV2
 		traceStmtBuilder, // traceStmtBuilder
 		nil,              // logStmtBuilder
 		nil,              // auditStmtBuilder

pkg/querybuilder/agg_rewrite.go

@@ -96,9 +96,9 @@ func (r *aggExprRewriter) Rewrite(
 	}
 
 	if visitor.isRate {
-		return fmt.Sprintf("%s/%d", sel.SelectItems[0].String(), rateInterval), visitor.chArgs, nil
+		return fmt.Sprintf("%s/%d", chparser.Format(sel.SelectItems[0]), rateInterval), visitor.chArgs, nil
 	}
-	return sel.SelectItems[0].String(), visitor.chArgs, nil
+	return chparser.Format(sel.SelectItems[0]), visitor.chArgs, nil
 }
 
 // RewriteMulti rewrites a slice of expressions.
@@ -207,7 +207,7 @@ func (v *exprVisitor) VisitFunctionExpr(fn *chparser.FunctionExpr) error {
 	// Handle *If functions with predicate + values
 	if aggFunc.FuncCombinator {
 		// Map the predicate (last argument)
-		origPred := args[len(args)-1].String()
+		origPred := chparser.Format(args[len(args)-1])
 		whereClause, err := PrepareWhereClause(
 			origPred,
 			FilterExprVisitorOpts{
@@ -242,7 +242,7 @@ func (v *exprVisitor) VisitFunctionExpr(fn *chparser.FunctionExpr) error {
 
 		// Map each value column argument
 		for i := 0; i < len(args)-1; i++ {
-			origVal := args[i].String()
+			origVal := chparser.Format(args[i])
 			fieldKey := telemetrytypes.GetFieldKeyFromKeyText(origVal)
 			expr, err := v.fieldMapper.ColumnExpressionFor(v.ctx, v.orgID, v.startNs, v.endNs, &fieldKey, dataType, v.fieldKeys)
 			if err != nil {
@@ -259,7 +259,7 @@ func (v *exprVisitor) VisitFunctionExpr(fn *chparser.FunctionExpr) error {
 	} else {
 		// Non-If functions: map every argument as a column/value
 		for i, arg := range args {
-			orig := arg.String()
+			orig := chparser.Format(arg)
 			fieldKey := telemetrytypes.GetFieldKeyFromKeyText(orig)
 			expr, err := v.fieldMapper.ColumnExpressionFor(v.ctx, v.orgID, v.startNs, v.endNs, &fieldKey, dataType, v.fieldKeys)
 			if err != nil {

pkg/querybuilder/clickhouse_sql.go

@@ -0,0 +1,76 @@
+package querybuilder
+
+import (
+	"context"
+	"log/slog"
+	"strings"
+
+	chparser "github.com/AfterShip/clickhouse-sql-parser/parser"
+	"github.com/SigNoz/signoz/pkg/errors"
+)
+
+// internalDatabases hold server metadata, credentials and grants rather than telemetry.
+var internalDatabases = map[string]struct{}{
+	"system":             {},
+	"information_schema": {},
+}
+
+// The parser's grammar has gaps against SQL that ClickHouse itself accepts. See TestErrIfStatementIsNotValid_ShouldPassButFails.
+func ErrIfStatementIsNotValid(query string) (err error) {
+	defer func() {
+		// The parser has a history of panicking on malformed input rather than returning an error.
+		if recovered := recover(); recovered != nil {
+			err = errors.NewInvalidInputf(errors.CodeInvalidInput, "invalid ClickHouse SQL (recovered): %v", recovered)
+		}
+	}()
+
+	stmts, parseErr := chparser.NewParser(query).ParseStmts()
+	if parseErr != nil {
+		// Wrapped rather than formatted in, so that callers can recover the parser's *ParseError and read the position off it.
+		return errors.WrapInvalidInputf(parseErr, errors.CodeInvalidInput, "invalid ClickHouse SQL: %s", parseErr.Error())
+	}
+
+	if len(stmts) != 1 {
+		return errors.NewInvalidInputf(errors.CodeInvalidInput, "ClickHouse SQL must contain exactly one statement, found %d statements", len(stmts))
+	}
+
+	selectQuery, ok := stmts[0].(*chparser.SelectQuery)
+	if !ok {
+		return errors.NewInvalidInputf(errors.CodeInvalidInput, "only SELECT statements are allowed in ClickHouse SQL queries")
+	}
+
+	visitor := &chparser.DefaultASTVisitor{Visit: func(node chparser.Expr) error {
+		switch expr := node.(type) {
+		case *chparser.TableFunctionExpr:
+			// Source table functions remain usable in ClickHouse read-only mode.
+			return errors.NewInvalidInputf(errors.CodeInvalidInput, "ClickHouse table functions are not allowed in SQL queries: %s", chparser.Format(expr.Name))
+
+		case *chparser.TableIdentifier:
+			// Reading these is unaffected by ClickHouse read-only mode.
+			if expr.Database == nil {
+				return nil
+			}
+
+			if _, ok := internalDatabases[strings.ToLower(expr.Database.Name)]; ok {
+				return errors.NewInvalidInputf(errors.CodeInvalidInput, "the ClickHouse %s database is not allowed in SQL queries", expr.Database.Name)
+			}
+
+		case *chparser.SettingExpr:
+			// A query-level setting takes precedence over the context setting.
+			if strings.EqualFold(expr.Name.Name, "readonly") {
+				return errors.NewInvalidInputf(errors.CodeInvalidInput, "the ClickHouse readonly setting cannot be overridden")
+			}
+		}
+
+		return nil
+	}}
+
+	return selectQuery.Accept(visitor)
+}
+
+// TODO(@therealpandey): remove this and move to ErrIfStatementIsNotValid.
+func LogIfStatementIsNotValid(ctx context.Context, logger *slog.Logger, query string) {
+	if err := ErrIfStatementIsNotValid(query); err != nil {
+		logger.WarnContext(ctx, "clickhouse sql is not valid", errors.Attr(err), slog.String("query", query))
+	}
+}

pkg/querybuilder/clickhouse_sql_test.go

@@ -0,0 +1,154 @@
+package querybuilder
+
+import (
+	"testing"
+
+	chparser "github.com/AfterShip/clickhouse-sql-parser/parser"
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+)
+
+func TestErrIfStatementIsNotValid_Pass(t *testing.T) {
+	testCases := []struct {
+		name  string
+		query string
+	}{
+		// Shapes a telemetry read is allowed to take.
+		{"Select", "SELECT region AS r, zone FROM metrics WHERE metric_name = 'cpu' GROUP BY region, zone"},
+		{"TrailingSemicolon", "SELECT count() FROM signoz_logs.distributed_logs_v2;"},
+		{"CommonTableExpression", "WITH t AS (SELECT fingerprint FROM signoz_metrics.time_series_v4) SELECT * FROM t"},
+		{"Join", "SELECT * FROM t1 LEFT JOIN t2 ON t1.a = t2.b"},
+		{"GlobalIn", "SELECT a FROM t WHERE a GLOBAL IN (SELECT b FROM t2)"},
+		{"Union", "SELECT * FROM t UNION ALL SELECT * FROM t2"},
+		{"Intersect", "SELECT * FROM t INTERSECT SELECT * FROM t2"},
+		{"WindowFunction", "SELECT sum(v) OVER (PARTITION BY a ORDER BY t) FROM t"},
+		{"UnrelatedSetting", "SELECT * FROM t SETTINGS max_threads = 4"},
+		{"TerminatedBlockComment", "SELECT /* keep me */ count() FROM t"},
+		{"BlockCommentMarkerInsideStringLiteral", "SELECT count() FROM t WHERE body = '/* not a comment'"},
+		// The parser used to loop forever on this; it now reads the comment to the end of
+		// the input, so this doubles as a canary for that regression.
+		{"TrailingUnterminatedBlockComment", "SELECT count() FROM t /* unterminated"},
+		// The rule keys on the database, not on the table name.
+		{"TableNamedSystemInTelemetryDatabase", "SELECT * FROM signoz_logs.system"},
+		{"SignedLiteralAfterClosingParen", "SELECT (toUnixTimestamp(now()) - 3600)*1000000000"},
+		// order by interval
+		{"OrderByInterval", "SELECT toStartOfInterval(timestamp, INTERVAL 1 MINUTE) AS interval ORDER BY interval"},
+		{"OrderByIntervalAndDirection", "SELECT toStartOfInterval(timestamp, INTERVAL 1 MINUTE) AS `interval` ORDER BY `interval` ASC"},
+		{"TrimFunction", "SELECT trimBoth('/api/endpoint/', '/');"},
+	}
+
+	for _, testCase := range testCases {
+		t.Run(testCase.name, func(t *testing.T) {
+			err := ErrIfStatementIsNotValid(testCase.query)
+			assert.NoError(t, err)
+		})
+	}
+}
+
+func TestErrIfStatementIsNotValid_Fail(t *testing.T) {
+	testCases := []struct {
+		name  string
+		query string
+	}{
+		// Not a single statement, or not a statement at all.
+		{"Empty", ""},
+		{"UnterminatedBlockCommentOnly", "/* x"},
+		{"Unparseable", "SELECT FROM WHERE"},
+		{"MultipleStatements", "SELECT 1; DROP TABLE signoz_logs.logs_v2"},
+		// Parses, but is not a SELECT.
+		{"Drop", "DROP TABLE signoz_logs.logs_v2"},
+		{"Insert", "INSERT INTO signoz_logs.logs_v2 SELECT * FROM signoz_logs.logs_v2"},
+		{"AlterDelete", "ALTER TABLE signoz_logs.logs_v2 DELETE WHERE 1 = 1"},
+		{"CreateTable", "CREATE TABLE evil (a Int) ENGINE = Memory"},
+		{"Grant", "GRANT ALL ON *.* TO admin"},
+		{"Set", "SET readonly = 0"},
+		// These the parser rejects outright rather than classifying.
+		{"ShowGrants", "SHOW GRANTS"},
+		{"IntoOutfile", "SELECT * FROM t INTO OUTFILE '/tmp/x.csv'"},
+		// Table functions, which read through something other than a telemetry table.
+		{"UrlTableFunction", "SELECT * FROM url('http://attacker.example/x', CSV, 'a String')"},
+		{"FileTableFunction", "SELECT * FROM file('/etc/passwd', CSV, 'a String')"},
+		{"ExecutableTableFunction", "SELECT * FROM executable('script.sh', CSV, 'a String')"},
+		{"TableFunctionInJoin", "SELECT * FROM t1 JOIN url('http://x', CSV, 'a String') u ON 1 = 1"},
+		{"TableFunctionInCommonTableExpression", "WITH c AS (SELECT * FROM url('http://x', CSV, 'a String')) SELECT * FROM c"},
+		{"TableFunctionInWhereSubquery", "SELECT * FROM t WHERE a IN (SELECT * FROM file('/etc/passwd', CSV, 'a String'))"},
+		{"TableFunctionInUnion", "SELECT * FROM t UNION ALL SELECT * FROM url('http://x', CSV, 'a String')"},
+		// Internal databases, which hold grants and server metadata rather than telemetry.
+		{"SystemUsers", "SELECT * FROM system.users"},
+		{"SystemUppercase", "SELECT * FROM SYSTEM.USERS"},
+		{"SystemQuoted", "SELECT count() FROM `system`.`tables`"},
+		{"SystemInSubquery", "SELECT * FROM (SELECT name FROM system.parts)"},
+		{"SystemInJoin", "SELECT * FROM signoz_logs.distributed_logs_v2 AS l JOIN system.users AS u ON 1 = 1"},
+		{"SystemInIntersect", "SELECT * FROM t INTERSECT SELECT * FROM system.users"},
+		{"InformationSchema", "SELECT * FROM information_schema.tables"},
+		// A query-level setting takes precedence over the one the caller applies.
+		{"ReadonlySettingOverride", "SELECT * FROM t SETTINGS readonly = 0"},
+		{"ReadonlySettingOverrideAmongOthers", "SELECT * FROM t SETTINGS max_threads = 4, readonly = 0"},
+	}
+
+	for _, testCase := range testCases {
+		t.Run(testCase.name, func(t *testing.T) {
+			err := ErrIfStatementIsNotValid(testCase.query)
+			assert.Error(t, err)
+		})
+	}
+}
+
+// Queries the parser cannot read. ClickHouse runs all of them.
+func TestErrIfStatementIsNotValid_ShouldPassButFails(t *testing.T) {
+	testCases := []struct {
+		name  string
+		query string
+		// The construct the parser stops after, which is the one it cannot read.
+		expectedStopsAfter string
+		// The same construct written so the parser accepts it.
+		fix string
+	}{
+		{
+			name:               "IntervalAliasInOrderBy",
+			query:              "SELECT toStartOfInterval(timestamp, INTERVAL 1 MINUTE) AS interval, resource_string_service$$name AS `service.name`, attributes_string['http.route'] AS `http.route`, quantile(0.95)(duration_nano) / 1000000000 AS value FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_string_service$$name = 'svc-a' AND resources_string['deployment.environment'] = 'dev' AND attributes_string['http.route'] = '/v1' AND http_method = 'POST' AND timestamp BETWEEN toDateTime(1784601720) AND toDateTime(1784602620) AND ts_bucket_start BETWEEN 1784601720 - 1800 AND 1784602620 GROUP BY `service.name`, `http.route`, interval ORDER BY interval ASC",
+			expectedStopsAfter: "ORDER BY interval ASC",
+			fix:                "SELECT count() AS interval FROM t ORDER BY `interval` ASC",
+		},
+		{
+			name:               "IntervalAliasInOrderByDesc",
+			query:              "SELECT count() AS value, toStartOfInterval(timestamp, INTERVAL 1 MINUTE) AS interval, serviceName, resourceTagsMap['deployment.environment'] AS environment, exceptionStacktrace FROM signoz_traces.distributed_signoz_error_index_v2 WHERE exceptionType != 'OSError' AND resourceTagsMap['deployment.environment'] = 'staging' AND timestamp BETWEEN toDateTime(1785186300) AND toDateTime(1785186600) GROUP BY serviceName, interval, environment, exceptionStacktrace ORDER BY interval DESC",
+			expectedStopsAfter: "ORDER BY interval DESC",
+			fix:                "SELECT count() AS interval FROM t ORDER BY `interval` DESC",
+		},
+		{
+			name:               "StandardTrimSyntax",
+			query:              "SELECT toStartOfInterval(fromUnixTimestamp64Nano(timestamp), INTERVAL 5 MINUTE) AS interval, resources_string['host.name'] as host_name, toFloat64(countIf( lower(trim(BOTH ' ' FROM replaceOne( JSONExtractString(body, 'Action'), 'health_status: ', '' ))) IN ('unhealthy','starting','failing') )) as value FROM signoz_logs.distributed_logs_v2 WHERE timestamp BETWEEN 1784602320000000000 AND 1784602620000000000 AND ts_bucket_start BETWEEN 1784602320 - 300 AND 1784602620 AND JSONExtractString(body, 'Type') = 'container' AND JSONExtractString(body, 'Actor', 'Attributes', 'name') IS NOT NULL AND resources_string['host.name'] IS NOT NULL AND resources_string['host.name'] = 'aihub-nightly' GROUP BY interval, host_name ORDER BY interval, host_name",
+			expectedStopsAfter: "trim(BOTH '",
+			fix:                "SELECT trimBoth(replaceOne( JSONExtractString(body, 'Action'), 'health_status: ', '' ), ' ')",
+		},
+		{
+			name:               "SignedLiteralAfterClosingParen",
+			query:              "SELECT now() AS ts, toFloat64(count()) AS value FROM ( SELECT attributes_string['TableName'] AS T, attributes_string['MissingId'] AS M, max(fromUnixTimestamp64Nano(timestamp)) AS last_seen, dateDiff('minute', min(fromUnixTimestamp64Nano(timestamp)), max(fromUnixTimestamp64Nano(timestamp))) AS age_min FROM signoz_logs.distributed_logs_v2 WHERE body='missing_map_record' AND timestamp >= (toUnixTimestamp(now())-3600)*1000000000 GROUP BY T, M ) WHERE age_min >= 20 AND last_seen >= now() - toIntervalMinute(8)",
+			expectedStopsAfter: "(toUnixTimestamp(now())",
+			fix:                "SELECT (toUnixTimestamp(now()) - 3600) * 1000000000",
+		},
+		{
+			name:               "SignedLiteralAfterClosingParenMinimal",
+			query:              "SELECT (1)-1",
+			expectedStopsAfter: "(1)",
+			fix:                "SELECT (1) - 1",
+		},
+	}
+
+	for _, testCase := range testCases {
+		t.Run(testCase.name, func(t *testing.T) {
+			err := ErrIfStatementIsNotValid(testCase.query)
+
+			var parseErr *chparser.ParseError
+			require.ErrorAs(t, err, &parseErr, "expected a parser failure rather than a rule violation")
+
+			// The parser reports the offset it stopped at, which sits just past the construct
+			// it choked on, so the text leading up to it is what needs looking at.
+			consumed := testCase.query[:parseErr.Pos]
+			assert.Equal(t, testCase.expectedStopsAfter, consumed[max(0, len(consumed)-len(testCase.expectedStopsAfter)):])
+
+			assert.NoError(t, ErrIfStatementIsNotValid(testCase.fix))
+		})
+	}
+}

pkg/querybuilder/init.go

@@ -1 +0,0 @@
-package querybuilder

pkg/queryparser/queryfilterextractor/clickhouse.go

@@ -316,17 +316,17 @@ func (e *ClickHouseFilterExtractor) extractColumnStrByExpr(expr clickhouse.Expr)
 	// FunctionExpr is a function call like "toDate(timestamp)"
 	case *clickhouse.FunctionExpr:
 		// For function expressions, return the complete function call string
-		return ex.String()
+		return clickhouse.Format(ex)
 	// ColumnExpr is a column expression like "m.region", "toDate(timestamp)"
 	case *clickhouse.ColumnExpr:
 		// ColumnExpr wraps another expression - extract the underlying expression
 		if ex.Expr != nil {
 			return e.extractColumnStrByExpr(ex.Expr)
 		}
-		return ex.String()
+		return clickhouse.Format(ex)
 	default:
 		// For other expression types, return the string representation
-		return expr.String()
+		return clickhouse.Format(expr)
 	}
 }
 
@@ -524,7 +524,7 @@ func (e *ClickHouseFilterExtractor) extractFullExpression(expr clickhouse.Expr)
 	if expr == nil {
 		return ""
 	}
-	return expr.String()
+	return clickhouse.Format(expr)
 }
 
 // isSimpleColumnReference checks if an expression is just a simple column reference
@@ -657,7 +657,7 @@ func (e *ClickHouseFilterExtractor) extractCTEName(cte *clickhouse.CTEStmt) stri
 	case *clickhouse.Ident:
 		return name.Name
 	default:
-		return cte.Expr.String()
+		return clickhouse.Format(cte.Expr)
 	}
 }
 

pkg/signoz/provider.go

@@ -45,6 +45,7 @@ import (
 	"github.com/SigNoz/signoz/pkg/pprof/nooppprof"
 	"github.com/SigNoz/signoz/pkg/prometheus"
 	"github.com/SigNoz/signoz/pkg/prometheus/clickhouseprometheus"
+	"github.com/SigNoz/signoz/pkg/prometheus/clickhouseprometheusv2"
 	"github.com/SigNoz/signoz/pkg/querier"
 	"github.com/SigNoz/signoz/pkg/querier/signozquerier"
 	"github.com/SigNoz/signoz/pkg/sharder"
@@ -244,6 +245,7 @@ func NewTelemetryStoreProviderFactories() factory.NamedMap[factory.ProviderFacto
 func NewPrometheusProviderFactories(telemetryStore telemetrystore.TelemetryStore) factory.NamedMap[factory.ProviderFactory[prometheus.Prometheus, prometheus.Config]] {
 	return factory.MustNewNamedMap(
 		clickhouseprometheus.NewFactory(telemetryStore),
+		clickhouseprometheusv2.NewFactory(telemetryStore),
 	)
 }
 
@@ -285,9 +287,9 @@ func NewStatsReporterProviderFactories(aggregator statsreporter.Aggregator, orgG
 	)
 }
 
-func NewQuerierProviderFactories(telemetryStore telemetrystore.TelemetryStore, prometheus prometheus.Prometheus, cache cache.Cache, flagger flagger.Flagger) factory.NamedMap[factory.ProviderFactory[querier.Querier, querier.Config]] {
+func NewQuerierProviderFactories(telemetryStore telemetrystore.TelemetryStore, prometheus prometheus.Prometheus, promV2 *clickhouseprometheusv2.Provider, cache cache.Cache, flagger flagger.Flagger) factory.NamedMap[factory.ProviderFactory[querier.Querier, querier.Config]] {
 	return factory.MustNewNamedMap(
-		signozquerier.NewFactory(telemetryStore, prometheus, cache, flagger),
+		signozquerier.NewFactory(telemetryStore, prometheus, promV2, cache, flagger),
 	)
 }
 

pkg/signoz/signoz.go

@@ -40,6 +40,7 @@ import (
 	"github.com/SigNoz/signoz/pkg/modules/tag/impltag"
 	"github.com/SigNoz/signoz/pkg/modules/user/impluser"
 	"github.com/SigNoz/signoz/pkg/prometheus"
+	"github.com/SigNoz/signoz/pkg/prometheus/clickhouseprometheusv2"
 	"github.com/SigNoz/signoz/pkg/querier"
 	"github.com/SigNoz/signoz/pkg/queryparser"
 	"github.com/SigNoz/signoz/pkg/ruler"
@@ -242,6 +243,11 @@ func New(
 
 	retentionGetter := implretention.NewGetter(implretention.NewStore(sqlstore))
 
+	// promV2 is the clickhousev2 provider handed to the querier for shadow
+	// comparison and pinned serving (declared before the serving provider,
+	// whose variable shadows the package name below).
+	var promV2 *clickhouseprometheusv2.Provider
+
 	// Initialize prometheus from the available prometheus provider factories
 	prometheus, err := factory.NewProviderFromNamedMap(
 		ctx,
@@ -254,12 +260,29 @@ func New(
 		return nil, err
 	}
 
+	// With the default provider, also stand up the clickhousev2 provider for
+	// the querier: PromQL queries shadow-compare against it behind the
+	// use_prometheus_clickhouse_v2 flag (see pkg/querier/promql_shadow.go).
+	// It never serves by default. An explicit
+	// prometheus::provider: clickhousev2 makes v2 the serving provider
+	// outright, so there is nothing to compare against.
+	if config.Prometheus.Provider() == "clickhouse" {
+		v2Config := config.Prometheus
+		// The v2 engine only evaluates shadow and pinned queries; disable its
+		// active query tracker so two trackers never share a file.
+		v2Config.ActiveQueryTrackerConfig.Enabled = false
+		promV2, err = clickhouseprometheusv2.New(ctx, providerSettings, v2Config, telemetrystore)
+		if err != nil {
+			return nil, err
+		}
+	}
+
 	// Initialize querier from the available querier provider factories
 	querier, err := factory.NewProviderFromNamedMap(
 		ctx,
 		providerSettings,
 		config.Querier,
-		NewQuerierProviderFactories(telemetrystore, prometheus, cache, flagger),
+		NewQuerierProviderFactories(telemetrystore, prometheus, promV2, cache, flagger),
 		config.Querier.Provider(),
 	)
 	if err != nil {

pkg/telemetrymetadata/metadata.go

@@ -2359,10 +2359,7 @@ func (t *telemetryMetaStore) fetchTemporalityTypeForTable(ctx context.Context, t
 		if temporality != metrictypes.Unknown {
 			temporalities[metricName] = append(temporalities[metricName], temporality)
 		}
-		// Monotonicity carries meaning only for cumulative sums: a non-monotonic
-		// cumulative sum is a gauge for all practical purposes. Delta sums are
-		// counters regardless of monotonicity.
-		if metricType == metrictypes.SumType && !isMonotonic && temporality == metrictypes.Cumulative {
+		if metricType == metrictypes.SumType && !isMonotonic {
 			metricType = metrictypes.GaugeType
 		}
 		types[metricName] = metricType
@@ -2415,9 +2412,7 @@ func (t *telemetryMetaStore) fetchMeterSourceMetricsTemporalityAndType(ctx conte
 		if err := rows.Scan(&metricName, &temporality, &metricType, &isMonotonic); err != nil {
 			return nil, nil, errors.Wrapf(err, errors.TypeInternal, errors.CodeInternal, "failed to scan temporality result")
 		}
-		// Non-monotonic cumulative sums are treated as gauges; delta sums are
-		// counters regardless of monotonicity.
-		if metricType == metrictypes.SumType && !isMonotonic && temporality == metrictypes.Cumulative {
+		if metricType == metrictypes.SumType && !isMonotonic {
 			metricType = metrictypes.GaugeType
 		}
 		temporalities[metricName] = temporality

pkg/telemetrymetadata/metadata_query_test.go

@@ -16,9 +16,7 @@ import (
 	"github.com/SigNoz/signoz/pkg/telemetrystore"
 	"github.com/SigNoz/signoz/pkg/telemetrystore/telemetrystoretest"
 	"github.com/SigNoz/signoz/pkg/telemetrytraces"
-	"github.com/SigNoz/signoz/pkg/types/metrictypes"
 	"github.com/SigNoz/signoz/pkg/types/telemetrytypes"
-	"github.com/SigNoz/signoz/pkg/valuer"
 	"github.com/stretchr/testify/assert"
 	"github.com/stretchr/testify/require"
 )
@@ -111,85 +109,3 @@ func TestGetFirstSeenFromMetricMetadata(t *testing.T) {
 		t.Errorf("there were unfulfilled expectations: %s", err)
 	}
 }
-
-// Sum metrics are classified as counters unless they are non-monotonic AND
-// cumulative; monotonicity carries no meaning for delta sums, so a
-// non-monotonic delta sum must remain a Sum (counter).
-func TestFetchTemporalityAndTypeMultiSumClassification(t *testing.T) {
-	mockTelemetryStore := telemetrystoretest.New(telemetrystore.Config{}, &regexMatcher{})
-	mock := mockTelemetryStore.Mock()
-
-	metadata := NewTelemetryMetaStore(
-		instrumentationtest.New().ToProviderSettings(),
-		mockTelemetryStore,
-		telemetrytraces.DBName,
-		telemetrytraces.TagAttributesV2TableName,
-		telemetrytraces.SpanAttributesKeysTblName,
-		telemetrytraces.SpanIndexV3TableName,
-		telemetrymetrics.DBName,
-		telemetrymetrics.AttributesMetadataTableName,
-		telemetrymeter.DBName,
-		telemetrymeter.SamplesAgg1dTableName,
-		telemetrylogs.DBName,
-		telemetrylogs.LogsV2TableName,
-		telemetrylogs.TagAttributesV2TableName,
-		telemetrylogs.LogAttributeKeysTblName,
-		telemetrylogs.LogResourceKeysTblName,
-		telemetryaudit.DBName,
-		telemetryaudit.AuditLogsTableName,
-		telemetryaudit.TagAttributesTableName,
-		telemetryaudit.LogAttributeKeysTblName,
-		telemetryaudit.LogResourceKeysTblName,
-		DBName,
-		AttributesMetadataTableName,
-		ColumnEvolutionMetadataTableName,
-		flaggertest.New(t),
-	)
-
-	metricNames := []string{
-		"delta.nonmono.sum",
-		"delta.mono.sum",
-		"cumulative.nonmono.sum",
-		"cumulative.mono.sum",
-	}
-
-	metadataCols := []cmock.ColumnType{
-		{Name: "metric_name", Type: "String"},
-		{Name: "temporality", Type: "String"},
-		{Name: "type", Type: "String"},
-		{Name: "is_monotonic", Type: "Bool"},
-	}
-
-	mock.ExpectQuery(`SELECT metric_name, temporality, any\(type\) AS type, argMax\(is_monotonic, unix_milli\) as is_monotonic FROM signoz_metrics\.`).
-		WithArgs(nil, nil, nil).
-		WillReturnRows(cmock.NewRows(metadataCols, [][]any{
-			{"delta.nonmono.sum", metrictypes.Delta, metrictypes.SumType, false},
-			{"delta.mono.sum", metrictypes.Delta, metrictypes.SumType, true},
-			{"cumulative.nonmono.sum", metrictypes.Cumulative, metrictypes.SumType, false},
-			{"cumulative.mono.sum", metrictypes.Cumulative, metrictypes.SumType, true},
-		}))
-	mock.ExpectQuery(`SELECT metric_name, argMax\(temporality, unix_milli\) as temporality, any\(type\) AS type, argMax\(is_monotonic, unix_milli\) as is_monotonic FROM signoz_meter\.`).
-		WithArgs(nil).
-		WillReturnRows(cmock.NewRows(metadataCols, [][]any{}))
-
-	temporalities, types, _, err := metadata.FetchTemporalityAndTypeMulti(
-		context.Background(),
-		valuer.GenerateUUID(),
-		1700000000000,
-		1700003600000,
-		metricNames...,
-	)
-	require.NoError(t, err)
-
-	assert.Equal(t, metrictypes.SumType, types["delta.nonmono.sum"])
-	assert.Equal(t, metrictypes.SumType, types["delta.mono.sum"])
-	assert.Equal(t, metrictypes.GaugeType, types["cumulative.nonmono.sum"])
-	assert.Equal(t, metrictypes.SumType, types["cumulative.mono.sum"])
-
-	assert.Equal(t, metrictypes.Delta, temporalities["delta.nonmono.sum"])
-	assert.Equal(t, metrictypes.Cumulative, temporalities["cumulative.nonmono.sum"])
-
-	if err := mock.ExpectationsWereMet(); err != nil {
-		t.Errorf("there were unfulfilled expectations: %s", err)
-	}
-}

pkg/telemetrymetadata/stmt_parse.go

@@ -27,7 +27,7 @@ func (v *TelemetryFieldVisitor) VisitColumnDef(expr *parser.ColumnDef) error {
 	}
 
 	// Parse column name to extract context and data type
-	columnName := expr.Name.String()
+	columnName := parser.Format(expr.Name)
 
 	// Remove backticks if present
 	columnName = strings.TrimPrefix(columnName, "`")
@@ -75,7 +75,7 @@ func (v *TelemetryFieldVisitor) VisitColumnDef(expr *parser.ColumnDef) error {
 	// Extract field name from the DEFAULT expression
 	// The DEFAULT expression should be something like: resources_string['k8s.cluster.name']
 	// We need to extract the key inside the square brackets
-	defaultExprStr := expr.DefaultExpr.String()
+	defaultExprStr := parser.Format(expr.DefaultExpr)
 
 	// Look for the pattern: map['key']
 	startIdx := strings.Index(defaultExprStr, "['")

pkg/types/querybuildertypes/querybuildertypesv5/req.go

@@ -370,6 +370,14 @@ type QueryRangeRequest struct {
 	// NoCache is a flag to disable caching for the request.
 	NoCache bool `json:"noCache,omitempty"`
 
+	// PromQLProvider serves this request's PromQL queries via the named
+	// prometheus provider ("clickhousev2") instead of the default — the same
+	// data read through a different implementation. It is set from the
+	// X-SigNoz-PromQL-Provider header by the API handler, never from the
+	// body: a rollout-scoped comparison hook for integration tests and
+	// support should not become part of the public request schema.
+	PromQLProvider string `json:"-"`
+
 	FormatOptions *FormatOptions `json:"formatOptions,omitempty"`
 }
 

pkg/variables/clickhouse/processor.go

@@ -69,7 +69,7 @@ func (qp *QueryProcessor) ProcessQuery(query string, transformer FilterTransform
 	// Reconstruct the query
 	var resultBuilder strings.Builder
 	for _, stmt := range stmts {
-		resultBuilder.WriteString(stmt.String())
+		resultBuilder.WriteString(parser.Format(stmt))
 		resultBuilder.WriteString(";")
 	}
 

scripts/promqltestcorpus/.gitignore

@@ -0,0 +1,2 @@
+# go build artifact (go run . is the supported entry)
+/promqltestcorpus

scripts/promqltestcorpus/corpus.go

@@ -0,0 +1,660 @@
+// SigNoz corpus policy: which upstream cases are representable through the
+// API, the grid variants that steer coarse-step code paths, the API's value
+// rounding, and the frozen JSON model. Nothing in this file mirrors
+// upstream code; it encodes what our conformance harness needs.
+package main
+
+import (
+	"context"
+	"encoding/json"
+	"fmt"
+	"math"
+	"os"
+	"path/filepath"
+	"strconv"
+	"strings"
+	"time"
+
+	"github.com/prometheus/common/model"
+	"github.com/prometheus/prometheus/model/labels"
+	"github.com/prometheus/prometheus/model/value"
+	"github.com/prometheus/prometheus/promql"
+	"github.com/prometheus/prometheus/promql/parser"
+	"github.com/prometheus/prometheus/tsdb/chunkenc"
+	"github.com/prometheus/prometheus/util/almost"
+	"github.com/prometheus/prometheus/util/teststorage"
+)
+
+// seriesDescParser is the slice of parser.Parser the loader needs.
+type seriesDescParser interface {
+	ParseSeriesDesc(input string) (labels.Labels, []parser.SequenceValue, error)
+}
+
+// Calendar and wall-clock functions are not invariant under time
+// translation, and the Python suite shifts every case to recent timestamps
+// (epoch-0 samples would sit 55 years past ClickHouse TTLs). Everything else
+// PromQL computes depends only on time differences.
+var timeDependentFuncs = map[string]bool{
+	"time": true, "timestamp": true, "month": true, "year": true,
+	"minute": true, "hour": true, "day_of_month": true, "day_of_week": true,
+	"day_of_year": true, "days_in_month": true,
+}
+
+const (
+	lookbackMs    = 300_000
+	instantStepMs = 1_000
+	maxSamples    = 50_000_000
+)
+
+type corpusSeries struct {
+	Labels  map[string]string `json:"labels"`
+	Samples [][2]any          `json:"samples"` // [offset_ms, value]
+}
+
+type corpusDataset struct {
+	ID     int            `json:"id"`
+	Source string         `json:"source"`
+	Series []corpusSeries `json:"series"`
+}
+
+type corpusPoint = [2]any // [offset_ms, value]
+
+type corpusResult struct {
+	Labels map[string]string `json:"labels"`
+	Points []corpusPoint     `json:"points"`
+}
+
+type corpusCase struct {
+	Dataset  int            `json:"dataset"`
+	Source   string         `json:"source"`
+	Variant  string         `json:"variant"`
+	Expr     string         `json:"expr"`
+	StartMs  int64          `json:"start_ms"`
+	EndMs    int64          `json:"end_ms"`
+	StepMs   int64          `json:"step_ms"`
+	Instant  bool           `json:"instant"`
+	Expected []corpusResult `json:"expected"`
+}
+
+type corpus struct {
+	Meta struct {
+		PrometheusVersion string `json:"prometheus_version"`
+		LookbackMs        int64  `json:"lookback_ms"`
+		InstantStepMs     int64  `json:"instant_step_ms"`
+		Note              string `json:"note"`
+	} `json:"meta"`
+	Datasets []corpusDataset `json:"datasets"`
+	Cases    []corpusCase    `json:"cases"`
+}
+
+func generate(files []string, engine *promql.Engine, seriesParser parser.Parser, exprParser parser.Parser, promVersion string) (*corpus, map[string]int, error) {
+	var c corpus
+	c.Meta.PrometheusVersion = promVersion
+	c.Meta.LookbackMs = lookbackMs
+	c.Meta.InstantStepMs = instantStepMs
+	c.Meta.Note = "expected values carry the API's 3-significant-decimal rounding (querybuildertypesv5 sanitizeValue); instant evals are encoded as start==end range queries"
+
+	skips := map[string]int{}
+	datasetIDs := map[string]int{}
+
+	for _, file := range files {
+		base := filepath.Base(file)
+		if base == "native_histograms.test" || base == "type_and_unit.test" {
+			// native histograms: the samples pipeline under test stores
+			// floats; type_and_unit: experimental __type__/__unit__ metadata
+			// labels our store does not materialize.
+			skips["file:"+strings.TrimSuffix(base, ".test")]++
+			continue
+		}
+		raw, err := os.ReadFile(file)
+		if err != nil {
+			return nil, nil, err
+		}
+		cmds := parseScript(string(raw))
+
+		segment := 0
+		var loads []command
+		segmentBad := "" // non-empty: reason the segment cannot be represented
+		for _, cmd := range cmds {
+			switch cmd.kind {
+			case "clear":
+				segment++
+				loads = nil
+				segmentBad = ""
+			case "skip":
+				skips["command:"+cmd.head]++
+			case "load":
+				if reason := checkLoad(seriesParser, cmd); reason != "" {
+					segmentBad = reason
+					skips["load:"+reason]++
+					continue
+				}
+				loads = append(loads, cmd)
+			case "eval":
+				if segmentBad != "" {
+					skips["segment:"+segmentBad]++
+					continue
+				}
+				if len(loads) == 0 {
+					skips["eval:no-data"]++
+					continue
+				}
+				ccs, reason, err := buildCases(engine, seriesParser, exprParser, cmd, loads, base, skips)
+				if err != nil {
+					return nil, nil, err
+				}
+				if reason != "" {
+					skips["eval:"+reason]++
+					continue
+				}
+				key := fmt.Sprintf("%s#%d#%d", base, segment, len(loads))
+				id, ok := datasetIDs[key]
+				if !ok {
+					ds, reason, err := dumpDataset(seriesParser, loads)
+					if err != nil {
+						return nil, nil, err
+					}
+					if reason != "" {
+						skips["dataset:"+reason]++
+						continue
+					}
+					id = len(c.Datasets)
+					datasetIDs[key] = id
+					ds.ID = id
+					ds.Source = key
+					c.Datasets = append(c.Datasets, *ds)
+				}
+				for _, cc := range ccs {
+					cc.Dataset = id
+					cc.Source = fmt.Sprintf("%s:%d", base, cmd.line)
+					c.Cases = append(c.Cases, cc)
+				}
+			}
+		}
+	}
+	if len(c.Cases) == 0 {
+		return nil, nil, fmt.Errorf("no corpus cases produced")
+	}
+	return &c, skips, nil
+}
+
+func writeCorpus(out string, c *corpus) error {
+	buf, err := json.MarshalIndent(c, "", " ")
+	if err != nil {
+		return err
+	}
+	return os.WriteFile(out, buf, 0o644)
+}
+
+// checkLoad validates a load block is representable: parsable series
+// notation, float samples only ("load_with_nhcb" and histogram literals are
+// out of scope — the samples pipeline under test stores float samples).
+func checkLoad(p parser.Parser, cmd command) string {
+	fields := strings.Fields(cmd.head)
+	if len(fields) != 2 || fields[0] != "load" {
+		return "unsupported-load-variant"
+	}
+	if _, err := model.ParseDuration(fields[1]); err != nil {
+		return "bad-interval"
+	}
+	for _, line := range cmd.body {
+		metric, vals, err := p.ParseSeriesDesc(line)
+		if err != nil {
+			return "unparsable-series"
+		}
+		if metric.Get(model.MetricNameLabel) == "" {
+			return "unnamed-series"
+		}
+		for _, v := range vals {
+			if v.Histogram != nil {
+				return "histogram-samples"
+			}
+		}
+	}
+	return ""
+}
+
+// durationExprUsesRange reports whether a duration-expression tree contains
+// range(). Instant evals are encoded as one-step range queries (the API
+// rejects start == end), which changes what range() evaluates to, so they
+// cannot carry it; range evals keep it — each variant's oracle is computed
+// on the exact window it requests.
+func durationExprUsesRange(e parser.Expr) bool {
+	d, ok := e.(*parser.DurationExpr)
+	if !ok || d == nil {
+		return false
+	}
+	if d.Op == parser.RANGE {
+		return true
+	}
+	return durationExprUsesRange(d.LHS) || durationExprUsesRange(d.RHS)
+}
+
+// buildCases parses one eval header, filters unservable expressions, and
+// emits the base case plus grid variants — each with expectations computed by
+// the reference engine over the loads. The variants exist because upstream's
+// own grids are fine-stepped: without them the coarse-step code paths (the
+// window-sliver filter, the disjoint over_time form, the lifted instant/last
+// gates) would pass through this corpus untouched. A variant is just another
+// grid over the same data and expression; the engine is the oracle either way.
+func buildCases(engine *promql.Engine, seriesParser parser.Parser, exprParser parser.Parser, cmd command, loads []command, sourceFile string, skips map[string]int) ([]corpusCase, string, error) {
+	for _, line := range cmd.body {
+		if patExpect.MatchString(line) && strings.HasPrefix(line, "expect fail") {
+			return nil, "expect-fail", nil
+		}
+	}
+
+	base := corpusCase{Variant: "base"}
+	if m := patEvalInstant.FindStringSubmatch(cmd.head); m != nil {
+		at, err := parseTestDuration(m[2])
+		if err != nil {
+			return nil, "bad-duration", nil
+		}
+		base.Instant = true
+		base.StartMs, base.EndMs, base.StepMs = at, at, instantStepMs
+		base.Expr = m[3]
+	} else if m := patEvalRange.FindStringSubmatch(cmd.head); m != nil {
+		from, err1 := parseTestDuration(m[2])
+		to, err2 := parseTestDuration(m[3])
+		step, err3 := parseTestDuration(m[4])
+		if err1 != nil || err2 != nil || err3 != nil {
+			return nil, "bad-duration", nil
+		}
+		if step <= 0 || to < from {
+			return nil, "bad-grid", nil
+		}
+		base.StartMs, base.EndMs, base.StepMs = from, to, step
+		base.Expr = m[5]
+	} else {
+		return nil, "unrecognized", nil
+	}
+
+	expr, err := exprParser.ParseExpr(base.Expr)
+	if err != nil {
+		return nil, "needs-experimental-parser", nil
+	}
+	if vt := expr.Type(); vt != parser.ValueTypeVector && vt != parser.ValueTypeScalar {
+		return nil, "non-instant-type", nil
+	}
+	unservable := ""
+	hasSelector := false
+	hasSubquery := false
+	var maxRangeMs int64
+	parser.Inspect(expr, func(node parser.Node, _ []parser.Node) error {
+		switch n := node.(type) {
+		case *parser.Call:
+			if timeDependentFuncs[n.Func.Name] {
+				unservable = "time-dependent"
+			}
+		case *parser.VectorSelector:
+			hasSelector = true
+			if n.Timestamp != nil || n.StartOrEnd != 0 {
+				unservable = "at-modifier"
+			}
+			if n.OriginalOffset < 0 {
+				// The server ships with negative offsets disabled.
+				unservable = "negative-offset"
+			}
+			if base.Instant && durationExprUsesRange(n.OriginalOffsetExpr) {
+				unservable = "range-duration-in-instant"
+			}
+			for _, m := range n.LabelMatchers {
+				if m.Name == "__type__" || m.Name == "__unit__" {
+					unservable = "type-unit-metadata"
+				}
+			}
+		case *parser.MatrixSelector:
+			if r := n.Range.Milliseconds(); r > maxRangeMs {
+				maxRangeMs = r
+			}
+			if base.Instant && durationExprUsesRange(n.RangeExpr) {
+				unservable = "range-duration-in-instant"
+			}
+		case *parser.SubqueryExpr:
+			hasSubquery = true
+			if n.Timestamp != nil || n.StartOrEnd != 0 {
+				unservable = "at-modifier"
+			}
+			if n.OriginalOffset < 0 {
+				unservable = "negative-offset"
+			}
+			if base.Instant && (durationExprUsesRange(n.RangeExpr) || durationExprUsesRange(n.StepExpr) || durationExprUsesRange(n.OriginalOffsetExpr)) {
+				unservable = "range-duration-in-instant"
+			}
+		}
+		return nil
+	})
+	if unservable != "" {
+		return nil, unservable, nil
+	}
+
+	stor, err := loadSeriesStorage(seriesParser, loads)
+	if err != nil {
+		return nil, "", err
+	}
+	defer func() { _ = stor.Close() }()
+
+	expected, reason := computeExpected(engine, stor, base.Expr, base.StartMs, base.EndMs, base.StepMs)
+	if reason != "" {
+		return nil, reason, nil
+	}
+	if err := crossCheckUpstream(engine, seriesParser, stor, cmd, base, sourceFile, skips); err != nil {
+		return nil, "", err
+	}
+	base.Expected = expected
+	out := []corpusCase{base}
+
+	// Grid variants. Subquery expressions keep their own inner grids; varying
+	// the outer grid there multiplies cases without steering the code paths
+	// the variants exist for, so they emit only the base.
+	if hasSubquery || !hasSelector {
+		return out, "", nil
+	}
+	type variant struct {
+		name                   string
+		startMs, endMs, stepMs int64
+	}
+	var variants []variant
+	if base.Instant {
+		// A coarse multi-point grid ending at the instant: step above the
+		// 5m lookback drives the lifted instant gate and the sliver filter.
+		const coarse = 600_000
+		variants = append(variants, variant{"instant-coarse", base.EndMs - 2*coarse, base.EndMs, coarse})
+	} else {
+		span := base.EndMs - base.StartMs
+		if maxRangeMs > 0 {
+			// Step wider than every window in the expression: the sliver
+			// filter and the disjoint over_time form become active.
+			if coarse := 2 * maxRangeMs; span >= coarse {
+				variants = append(variants, variant{"coarse-step", base.StartMs, base.EndMs, coarse})
+			}
+			// Whole-bucket tiling (range == 2 steps) drives the windowed
+			// over_time slide with W = 2.
+			if tiled := maxRangeMs / 2; tiled >= 1000 && maxRangeMs%2000 == 0 && tiled != base.StepMs && span >= tiled {
+				variants = append(variants, variant{"tiled", base.StartMs, base.EndMs, tiled})
+			}
+		}
+		// A start off every natural alignment shifts which samples each
+		// window sees; an end short of the lattice exercises the
+		// last-grid-point handling.
+		if span > 17_000 {
+			variants = append(variants, variant{"unaligned-start", base.StartMs + 17_000, base.EndMs, base.StepMs})
+		}
+		if lastIdx := span / base.StepMs; lastIdx >= 2 {
+			offEnd := base.StartMs + lastIdx*base.StepMs - base.StepMs/3
+			variants = append(variants, variant{"off-lattice-end", base.StartMs, offEnd, base.StepMs})
+		}
+	}
+	for _, v := range variants {
+		if v.stepMs <= 0 || v.endMs <= v.startMs || v.stepMs%1000 != 0 {
+			continue
+		}
+		expected, reason := computeExpected(engine, stor, base.Expr, v.startMs, v.endMs, v.stepMs)
+		if reason != "" {
+			continue
+		}
+		out = append(out, corpusCase{
+			Variant: v.name, Expr: base.Expr,
+			StartMs: v.startMs, EndMs: v.endMs, StepMs: v.stepMs,
+			Expected: expected,
+		})
+	}
+	return out, "", nil
+}
+
+// computeExpected evaluates the expression on one grid with the reference
+// engine and serializes the result with the API's value rounding.
+func computeExpected(engine *promql.Engine, stor *teststorage.TestStorage, expr string, startMs, endMs, stepMs int64) ([]corpusResult, string) {
+	qry, err := engine.NewRangeQuery(context.Background(), stor, nil, expr,
+		time.UnixMilli(startMs), time.UnixMilli(endMs), time.Duration(stepMs)*time.Millisecond)
+	if err != nil {
+		return nil, "engine-parse"
+	}
+	defer qry.Close()
+	res := qry.Exec(context.Background())
+	if res.Err != nil {
+		// Covers upstream's expected-error cases and engine features the
+		// range form cannot evaluate; a case we cannot compute is a case we
+		// cannot assert.
+		return nil, "engine-error"
+	}
+	matrix, ok := res.Value.(promql.Matrix)
+	if !ok {
+		return nil, "non-matrix-result"
+	}
+	expected := []corpusResult{}
+	for _, s := range matrix {
+		if len(s.Histograms) > 0 {
+			return nil, "histogram-result"
+		}
+		r := corpusResult{Labels: s.Metric.Map(), Points: []corpusPoint{}}
+		for _, p := range s.Floats {
+			r.Points = append(r.Points, corpusPoint{p.T, encodeFloat(roundToNonZeroDecimals(p.F, 3))})
+		}
+		expected = append(expected, r)
+	}
+	return expected, ""
+}
+
+// dumpDataset walks the loaded storage and serializes every float sample.
+func dumpDataset(seriesParser parser.Parser, loads []command) (*corpusDataset, string, error) {
+	stor, err := loadSeriesStorage(seriesParser, loads)
+	if err != nil {
+		return nil, "", err
+	}
+	defer func() { _ = stor.Close() }()
+
+	q, err := stor.Querier(math.MinInt64/2, math.MaxInt64/2)
+	if err != nil {
+		return nil, "querier", nil
+	}
+	defer q.Close()
+
+	ds := &corpusDataset{}
+	ss := q.Select(context.Background(), true, nil, labels.MustNewMatcher(labels.MatchRegexp, model.MetricNameLabel, ".*"))
+	var it chunkenc.Iterator
+	for ss.Next() {
+		s := ss.At()
+		cs := corpusSeries{Labels: s.Labels().Map(), Samples: [][2]any{}}
+		it = s.Iterator(it)
+		for vt := it.Next(); vt != chunkenc.ValNone; vt = it.Next() {
+			if vt != chunkenc.ValFloat {
+				return nil, "histogram-samples", nil
+			}
+			ts, v := it.At()
+			if value.IsStaleNaN(v) {
+				cs.Samples = append(cs.Samples, [2]any{ts, "stale"})
+				continue
+			}
+			cs.Samples = append(cs.Samples, [2]any{ts, encodeFloat(v)})
+		}
+		ds.Series = append(ds.Series, cs)
+	}
+	if err := ss.Err(); err != nil {
+		return nil, "series-set", nil
+	}
+	return ds, "", nil
+}
+
+// parseTestDuration accepts promqltest's time notation: a Prometheus
+// duration ("5m", "1m30s"), a bare "0", or bare seconds.
+func parseTestDuration(s string) (int64, error) {
+	if d, err := model.ParseDuration(s); err == nil {
+		return int64(time.Duration(d) / time.Millisecond), nil
+	}
+	if n, err := strconv.ParseFloat(s, 64); err == nil {
+		return int64(n * 1000), nil
+	}
+	return 0, fmt.Errorf("unparsable duration %q", s)
+}
+
+func encodeFloat(f float64) any {
+	switch {
+	case math.IsNaN(f):
+		return "NaN"
+	case math.IsInf(f, 1):
+		return "Inf"
+	case math.IsInf(f, -1):
+		return "-Inf"
+	default:
+		return f
+	}
+}
+
+// roundToNonZeroDecimals mirrors querybuildertypesv5's sanitizeValue rounding
+// (pkg/types/querybuildertypes/querybuildertypesv5/resp.go) so the frozen
+// expectations equal what the API emits for the same float.
+func roundToNonZeroDecimals(val float64, n int) float64 {
+	if val == 0 || math.IsNaN(val) || math.IsInf(val, 0) {
+		return val
+	}
+	absVal := math.Abs(val)
+	if absVal >= 1 {
+		multiplier := math.Pow(10, float64(n))
+		rounded := math.Round(val*multiplier) / multiplier
+		if math.IsInf(rounded, 0) {
+			// Mirrors the overflow guard in querybuildertypesv5.
+			return val
+		}
+		if rounded == math.Trunc(rounded) {
+			return rounded
+		}
+		str := strconv.FormatFloat(rounded, 'f', -1, 64)
+		result, _ := strconv.ParseFloat(str, 64)
+		return result
+	}
+	order := math.Floor(math.Log10(absVal))
+	scale := math.Pow(10, -order+float64(n)-1)
+	rounded := math.Round(val*scale) / scale
+	str := strconv.FormatFloat(rounded, 'f', -1, 64)
+	result, _ := strconv.ParseFloat(str, 64)
+	return result
+}
+
+// crossCheckFileAllowlist names files whose written expectations assume
+// engine options we deliberately run differently, with the reason. Every
+// other mismatch between our engine-computed expectations and upstream's
+// hand-written ones aborts generation: the corpus must never contradict
+// the testdata it claims to represent.
+var crossCheckFileAllowlist = map[string]string{
+	"name_label_dropping.test": "expectations written for EnableDelayedNameRemoval; our engine matches the server default (off)",
+}
+
+// crossCheckUpstream validates the transcription chain — load parsing,
+// eval parsing, storage loading — by comparing the reference engine's raw
+// output on the base grid against the expectations upstream wrote under the
+// same eval, with upstream's own tolerance (almost.Equal, 1e-6 relative).
+// The corpus's authority is "what the reference engine computes over
+// upstream's data"; this pins that computation to upstream's own record of
+// it.
+func crossCheckUpstream(engine *promql.Engine, seriesParser parser.Parser, stor *teststorage.TestStorage, cmd command, base corpusCase, sourceFile string, skips map[string]int) error {
+	type expSeries struct {
+		labels labels.Labels
+		points map[int64]float64
+	}
+	var expected []expSeries
+	scalarOnly := false
+	var scalarValue float64
+	for _, line := range cmd.body {
+		if strings.HasPrefix(line, "expect") {
+			// expect fail/warn/info/ordered directives and "expect range
+			// vector"/"expect string" annotations, not series expectations.
+			continue
+		}
+		if f, err := strconv.ParseFloat(line, 64); err == nil && len(cmd.body) == 1 {
+			scalarOnly, scalarValue = true, f
+			break
+		}
+		metric, vals, err := seriesParser.ParseSeriesDesc(line)
+		if err != nil {
+			skips["crosscheck-skip:unparsable-expectation"]++
+			return nil
+		}
+		points := map[int64]float64{}
+		for k, v := range vals {
+			if v.Histogram != nil {
+				skips["crosscheck-skip:histogram-expectation"]++
+				return nil
+			}
+			if v.Omitted {
+				continue
+			}
+			points[base.StartMs+int64(k)*base.StepMs] = v.Value
+		}
+		expected = append(expected, expSeries{labels: metric, points: points})
+	}
+
+	qry, err := engine.NewRangeQuery(context.Background(), stor, nil, base.Expr,
+		time.UnixMilli(base.StartMs), time.UnixMilli(base.EndMs), time.Duration(base.StepMs)*time.Millisecond)
+	if err != nil {
+		return fmt.Errorf("crosscheck parse %q: %w", base.Expr, err)
+	}
+	defer qry.Close()
+	res := qry.Exec(context.Background())
+	if res.Err != nil {
+		return fmt.Errorf("crosscheck eval %q: %w", base.Expr, res.Err)
+	}
+	matrix, ok := res.Value.(promql.Matrix)
+	if !ok {
+		skips["crosscheck-skip:non-matrix"]++
+		return nil
+	}
+
+	mismatch := func(format string, args ...any) error {
+		if reason, ok := crossCheckFileAllowlist[sourceFile]; ok {
+			skips["crosscheck-allowlisted:"+sourceFile]++
+			_ = reason
+			return nil
+		}
+		return fmt.Errorf("%s:%d: corpus contradicts upstream expectation for %q: %s",
+			sourceFile, cmd.line, base.Expr, fmt.Sprintf(format, args...))
+	}
+
+	if scalarOnly {
+		if len(matrix) != 1 || matrix[0].Metric.Len() != 0 {
+			return mismatch("scalar expectation but %d series", len(matrix))
+		}
+		if len(matrix[0].Floats) == 0 || !almost.Equal(matrix[0].Floats[len(matrix[0].Floats)-1].F, scalarValue, defaultEpsilon) {
+			return mismatch("scalar %v != expected %v", matrix[0].Floats, scalarValue)
+		}
+		skips["crosscheck-ok"]++
+		return nil
+	}
+
+	if len(matrix) != len(expected) {
+		return mismatch("engine returned %d series, upstream wrote %d", len(matrix), len(expected))
+	}
+	for _, exp := range expected {
+		var got *promql.Series
+		for i := range matrix {
+			if labels.Equal(matrix[i].Metric, exp.labels) {
+				got = &matrix[i]
+				break
+			}
+		}
+		if got == nil {
+			return mismatch("series %s missing from engine result", exp.labels)
+		}
+		gotPoints := map[int64]float64{}
+		for _, p := range got.Floats {
+			gotPoints[p.T] = p.F
+		}
+		if len(gotPoints) != len(exp.points) {
+			return mismatch("series %s: %d points, upstream wrote %d", exp.labels, len(gotPoints), len(exp.points))
+		}
+		for ts, want := range exp.points {
+			gotV, ok := gotPoints[ts]
+			if !ok {
+				return mismatch("series %s: no point at %d", exp.labels, ts)
+			}
+			if math.IsNaN(want) && math.IsNaN(gotV) {
+				continue
+			}
+			if !almost.Equal(gotV, want, defaultEpsilon) {
+				return mismatch("series %s at %d: engine %v, upstream wrote %v", exp.labels, ts, gotV, want)
+			}
+		}
+	}
+	skips["crosscheck-ok"]++
+	return nil
+}

scripts/promqltestcorpus/go.mod

@@ -0,0 +1,103 @@
+module github.com/SigNoz/signoz/scripts/promqltestcorpus
+
+go 1.25.7
+
+require (
+	github.com/prometheus/common v0.67.5
+	github.com/prometheus/prometheus v0.311.3
+)
+
+require (
+	cloud.google.com/go/auth v0.18.2 // indirect
+	cloud.google.com/go/auth/oauth2adapt v0.2.8 // indirect
+	cloud.google.com/go/compute/metadata v0.9.0 // indirect
+	github.com/Azure/azure-sdk-for-go/sdk/azcore v1.21.0 // indirect
+	github.com/Azure/azure-sdk-for-go/sdk/azidentity v1.13.1 // indirect
+	github.com/Azure/azure-sdk-for-go/sdk/internal v1.11.2 // indirect
+	github.com/AzureAD/microsoft-authentication-library-for-go v1.6.0 // indirect
+	github.com/alecthomas/units v0.0.0-20240927000941-0f3dac36c52b // indirect
+	github.com/aws/aws-sdk-go-v2 v1.41.4 // indirect
+	github.com/aws/aws-sdk-go-v2/config v1.32.12 // indirect
+	github.com/aws/aws-sdk-go-v2/credentials v1.19.12 // indirect
+	github.com/aws/aws-sdk-go-v2/feature/ec2/imds v1.18.20 // indirect
+	github.com/aws/aws-sdk-go-v2/internal/configsources v1.4.20 // indirect
+	github.com/aws/aws-sdk-go-v2/internal/endpoints/v2 v2.7.20 // indirect
+	github.com/aws/aws-sdk-go-v2/internal/ini v1.8.6 // indirect
+	github.com/aws/aws-sdk-go-v2/service/internal/accept-encoding v1.13.7 // indirect
+	github.com/aws/aws-sdk-go-v2/service/internal/presigned-url v1.13.20 // indirect
+	github.com/aws/aws-sdk-go-v2/service/signin v1.0.8 // indirect
+	github.com/aws/aws-sdk-go-v2/service/sso v1.30.13 // indirect
+	github.com/aws/aws-sdk-go-v2/service/ssooidc v1.35.17 // indirect
+	github.com/aws/aws-sdk-go-v2/service/sts v1.41.9 // indirect
+	github.com/aws/smithy-go v1.24.2 // indirect
+	github.com/bboreham/go-loser v0.0.0-20230920113527-fcc2c21820a3 // indirect
+	github.com/beorn7/perks v1.0.1 // indirect
+	github.com/cespare/xxhash/v2 v2.3.0 // indirect
+	github.com/davecgh/go-spew v1.1.2-0.20180830191138-d8f796af33cc // indirect
+	github.com/dennwc/varint v1.0.0 // indirect
+	github.com/edsrzf/mmap-go v1.2.0 // indirect
+	github.com/facette/natsort v0.0.0-20181210072756-2cd4dd1e2dcb // indirect
+	github.com/felixge/httpsnoop v1.0.4 // indirect
+	github.com/fxamacker/cbor/v2 v2.9.0 // indirect
+	github.com/go-logr/logr v1.4.3 // indirect
+	github.com/go-logr/stdr v1.2.2 // indirect
+	github.com/golang-jwt/jwt/v5 v5.3.1 // indirect
+	github.com/golang/snappy v1.0.0 // indirect
+	github.com/google/go-cmp v0.7.0 // indirect
+	github.com/google/s2a-go v0.1.9 // indirect
+	github.com/google/uuid v1.6.0 // indirect
+	github.com/googleapis/enterprise-certificate-proxy v0.3.14 // indirect
+	github.com/googleapis/gax-go/v2 v2.18.0 // indirect
+	github.com/grafana/regexp v0.0.0-20250905093917-f7b3be9d1853 // indirect
+	github.com/jpillora/backoff v1.0.0 // indirect
+	github.com/json-iterator/go v1.1.12 // indirect
+	github.com/klauspost/compress v1.18.5 // indirect
+	github.com/kylelemons/godebug v1.1.0 // indirect
+	github.com/modern-go/concurrent v0.0.0-20180306012644-bacd9c7ef1dd // indirect
+	github.com/modern-go/reflect2 v1.0.3-0.20250322232337-35a7c28c31ee // indirect
+	github.com/munnerz/goautoneg v0.0.0-20191010083416-a7dc8b61c822 // indirect
+	github.com/mwitkow/go-conntrack v0.0.0-20190716064945-2f068394615f // indirect
+	github.com/oklog/ulid/v2 v2.1.1 // indirect
+	github.com/pkg/browser v0.0.0-20240102092130-5ac0b6a4141c // indirect
+	github.com/pmezard/go-difflib v1.0.1-0.20181226105442-5d4384ee4fb2 // indirect
+	github.com/prometheus/client_golang v1.23.2 // indirect
+	github.com/prometheus/client_golang/exp v0.0.0-20260325093428-d8591d0db856 // indirect
+	github.com/prometheus/client_model v0.6.2 // indirect
+	github.com/prometheus/otlptranslator v1.0.0 // indirect
+	github.com/prometheus/procfs v0.16.1 // indirect
+	github.com/prometheus/sigv4 v0.4.1 // indirect
+	github.com/stretchr/testify v1.11.1 // indirect
+	github.com/x448/float16 v0.8.4 // indirect
+	go.opentelemetry.io/auto/sdk v1.2.1 // indirect
+	go.opentelemetry.io/contrib/instrumentation/net/http/otelhttp v0.67.0 // indirect
+	go.opentelemetry.io/otel v1.42.0 // indirect
+	go.opentelemetry.io/otel/metric v1.42.0 // indirect
+	go.opentelemetry.io/otel/trace v1.42.0 // indirect
+	go.uber.org/atomic v1.11.0 // indirect
+	go.uber.org/goleak v1.3.0 // indirect
+	go.yaml.in/yaml/v2 v2.4.4 // indirect
+	golang.org/x/crypto v0.49.0 // indirect
+	golang.org/x/exp v0.0.0-20260218203240-3dfff04db8fa // indirect
+	golang.org/x/net v0.52.0 // indirect
+	golang.org/x/oauth2 v0.36.0 // indirect
+	golang.org/x/sync v0.20.0 // indirect
+	golang.org/x/sys v0.42.0 // indirect
+	golang.org/x/term v0.41.0 // indirect
+	golang.org/x/text v0.35.0 // indirect
+	golang.org/x/time v0.15.0 // indirect
+	google.golang.org/api v0.272.0 // indirect
+	google.golang.org/genproto/googleapis/rpc v0.0.0-20260311181403-84a4fc48630c // indirect
+	google.golang.org/grpc v1.79.3 // indirect
+	google.golang.org/protobuf v1.36.11 // indirect
+	gopkg.in/inf.v0 v0.9.1 // indirect
+	gopkg.in/yaml.v3 v3.0.1 // indirect
+	k8s.io/apimachinery v0.35.3 // indirect
+	k8s.io/client-go v0.35.3 // indirect
+	k8s.io/klog/v2 v2.140.0 // indirect
+	k8s.io/kube-openapi v0.0.0-20250910181357-589584f1c912 // indirect
+	k8s.io/utils v0.0.0-20251002143259-bc988d571ff4 // indirect
+	sigs.k8s.io/json v0.0.0-20250730193827-2d320260d730 // indirect
+	sigs.k8s.io/randfill v1.0.0 // indirect
+	sigs.k8s.io/structured-merge-diff/v6 v6.3.0 // indirect
+	sigs.k8s.io/yaml v1.6.0 // indirect
+)