Compare commits

..

14 Commits

Author SHA1 Message Date
Srikanth Chekuri
7aa0eff6f4 Merge branch 'main' into tvats-export-traces 2026-02-27 15:23:28 +05:30
Tushar Vats
0e30a11231 fix: address comments 2026-02-27 03:39:08 +05:30
Srikanth Chekuri
5901074385 Merge branch 'main' into tvats-export-traces 2026-02-26 06:57:14 +05:30
Tushar Vats
5b27121530 fix: ran yarn generate:api 2026-02-26 02:56:37 +05:30
Tushar Vats
7090d1af17 fix: lint error 2026-02-26 02:39:43 +05:30
Tushar Vats
31152e01f5 fix: lint error 2026-02-25 21:54:25 +05:30
Tushar Vats
e19b37620d Merge branch 'main' into tvats-export-traces 2026-02-25 21:46:49 +05:30
Tushar Vats
820f0723a5 fix: address comments 2026-02-25 21:44:32 +05:30
Tushar Vats
b8c101e95a fix: renamed method 2026-02-19 10:53:33 +05:30
Tushar Vats
85c9daab12 fix: rebased main and ran generate cmd 2026-02-19 10:50:00 +05:30
Tushar Vats
4a80ef8f43 fix: go fmt 2026-02-19 10:47:33 +05:30
Tushar Vats
ac8600f5da fix: removed nits 2026-02-19 10:47:33 +05:30
Tushar Vats
b2655634fe fix: address comments 2026-02-19 10:47:25 +05:30
Tushar Vats
b2f20f7a64 feat: added trace export
feat: added types for export

feat: added support for complex queries

fix: added correct open api spec

fix: updated unit tests

fix: type handling logic

fix: improve order by

feat: added integration tests

fix: address comments
2026-02-19 10:44:45 +05:30
25 changed files with 3589 additions and 1173 deletions

View File

@@ -17,5 +17,7 @@
},
"[html]": {
"editor.defaultFormatter": "vscode.html-language-features"
}
},
"python-envs.defaultEnvManager": "ms-python.python:system",
"python-envs.pythonProjects": []
}

View File

@@ -2858,6 +2858,186 @@ paths:
summary: Update auth domain
tags:
- authdomains
/api/v1/export_raw_data:
get:
deprecated: false
description: This endpoints allows simple query exporting raw data for traces
and logs
operationId: HandleExportRawDataGET
parameters:
- in: query
name: format
schema:
default: csv
enum:
- csv
- jsonl
type: string
- in: query
name: signal
required: true
schema:
$ref: '#/components/schemas/TelemetrytypesSignal'
- deprecated: true
in: query
name: source
schema:
deprecated: true
type: string
- in: query
name: start
schema:
minimum: 0
type: integer
- in: query
name: end
schema:
minimum: 0
type: integer
- in: query
name: limit
schema:
default: 10000
maximum: 50000
minimum: 1
type: integer
- deprecated: true
in: query
name: filter
schema:
deprecated: true
type: string
- content:
application/json:
schema:
$ref: '#/components/schemas/Querybuildertypesv5Filter'
in: query
name: filterExpression
- deprecated: true
in: query
name: columns
schema:
deprecated: true
items:
type: string
type: array
- in: query
name: selectFields
schema:
items:
$ref: '#/components/schemas/TelemetrytypesTelemetryFieldKey'
type: array
- deprecated: true
in: query
name: order_by
schema:
deprecated: true
type: string
- in: query
name: order
schema:
items:
$ref: '#/components/schemas/Querybuildertypesv5OrderBy'
type: array
responses:
"200":
content:
application/json:
schema:
type: string
description: OK
"400":
content:
application/json:
schema:
$ref: '#/components/schemas/RenderErrorResponse'
description: Bad Request
"401":
content:
application/json:
schema:
$ref: '#/components/schemas/RenderErrorResponse'
description: Unauthorized
"403":
content:
application/json:
schema:
$ref: '#/components/schemas/RenderErrorResponse'
description: Forbidden
"500":
content:
application/json:
schema:
$ref: '#/components/schemas/RenderErrorResponse'
description: Internal Server Error
security:
- api_key:
- VIEWER
- tokenizer:
- VIEWER
summary: Export raw data
tags:
- logs
- traces
post:
deprecated: false
description: This endpoints allows complex query exporting raw data for traces
and logs
operationId: HandleExportRawDataPOST
parameters:
- in: query
name: format
schema:
default: csv
enum:
- csv
- jsonl
type: string
requestBody:
content:
application/json:
schema:
$ref: '#/components/schemas/Querybuildertypesv5QueryRangeRequest'
responses:
"200":
content:
application/json:
schema:
type: string
description: OK
"400":
content:
application/json:
schema:
$ref: '#/components/schemas/RenderErrorResponse'
description: Bad Request
"401":
content:
application/json:
schema:
$ref: '#/components/schemas/RenderErrorResponse'
description: Unauthorized
"403":
content:
application/json:
schema:
$ref: '#/components/schemas/RenderErrorResponse'
description: Forbidden
"500":
content:
application/json:
schema:
$ref: '#/components/schemas/RenderErrorResponse'
description: Internal Server Error
security:
- api_key:
- VIEWER
- tokenizer:
- VIEWER
summary: Export raw data
tags:
- logs
- traces
/api/v1/fields/keys:
get:
deprecated: false

View File

@@ -1,127 +0,0 @@
# Abstractions
This document provides rules for deciding when a new type, interface, or intermediate representation is warranted in Go code. The goal is to keep the codebase navigable by ensuring every abstraction earns its place.
## The cost of a new abstraction
Every exported type, interface, or wrapper is a permanent commitment. It must be named, documented, tested, and understood by every future contributor. It creates a new concept in the codebase vocabulary. Before introducing one, verify that the cost is justified by a concrete benefit that cannot be achieved with existing mechanisms.
## Before you introduce anything new
Answer these four questions. If writing a PR, include the answers in the description.
1. **What already exists?** Name the specific type, function, interface, or library that covers this ground today.
2. **What does the new abstraction add?** Name the concrete operation, guarantee, or capability. "Cleaner" or "more reusable" are not sufficient; name what the caller can do that it could not do before.
3. **What does the new abstraction drop?** If it wraps or mirrors an existing structure, list what it cannot represent. Every gap must be either justified or handled with an explicit error.
4. **Who consumes it?** List the call sites. If there is only one producer and one consumer in the same call chain, you likely need a function, not a type.
## Rules
### 1. Prefer functions over types
If a piece of logic has one input and one output, write a function. Do not create a struct to hold intermediate state that is built in one place and read in one place. A function is easier to test, easier to inline, and does not expand the vocabulary of the codebase.
```go
// Prefer this:
func ConvertConfig(src ExternalConfig) (InternalConfig, error)
// Over this:
type ConfigAdapter struct { ... }
func NewConfigAdapter(src ExternalConfig) *ConfigAdapter
func (a *ConfigAdapter) ToInternal() (InternalConfig, error)
```
The two-step version is only justified when `ConfigAdapter` has multiple distinct consumers that use it in different ways.
### 2. Do not duplicate structures you do not own
When a library or external package produces a structured output, operate on that output directly. Do not create a parallel type that mirrors a subset of its fields.
A partial copy will:
- **Silently lose data** when the source has fields or variants the copy does not account for.
- **Drift** when the source evolves and the copy is not updated in lockstep.
- **Add a conversion step** that doubles the code surface and the opportunity for bugs.
If you need to shield consumers from a dependency, define a narrow interface over the dependency's type rather than copying its shape into a new struct.
### 3. Never silently discard input
If your code receives structured input and cannot handle part of it, return an error. Do not silently return nil, skip the element, or produce a partial result. Silent data loss is the hardest class of bug to detect because the code appears to work, it just produces wrong results.
```go
// Wrong: silently ignores the unrecognized case.
default:
return nil
// Right: makes the gap visible.
default:
return nil, fmt.Errorf("unsupported %T value: %v", v, v)
```
This applies broadly: type switches, format conversions, data migrations, enum mappings, configuration parsing. Anywhere a `default` or `else` branch can swallow input, it should surface an error instead.
### 4. Do not expose methods that lose information
A method on a structured type should not strip meaning from the structure it belongs to. If a caller needs to iterate over elements for a specific purpose (validation, aggregation, logging), write that logic as a standalone function that operates on the structure with full context, rather than adding a method that returns a reduced view.
```go
// Problematic: callers cannot distinguish how items were related.
func (o *Order) AllLineItems() []LineItem { ... }
// Better: the validation logic operates on the full structure.
func ValidateOrder(o *Order) error { ... }
```
Public methods shape how a type is used. Once a lossy accessor exists, callers will depend on it, and the lost information becomes unrecoverable at those call sites.
### 5. Interfaces should be discovered, not predicted
Do not define an interface before you have at least two concrete implementations that need it. An interface with one implementation is not abstraction; it is indirection that makes it harder to navigate from call site to implementation.
The exception is interfaces required for testing (e.g., for mocking an external dependency). In that case, define the interface in the **consuming** package, not the providing package, following the Go convention of [accepting interfaces and returning structs](https://go.dev/wiki/CodeReviewComments#interfaces).
### 6. Wrappers must add semantics, not just rename
A wrapper type is justified when it adds meaning, validation, or invariants that the underlying type does not carry. It is not justified when it merely renames fields or reorganizes the same data into a different shape.
```go
// Justified: adds validation that the underlying string does not carry.
type OrgID struct{ value string }
func NewOrgID(s string) (OrgID, error) { /* validates format */ }
// Not justified: renames fields with no new invariant or behavior.
type UserInfo struct {
Name string // same as source.Name
Email string // same as source.Email
}
```
Ask: what does the wrapper guarantee that the underlying type does not? If the answer is nothing, use the underlying type directly.
## When a new type IS warranted
A new type earns its place when it meets **at least one** of these criteria:
- **Serialization boundary**: It must be persisted, sent over the wire, or written to config. The source type is unsuitable (unexported fields, function pointers, cycles).
- **Invariant enforcement**: The constructor or methods enforce constraints that raw data does not carry (e.g., non-empty, validated format, bounded range).
- **Multiple distinct consumers**: Three or more call sites use the type in meaningfully different ways. The type is the shared vocabulary between them.
- **Dependency firewall**: The type lives in a lightweight package so that consumers avoid importing a heavy dependency.
## What should I remember?
- A function is almost always simpler than a type. Start with a function; promote to a type only when you have evidence of need.
- Never silently drop data. If you cannot handle it, error.
- If your new type mirrors an existing one, you need a strong reason beyond "nicer to work with".
- If your type has one producer and one consumer, it is indirection, not abstraction.
- Interfaces come from need (multiple implementations), not from prediction.
- When in doubt, do not add it. It is easier to add an abstraction later when the need is clear than to remove one after it has spread through the codebase.
## Further reading
These works and our own lessions shaped the above guidelines
- [The Wrong Abstraction](https://sandimetz.com/blog/2016/1/20/the-wrong-abstraction) - Sandi Metz. The wrong abstraction is worse than duplication. If you find yourself passing parameters and adding conditional paths through shared code, inline it back into every caller and let the duplication show you what the right abstraction is.
- [Write code that is easy to delete, not easy to extend](https://programmingisterrible.com/post/139222674273/write-code-that-is-easy-to-delete-not-easy-to) - tef. Every abstraction is a bet on the future. Optimize for how cheaply you can remove code when the bet is wrong, not for how easily you can extend it when the bet is right.
- [Goodbye, Clean Code](https://overreacted.io/goodbye-clean-code/) - Dan Abramov. A refactoring that removes duplication can look cleaner while making the code harder to change. Clean-looking and easy-to-change are not the same thing.
- [A Philosophy of Software Design](https://www.amazon.com/Philosophy-Software-Design-John-Ousterhout/dp/1732102201) - John Ousterhout. Good abstractions are deep: simple interface, complex implementation. A "false abstraction" omits important details while appearing simple, and is worse than no abstraction at all. ([Summary by Pragmatic Engineer](https://blog.pragmaticengineer.com/a-philosophy-of-software-design-review/))
- [Simplicity is Complicated](https://go.dev/talks/2015/simplicity-is-complicated.slide) - Rob Pike. Go-specific. Fewer orthogonal concepts that compose predictably beat many overlapping ones. Features were left out of Go deliberately; the same discipline applies to your own code.

View File

@@ -123,6 +123,7 @@ if err := router.Handle("/api/v1/things", handler.New(
Description: "This endpoint creates a thing",
Request: new(types.PostableThing),
RequestContentType: "application/json",
RequestQuery: new(types.QueryableThing),
Response: new(types.GettableThing),
ResponseContentType: "application/json",
SuccessStatusCode: http.StatusCreated,
@@ -155,6 +156,8 @@ The `handler.New` function ties the HTTP handler to OpenAPI metadata via `OpenAP
- **Request / RequestContentType**:
- `Request` is a Go type that describes the request body or form.
- `RequestContentType` is usually `"application/json"` or `"application/x-www-form-urlencoded"` (for callbacks like SAML).
- **RequestQuery**:
- `RequestQuery` is a Go type that descirbes query url params.
- **RequestExamples**: An array of `handler.OpenAPIExample` that provide concrete request payloads in the generated spec. See [Adding request examples](#adding-request-examples) below.
- **Response / ResponseContentType**:
- `Response` is the Go type for the successful response payload.

View File

@@ -1,126 +0,0 @@
# Packages
All shared Go code in SigNoz lives under `pkg/`. Each package represents a distinct domain concept and exposes a clear public interface. This guide covers the conventions for creating, naming, and organising packages so the codebase stays consistent as it grows.
## How should I name a package?
Use short, lowercase, single-word names. No underscores or camelCase (`querier`, `cache`, `authz`, not `query_builder` or `dataStore`).
Names must be **domain-specific**. A package name should tell you what problem domain it deals with, not what data structure it wraps. Prefer `alertmanager` over `manager`, `licensing` over `checker`.
Avoid generic names like `util`, `helpers`, `common`, `misc`, or `base`. If you can't name it, the code probably belongs in an existing package.
## When should I create a new package?
Create a new package when:
- The functionality represents a **distinct domain concept** (e.g., `authz`, `licensing`, `cache`).
- Two or more other packages would import it; it serves as shared infrastructure.
- The code has a clear public interface that can stand on its own.
Do **not** create a new package when:
- There is already a package that covers the same domain. Extend the existing package instead.
- The code is only used in one place. Keep it local to the caller.
- You are splitting purely for file size. Use multiple files within the same package instead.
## How should I lay out a package?
A typical package looks like:
```
pkg/cache/
├── cache.go # Public interface + exported types
├── config.go # Configuration types if needed
├── memorycache/ # Implementation sub-package
├── rediscache/ # Another implementation
└── cachetest/ # Test helpers for consumers
```
Follow these rules:
1. **Interface-first file**: The file matching the package name (e.g., `cache.go` in `pkg/cache/`) should define the public interface and core exported types. Keep implementation details out of this file.
2. **One responsibility per file**: Name files after what they contain (`config.go`, `handler.go`, `service.go`), not after the package name. If a package merges two concerns, prefix files to group them (e.g., `memory_store.go`, `redis_store.go` in a storage package).
3. **Sub-packages for implementations**: When a package defines an interface with multiple implementations, put each implementation in its own sub-package (`memorycache/`, `rediscache/`). This keeps the parent package import-free of implementation dependencies.
4. **Test helpers in `{pkg}test/`**: If consumers need test mocks or builders, put them in a `{pkg}test/` sub-package (e.g., `cachetest/`, `sqlstoretest/`). This avoids polluting the main package with test-only code.
5. **Test files stay alongside source**: Unit tests go in `_test.go` files next to the code they test, in the same package.
## How should I name symbols?
### Exported symbols
- **Interfaces**: For single-method interfaces, follow the standard `-er` suffix convention (`Reader`, `Writer`, `Closer`). For multi-method interfaces, use clear nouns (`Cache`, `Store`, `Provider`).
- **Constructors**: `New<Type>(...)` (e.g., `NewMemoryCache()`).
- **Avoid stutter**: Since callers qualify with the package name, don't repeat it. Write `cache.Cache`, not `cache.CacheInterface`. Write `authz.FromRole`, not `authz.AuthzFromRole`.
### Unexported symbols
- Struct receivers: one or two characters (`c`, `f`, `br`).
- Helper functions: descriptive lowercase names (`parseToken`, `buildQuery`).
### Constants
- Use `PascalCase` for exported constants.
- When merging files from different origins into one package, watch out for **name collisions** across files. Prefix to disambiguate when two types share a natural name.
## How should I organise imports?
Group imports in three blocks separated by blank lines:
```go
import (
// 1. Standard library
"fmt"
"net/http"
// 2. External dependencies
"github.com/gorilla/mux"
// 3. Internal
"github.com/SigNoz/signoz/pkg/errors"
"github.com/SigNoz/signoz/pkg/types"
)
```
Never introduce circular imports. If package A needs package B and B needs A, extract the shared types into a third package (often under `pkg/types/`).
## Where do shared types go?
Most types belong in `pkg/types/` under a domain-specific sub-package (e.g., `pkg/types/ruletypes`, `pkg/types/authtypes`).
Do not put domain logic in `pkg/types/`. Only data structures, constants, and simple methods.
## How do I merge or move packages?
When two packages are tightly coupled (one imports the other's constants, they cover the same domain), merge them:
1. Pick a domain-specific name for the combined package.
2. Prefix files to preserve origin (e.g., `memory_store.go`, `redis_store.go`).
3. Resolve symbol conflicts explicitly; rename with a prefix rather than silently shadowing.
4. Update all consumers in a single change.
5. Delete the old packages. Do not leave behind re-export shims.
6. Verify with `go build ./...`, `go test ./<new-pkg>/...`, and `go vet ./...`.
## When should I add documentation?
Add a `doc.go` with a package-level comment for any package that is non-trivial or has multiple consumers. Keep it to 13 sentences:
```go
// Package cache provides a caching interface with pluggable backends
// for in-memory and Redis-based storage.
package cache
```
## What should I remember?
- Package names are domain-specific and lowercase. Never generic names like `util` or `common`.
- The file matching the package name (e.g., `cache.go`) defines the public interface. Implementation details go elsewhere.
- Never introduce circular imports. Extract shared types into `pkg/types/` when needed.
- Watch for symbol name collisions when merging packages, prefix to disambiguate.
- Put test helpers in a `{pkg}test/` sub-package, not in the main package.
- Before submitting, verify with `go build ./...`, `go test ./<your-pkg>/...`, and `go vet ./...`.
- Update all consumers when you rename or move symbols.

View File

@@ -8,14 +8,4 @@ We adhere to three primary style guides as our foundation:
- [Code Review Comments](https://go.dev/wiki/CodeReviewComments) - For understanding common comments in code reviews
- [Google Style Guide](https://google.github.io/styleguide/go/) - Additional practices from Google
We **recommend** (almost enforce) reviewing these guides before contributing to the codebase. They provide valuable insights into writing idiomatic Go code and will help you understand our approach to backend development. In addition, we have a few additional rules that make certain areas stricter than the above which can be found in area-specific files in this package:
- [Abstractions](abstractions.md) - When to introduce new types and intermediate representations
- [Errors](errors.md) - Structured error handling
- [Endpoint](endpoint.md) - HTTP endpoint patterns
- [Flagger](flagger.md) - Feature flag patterns
- [Handler](handler.md) - HTTP handler patterns
- [Integration](integration.md) - Integration testing
- [Provider](provider.md) - Dependency injection and provider patterns
- [Packages](packages.md) — Naming, layout, and conventions for `pkg/` packages
- [SQL](sql.md) - Database and SQL patterns
We **recommend** (almost enforce) reviewing these guides before contributing to the codebase. They provide valuable insights into writing idiomatic Go code and will help you understand our approach to backend development. In addition, we have a few additional rules that make certain areas stricter than the above which can be found in area-specific files in this package.

View File

@@ -20,8 +20,11 @@ import { useMutation, useQuery } from 'react-query';
import type { BodyType, ErrorType } from '../../../generatedAPIInstance';
import { GeneratedAPIInstance } from '../../../generatedAPIInstance';
import type {
HandleExportRawDataGETParams,
HandleExportRawDataPOSTParams,
ListPromotedAndIndexedPaths200,
PromotetypesPromotePathDTO,
Querybuildertypesv5QueryRangeRequestDTO,
RenderErrorResponseDTO,
} from '../sigNoz.schemas';
@@ -29,6 +32,206 @@ type AwaitedInput<T> = PromiseLike<T> | T;
type Awaited<O> = O extends AwaitedInput<infer T> ? T : never;
/**
* This endpoints allows simple query exporting raw data for traces and logs
* @summary Export raw data
*/
export const handleExportRawDataGET = (
params?: HandleExportRawDataGETParams,
signal?: AbortSignal,
) => {
return GeneratedAPIInstance<string>({
url: `/api/v1/export_raw_data`,
method: 'GET',
params,
signal,
});
};
export const getHandleExportRawDataGETQueryKey = (
params?: HandleExportRawDataGETParams,
) => {
return [`/api/v1/export_raw_data`, ...(params ? [params] : [])] as const;
};
export const getHandleExportRawDataGETQueryOptions = <
TData = Awaited<ReturnType<typeof handleExportRawDataGET>>,
TError = ErrorType<RenderErrorResponseDTO>
>(
params?: HandleExportRawDataGETParams,
options?: {
query?: UseQueryOptions<
Awaited<ReturnType<typeof handleExportRawDataGET>>,
TError,
TData
>;
},
) => {
const { query: queryOptions } = options ?? {};
const queryKey =
queryOptions?.queryKey ?? getHandleExportRawDataGETQueryKey(params);
const queryFn: QueryFunction<
Awaited<ReturnType<typeof handleExportRawDataGET>>
> = ({ signal }) => handleExportRawDataGET(params, signal);
return { queryKey, queryFn, ...queryOptions } as UseQueryOptions<
Awaited<ReturnType<typeof handleExportRawDataGET>>,
TError,
TData
> & { queryKey: QueryKey };
};
export type HandleExportRawDataGETQueryResult = NonNullable<
Awaited<ReturnType<typeof handleExportRawDataGET>>
>;
export type HandleExportRawDataGETQueryError = ErrorType<RenderErrorResponseDTO>;
/**
* @summary Export raw data
*/
export function useHandleExportRawDataGET<
TData = Awaited<ReturnType<typeof handleExportRawDataGET>>,
TError = ErrorType<RenderErrorResponseDTO>
>(
params?: HandleExportRawDataGETParams,
options?: {
query?: UseQueryOptions<
Awaited<ReturnType<typeof handleExportRawDataGET>>,
TError,
TData
>;
},
): UseQueryResult<TData, TError> & { queryKey: QueryKey } {
const queryOptions = getHandleExportRawDataGETQueryOptions(params, options);
const query = useQuery(queryOptions) as UseQueryResult<TData, TError> & {
queryKey: QueryKey;
};
query.queryKey = queryOptions.queryKey;
return query;
}
/**
* @summary Export raw data
*/
export const invalidateHandleExportRawDataGET = async (
queryClient: QueryClient,
params?: HandleExportRawDataGETParams,
options?: InvalidateOptions,
): Promise<QueryClient> => {
await queryClient.invalidateQueries(
{ queryKey: getHandleExportRawDataGETQueryKey(params) },
options,
);
return queryClient;
};
/**
* This endpoints allows complex query exporting raw data for traces and logs
* @summary Export raw data
*/
export const handleExportRawDataPOST = (
querybuildertypesv5QueryRangeRequestDTO: BodyType<Querybuildertypesv5QueryRangeRequestDTO>,
params?: HandleExportRawDataPOSTParams,
signal?: AbortSignal,
) => {
return GeneratedAPIInstance<string>({
url: `/api/v1/export_raw_data`,
method: 'POST',
headers: { 'Content-Type': 'application/json' },
data: querybuildertypesv5QueryRangeRequestDTO,
params,
signal,
});
};
export const getHandleExportRawDataPOSTMutationOptions = <
TError = ErrorType<RenderErrorResponseDTO>,
TContext = unknown
>(options?: {
mutation?: UseMutationOptions<
Awaited<ReturnType<typeof handleExportRawDataPOST>>,
TError,
{
data: BodyType<Querybuildertypesv5QueryRangeRequestDTO>;
params?: HandleExportRawDataPOSTParams;
},
TContext
>;
}): UseMutationOptions<
Awaited<ReturnType<typeof handleExportRawDataPOST>>,
TError,
{
data: BodyType<Querybuildertypesv5QueryRangeRequestDTO>;
params?: HandleExportRawDataPOSTParams;
},
TContext
> => {
const mutationKey = ['handleExportRawDataPOST'];
const { mutation: mutationOptions } = options
? options.mutation &&
'mutationKey' in options.mutation &&
options.mutation.mutationKey
? options
: { ...options, mutation: { ...options.mutation, mutationKey } }
: { mutation: { mutationKey } };
const mutationFn: MutationFunction<
Awaited<ReturnType<typeof handleExportRawDataPOST>>,
{
data: BodyType<Querybuildertypesv5QueryRangeRequestDTO>;
params?: HandleExportRawDataPOSTParams;
}
> = (props) => {
const { data, params } = props ?? {};
return handleExportRawDataPOST(data, params);
};
return { mutationFn, ...mutationOptions };
};
export type HandleExportRawDataPOSTMutationResult = NonNullable<
Awaited<ReturnType<typeof handleExportRawDataPOST>>
>;
export type HandleExportRawDataPOSTMutationBody = BodyType<Querybuildertypesv5QueryRangeRequestDTO>;
export type HandleExportRawDataPOSTMutationError = ErrorType<RenderErrorResponseDTO>;
/**
* @summary Export raw data
*/
export const useHandleExportRawDataPOST = <
TError = ErrorType<RenderErrorResponseDTO>,
TContext = unknown
>(options?: {
mutation?: UseMutationOptions<
Awaited<ReturnType<typeof handleExportRawDataPOST>>,
TError,
{
data: BodyType<Querybuildertypesv5QueryRangeRequestDTO>;
params?: HandleExportRawDataPOSTParams;
},
TContext
>;
}): UseMutationResult<
Awaited<ReturnType<typeof handleExportRawDataPOST>>,
TError,
{
data: BodyType<Querybuildertypesv5QueryRangeRequestDTO>;
params?: HandleExportRawDataPOSTParams;
},
TContext
> => {
const mutationOptions = getHandleExportRawDataPOSTMutationOptions(options);
return useMutation(mutationOptions);
};
/**
* This endpoints promotes and indexes paths
* @summary Promote and index paths

View File

@@ -2732,6 +2732,76 @@ export type DeleteAuthDomainPathParameters = {
export type UpdateAuthDomainPathParameters = {
id: string;
};
export type HandleExportRawDataGETParams = {
/**
* @enum csv,jsonl
* @type string
* @description undefined
*/
format?: HandleExportRawDataGETFormat;
/**
* @enum logs,traces
* @type string
* @description undefined
*/
source?: HandleExportRawDataGETSource;
/**
* @type integer
* @minimum 0
* @description undefined
*/
start?: number;
/**
* @type integer
* @minimum 0
* @description undefined
*/
end?: number;
/**
* @type integer
* @maximum 50000
* @minimum 1
* @description undefined
*/
limit?: number;
/**
* @type string
* @description undefined
*/
filter?: string;
/**
* @type array
* @description undefined
*/
columns?: string[];
/**
* @type string
* @description undefined
*/
order_by?: string;
};
export enum HandleExportRawDataGETFormat {
csv = 'csv',
jsonl = 'jsonl',
}
export enum HandleExportRawDataGETSource {
logs = 'logs',
traces = 'traces',
}
export type HandleExportRawDataPOSTParams = {
/**
* @enum csv,jsonl
* @type string
* @description undefined
*/
format?: HandleExportRawDataPOSTFormat;
};
export enum HandleExportRawDataPOSTFormat {
csv = 'csv',
jsonl = 'jsonl',
}
export type GetFieldsKeysParams = {
/**
* @description undefined

View File

@@ -18,6 +18,7 @@ import (
"github.com/SigNoz/signoz/pkg/modules/organization"
"github.com/SigNoz/signoz/pkg/modules/preference"
"github.com/SigNoz/signoz/pkg/modules/promote"
"github.com/SigNoz/signoz/pkg/modules/rawdataexport"
"github.com/SigNoz/signoz/pkg/modules/session"
"github.com/SigNoz/signoz/pkg/modules/user"
"github.com/SigNoz/signoz/pkg/querier"
@@ -46,6 +47,7 @@ type provider struct {
gatewayHandler gateway.Handler
fieldsHandler fields.Handler
authzHandler authz.Handler
rawDataExportHandler rawdataexport.Handler
zeusHandler zeus.Handler
querierHandler querier.Handler
}
@@ -67,6 +69,7 @@ func NewFactory(
gatewayHandler gateway.Handler,
fieldsHandler fields.Handler,
authzHandler authz.Handler,
rawDataExportHandler rawdataexport.Handler,
zeusHandler zeus.Handler,
querierHandler querier.Handler,
) factory.ProviderFactory[apiserver.APIServer, apiserver.Config] {
@@ -91,6 +94,7 @@ func NewFactory(
gatewayHandler,
fieldsHandler,
authzHandler,
rawDataExportHandler,
zeusHandler,
querierHandler,
)
@@ -117,6 +121,7 @@ func newProvider(
gatewayHandler gateway.Handler,
fieldsHandler fields.Handler,
authzHandler authz.Handler,
rawDataExportHandler rawdataexport.Handler,
zeusHandler zeus.Handler,
querierHandler querier.Handler,
) (apiserver.APIServer, error) {
@@ -141,6 +146,7 @@ func newProvider(
gatewayHandler: gatewayHandler,
fieldsHandler: fieldsHandler,
authzHandler: authzHandler,
rawDataExportHandler: rawDataExportHandler,
zeusHandler: zeusHandler,
querierHandler: querierHandler,
}
@@ -215,6 +221,10 @@ func (provider *provider) AddToRouter(router *mux.Router) error {
return err
}
if err := provider.addRawDataExportRoutes(router); err != nil {
return err
}
if err := provider.addZeusRoutes(router); err != nil {
return err
}

View File

@@ -0,0 +1,47 @@
package signozapiserver
import (
"net/http"
"github.com/SigNoz/signoz/pkg/http/handler"
"github.com/SigNoz/signoz/pkg/types"
"github.com/SigNoz/signoz/pkg/types/exporttypes"
v5 "github.com/SigNoz/signoz/pkg/types/querybuildertypes/querybuildertypesv5"
"github.com/gorilla/mux"
)
func (provider *provider) addRawDataExportRoutes(router *mux.Router) error {
if err := router.Handle("/api/v1/export_raw_data", handler.New(provider.authZ.ViewAccess(provider.rawDataExportHandler.ExportRawData), handler.OpenAPIDef{
ID: "HandleExportRawDataGET",
Tags: []string{"logs", "traces"},
Summary: "Export raw data",
Description: "This endpoints allows simple query exporting raw data for traces and logs",
RequestQuery: new(exporttypes.ExportRawDataQueryParams),
RequestContentType: "",
Response: nil,
ResponseContentType: "application/json",
SuccessStatusCode: http.StatusOK,
ErrorStatusCodes: []int{http.StatusBadRequest},
SecuritySchemes: newSecuritySchemes(types.RoleViewer),
})).Methods(http.MethodGet).GetError(); err != nil {
return err
}
if err := router.Handle("/api/v1/export_raw_data", handler.New(provider.authZ.ViewAccess(provider.rawDataExportHandler.ExportRawData), handler.OpenAPIDef{
ID: "HandleExportRawDataPOST",
Tags: []string{"logs", "traces"},
Summary: "Export raw data",
Description: "This endpoints allows complex query exporting raw data for traces and logs",
Request: new(v5.QueryRangeRequest),
RequestQuery: new(exporttypes.ExportRawDataFormatQueryParam),
RequestContentType: "application/json",
Response: nil,
ResponseContentType: "application/json",
SuccessStatusCode: http.StatusOK,
ErrorStatusCodes: []int{http.StatusBadRequest},
SecuritySchemes: newSecuritySchemes(types.RoleViewer),
})).Methods(http.MethodPost).GetError(); err != nil {
return err
}
return nil
}

View File

@@ -6,18 +6,18 @@ import (
"fmt"
"io"
"net/http"
"net/url"
"slices"
"strconv"
"strings"
"time"
"unicode"
"unicode/utf8"
"github.com/SigNoz/signoz/pkg/errors"
"github.com/SigNoz/signoz/pkg/http/binding"
"github.com/SigNoz/signoz/pkg/http/render"
"github.com/SigNoz/signoz/pkg/modules/rawdataexport"
"github.com/SigNoz/signoz/pkg/telemetrylogs"
"github.com/SigNoz/signoz/pkg/types/authtypes"
"github.com/SigNoz/signoz/pkg/types/exporttypes"
qbtypes "github.com/SigNoz/signoz/pkg/types/querybuildertypes/querybuildertypesv5"
"github.com/SigNoz/signoz/pkg/types/telemetrytypes"
"github.com/SigNoz/signoz/pkg/valuer"
@@ -31,129 +31,52 @@ func NewHandler(module rawdataexport.Module) rawdataexport.Handler {
return &handler{module: module}
}
// ExportRawData handles data export requests.
//
// API Documentation:
// Endpoint: GET /api/v1/export_raw_data
//
// Query Parameters:
//
// - source (optional): Type of data to export ["logs" (default), "metrics", "traces"]
// Note: Currently only "logs" is fully supported
//
// - format (optional): Output format ["csv" (default), "jsonl"]
//
// - start (required): Start time for query (Unix timestamp in nanoseconds)
//
// - end (required): End time for query (Unix timestamp in nanoseconds)
//
// - limit (optional): Maximum number of rows to export
// Constraints: Must be positive and cannot exceed MAX_EXPORT_ROW_COUNT_LIMIT
//
// - filter (optional): Filter expression to apply to the query
//
// - columns (optional): Specific columns to include in export
// Default: all columns are returned
// Format: ["context.field:type", "context.field", "field"]
//
// - order_by (optional): Sorting specification ["column:direction" or "context.field:type:direction"]
// Direction: "asc" or "desc"
// Default: ["timestamp:desc", "id:desc"]
//
// Response Headers:
// - Content-Type: "text/csv" or "application/x-ndjson"
// - Content-Encoding: "gzip" (handled by HTTP middleware)
// - Content-Disposition: "attachment; filename=\"data_exported.[format]\""
// - Cache-Control: "no-cache"
// - Vary: "Accept-Encoding"
// - Transfer-Encoding: "chunked"
// - Trailers: X-Response-Complete
//
// Response Format:
//
// CSV: Headers in first row, data in subsequent rows
// JSONL: One JSON object per line
//
// Example Usage:
//
// Basic CSV export:
// GET /api/v1/export_raw_data?start=1693612800000000000&end=1693699199000000000
//
// Export with columns and format:
// GET /api/v1/export_raw_data?start=1693612800000000000&end=1693699199000000000&format=jsonl
// &columns=timestamp&columns=severity&columns=message
//
// Export with filter and ordering:
// GET /api/v1/export_raw_data?start=1693612800000000000&end=1693699199000000000
// &filter=severity="error"&order_by=timestamp:desc&limit=1000
func (handler *handler) ExportRawData(rw http.ResponseWriter, r *http.Request) {
source, err := getExportQuerySource(r.URL.Query())
if err != nil {
render.Error(rw, err)
return
}
var queryRangeRequest qbtypes.QueryRangeRequest
var format string
switch source {
case "logs":
handler.exportLogs(rw, r)
case "traces":
handler.exportTraces(rw, r)
case "metrics":
handler.exportMetrics(rw, r)
switch r.Method {
case http.MethodGet:
var params exporttypes.ExportRawDataQueryParams
if err := binding.Query.BindQuery(r.URL.Query(), &params); err != nil {
render.Error(rw, err)
return
}
params.Normalize()
if err := params.Validate(); err != nil {
render.Error(rw, err)
return
}
format = params.Format
queryRangeRequest = buildQueryRangeRequest(&params)
case http.MethodPost:
var formatParam exporttypes.ExportRawDataFormatQueryParam
if err := binding.Query.BindQuery(r.URL.Query(), &formatParam); err != nil {
render.Error(rw, err)
return
}
format = formatParam.Format
if err := json.NewDecoder(r.Body).Decode(&queryRangeRequest); err != nil {
render.Error(rw, errors.NewInvalidInputf(errors.CodeInvalidInput, "invalid request body: %v", err))
return
}
default:
render.Error(rw, errors.NewInvalidInputf(errors.CodeInvalidInput, "invalid source: must be logs"))
render.Error(rw, errors.Newf(errors.TypeMethodNotAllowed, errors.CodeMethodNotAllowed, "method not allowed, only GET/POST supported"))
return
}
}
func (handler *handler) exportMetrics(rw http.ResponseWriter, r *http.Request) {
render.Error(rw, errors.Newf(errors.TypeUnsupported, errors.CodeUnsupported, "metrics export is not yet supported"))
}
func (handler *handler) exportTraces(rw http.ResponseWriter, r *http.Request) {
render.Error(rw, errors.Newf(errors.TypeUnsupported, errors.CodeUnsupported, "traces export is not yet supported"))
}
func (handler *handler) exportLogs(rw http.ResponseWriter, r *http.Request) {
// Set up response headers
rw.Header().Set("Cache-Control", "no-cache")
rw.Header().Set("Vary", "Accept-Encoding") // Indicate that response varies based on Accept-Encoding
rw.Header().Set("Access-Control-Expose-Headers", "Content-Disposition, X-Response-Complete")
rw.Header().Set("Trailer", "X-Response-Complete")
rw.Header().Set("Transfer-Encoding", "chunked")
queryParams := r.URL.Query()
startTime, endTime, err := getExportQueryTimeRange(queryParams)
if err != nil {
if err := validateSpecForExport(&queryRangeRequest); err != nil {
render.Error(rw, err)
return
}
limit, err := getExportQueryLimit(queryParams)
if err != nil {
if err := validateAndApplyDefaultExportLimits(queryRangeRequest.CompositeQuery.Queries); err != nil {
render.Error(rw, err)
return
}
format, err := getExportQueryFormat(queryParams)
if err != nil {
render.Error(rw, err)
return
}
// Set appropriate content type and filename
filename := fmt.Sprintf("data_exported_%s.%s", time.Now().Format("2006-01-02_150405"), format)
rw.Header().Set("Content-Disposition", fmt.Sprintf("attachment; filename=\"%s\"", filename))
filterExpression := queryParams.Get("filter")
orderByExpression, err := getExportQueryOrderBy(queryParams)
if err != nil {
render.Error(rw, err)
return
}
columns := getExportQueryColumns(queryParams)
// Use default OrderBy if not specified
queryRangeRequest.UseDefaultOrderBy()
claims, err := authtypes.ClaimsFromContext(r.Context())
if err != nil {
@@ -167,70 +90,134 @@ func (handler *handler) exportLogs(rw http.ResponseWriter, r *http.Request) {
return
}
queryRangeRequest := qbtypes.QueryRangeRequest{
Start: startTime,
End: endTime,
RequestType: qbtypes.RequestTypeRaw,
CompositeQuery: qbtypes.CompositeQuery{
Queries: []qbtypes.QueryEnvelope{
{
Type: qbtypes.QueryTypeBuilder,
Spec: nil,
},
},
},
}
setExportResponseHeaders(rw, format)
spec := qbtypes.QueryBuilderQuery[qbtypes.LogAggregation]{
Signal: telemetrytypes.SignalLogs,
Name: "raw",
Filter: &qbtypes.Filter{
Expression: filterExpression,
},
Limit: limit,
Order: orderByExpression,
}
spec.SelectFields = columns
queryRangeRequest.CompositeQuery.Queries[0].Spec = spec
// This will signal Export module to stop sending data
doneChan := make(chan any)
defer close(doneChan)
rowChan, errChan := handler.module.ExportRawData(r.Context(), orgID, &queryRangeRequest, doneChan)
var isComplete bool
isComplete, err := handler.executeExport(rowChan, errChan, format, rw)
if err != nil {
render.Error(rw, err)
return
}
rw.Header().Set("X-Response-Complete", strconv.FormatBool(isComplete))
}
// validateSpecForExport validates query specs
func validateSpecForExport(req *qbtypes.QueryRangeRequest) error {
queries := req.CompositeQuery.Queries
// If the trace operator query is not present, and there are multiple queries, return an error
if req.TraceOperatorQueryIndex() == -1 && len(queries) > 1 {
return errors.NewInvalidInputf(errors.CodeInvalidInput, "multiple queries not allowed without a trace operator query")
}
for idx := range queries {
switch spec := queries[idx].Spec.(type) {
case qbtypes.QueryBuilderQuery[qbtypes.LogAggregation],
qbtypes.QueryBuilderQuery[qbtypes.TraceAggregation],
qbtypes.QueryBuilderTraceOperator:
// Supported spec types
default:
return errors.NewInvalidInputf(errors.CodeInvalidInput, "unsupported query at index %d type: %T", idx, spec)
}
}
err := req.Validate(qbtypes.WithSkipLimitValidation())
if err != nil {
return err
}
return nil
}
func validateAndApplyDefaultExportLimits(queries []qbtypes.QueryEnvelope) error {
for idx := range queries {
limit := queries[idx].GetLimit()
if limit == 0 {
limit = DefaultExportRowCountLimit
} else if limit < 0 {
return errors.NewInvalidInputf(errors.CodeInvalidInput, "limit must be positive")
} else if limit > MaxExportRowCountLimit {
return errors.NewInvalidInputf(errors.CodeInvalidInput, "limit cannot be more than %d", MaxExportRowCountLimit)
}
queries[idx].SetLimit(limit)
}
return nil
}
// buildQueryEnvelope creates a QueryEnvelope with a QueryBuilderQuery for the given aggregation type.
func buildQueryEnvelope[T any](signal telemetrytypes.Signal, filter *qbtypes.Filter, limit int, order []qbtypes.OrderBy, selectFields []telemetrytypes.TelemetryFieldKey) qbtypes.QueryEnvelope {
return qbtypes.QueryEnvelope{
Type: qbtypes.QueryTypeBuilder,
Spec: qbtypes.QueryBuilderQuery[T]{
Signal: signal,
Filter: filter,
Limit: limit,
Order: order,
SelectFields: selectFields,
},
}
}
// buildQueryRangeRequest builds a QueryRangeRequest from already-bound and validated GET query params.
func buildQueryRangeRequest(params *exporttypes.ExportRawDataQueryParams) qbtypes.QueryRangeRequest {
var query qbtypes.QueryEnvelope
switch params.Signal {
case telemetrytypes.SignalLogs:
query = buildQueryEnvelope[qbtypes.LogAggregation](params.Signal, &params.Filter, params.Limit, params.Order, params.SelectFields)
case telemetrytypes.SignalTraces:
query = buildQueryEnvelope[qbtypes.TraceAggregation](params.Signal, &params.Filter, params.Limit, params.Order, params.SelectFields)
}
return qbtypes.QueryRangeRequest{
Start: params.Start,
End: params.End,
RequestType: qbtypes.RequestTypeRaw,
CompositeQuery: qbtypes.CompositeQuery{
Queries: []qbtypes.QueryEnvelope{query},
},
}
}
// setExportResponseHeaders sets common HTTP headers for export responses.
func setExportResponseHeaders(rw http.ResponseWriter, format string) {
rw.Header().Set("Cache-Control", "no-cache")
rw.Header().Set("Vary", "Accept-Encoding")
rw.Header().Set("Access-Control-Expose-Headers", "Content-Disposition, X-Response-Complete")
rw.Header().Set("Trailer", "X-Response-Complete")
rw.Header().Set("Transfer-Encoding", "chunked")
filename := fmt.Sprintf("data_exported_%s.%s", time.Now().Format("2006-01-02_150405"), format)
rw.Header().Set("Content-Disposition", fmt.Sprintf("attachment; filename=\"%s\"", filename))
}
// executeExport streams data from rowChan to the response writer in the specified format.
func (handler *handler) executeExport(rowChan <-chan *qbtypes.RawRow, errChan <-chan error, format string, rw http.ResponseWriter) (bool, error) {
switch format {
case "csv", "":
rw.Header().Set("Content-Type", "text/csv")
csvWriter := csv.NewWriter(rw)
isComplete, err = handler.exportLogsCSV(rowChan, errChan, csvWriter)
isComplete, err := handler.exportRawDataCSV(rowChan, errChan, csvWriter)
if err != nil {
render.Error(rw, err)
return
return false, err
}
csvWriter.Flush()
return isComplete, nil
case "jsonl":
rw.Header().Set("Content-Type", "application/x-ndjson")
isComplete, err = handler.exportLogsJSONL(rowChan, errChan, rw)
if err != nil {
render.Error(rw, err)
return
}
return handler.exportRawDataJSONL(rowChan, errChan, rw)
default:
render.Error(rw, errors.NewInvalidInputf(errors.CodeInvalidInput, "invalid format: must be csv or jsonl"))
return
return false, errors.NewInvalidInputf(errors.CodeInvalidInput, "invalid format: must be csv or jsonl")
}
rw.Header().Set("X-Response-Complete", strconv.FormatBool(isComplete))
}
func (handler *handler) exportLogsCSV(rowChan <-chan *qbtypes.RawRow, errChan <-chan error, csvWriter *csv.Writer) (bool, error) {
var header []string
// exportRawDataCSV is a generic CSV export function that works with any raw data (logs, traces, etc.)
func (handler *handler) exportRawDataCSV(rowChan <-chan *qbtypes.RawRow, errChan <-chan error, csvWriter *csv.Writer) (bool, error) {
headerToIndexMapping := make(map[string]int, len(header))
var header []string
headerToIndexMapping := make(map[string]int)
totalBytes := uint64(0)
for {
@@ -268,8 +255,8 @@ func (handler *handler) exportLogsCSV(rowChan <-chan *qbtypes.RawRow, errChan <-
}
}
func (handler *handler) exportLogsJSONL(rowChan <-chan *qbtypes.RawRow, errChan <-chan error, writer io.Writer) (bool, error) {
// exportRawDataJSONL is a generic JSONL export function that works with any raw data (logs, traces, etc.)
func (handler *handler) exportRawDataJSONL(rowChan <-chan *qbtypes.RawRow, errChan <-chan error, writer io.Writer) (bool, error) {
totalBytes := uint64(0)
for {
select {
@@ -277,9 +264,11 @@ func (handler *handler) exportLogsJSONL(rowChan <-chan *qbtypes.RawRow, errChan
if !ok {
return true, nil
}
// Handle JSON format (JSONL - one object per line)
jsonBytes, _ := json.Marshal(row.Data)
totalBytes += uint64(len(jsonBytes)) + 1 // +1 for newline
jsonBytes, err := json.Marshal(row.Data)
if err != nil {
return false, errors.NewUnexpectedf(errors.CodeInternal, "error marshaling JSON: %s", err)
}
totalBytes += uint64(len(jsonBytes)) + 1
if _, err := writer.Write(jsonBytes); err != nil {
return false, errors.NewUnexpectedf(errors.CodeInternal, "error writing JSON: %s", err)
@@ -299,74 +288,33 @@ func (handler *handler) exportLogsJSONL(rowChan <-chan *qbtypes.RawRow, errChan
}
}
func getExportQuerySource(queryParams url.Values) (string, error) {
switch queryParams.Get("source") {
case "logs", "":
return "logs", nil
case "metrics":
return "metrics", errors.NewInvalidInputf(errors.CodeInvalidInput, "metrics export not yet supported")
case "traces":
return "traces", errors.NewInvalidInputf(errors.CodeInvalidInput, "traces export not yet supported")
default:
return "", errors.NewInvalidInputf(errors.CodeInvalidInput, "invalid source: must be logs, metrics or traces")
}
}
func getExportQueryFormat(queryParams url.Values) (string, error) {
switch queryParams.Get("format") {
case "csv", "":
return "csv", nil
case "jsonl":
return "jsonl", nil
default:
return "", errors.NewInvalidInputf(errors.CodeInvalidInput, "invalid format: must be csv or jsonl")
}
}
func getExportQueryLimit(queryParams url.Values) (int, error) {
limitStr := queryParams.Get("limit")
if limitStr == "" {
return DefaultExportRowCountLimit, nil
} else {
limit, err := strconv.Atoi(limitStr)
if err != nil {
return 0, errors.NewInvalidInputf(errors.CodeInvalidInput, "invalid limit format: %s", err.Error())
}
if limit <= 0 {
return 0, errors.NewInvalidInputf(errors.CodeInvalidInput, "limit must be positive")
}
if limit > MaxExportRowCountLimit {
return 0, errors.NewInvalidInputf(errors.CodeInvalidInput, "limit cannot be more than %d", MaxExportRowCountLimit)
}
return limit, nil
}
}
func getExportQueryTimeRange(queryParams url.Values) (uint64, uint64, error) {
startTimeStr := queryParams.Get("start")
endTimeStr := queryParams.Get("end")
if startTimeStr == "" || endTimeStr == "" {
return 0, 0, errors.NewInvalidInputf(errors.CodeInvalidInput, "start and end time are required")
}
startTime, err := strconv.ParseUint(startTimeStr, 10, 64)
if err != nil {
return 0, 0, errors.NewInvalidInputf(errors.CodeInvalidInput, "invalid start time format: %s", err.Error())
}
endTime, err := strconv.ParseUint(endTimeStr, 10, 64)
if err != nil {
return 0, 0, errors.NewInvalidInputf(errors.CodeInvalidInput, "invalid end time format: %s", err.Error())
}
return startTime, endTime, nil
}
// priorityColumns defines the columns that should appear first in the CSV output, in order.
var priorityColumns = []string{"timestamp", "id"}
func constructCSVHeaderFromQueryResponse(data map[string]any) []string {
header := make([]string, 0, len(data))
for key := range data {
header = append(header, key)
}
// This is to ensure CSV output is consistent across multiple queries
slices.SortFunc(header, func(a, b string) int {
ai, bi := slices.Index(priorityColumns, a), slices.Index(priorityColumns, b)
switch {
case ai != -1 && bi != -1:
return ai - bi
case ai != -1:
return -1
case bi != -1:
return 1
default:
if a < b {
return -1
} else if a > b {
return 1
}
return 0
}
})
return header
}
@@ -427,9 +375,12 @@ func constructCSVRecordFromQueryResponse(data map[string]any, headerToIndexMappi
valueStr = v.String()
default:
// For all other complex types (maps, structs, etc.)
jsonBytes, _ := json.Marshal(v)
valueStr = string(jsonBytes)
jsonBytes, err := json.Marshal(v)
if err != nil {
valueStr = fmt.Sprintf("%v", v)
} else {
valueStr = string(jsonBytes)
}
}
record[index] = sanitizeForCSV(valueStr)
@@ -438,26 +389,6 @@ func constructCSVRecordFromQueryResponse(data map[string]any, headerToIndexMappi
return record
}
// getExportQueryColumns parses the "columns" query parameters and returns a slice of TelemetryFieldKey structs.
// Each column should be a valid telemetry field key in the format "context.field:type" or "context.field" or "field"
func getExportQueryColumns(queryParams url.Values) []telemetrytypes.TelemetryFieldKey {
columnParams := queryParams["columns"]
columns := make([]telemetrytypes.TelemetryFieldKey, 0, len(columnParams))
for _, columnStr := range columnParams {
// Skip empty strings
columnStr = strings.TrimSpace(columnStr)
if columnStr == "" {
continue
}
columns = append(columns, telemetrytypes.GetFieldKeyFromKeyText(columnStr))
}
return columns
}
func getsizeOfStringSlice(slice []string) uint64 {
var totalBytes uint64
for _, str := range slice {
@@ -465,52 +396,3 @@ func getsizeOfStringSlice(slice []string) uint64 {
}
return totalBytes
}
// getExportQueryOrderBy parses the "order_by" query parameters and returns a slice of OrderBy structs.
// Each "order_by" parameter should be in the format "column:direction"
// Each "column" should be a valid telemetry field key in the format "context.field:type" or "context.field" or "field"
func getExportQueryOrderBy(queryParams url.Values) ([]qbtypes.OrderBy, error) {
orderByParam := queryParams.Get("order_by")
orderByParam = strings.TrimSpace(orderByParam)
if orderByParam == "" {
return telemetrylogs.DefaultLogsV2SortingOrder, nil
}
parts := strings.Split(orderByParam, ":")
if len(parts) != 2 && len(parts) != 3 {
return nil, errors.NewInvalidInputf(errors.CodeInvalidInput, "invalid order_by format: %s, should be <column>:<direction>", orderByParam)
}
column := strings.Join(parts[:len(parts)-1], ":")
direction := parts[len(parts)-1]
orderDirection, ok := qbtypes.OrderDirectionMap[direction]
if !ok {
return nil, errors.NewInvalidInputf(errors.CodeInvalidInput, "invalid order_by direction: %s, should be one of %s, %s", direction, qbtypes.OrderDirectionAsc, qbtypes.OrderDirectionDesc)
}
orderByKey := telemetrytypes.GetFieldKeyFromKeyText(column)
orderBy := []qbtypes.OrderBy{
{
Key: qbtypes.OrderByKey{
TelemetryFieldKey: orderByKey,
},
Direction: orderDirection,
},
}
// If we are ordering by the timestamp column, also order by the ID column
if orderByKey.Name == telemetrylogs.LogsV2TimestampColumn {
orderBy = append(orderBy, qbtypes.OrderBy{
Key: qbtypes.OrderByKey{
TelemetryFieldKey: telemetrytypes.TelemetryFieldKey{
Name: telemetrylogs.LogsV2IDColumn,
},
},
Direction: orderDirection,
})
}
return orderBy, nil
}

View File

@@ -2,162 +2,85 @@ package implrawdataexport
import (
"net/url"
"strconv"
"testing"
"github.com/SigNoz/signoz/pkg/telemetrylogs"
"github.com/SigNoz/signoz/pkg/http/binding"
"github.com/SigNoz/signoz/pkg/types/exporttypes"
qbtypes "github.com/SigNoz/signoz/pkg/types/querybuildertypes/querybuildertypesv5"
"github.com/SigNoz/signoz/pkg/types/telemetrytypes"
"github.com/stretchr/testify/assert"
)
func TestGetExportQuerySource(t *testing.T) {
tests := []struct {
name string
queryParams url.Values
expectedSource string
expectedError bool
}{
{
name: "default logs source",
queryParams: url.Values{},
expectedSource: "logs",
expectedError: false,
},
{
name: "explicit logs source",
queryParams: url.Values{"source": {"logs"}},
expectedSource: "logs",
expectedError: false,
},
{
name: "metrics source - not supported",
queryParams: url.Values{"source": {"metrics"}},
expectedSource: "metrics",
expectedError: true,
},
{
name: "traces source - not supported",
queryParams: url.Values{"source": {"traces"}},
expectedSource: "traces",
expectedError: true,
},
{
name: "invalid source",
queryParams: url.Values{"source": {"invalid"}},
expectedSource: "",
expectedError: true,
},
}
func TestExportRawDataQueryParams_BindingDefaults(t *testing.T) {
var params exporttypes.ExportRawDataQueryParams
err := binding.Query.BindQuery(url.Values{}, &params)
assert.NoError(t, err)
assert.Equal(t, "csv", params.Format)
assert.Equal(t, DefaultExportRowCountLimit, params.Limit)
}
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
source, err := getExportQuerySource(tt.queryParams)
assert.Equal(t, tt.expectedSource, source)
if tt.expectedError {
assert.Error(t, err)
} else {
assert.NoError(t, err)
}
})
func logQuery(limit int) qbtypes.QueryEnvelope {
return qbtypes.QueryEnvelope{
Type: qbtypes.QueryTypeBuilder,
Spec: qbtypes.QueryBuilderQuery[qbtypes.LogAggregation]{Limit: limit},
}
}
func TestGetExportQueryFormat(t *testing.T) {
tests := []struct {
name string
queryParams url.Values
expectedFormat string
expectedError bool
}{
{
name: "default csv format",
queryParams: url.Values{},
expectedFormat: "csv",
expectedError: false,
},
{
name: "explicit csv format",
queryParams: url.Values{"format": {"csv"}},
expectedFormat: "csv",
expectedError: false,
},
{
name: "jsonl format",
queryParams: url.Values{"format": {"jsonl"}},
expectedFormat: "jsonl",
expectedError: false,
},
{
name: "invalid format",
queryParams: url.Values{"format": {"xml"}},
expectedFormat: "",
expectedError: true,
},
}
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
format, err := getExportQueryFormat(tt.queryParams)
assert.Equal(t, tt.expectedFormat, format)
if tt.expectedError {
assert.Error(t, err)
} else {
assert.NoError(t, err)
}
})
func traceQuery(limit int) qbtypes.QueryEnvelope {
return qbtypes.QueryEnvelope{
Type: qbtypes.QueryTypeBuilder,
Spec: qbtypes.QueryBuilderQuery[qbtypes.TraceAggregation]{Limit: limit},
}
}
func TestGetExportQueryLimit(t *testing.T) {
func traceOperatorQuery(limit int) qbtypes.QueryEnvelope {
return qbtypes.QueryEnvelope{
Type: qbtypes.QueryTypeTraceOperator,
Spec: qbtypes.QueryBuilderTraceOperator{Limit: limit, Expression: "A"},
}
}
func makeRequest(queries ...qbtypes.QueryEnvelope) qbtypes.QueryRangeRequest {
return qbtypes.QueryRangeRequest{
Start: 1000000000000,
End: 1000003600000,
RequestType: qbtypes.RequestTypeRaw,
CompositeQuery: qbtypes.CompositeQuery{Queries: queries},
}
}
func TestValidateSpecForExport(t *testing.T) {
tests := []struct {
name string
queryParams url.Values
expectedLimit int
req qbtypes.QueryRangeRequest
expectedError bool
}{
{
name: "default limit",
queryParams: url.Values{},
expectedLimit: DefaultExportRowCountLimit,
expectedError: false,
name: "single log query",
req: makeRequest(logQuery(0)),
},
{
name: "valid limit",
queryParams: url.Values{"limit": {"5000"}},
expectedLimit: 5000,
expectedError: false,
name: "single trace query",
req: makeRequest(traceQuery(0)),
},
{
name: "maximum limit",
queryParams: url.Values{"limit": {strconv.Itoa(MaxExportRowCountLimit)}},
expectedLimit: MaxExportRowCountLimit,
expectedError: false,
name: "trace operator alone",
req: makeRequest(traceOperatorQuery(0)),
},
{
name: "limit exceeds maximum",
queryParams: url.Values{"limit": {"100000"}},
expectedLimit: 0,
name: "multiple queries without trace operator",
req: makeRequest(logQuery(0), traceQuery(0)),
expectedError: true,
},
{
name: "invalid limit format",
queryParams: url.Values{"limit": {"invalid"}},
expectedLimit: 0,
expectedError: true,
},
{
name: "negative limit",
queryParams: url.Values{"limit": {"-100"}},
expectedLimit: 0,
name: "unsupported query type",
req: makeRequest(qbtypes.QueryEnvelope{Type: qbtypes.QueryTypeBuilder, Spec: qbtypes.QueryBuilderQuery[qbtypes.MetricAggregation]{}}),
expectedError: true,
},
}
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
limit, err := getExportQueryLimit(tt.queryParams)
assert.Equal(t, tt.expectedLimit, limit)
err := validateSpecForExport(&tt.req)
if tt.expectedError {
assert.Error(t, err)
} else {
@@ -167,352 +90,69 @@ func TestGetExportQueryLimit(t *testing.T) {
}
}
func TestGetExportQueryTimeRange(t *testing.T) {
tests := []struct {
name string
queryParams url.Values
expectedStartTime uint64
expectedEndTime uint64
expectedError bool
}{
{
name: "valid time range",
queryParams: url.Values{
"start": {"1640995200"},
"end": {"1641081600"},
},
expectedStartTime: 1640995200,
expectedEndTime: 1641081600,
expectedError: false,
},
{
name: "missing start time",
queryParams: url.Values{"end": {"1641081600"}},
expectedError: true,
},
{
name: "missing end time",
queryParams: url.Values{"start": {"1640995200"}},
expectedError: true,
},
{
name: "missing both times",
queryParams: url.Values{},
expectedError: true,
},
{
name: "invalid start time format",
queryParams: url.Values{
"start": {"invalid"},
"end": {"1641081600"},
},
expectedError: true,
},
{
name: "invalid end time format",
queryParams: url.Values{
"start": {"1640995200"},
"end": {"invalid"},
},
expectedError: true,
},
}
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
startTime, endTime, err := getExportQueryTimeRange(tt.queryParams)
if tt.expectedError {
assert.Error(t, err)
} else {
assert.NoError(t, err)
assert.Equal(t, tt.expectedStartTime, startTime)
assert.Equal(t, tt.expectedEndTime, endTime)
}
})
}
}
func TestGetExportQueryColumns(t *testing.T) {
tests := []struct {
name string
queryParams url.Values
expectedColumns []telemetrytypes.TelemetryFieldKey
}{
{
name: "no columns specified",
queryParams: url.Values{},
expectedColumns: []telemetrytypes.TelemetryFieldKey{},
},
{
name: "single column",
queryParams: url.Values{
"columns": {"timestamp"},
},
expectedColumns: []telemetrytypes.TelemetryFieldKey{
{Name: "timestamp"},
},
},
{
name: "multiple columns",
queryParams: url.Values{
"columns": {"timestamp", "message", "level"},
},
expectedColumns: []telemetrytypes.TelemetryFieldKey{
{Name: "timestamp"},
{Name: "message"},
{Name: "level"},
},
},
{
name: "empty column name (should be skipped)",
queryParams: url.Values{
"columns": {"timestamp", "", "level"},
},
expectedColumns: []telemetrytypes.TelemetryFieldKey{
{Name: "timestamp"},
{Name: "level"},
},
},
{
name: "whitespace column name (should be skipped)",
queryParams: url.Values{
"columns": {"timestamp", " ", "level"},
},
expectedColumns: []telemetrytypes.TelemetryFieldKey{
{Name: "timestamp"},
{Name: "level"},
},
},
{
name: "valid column name with data type",
queryParams: url.Values{
"columns": {"timestamp", "attribute.user:string", "level"},
},
expectedColumns: []telemetrytypes.TelemetryFieldKey{
{Name: "timestamp"},
{Name: "user", FieldContext: telemetrytypes.FieldContextAttribute, FieldDataType: telemetrytypes.FieldDataTypeString},
{Name: "level"},
},
},
{
name: "valid column name with dot notation",
queryParams: url.Values{
"columns": {"timestamp", "attribute.user.string", "level"},
},
expectedColumns: []telemetrytypes.TelemetryFieldKey{
{Name: "timestamp"},
{Name: "user.string", FieldContext: telemetrytypes.FieldContextAttribute},
{Name: "level"},
},
},
}
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
columns := getExportQueryColumns(tt.queryParams)
assert.Equal(t, len(tt.expectedColumns), len(columns))
for i, expectedCol := range tt.expectedColumns {
assert.Equal(t, expectedCol, columns[i])
}
})
}
}
func TestGetExportQueryOrderBy(t *testing.T) {
func TestValidateAndApplyDefaultExportLimits(t *testing.T) {
tests := []struct {
name string
queryParams url.Values
expectedOrder []qbtypes.OrderBy
queries []qbtypes.QueryEnvelope
expectedError bool
checkQueries func(t *testing.T, queries []qbtypes.QueryEnvelope)
}{
{
name: "no order specified",
queryParams: url.Values{},
expectedOrder: []qbtypes.OrderBy{
{
Direction: qbtypes.OrderDirectionDesc,
Key: qbtypes.OrderByKey{
TelemetryFieldKey: telemetrytypes.TelemetryFieldKey{
Name: telemetrylogs.LogsV2TimestampColumn,
},
},
},
{
Direction: qbtypes.OrderDirectionDesc,
Key: qbtypes.OrderByKey{
TelemetryFieldKey: telemetrytypes.TelemetryFieldKey{
Name: telemetrylogs.LogsV2IDColumn,
},
},
},
name: "single log query, zero limit gets default",
queries: makeRequest(logQuery(0)).CompositeQuery.Queries,
checkQueries: func(t *testing.T, q []qbtypes.QueryEnvelope) {
assert.Equal(t, DefaultExportRowCountLimit, q[0].GetLimit())
},
expectedError: false,
},
{
name: "single order error, direction not specified",
queryParams: url.Values{
"order_by": {"timestamp"},
name: "single log query, valid limit kept",
queries: makeRequest(logQuery(1000)).CompositeQuery.Queries,
checkQueries: func(t *testing.T, q []qbtypes.QueryEnvelope) {
assert.Equal(t, 1000, q[0].GetLimit())
},
expectedOrder: nil,
},
{
name: "single log query, max limit kept",
queries: makeRequest(logQuery(MaxExportRowCountLimit)).CompositeQuery.Queries,
checkQueries: func(t *testing.T, q []qbtypes.QueryEnvelope) {
assert.Equal(t, MaxExportRowCountLimit, q[0].GetLimit())
},
},
{
name: "single log query, limit exceeds max",
queries: makeRequest(logQuery(MaxExportRowCountLimit + 1)).CompositeQuery.Queries,
expectedError: true,
},
{
name: "single order no error",
queryParams: url.Values{
"order_by": {"timestamp:asc"},
},
expectedOrder: []qbtypes.OrderBy{
{
Direction: qbtypes.OrderDirectionAsc,
Key: qbtypes.OrderByKey{
TelemetryFieldKey: telemetrytypes.TelemetryFieldKey{
Name: telemetrylogs.LogsV2TimestampColumn,
},
},
},
{
Direction: qbtypes.OrderDirectionAsc,
Key: qbtypes.OrderByKey{
TelemetryFieldKey: telemetrytypes.TelemetryFieldKey{
Name: telemetrylogs.LogsV2IDColumn,
},
},
},
},
expectedError: false,
},
{
name: "multiple orders",
queryParams: url.Values{
"order_by": {"timestamp:asc", "body:desc", "id:asc"},
},
expectedOrder: []qbtypes.OrderBy{
{
Direction: qbtypes.OrderDirectionAsc,
Key: qbtypes.OrderByKey{
TelemetryFieldKey: telemetrytypes.TelemetryFieldKey{
Name: telemetrylogs.LogsV2TimestampColumn,
},
},
},
{
Direction: qbtypes.OrderDirectionAsc,
Key: qbtypes.OrderByKey{
TelemetryFieldKey: telemetrytypes.TelemetryFieldKey{
Name: telemetrylogs.LogsV2IDColumn,
},
},
},
},
expectedError: false,
},
{
name: "empty order name (should be skipped)",
queryParams: url.Values{
"order_by": {"timestamp:asc", "", "id:asc"},
},
expectedOrder: []qbtypes.OrderBy{
{
Direction: qbtypes.OrderDirectionAsc,
Key: qbtypes.OrderByKey{
TelemetryFieldKey: telemetrytypes.TelemetryFieldKey{
Name: telemetrylogs.LogsV2TimestampColumn,
},
},
},
{
Direction: qbtypes.OrderDirectionAsc,
Key: qbtypes.OrderByKey{
TelemetryFieldKey: telemetrytypes.TelemetryFieldKey{
Name: telemetrylogs.LogsV2IDColumn,
},
},
},
},
expectedError: false,
},
{
name: "whitespace order name (should be skipped)",
queryParams: url.Values{
"order_by": {"timestamp:asc", " ", "id:asc"},
},
expectedOrder: []qbtypes.OrderBy{
{
Direction: qbtypes.OrderDirectionAsc,
Key: qbtypes.OrderByKey{
TelemetryFieldKey: telemetrytypes.TelemetryFieldKey{
Name: telemetrylogs.LogsV2TimestampColumn,
},
},
},
{
Direction: qbtypes.OrderDirectionAsc,
Key: qbtypes.OrderByKey{
TelemetryFieldKey: telemetrytypes.TelemetryFieldKey{
Name: telemetrylogs.LogsV2IDColumn,
},
},
},
},
expectedError: false,
},
{
name: "invalid order name (should error out)",
queryParams: url.Values{
"order_by": {"attributes.user:", "id:asc"},
},
expectedOrder: nil,
name: "single log query, negative limit",
queries: makeRequest(logQuery(-1)).CompositeQuery.Queries,
expectedError: true,
},
{
name: "valid order name (should be included)",
queryParams: url.Values{
"order_by": {"attribute.user:string:desc", "id:asc"},
name: "single trace query, zero limit gets default",
queries: makeRequest(traceQuery(0)).CompositeQuery.Queries,
checkQueries: func(t *testing.T, q []qbtypes.QueryEnvelope) {
assert.Equal(t, DefaultExportRowCountLimit, q[0].GetLimit())
},
expectedOrder: []qbtypes.OrderBy{
{
Direction: qbtypes.OrderDirectionDesc,
Key: qbtypes.OrderByKey{
TelemetryFieldKey: telemetrytypes.TelemetryFieldKey{
Name: "user",
FieldContext: telemetrytypes.FieldContextAttribute,
FieldDataType: telemetrytypes.FieldDataTypeString,
},
},
},
},
expectedError: false,
},
{
name: "valid order name (should be included)",
queryParams: url.Values{
"order_by": {"attribute.user.string:desc", "id:asc"},
name: "trace operator alone, zero limit gets default",
queries: makeRequest(traceOperatorQuery(0)).CompositeQuery.Queries,
checkQueries: func(t *testing.T, q []qbtypes.QueryEnvelope) {
assert.Equal(t, DefaultExportRowCountLimit, q[0].GetLimit())
},
expectedOrder: []qbtypes.OrderBy{
{
Direction: qbtypes.OrderDirectionDesc,
Key: qbtypes.OrderByKey{
TelemetryFieldKey: telemetrytypes.TelemetryFieldKey{
Name: "user.string",
FieldContext: telemetrytypes.FieldContextAttribute,
},
},
},
},
expectedError: false,
},
}
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
order, err := getExportQueryOrderBy(tt.queryParams)
err := validateAndApplyDefaultExportLimits(tt.queries)
if tt.expectedError {
assert.Error(t, err)
} else {
assert.NoError(t, err)
assert.Equal(t, len(tt.expectedOrder), len(order))
for i, expectedOrd := range tt.expectedOrder {
assert.Equal(t, expectedOrd, order[i])
if tt.checkQueries != nil {
tt.checkQueries(t, tt.queries)
}
}
})
@@ -529,13 +169,8 @@ func TestConstructCSVHeaderFromQueryResponse(t *testing.T) {
header := constructCSVHeaderFromQueryResponse(data)
// Since map iteration order is not guaranteed, check that all expected keys are present
expectedKeys := []string{"timestamp", "message", "level", "id"}
assert.Equal(t, len(expectedKeys), len(header))
for _, key := range expectedKeys {
assert.Contains(t, header, key)
}
// Priority columns come first in order, then the rest alphabetically.
assert.Equal(t, []string{"timestamp", "id", "level", "message"}, header)
}
func TestConstructCSVRecordFromQueryResponse(t *testing.T) {

View File

@@ -23,8 +23,18 @@ func NewModule(querier querier.Querier) rawdataexport.Module {
func (m *Module) ExportRawData(ctx context.Context, orgID valuer.UUID, rangeRequest *qbtypes.QueryRangeRequest, doneChan chan any) (chan *qbtypes.RawRow, chan error) {
spec := rangeRequest.CompositeQuery.Queries[0].Spec.(qbtypes.QueryBuilderQuery[qbtypes.LogAggregation])
rowCountLimit := spec.Limit
traceOperatorQueryIndex := rangeRequest.TraceOperatorQueryIndex()
queries := rangeRequest.CompositeQuery.Queries
// If the trace operator query is present, mark the queries other than trace operator as disabled
if traceOperatorQueryIndex > -1 {
for idx := range len(queries) {
if idx != traceOperatorQueryIndex {
queries[idx].SetDisabled(true)
}
}
}
rowChan := make(chan *qbtypes.RawRow, 1)
errChan := make(chan error, 1)
@@ -38,52 +48,62 @@ func (m *Module) ExportRawData(ctx context.Context, orgID valuer.UUID, rangeRequ
defer close(errChan)
defer close(rowChan)
rowCount := 0
for rowCount < rowCountLimit {
spec.Limit = min(ChunkSize, rowCountLimit-rowCount)
spec.Offset = rowCount
rangeRequest.CompositeQuery.Queries[0].Spec = spec
response, err := m.querier.QueryRange(contextWithTimeout, orgID, rangeRequest)
if err != nil {
errChan <- err
return
}
newRowsCount := 0
for _, result := range response.Data.Results {
resultData, ok := result.(*qbtypes.RawData)
if !ok {
errChan <- errors.NewInternalf(errors.CodeInternal, "expected RawData, got %T", result)
return
}
newRowsCount += len(resultData.Rows)
for _, row := range resultData.Rows {
select {
case rowChan <- row:
case <-doneChan:
return
case <-ctx.Done():
errChan <- ctx.Err()
return
}
}
}
// Break if we did not receive any new rows
if newRowsCount == 0 {
return
}
rowCount += newRowsCount
if traceOperatorQueryIndex > -1 {
// If the trace operator query is present, we need to export the data for the trace operator query only
exportRawDataForSingleQuery(m.querier, contextWithTimeout, orgID, rangeRequest, rowChan, errChan, doneChan, traceOperatorQueryIndex)
} else {
// If the trace operator query is not present, we need to export the data for the first query only
exportRawDataForSingleQuery(m.querier, contextWithTimeout, orgID, rangeRequest, rowChan, errChan, doneChan, 0)
}
}()
return rowChan, errChan
}
func exportRawDataForSingleQuery(querier querier.Querier, ctx context.Context, orgID valuer.UUID, rangeRequest *qbtypes.QueryRangeRequest, rowChan chan *qbtypes.RawRow, errChan chan error, doneChan chan any, queryIndex int) {
queries := rangeRequest.CompositeQuery.Queries
rowCountLimit := queries[queryIndex].GetLimit()
rowCount := 0
for rowCount < rowCountLimit {
chunkSize := min(ChunkSize, rowCountLimit-rowCount)
queries[queryIndex].SetLimit(chunkSize)
queries[queryIndex].SetOffset(rowCount)
response, err := querier.QueryRange(ctx, orgID, rangeRequest)
if err != nil {
errChan <- err
return
}
newRowsCount := 0
for _, result := range response.Data.Results {
resultData, ok := result.(*qbtypes.RawData)
if !ok {
errChan <- errors.NewInternalf(errors.CodeInternal, "expected RawData, got %T", result)
return
}
newRowsCount += len(resultData.Rows)
for _, row := range resultData.Rows {
select {
case rowChan <- row:
case <-doneChan:
return
case <-ctx.Done():
errChan <- ctx.Err()
return
}
}
}
rowCount += newRowsCount
// Stop if we received fewer rows than requested — no more data available
if newRowsCount < chunkSize {
return
}
}
}

View File

@@ -572,9 +572,6 @@ func (aH *APIHandler) RegisterRoutes(router *mux.Router, am *middleware.AuthZ) {
aH.LicensingAPI.Activate(rw, req)
})).Methods(http.MethodGet)
// Export
router.HandleFunc("/api/v1/export_raw_data", am.ViewAccess(aH.Signoz.Handlers.RawDataExport.ExportRawData)).Methods(http.MethodGet)
router.HandleFunc("/api/v1/span_percentile", am.ViewAccess(aH.Signoz.Handlers.SpanPercentile.GetSpanPercentileDetails)).Methods(http.MethodPost)
// Query Filter Analyzer api used to extract metric names and grouping columns from a query

View File

@@ -22,6 +22,7 @@ import (
"github.com/SigNoz/signoz/pkg/modules/organization"
"github.com/SigNoz/signoz/pkg/modules/preference"
"github.com/SigNoz/signoz/pkg/modules/promote"
"github.com/SigNoz/signoz/pkg/modules/rawdataexport"
"github.com/SigNoz/signoz/pkg/modules/session"
"github.com/SigNoz/signoz/pkg/modules/user"
"github.com/SigNoz/signoz/pkg/querier"
@@ -57,6 +58,7 @@ func NewOpenAPI(ctx context.Context, instrumentation instrumentation.Instrumenta
struct{ gateway.Handler }{},
struct{ fields.Handler }{},
struct{ authz.Handler }{},
struct{ rawdataexport.Handler }{},
struct{ zeus.Handler }{},
struct{ querier.Handler }{},
).New(ctx, instrumentation.ToProviderSettings(), apiserver.Config{})

View File

@@ -253,6 +253,7 @@ func NewAPIServerProviderFactories(orgGetter organization.Getter, authz authz.Au
handlers.GatewayHandler,
handlers.Fields,
handlers.AuthzHandler,
handlers.RawDataExport,
handlers.ZeusHandler,
handlers.QuerierHandler,
),

View File

@@ -0,0 +1,120 @@
package exporttypes
import (
"strings"
"github.com/SigNoz/signoz/pkg/errors"
qbtypes "github.com/SigNoz/signoz/pkg/types/querybuildertypes/querybuildertypesv5"
"github.com/SigNoz/signoz/pkg/types/telemetrytypes"
"github.com/SigNoz/signoz/pkg/valuer"
)
// ExportRawDataQueryParams represents the query parameters for the export raw data endpoint
type ExportRawDataQueryParams struct {
ExportRawDataFormatQueryParam
// Signal specifies the type of data to export: "logs" or "traces"
Signal telemetrytypes.Signal `query:"signal" enum:"logs,traces" required:"true"`
// Source specifies the type of data to export: "logs" or "traces"
// Deprecated: Use Signal instead.
Source string `query:"source" deprecated:"true"`
// Start is the start time for the query (Unix timestamp in nanoseconds)
Start uint64 `query:"start"`
// End is the end time for the query (Unix timestamp in nanoseconds)
End uint64 `query:"end"`
// Limit specifies the maximum number of rows to export
Limit int `query:"limit,default=10000" default:"10000" minimum:"1" maximum:"50000"`
// Filter is a filter expression to apply to the query
FilterString string `query:"filter" deprecated:"true"`
Filter qbtypes.Filter `query:"filterExpression"`
// Columns specifies the columns to include in the export
// Format: ["context.field:type", "context.field", "field"]
Columns []string `query:"columns" deprecated:"true"`
// SelectFields specifies the columns to include in the export
SelectFields []telemetrytypes.TelemetryFieldKey `query:"selectFields"`
// OrderBy specifies the sorting order
// Format: "column:direction" or "context.field:type:direction"
// Direction can be "asc" or "desc"
// ** Deprecated **
OrderBy string `query:"order_by" deprecated:"true"`
// order by keys and directions
Order []qbtypes.OrderBy `query:"order"`
}
type ExportRawDataFormatQueryParam struct {
// Format specifies the output format: "csv" or "jsonl"
Format string `query:"format,default=csv" default:"csv" enum:"csv,jsonl"`
}
func (p *ExportRawDataQueryParams) Normalize() {
if len(p.Order) == 0 && len(p.OrderBy) > 0 {
p.Order = parseExportQueryOrderBy(p.OrderBy)
}
if len(p.SelectFields) == 0 && len(p.Columns) != 0 {
p.SelectFields = parseExportQueryColumns(p.Columns)
}
if len(p.Filter.Expression) == 0 && len(p.FilterString) > 0 {
p.Filter = qbtypes.Filter{Expression: p.FilterString}
}
if p.Signal == telemetrytypes.SignalUnspecified && p.Source != "" {
p.Signal = telemetrytypes.Signal{String: valuer.NewString(p.Source)}
}
}
func (p *ExportRawDataQueryParams) Validate() error {
if p.Signal != telemetrytypes.SignalLogs && p.Signal != telemetrytypes.SignalTraces {
return errors.NewInvalidInputf(errors.CodeInvalidInput, "invalid signal %s", p.Signal).WithAdditional("Allowed values: [logs, traces]")
}
return nil
}
// parseExportQueryColumns converts bound column strings to TelemetryFieldKey structs.
// Each column should be in the format "context.field:type" or "context.field" or "field"
func parseExportQueryColumns(columnParams []string) []telemetrytypes.TelemetryFieldKey {
columns := make([]telemetrytypes.TelemetryFieldKey, 0, len(columnParams))
for _, columnStr := range columnParams {
columnStr = strings.TrimSpace(columnStr)
if columnStr == "" {
continue
}
columns = append(columns, telemetrytypes.GetFieldKeyFromKeyText(columnStr))
}
return columns
}
// parseExportQueryOrderBy converts a bound order_by string to an OrderBy slice.
// The string should be in the format "column:direction" and is assumed already validated.
func parseExportQueryOrderBy(orderByParam string) []qbtypes.OrderBy {
orderByParam = strings.TrimSpace(orderByParam)
if orderByParam == "" {
return []qbtypes.OrderBy{}
}
parts := strings.Split(orderByParam, ":")
column := strings.Join(parts[:len(parts)-1], ":")
direction := parts[len(parts)-1]
return []qbtypes.OrderBy{
{
Key: qbtypes.OrderByKey{
TelemetryFieldKey: telemetrytypes.GetFieldKeyFromKeyText(column),
},
Direction: qbtypes.OrderDirectionMap[direction],
},
}
}

View File

@@ -0,0 +1,174 @@
package exporttypes
import (
"testing"
qbtypes "github.com/SigNoz/signoz/pkg/types/querybuildertypes/querybuildertypesv5"
"github.com/SigNoz/signoz/pkg/types/telemetrytypes"
"github.com/stretchr/testify/assert"
)
func TestParseExportQueryColumns(t *testing.T) {
tests := []struct {
name string
input []string
expectedColumns []telemetrytypes.TelemetryFieldKey
}{
{
name: "empty input",
input: []string{},
expectedColumns: []telemetrytypes.TelemetryFieldKey{},
},
{
name: "single column",
input: []string{"timestamp"},
expectedColumns: []telemetrytypes.TelemetryFieldKey{
{Name: "timestamp"},
},
},
{
name: "multiple columns",
input: []string{"timestamp", "message", "level"},
expectedColumns: []telemetrytypes.TelemetryFieldKey{
{Name: "timestamp"},
{Name: "message"},
{Name: "level"},
},
},
{
name: "empty entry is skipped",
input: []string{"timestamp", "", "level"},
expectedColumns: []telemetrytypes.TelemetryFieldKey{
{Name: "timestamp"},
{Name: "level"},
},
},
{
name: "whitespace-only entry is skipped",
input: []string{"timestamp", " ", "level"},
expectedColumns: []telemetrytypes.TelemetryFieldKey{
{Name: "timestamp"},
{Name: "level"},
},
},
{
name: "column with context and type",
input: []string{"attribute.user:string"},
expectedColumns: []telemetrytypes.TelemetryFieldKey{
{Name: "user", FieldContext: telemetrytypes.FieldContextAttribute, FieldDataType: telemetrytypes.FieldDataTypeString},
},
},
{
name: "column with context, dot-notation name",
input: []string{"attribute.user.string"},
expectedColumns: []telemetrytypes.TelemetryFieldKey{
{Name: "user.string", FieldContext: telemetrytypes.FieldContextAttribute},
},
},
}
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
columns := parseExportQueryColumns(tt.input)
assert.Equal(t, len(tt.expectedColumns), len(columns))
for i, expected := range tt.expectedColumns {
assert.Equal(t, expected, columns[i])
}
})
}
}
func TestParseExportQueryOrderBy(t *testing.T) {
tests := []struct {
name string
input string
expectedOrder []qbtypes.OrderBy
}{
{
name: "empty string returns empty slice",
input: "",
expectedOrder: []qbtypes.OrderBy{},
},
{
name: "simple column asc",
input: "timestamp:asc",
expectedOrder: []qbtypes.OrderBy{
{
Direction: qbtypes.OrderDirectionAsc,
Key: qbtypes.OrderByKey{
TelemetryFieldKey: telemetrytypes.TelemetryFieldKey{Name: "timestamp"},
},
},
},
},
{
name: "simple column desc",
input: "timestamp:desc",
expectedOrder: []qbtypes.OrderBy{
{
Direction: qbtypes.OrderDirectionDesc,
Key: qbtypes.OrderByKey{
TelemetryFieldKey: telemetrytypes.TelemetryFieldKey{Name: "timestamp"},
},
},
},
},
{
name: "column with context and type qualifier",
input: "attribute.user:string:desc",
expectedOrder: []qbtypes.OrderBy{
{
Direction: qbtypes.OrderDirectionDesc,
Key: qbtypes.OrderByKey{
TelemetryFieldKey: telemetrytypes.TelemetryFieldKey{
Name: "user",
FieldContext: telemetrytypes.FieldContextAttribute,
FieldDataType: telemetrytypes.FieldDataTypeString,
},
},
},
},
},
{
name: "column with context, dot-notation name",
input: "attribute.user.string:desc",
expectedOrder: []qbtypes.OrderBy{
{
Direction: qbtypes.OrderDirectionDesc,
Key: qbtypes.OrderByKey{
TelemetryFieldKey: telemetrytypes.TelemetryFieldKey{
Name: "user.string",
FieldContext: telemetrytypes.FieldContextAttribute,
},
},
},
},
},
{
name: "resource with context and type",
input: "resource.service.name:string:asc",
expectedOrder: []qbtypes.OrderBy{
{
Direction: qbtypes.OrderDirectionAsc,
Key: qbtypes.OrderByKey{
TelemetryFieldKey: telemetrytypes.TelemetryFieldKey{
Name: "service.name",
FieldContext: telemetrytypes.FieldContextResource,
FieldDataType: telemetrytypes.FieldDataTypeString,
},
},
},
},
},
}
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
order := parseExportQueryOrderBy(tt.input)
assert.Equal(t, len(tt.expectedOrder), len(order))
for i, expected := range tt.expectedOrder {
assert.Equal(t, expected, order[i])
}
})
}
}

View File

@@ -393,6 +393,59 @@ func (r *QueryRangeRequest) HasOrderSpecified() bool {
return false
}
// UseDefaultOrderBy applies a default order unless query has an explicit order provided.
func (r *QueryRangeRequest) UseDefaultOrderBy() {
queries := r.CompositeQuery.Queries
for idx := range queries {
switch queries[idx].Spec.(type) {
case QueryBuilderQuery[TraceAggregation],
QueryBuilderTraceOperator:
if len(queries[idx].GetOrder()) == 0 {
queries[idx].SetOrder(
[]OrderBy{
{
Key: OrderByKey{
TelemetryFieldKey: telemetrytypes.TelemetryFieldKey{
Name: "timestamp",
Signal: telemetrytypes.SignalTraces,
FieldContext: telemetrytypes.FieldContextSpan,
FieldDataType: telemetrytypes.FieldDataTypeNumber,
},
},
Direction: OrderDirectionDesc,
},
},
)
}
case QueryBuilderQuery[LogAggregation]:
if len(queries[idx].GetOrder()) == 0 {
queries[idx].SetOrder(
[]OrderBy{
{
Key: OrderByKey{
TelemetryFieldKey: telemetrytypes.TelemetryFieldKey{Name: "timestamp",
Signal: telemetrytypes.SignalLogs,
FieldContext: telemetrytypes.FieldContextLog,
FieldDataType: telemetrytypes.FieldDataTypeNumber},
},
Direction: OrderDirectionDesc,
},
{
Key: OrderByKey{
TelemetryFieldKey: telemetrytypes.TelemetryFieldKey{Name: "id",
Signal: telemetrytypes.SignalLogs,
FieldContext: telemetrytypes.FieldContextLog,
FieldDataType: telemetrytypes.FieldDataTypeString},
},
Direction: OrderDirectionDesc,
},
},
)
}
}
}
}
func (r *QueryRangeRequest) FuncsForQuery(name string) []Function {
funcs := []Function{}
for _, query := range r.CompositeQuery.Queries {
@@ -437,6 +490,16 @@ func (r *QueryRangeRequest) IsAnomalyRequest() (*QueryBuilderQuery[MetricAggrega
return &q, hasAnomaly
}
func (r *QueryRangeRequest) TraceOperatorQueryIndex() int {
for idx, query := range r.CompositeQuery.Queries {
switch query.Spec.(type) {
case TraceOperatorType:
return idx
}
}
return -1
}
// We do not support fill gaps for these queries. Maybe support in future?
func (r *QueryRangeRequest) SkipFillGaps(name string) bool {
for _, query := range r.CompositeQuery.Queries {

View File

@@ -0,0 +1,379 @@
package querybuildertypesv5
import "github.com/SigNoz/signoz/pkg/types/telemetrytypes"
// GetExpression returns the expression string.
func (q *QueryEnvelope) GetExpression() string {
switch spec := q.Spec.(type) {
case QueryBuilderTraceOperator:
return spec.Expression
case QueryBuilderFormula:
return spec.Expression
}
return ""
}
// GetReturnSpansFrom returns the return-spans-from value.
func (q *QueryEnvelope) GetReturnSpansFrom() string {
switch spec := q.Spec.(type) {
case QueryBuilderTraceOperator:
return spec.ReturnSpansFrom
}
return ""
}
// GetSignal returns the signal.
func (q *QueryEnvelope) GetSignal() telemetrytypes.Signal {
switch spec := q.Spec.(type) {
case QueryBuilderQuery[TraceAggregation]:
return spec.Signal
case QueryBuilderQuery[LogAggregation]:
return spec.Signal
case QueryBuilderQuery[MetricAggregation]:
return spec.Signal
}
return telemetrytypes.SignalUnspecified
}
// GetSource returns the source.
func (q *QueryEnvelope) GetSource() telemetrytypes.Source {
switch spec := q.Spec.(type) {
case QueryBuilderQuery[TraceAggregation]:
return spec.Source
case QueryBuilderQuery[LogAggregation]:
return spec.Source
case QueryBuilderQuery[MetricAggregation]:
return spec.Source
}
return telemetrytypes.SourceUnspecified
}
// GetQuery returns the raw query string.
func (q *QueryEnvelope) GetQuery() string {
switch spec := q.Spec.(type) {
case PromQuery:
return spec.Query
case ClickHouseQuery:
return spec.Query
}
return ""
}
// GetStats returns the PromQL stats flag.
func (q *QueryEnvelope) GetStats() bool {
switch spec := q.Spec.(type) {
case PromQuery:
return spec.Stats
}
return false
}
// GetLeft returns the left query reference of a join.
func (q *QueryEnvelope) GetLeft() QueryRef {
switch spec := q.Spec.(type) {
case QueryBuilderJoin:
return spec.Left
}
return QueryRef{}
}
// GetRight returns the right query reference of a join.
func (q *QueryEnvelope) GetRight() QueryRef {
switch spec := q.Spec.(type) {
case QueryBuilderJoin:
return spec.Right
}
return QueryRef{}
}
// GetJoinType returns the join type.
func (q *QueryEnvelope) GetJoinType() JoinType {
switch spec := q.Spec.(type) {
case QueryBuilderJoin:
return spec.Type
}
return JoinType{}
}
// GetOn returns the join ON condition.
func (q *QueryEnvelope) GetOn() string {
switch spec := q.Spec.(type) {
case QueryBuilderJoin:
return spec.On
}
return ""
}
// GetQueryName returns the name of the spec.
func (q *QueryEnvelope) GetQueryName() string {
switch spec := q.Spec.(type) {
case QueryBuilderTraceOperator:
return spec.Name
case QueryBuilderQuery[TraceAggregation]:
return spec.Name
case QueryBuilderQuery[LogAggregation]:
return spec.Name
case QueryBuilderQuery[MetricAggregation]:
return spec.Name
case QueryBuilderFormula:
return spec.Name
case QueryBuilderJoin:
return spec.Name
case PromQuery:
return spec.Name
case ClickHouseQuery:
return spec.Name
}
return ""
}
// IsDisabled returns whether the spec is disabled.
func (q *QueryEnvelope) IsDisabled() bool {
switch spec := q.Spec.(type) {
case QueryBuilderTraceOperator:
return spec.Disabled
case QueryBuilderQuery[TraceAggregation]:
return spec.Disabled
case QueryBuilderQuery[LogAggregation]:
return spec.Disabled
case QueryBuilderQuery[MetricAggregation]:
return spec.Disabled
case QueryBuilderFormula:
return spec.Disabled
case QueryBuilderJoin:
return spec.Disabled
case PromQuery:
return spec.Disabled
case ClickHouseQuery:
return spec.Disabled
}
return false
}
// GetLimit returns the row limit.
func (q *QueryEnvelope) GetLimit() int {
switch spec := q.Spec.(type) {
case QueryBuilderTraceOperator:
return spec.Limit
case QueryBuilderQuery[TraceAggregation]:
return spec.Limit
case QueryBuilderQuery[LogAggregation]:
return spec.Limit
case QueryBuilderQuery[MetricAggregation]:
return spec.Limit
case QueryBuilderFormula:
return spec.Limit
case QueryBuilderJoin:
return spec.Limit
}
return 0
}
// GetOffset returns the row offset.
func (q *QueryEnvelope) GetOffset() int {
switch spec := q.Spec.(type) {
case QueryBuilderTraceOperator:
return spec.Offset
case QueryBuilderQuery[TraceAggregation]:
return spec.Offset
case QueryBuilderQuery[LogAggregation]:
return spec.Offset
case QueryBuilderQuery[MetricAggregation]:
return spec.Offset
}
return 0
}
// GetType returns the QueryType of the envelope.
func (q *QueryEnvelope) GetType() QueryType {
return q.Type
}
// GetOrder returns the order-by clauses.
func (q *QueryEnvelope) GetOrder() []OrderBy {
switch spec := q.Spec.(type) {
case QueryBuilderTraceOperator:
return spec.Order
case QueryBuilderQuery[TraceAggregation]:
return spec.Order
case QueryBuilderQuery[LogAggregation]:
return spec.Order
case QueryBuilderQuery[MetricAggregation]:
return spec.Order
case QueryBuilderFormula:
return spec.Order
case QueryBuilderJoin:
return spec.Order
}
return nil
}
// GetGroupBy returns the group-by keys.
func (q *QueryEnvelope) GetGroupBy() []GroupByKey {
switch spec := q.Spec.(type) {
case QueryBuilderTraceOperator:
return spec.GroupBy
case QueryBuilderQuery[TraceAggregation]:
return spec.GroupBy
case QueryBuilderQuery[LogAggregation]:
return spec.GroupBy
case QueryBuilderQuery[MetricAggregation]:
return spec.GroupBy
case QueryBuilderJoin:
return spec.GroupBy
}
return nil
}
// GetFilter returns the filter.
func (q *QueryEnvelope) GetFilter() *Filter {
switch spec := q.Spec.(type) {
case QueryBuilderTraceOperator:
return spec.Filter
case QueryBuilderQuery[TraceAggregation]:
return spec.Filter
case QueryBuilderQuery[LogAggregation]:
return spec.Filter
case QueryBuilderQuery[MetricAggregation]:
return spec.Filter
case QueryBuilderJoin:
return spec.Filter
}
return nil
}
// GetHaving returns the having clause.
func (q *QueryEnvelope) GetHaving() *Having {
switch spec := q.Spec.(type) {
case QueryBuilderTraceOperator:
return spec.Having
case QueryBuilderQuery[TraceAggregation]:
return spec.Having
case QueryBuilderQuery[LogAggregation]:
return spec.Having
case QueryBuilderQuery[MetricAggregation]:
return spec.Having
case QueryBuilderFormula:
return spec.Having
case QueryBuilderJoin:
return spec.Having
}
return nil
}
// GetFunctions returns the post-processing functions.
func (q *QueryEnvelope) GetFunctions() []Function {
switch spec := q.Spec.(type) {
case QueryBuilderTraceOperator:
return spec.Functions
case QueryBuilderQuery[TraceAggregation]:
return spec.Functions
case QueryBuilderQuery[LogAggregation]:
return spec.Functions
case QueryBuilderQuery[MetricAggregation]:
return spec.Functions
case QueryBuilderFormula:
return spec.Functions
case QueryBuilderJoin:
return spec.Functions
}
return nil
}
// GetSelectFields returns the selected fields.
func (q *QueryEnvelope) GetSelectFields() []telemetrytypes.TelemetryFieldKey {
switch spec := q.Spec.(type) {
case QueryBuilderTraceOperator:
return spec.SelectFields
case QueryBuilderQuery[TraceAggregation]:
return spec.SelectFields
case QueryBuilderQuery[LogAggregation]:
return spec.SelectFields
case QueryBuilderQuery[MetricAggregation]:
return spec.SelectFields
case QueryBuilderJoin:
return spec.SelectFields
}
return nil
}
// GetLegend returns the legend label.
func (q *QueryEnvelope) GetLegend() string {
switch spec := q.Spec.(type) {
case QueryBuilderTraceOperator:
return spec.Legend
case QueryBuilderQuery[TraceAggregation]:
return spec.Legend
case QueryBuilderQuery[LogAggregation]:
return spec.Legend
case QueryBuilderQuery[MetricAggregation]:
return spec.Legend
case QueryBuilderFormula:
return spec.Legend
case PromQuery:
return spec.Legend
case ClickHouseQuery:
return spec.Legend
}
return ""
}
// GetCursor returns the pagination cursor.
func (q *QueryEnvelope) GetCursor() string {
switch spec := q.Spec.(type) {
case QueryBuilderTraceOperator:
return spec.Cursor
case QueryBuilderQuery[TraceAggregation]:
return spec.Cursor
case QueryBuilderQuery[LogAggregation]:
return spec.Cursor
case QueryBuilderQuery[MetricAggregation]:
return spec.Cursor
}
return ""
}
// GetStepInterval returns the step interval.
func (q *QueryEnvelope) GetStepInterval() Step {
switch spec := q.Spec.(type) {
case QueryBuilderTraceOperator:
return spec.StepInterval
case QueryBuilderQuery[TraceAggregation]:
return spec.StepInterval
case QueryBuilderQuery[LogAggregation]:
return spec.StepInterval
case QueryBuilderQuery[MetricAggregation]:
return spec.StepInterval
case PromQuery:
return spec.Step
}
return Step{}
}
// GetSecondaryAggregations returns the secondary aggregations.
func (q *QueryEnvelope) GetSecondaryAggregations() []SecondaryAggregation {
switch spec := q.Spec.(type) {
case QueryBuilderQuery[TraceAggregation]:
return spec.SecondaryAggregations
case QueryBuilderQuery[LogAggregation]:
return spec.SecondaryAggregations
case QueryBuilderQuery[MetricAggregation]:
return spec.SecondaryAggregations
case QueryBuilderJoin:
return spec.SecondaryAggregations
}
return nil
}
// GetLimitBy returns the limit-by configuration.
func (q *QueryEnvelope) GetLimitBy() *LimitBy {
switch spec := q.Spec.(type) {
case QueryBuilderQuery[TraceAggregation]:
return spec.LimitBy
case QueryBuilderQuery[LogAggregation]:
return spec.LimitBy
case QueryBuilderQuery[MetricAggregation]:
return spec.LimitBy
}
return nil
}

View File

@@ -0,0 +1,452 @@
package querybuildertypesv5
import "github.com/SigNoz/signoz/pkg/types/telemetrytypes"
// SetExpression sets the expression string of the spec, if applicable.
func (q *QueryEnvelope) SetExpression(expression string) {
switch spec := q.Spec.(type) {
case QueryBuilderTraceOperator:
spec.Expression = expression
q.Spec = spec
case QueryBuilderFormula:
spec.Expression = expression
q.Spec = spec
}
}
// SetReturnSpansFrom sets the return-spans-from value, if applicable.
func (q *QueryEnvelope) SetReturnSpansFrom(returnSpansFrom string) {
switch spec := q.Spec.(type) {
case QueryBuilderTraceOperator:
spec.ReturnSpansFrom = returnSpansFrom
q.Spec = spec
}
}
// SetSignal sets the signal of the spec, if applicable.
func (q *QueryEnvelope) SetSignal(signal telemetrytypes.Signal) {
switch spec := q.Spec.(type) {
case QueryBuilderQuery[TraceAggregation]:
spec.Signal = signal
q.Spec = spec
case QueryBuilderQuery[LogAggregation]:
spec.Signal = signal
q.Spec = spec
case QueryBuilderQuery[MetricAggregation]:
spec.Signal = signal
q.Spec = spec
}
}
// SetSource sets the source of the spec, if applicable.
func (q *QueryEnvelope) SetSource(source telemetrytypes.Source) {
switch spec := q.Spec.(type) {
case QueryBuilderQuery[TraceAggregation]:
spec.Source = source
q.Spec = spec
case QueryBuilderQuery[LogAggregation]:
spec.Source = source
q.Spec = spec
case QueryBuilderQuery[MetricAggregation]:
spec.Source = source
q.Spec = spec
}
}
// SetQuery sets the raw query string of the spec, if applicable.
func (q *QueryEnvelope) SetQuery(query string) {
switch spec := q.Spec.(type) {
case PromQuery:
spec.Query = query
q.Spec = spec
case ClickHouseQuery:
spec.Query = query
q.Spec = spec
}
}
// SetStats sets the PromQL stats flag, if applicable.
func (q *QueryEnvelope) SetStats(stats bool) {
switch spec := q.Spec.(type) {
case PromQuery:
spec.Stats = stats
q.Spec = spec
}
}
// SetLeft sets the left query reference of a join, if applicable.
func (q *QueryEnvelope) SetLeft(left QueryRef) {
switch spec := q.Spec.(type) {
case QueryBuilderJoin:
spec.Left = left
q.Spec = spec
}
}
// SetRight sets the right query reference of a join, if applicable.
func (q *QueryEnvelope) SetRight(right QueryRef) {
switch spec := q.Spec.(type) {
case QueryBuilderJoin:
spec.Right = right
q.Spec = spec
}
}
// SetJoinType sets the join type, if applicable.
func (q *QueryEnvelope) SetJoinType(joinType JoinType) {
switch spec := q.Spec.(type) {
case QueryBuilderJoin:
spec.Type = joinType
q.Spec = spec
}
}
// SetOn sets the join ON condition, if applicable.
func (q *QueryEnvelope) SetOn(on string) {
switch spec := q.Spec.(type) {
case QueryBuilderJoin:
spec.On = on
q.Spec = spec
}
}
// SetQueryName sets the name of the spec, if applicable.
func (q *QueryEnvelope) SetQueryName(name string) {
switch spec := q.Spec.(type) {
case QueryBuilderTraceOperator:
spec.Name = name
q.Spec = spec
case QueryBuilderQuery[TraceAggregation]:
spec.Name = name
q.Spec = spec
case QueryBuilderQuery[LogAggregation]:
spec.Name = name
q.Spec = spec
case QueryBuilderQuery[MetricAggregation]:
spec.Name = name
q.Spec = spec
case QueryBuilderFormula:
spec.Name = name
q.Spec = spec
case QueryBuilderJoin:
spec.Name = name
q.Spec = spec
case PromQuery:
spec.Name = name
q.Spec = spec
case ClickHouseQuery:
spec.Name = name
q.Spec = spec
}
}
// SetDisabled sets the disabled flag of the spec, if applicable.
func (q *QueryEnvelope) SetDisabled(disabled bool) {
switch spec := q.Spec.(type) {
case QueryBuilderTraceOperator:
spec.Disabled = disabled
q.Spec = spec
case QueryBuilderQuery[TraceAggregation]:
spec.Disabled = disabled
q.Spec = spec
case QueryBuilderQuery[LogAggregation]:
spec.Disabled = disabled
q.Spec = spec
case QueryBuilderQuery[MetricAggregation]:
spec.Disabled = disabled
q.Spec = spec
case QueryBuilderFormula:
spec.Disabled = disabled
q.Spec = spec
case QueryBuilderJoin:
spec.Disabled = disabled
q.Spec = spec
case PromQuery:
spec.Disabled = disabled
q.Spec = spec
case ClickHouseQuery:
spec.Disabled = disabled
q.Spec = spec
}
}
// SetLimit sets the row limit of the spec, if applicable.
func (q *QueryEnvelope) SetLimit(limit int) {
switch spec := q.Spec.(type) {
case QueryBuilderTraceOperator:
spec.Limit = limit
q.Spec = spec
case QueryBuilderQuery[TraceAggregation]:
spec.Limit = limit
q.Spec = spec
case QueryBuilderQuery[LogAggregation]:
spec.Limit = limit
q.Spec = spec
case QueryBuilderQuery[MetricAggregation]:
spec.Limit = limit
q.Spec = spec
case QueryBuilderFormula:
spec.Limit = limit
q.Spec = spec
case QueryBuilderJoin:
spec.Limit = limit
q.Spec = spec
}
}
// SetOffset sets the row offset of the spec, if applicable.
func (q *QueryEnvelope) SetOffset(offset int) {
switch spec := q.Spec.(type) {
case QueryBuilderTraceOperator:
spec.Offset = offset
q.Spec = spec
case QueryBuilderQuery[TraceAggregation]:
spec.Offset = offset
q.Spec = spec
case QueryBuilderQuery[LogAggregation]:
spec.Offset = offset
q.Spec = spec
case QueryBuilderQuery[MetricAggregation]:
spec.Offset = offset
q.Spec = spec
}
}
// SetType sets the QueryType of the envelope.
func (q *QueryEnvelope) SetType(t QueryType) {
q.Type = t
}
// SetOrder sets the order-by clauses of the spec, if applicable.
func (q *QueryEnvelope) SetOrder(order []OrderBy) {
switch spec := q.Spec.(type) {
case QueryBuilderTraceOperator:
spec.Order = order
q.Spec = spec
case QueryBuilderQuery[TraceAggregation]:
spec.Order = order
q.Spec = spec
case QueryBuilderQuery[LogAggregation]:
spec.Order = order
q.Spec = spec
case QueryBuilderQuery[MetricAggregation]:
spec.Order = order
q.Spec = spec
case QueryBuilderFormula:
spec.Order = order
q.Spec = spec
case QueryBuilderJoin:
spec.Order = order
q.Spec = spec
}
}
// SetGroupBy sets the group-by keys of the spec, if applicable.
func (q *QueryEnvelope) SetGroupBy(groupBy []GroupByKey) {
switch spec := q.Spec.(type) {
case QueryBuilderTraceOperator:
spec.GroupBy = groupBy
q.Spec = spec
case QueryBuilderQuery[TraceAggregation]:
spec.GroupBy = groupBy
q.Spec = spec
case QueryBuilderQuery[LogAggregation]:
spec.GroupBy = groupBy
q.Spec = spec
case QueryBuilderQuery[MetricAggregation]:
spec.GroupBy = groupBy
q.Spec = spec
case QueryBuilderJoin:
spec.GroupBy = groupBy
q.Spec = spec
}
}
// SetFilter sets the filter of the spec, if applicable.
func (q *QueryEnvelope) SetFilter(filter *Filter) {
switch spec := q.Spec.(type) {
case QueryBuilderTraceOperator:
spec.Filter = filter
q.Spec = spec
case QueryBuilderQuery[TraceAggregation]:
spec.Filter = filter
q.Spec = spec
case QueryBuilderQuery[LogAggregation]:
spec.Filter = filter
q.Spec = spec
case QueryBuilderQuery[MetricAggregation]:
spec.Filter = filter
q.Spec = spec
case QueryBuilderJoin:
spec.Filter = filter
q.Spec = spec
}
}
// SetHaving sets the having clause of the spec, if applicable.
func (q *QueryEnvelope) SetHaving(having *Having) {
switch spec := q.Spec.(type) {
case QueryBuilderTraceOperator:
spec.Having = having
q.Spec = spec
case QueryBuilderQuery[TraceAggregation]:
spec.Having = having
q.Spec = spec
case QueryBuilderQuery[LogAggregation]:
spec.Having = having
q.Spec = spec
case QueryBuilderQuery[MetricAggregation]:
spec.Having = having
q.Spec = spec
case QueryBuilderFormula:
spec.Having = having
q.Spec = spec
case QueryBuilderJoin:
spec.Having = having
q.Spec = spec
}
}
// SetFunctions sets the post-processing functions of the spec, if applicable.
func (q *QueryEnvelope) SetFunctions(functions []Function) {
switch spec := q.Spec.(type) {
case QueryBuilderTraceOperator:
spec.Functions = functions
q.Spec = spec
case QueryBuilderQuery[TraceAggregation]:
spec.Functions = functions
q.Spec = spec
case QueryBuilderQuery[LogAggregation]:
spec.Functions = functions
q.Spec = spec
case QueryBuilderQuery[MetricAggregation]:
spec.Functions = functions
q.Spec = spec
case QueryBuilderFormula:
spec.Functions = functions
q.Spec = spec
case QueryBuilderJoin:
spec.Functions = functions
q.Spec = spec
}
}
// SetSelectFields sets the selected fields of the spec, if applicable.
func (q *QueryEnvelope) SetSelectFields(fields []telemetrytypes.TelemetryFieldKey) {
switch spec := q.Spec.(type) {
case QueryBuilderTraceOperator:
spec.SelectFields = fields
q.Spec = spec
case QueryBuilderQuery[TraceAggregation]:
spec.SelectFields = fields
q.Spec = spec
case QueryBuilderQuery[LogAggregation]:
spec.SelectFields = fields
q.Spec = spec
case QueryBuilderQuery[MetricAggregation]:
spec.SelectFields = fields
q.Spec = spec
case QueryBuilderJoin:
spec.SelectFields = fields
q.Spec = spec
}
}
// SetLegend sets the legend label of the spec, if applicable.
func (q *QueryEnvelope) SetLegend(legend string) {
switch spec := q.Spec.(type) {
case QueryBuilderTraceOperator:
spec.Legend = legend
q.Spec = spec
case QueryBuilderQuery[TraceAggregation]:
spec.Legend = legend
q.Spec = spec
case QueryBuilderQuery[LogAggregation]:
spec.Legend = legend
q.Spec = spec
case QueryBuilderQuery[MetricAggregation]:
spec.Legend = legend
q.Spec = spec
case QueryBuilderFormula:
spec.Legend = legend
q.Spec = spec
case PromQuery:
spec.Legend = legend
q.Spec = spec
case ClickHouseQuery:
spec.Legend = legend
q.Spec = spec
}
}
// SetCursor sets the pagination cursor of the spec, if applicable.
func (q *QueryEnvelope) SetCursor(cursor string) {
switch spec := q.Spec.(type) {
case QueryBuilderTraceOperator:
spec.Cursor = cursor
q.Spec = spec
case QueryBuilderQuery[TraceAggregation]:
spec.Cursor = cursor
q.Spec = spec
case QueryBuilderQuery[LogAggregation]:
spec.Cursor = cursor
q.Spec = spec
case QueryBuilderQuery[MetricAggregation]:
spec.Cursor = cursor
q.Spec = spec
}
}
// SetStepInterval sets the step interval of the spec, if applicable.
func (q *QueryEnvelope) SetStepInterval(step Step) {
switch spec := q.Spec.(type) {
case QueryBuilderTraceOperator:
spec.StepInterval = step
q.Spec = spec
case QueryBuilderQuery[TraceAggregation]:
spec.StepInterval = step
q.Spec = spec
case QueryBuilderQuery[LogAggregation]:
spec.StepInterval = step
q.Spec = spec
case QueryBuilderQuery[MetricAggregation]:
spec.StepInterval = step
q.Spec = spec
case PromQuery:
spec.Step = step
q.Spec = spec
}
}
// SetSecondaryAggregations sets the secondary aggregations of the spec, if applicable.
func (q *QueryEnvelope) SetSecondaryAggregations(secondaryAggregations []SecondaryAggregation) {
switch spec := q.Spec.(type) {
case QueryBuilderQuery[TraceAggregation]:
spec.SecondaryAggregations = secondaryAggregations
q.Spec = spec
case QueryBuilderQuery[LogAggregation]:
spec.SecondaryAggregations = secondaryAggregations
q.Spec = spec
case QueryBuilderQuery[MetricAggregation]:
spec.SecondaryAggregations = secondaryAggregations
q.Spec = spec
case QueryBuilderJoin:
spec.SecondaryAggregations = secondaryAggregations
q.Spec = spec
}
}
// SetLimitBy sets the limit-by configuration of the spec, if applicable.
func (q *QueryEnvelope) SetLimitBy(limitBy *LimitBy) {
switch spec := q.Spec.(type) {
case QueryBuilderQuery[TraceAggregation]:
spec.LimitBy = limitBy
q.Spec = spec
case QueryBuilderQuery[LogAggregation]:
spec.LimitBy = limitBy
q.Spec = spec
case QueryBuilderQuery[MetricAggregation]:
spec.LimitBy = limitBy
q.Spec = spec
}
}

View File

@@ -10,55 +10,9 @@ import (
"github.com/SigNoz/signoz/pkg/types/telemetrytypes"
)
// queryName returns the name from any query envelope spec type.
func (e QueryEnvelope) queryName() string {
switch spec := e.Spec.(type) {
case QueryBuilderQuery[TraceAggregation]:
return spec.Name
case QueryBuilderQuery[LogAggregation]:
return spec.Name
case QueryBuilderQuery[MetricAggregation]:
return spec.Name
case QueryBuilderFormula:
return spec.Name
case QueryBuilderTraceOperator:
return spec.Name
case QueryBuilderJoin:
return spec.Name
case PromQuery:
return spec.Name
case ClickHouseQuery:
return spec.Name
}
return ""
}
// isDisabled returns the disabled status from any query envelope spec type.
func (e QueryEnvelope) isDisabled() bool {
switch spec := e.Spec.(type) {
case QueryBuilderQuery[TraceAggregation]:
return spec.Disabled
case QueryBuilderQuery[LogAggregation]:
return spec.Disabled
case QueryBuilderQuery[MetricAggregation]:
return spec.Disabled
case QueryBuilderFormula:
return spec.Disabled
case QueryBuilderTraceOperator:
return spec.Disabled
case QueryBuilderJoin:
return spec.Disabled
case PromQuery:
return spec.Disabled
case ClickHouseQuery:
return spec.Disabled
}
return false
}
// getQueryIdentifier returns a friendly identifier for a query based on its type and name/content
func getQueryIdentifier(envelope QueryEnvelope, index int) string {
name := envelope.queryName()
name := envelope.GetQueryName()
var typeLabel string
switch envelope.Type {
@@ -89,50 +43,115 @@ const (
MaxQueryLimit = 10000
)
// Validate performs preliminary validation on QueryBuilderQuery
func (q *QueryBuilderQuery[T]) Validate(requestType RequestType) error {
// Validate signal
// ValidationOption is a functional option for configuring validation behaviour.
type ValidationOption func(*validationConfig)
type validationConfig struct {
skipLimitValidation bool
skipAggregationValidation bool
skipHavingValidation bool
skipAggregationOrderBy bool
skipSelectFieldValidation bool
skipGroupByValidation bool
}
func applyValidationOptions(opts []ValidationOption) validationConfig {
cfg := validationConfig{}
for _, opt := range opts {
opt(&cfg)
}
return cfg
}
// SkipLimitValidation returns a ValidationOption that skips the limit range check.
// Use this when the caller has already validated limits with different constraints.
func WithSkipLimitValidation() ValidationOption {
return func(cfg *validationConfig) {
cfg.skipLimitValidation = true
}
}
// SkipAggregationValidation skips aggregation validation.
// Used for raw/trace request types where aggregations are not required.
func WithSkipAggregationValidation() ValidationOption {
return func(cfg *validationConfig) {
cfg.skipAggregationValidation = true
}
}
// SkipHavingValidation skips having-clause validation.
// Used for raw/trace request types where having clauses do not apply.
func WithSkipHavingValidation() ValidationOption {
return func(cfg *validationConfig) {
cfg.skipHavingValidation = true
}
}
// SkipAggregationOrderBy skips the aggregation-specific order-by key validation.
// Used for raw/trace request types where order-by keys are not restricted to group-by or aggregation keys.
func WithSkipAggregationOrderBy() ValidationOption {
return func(cfg *validationConfig) {
cfg.skipAggregationOrderBy = true
}
}
// SkipSelectFieldValidation skips select-field validation.
// Used for aggregation request types where select fields do not apply.
func WithSkipSelectFieldValidation() ValidationOption {
return func(cfg *validationConfig) {
cfg.skipSelectFieldValidation = true
}
}
// SkipGroupByValidation skips group-by validation.
// Used for raw/trace request types where group-by does not apply.
func WithSkipGroupByValidation() ValidationOption {
return func(cfg *validationConfig) {
cfg.skipGroupByValidation = true
}
}
// Validate performs preliminary validation on QueryBuilderQuery.
func (q *QueryBuilderQuery[T]) Validate(opts ...ValidationOption) error {
cfg := applyValidationOptions(opts)
if err := q.validateSignal(); err != nil {
return err
}
if err := q.validateAggregations(requestType); err != nil {
if err := q.validateAggregations(cfg); err != nil {
return err
}
if err := q.validateGroupBy(requestType); err != nil {
if err := q.validateGroupBy(cfg); err != nil {
return err
}
// Validate limit and pagination
if err := q.validateLimitAndPagination(); err != nil {
if err := q.validateLimitAndPagination(cfg); err != nil {
return err
}
// Validate functions
if err := q.validateFunctions(); err != nil {
return err
}
// Validate secondary aggregations
if err := q.validateSecondaryAggregations(); err != nil {
return err
}
if err := q.validateOrderBy(requestType); err != nil {
if err := q.validateOrderBy(cfg); err != nil {
return err
}
if err := q.validateSelectFields(requestType); err != nil {
if err := q.validateSelectFields(cfg); err != nil {
return err
}
return nil
}
func (q *QueryBuilderQuery[T]) validateSelectFields(requestType RequestType) error {
// selectFields don't apply to aggregation queries, skip validation
if requestType.IsAggregation() {
func (q *QueryBuilderQuery[T]) validateSelectFields(cfg validationConfig) error {
if cfg.skipSelectFieldValidation {
return nil
}
@@ -148,9 +167,8 @@ func (q *QueryBuilderQuery[T]) validateSelectFields(requestType RequestType) err
return nil
}
func (q *QueryBuilderQuery[T]) validateGroupBy(requestType RequestType) error {
// groupBy doesn't apply to non-aggregation queries, skip validation
if !requestType.IsAggregation() {
func (q *QueryBuilderQuery[T]) validateGroupBy(cfg validationConfig) error {
if cfg.skipGroupByValidation {
return nil
}
for idx, item := range q.GroupBy {
@@ -183,9 +201,8 @@ func (q *QueryBuilderQuery[T]) validateSignal() error {
}
}
func (q *QueryBuilderQuery[T]) validateAggregations(requestType RequestType) error {
// aggregations don't apply to non-aggregation queries, skip validation
if !requestType.IsAggregation() {
func (q *QueryBuilderQuery[T]) validateAggregations(cfg validationConfig) error {
if cfg.skipAggregationValidation {
return nil
}
@@ -265,24 +282,25 @@ func (q *QueryBuilderQuery[T]) validateAggregations(requestType RequestType) err
return nil
}
func (q *QueryBuilderQuery[T]) validateLimitAndPagination() error {
// Validate limit
if q.Limit < 0 {
return errors.NewInvalidInputf(
errors.CodeInvalidInput,
"limit must be non-negative, got %d",
q.Limit,
)
}
func (q *QueryBuilderQuery[T]) validateLimitAndPagination(cfg validationConfig) error {
if !cfg.skipLimitValidation {
if q.Limit < 0 {
return errors.NewInvalidInputf(
errors.CodeInvalidInput,
"limit must be non-negative, got %d",
q.Limit,
)
}
if q.Limit > MaxQueryLimit {
return errors.NewInvalidInputf(
errors.CodeInvalidInput,
"limit exceeds maximum allowed value of %d",
MaxQueryLimit,
).WithAdditional(
fmt.Sprintf("Provided limit: %d", q.Limit),
)
if q.Limit > MaxQueryLimit {
return errors.NewInvalidInputf(
errors.CodeInvalidInput,
"limit exceeds maximum allowed value of %d",
MaxQueryLimit,
).WithAdditional(
fmt.Sprintf("Provided limit: %d", q.Limit),
)
}
}
// Validate offset
@@ -329,7 +347,7 @@ func (q *QueryBuilderQuery[T]) validateSecondaryAggregations() error {
return nil
}
func (q *QueryBuilderQuery[T]) validateOrderBy(requestType RequestType) error {
func (q *QueryBuilderQuery[T]) validateOrderBy(cfg validationConfig) error {
for i, order := range q.Order {
// Direction validation is handled by the OrderDirection type
if order.Direction != OrderDirectionAsc && order.Direction != OrderDirectionDesc {
@@ -348,8 +366,7 @@ func (q *QueryBuilderQuery[T]) validateOrderBy(requestType RequestType) error {
}
}
// aggregation-specific order key validation only applies to aggregation queries
if requestType.IsAggregation() {
if !cfg.skipAggregationOrderBy {
return q.validateOrderByForAggregation()
}
@@ -431,8 +448,8 @@ func (q *QueryBuilderQuery[T]) validateOrderByForAggregation() error {
return nil
}
// ValidateQueryRangeRequest validates the entire query range request
func (r *QueryRangeRequest) Validate() error {
// Validate validates the entire query range request.
func (r *QueryRangeRequest) Validate(opts ...ValidationOption) error {
// Validate time range
if r.RequestType != RequestTypeRawStream && r.Start >= r.End {
return errors.NewInvalidInputf(
@@ -443,8 +460,10 @@ func (r *QueryRangeRequest) Validate() error {
// Validate request type
switch r.RequestType {
case RequestTypeRaw, RequestTypeRawStream, RequestTypeTimeSeries, RequestTypeScalar, RequestTypeTrace:
// Valid request types
case RequestTypeRaw, RequestTypeRawStream, RequestTypeTrace:
opts = append(opts, getValidationOptions(false)...)
case RequestTypeTimeSeries, RequestTypeScalar:
opts = append(opts, getValidationOptions(true)...)
default:
return errors.NewInvalidInputf(
errors.CodeInvalidInput,
@@ -456,7 +475,7 @@ func (r *QueryRangeRequest) Validate() error {
}
// Validate composite query
if err := r.validateCompositeQuery(); err != nil {
if err := r.CompositeQuery.Validate(opts...); err != nil {
return err
}
@@ -471,7 +490,7 @@ func (r *QueryRangeRequest) Validate() error {
// validateAllQueriesNotDisabled validates that at least one query in the composite query is enabled
func (r *QueryRangeRequest) validateAllQueriesNotDisabled() error {
for _, envelope := range r.CompositeQuery.Queries {
if !envelope.isDisabled() {
if !envelope.IsDisabled() {
return nil
}
}
@@ -482,12 +501,8 @@ func (r *QueryRangeRequest) validateAllQueriesNotDisabled() error {
)
}
func (r *QueryRangeRequest) validateCompositeQuery() error {
return r.CompositeQuery.Validate(r.RequestType)
}
// Validate performs validation on CompositeQuery
func (c *CompositeQuery) Validate(requestType RequestType) error {
func (c *CompositeQuery) Validate(opts ...ValidationOption) error {
if len(c.Queries) == 0 {
return errors.NewInvalidInputf(
errors.CodeInvalidInput,
@@ -499,14 +514,14 @@ func (c *CompositeQuery) Validate(requestType RequestType) error {
queryNames := make(map[string]bool)
for i, envelope := range c.Queries {
if err := validateQueryEnvelope(envelope, requestType); err != nil {
if err := validateQueryEnvelope(envelope, opts...); err != nil {
queryId := getQueryIdentifier(envelope, i)
return wrapValidationError(err, queryId, "invalid %s: %s")
}
// Check name uniqueness for builder queries
if envelope.Type == QueryTypeBuilder || envelope.Type == QueryTypeSubQuery {
name := envelope.queryName()
name := envelope.GetQueryName()
if name != "" {
if queryNames[name] {
return errors.NewInvalidInputf(
@@ -523,16 +538,16 @@ func (c *CompositeQuery) Validate(requestType RequestType) error {
return nil
}
func validateQueryEnvelope(envelope QueryEnvelope, requestType RequestType) error {
func validateQueryEnvelope(envelope QueryEnvelope, opts ...ValidationOption) error {
switch envelope.Type {
case QueryTypeBuilder, QueryTypeSubQuery:
switch spec := envelope.Spec.(type) {
case QueryBuilderQuery[TraceAggregation]:
return spec.Validate(requestType)
return spec.Validate(opts...)
case QueryBuilderQuery[LogAggregation]:
return spec.Validate(requestType)
return spec.Validate(opts...)
case QueryBuilderQuery[MetricAggregation]:
return spec.Validate(requestType)
return spec.Validate(opts...)
default:
return errors.NewInvalidInputf(
errors.CodeInvalidInput,
@@ -618,3 +633,11 @@ func validateQueryEnvelope(envelope QueryEnvelope, requestType RequestType) erro
)
}
}
func getValidationOptions(isAggregationQuery bool) []ValidationOption {
if isAggregationQuery {
return []ValidationOption{WithSkipSelectFieldValidation()}
}
return []ValidationOption{WithSkipAggregationValidation(), WithSkipHavingValidation(), WithSkipAggregationOrderBy(), WithSkipGroupByValidation()}
}

View File

@@ -743,7 +743,7 @@ func TestValidateQueryEnvelope(t *testing.T) {
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
err := validateQueryEnvelope(tt.envelope, tt.requestType)
err := validateQueryEnvelope(tt.envelope)
if tt.wantErr {
if err == nil {
t.Errorf("validateQueryEnvelope() expected error but got none")
@@ -816,7 +816,7 @@ func TestQueryEnvelope_Helpers(t *testing.T) {
}
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
got := tt.envelope.queryName()
got := tt.envelope.GetQueryName()
if got != tt.want {
t.Errorf("queryName() = %q, want %q", got, tt.want)
}
@@ -868,7 +868,7 @@ func TestQueryEnvelope_Helpers(t *testing.T) {
}
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
got := tt.envelope.isDisabled()
got := tt.envelope.IsDisabled()
if got != tt.want {
t.Errorf("isDisabled() = %v, want %v", got, tt.want)
}
@@ -1107,7 +1107,7 @@ func TestQueryRangeRequest_ValidateOrderByForAggregation(t *testing.T) {
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
err := tt.query.Validate(RequestTypeTimeSeries)
err := tt.query.Validate(getValidationOptions(true)...)
if tt.wantErr {
if err == nil {
t.Errorf("validateOrderByForAggregation() expected error but got none")
@@ -1161,7 +1161,7 @@ func TestNonAggregationFieldsSkipped(t *testing.T) {
{TelemetryFieldKey: telemetrytypes.TelemetryFieldKey{Name: "service.name"}},
},
}
err := query.Validate(RequestTypeRaw)
err := query.Validate(getValidationOptions(false)...)
if err != nil {
t.Errorf("expected no error for groupBy with raw request type, got: %v", err)
}
@@ -1178,7 +1178,7 @@ func TestNonAggregationFieldsSkipped(t *testing.T) {
{TelemetryFieldKey: telemetrytypes.TelemetryFieldKey{Name: ""}},
},
}
err := query.Validate(RequestTypeTimeSeries)
err := query.Validate(getValidationOptions(true)...)
if err == nil {
t.Errorf("expected error for empty groupBy key with timeseries request type")
}
@@ -1190,7 +1190,7 @@ func TestNonAggregationFieldsSkipped(t *testing.T) {
Signal: telemetrytypes.SignalLogs,
Having: &Having{Expression: "count() > 10"},
}
err := query.Validate(RequestTypeRaw)
err := query.Validate(getValidationOptions(false)...)
if err != nil {
t.Errorf("expected no error for having with raw request type, got: %v", err)
}
@@ -1202,7 +1202,7 @@ func TestNonAggregationFieldsSkipped(t *testing.T) {
Signal: telemetrytypes.SignalTraces,
Having: &Having{Expression: "count() > 10"},
}
err := query.Validate(RequestTypeTrace)
err := query.Validate(getValidationOptions(false)...)
if err != nil {
t.Errorf("expected no error for having with trace request type, got: %v", err)
}
@@ -1216,7 +1216,7 @@ func TestNonAggregationFieldsSkipped(t *testing.T) {
{Expression: "count()"},
},
}
err := query.Validate(RequestTypeRaw)
err := query.Validate(getValidationOptions(false)...)
if err != nil {
t.Errorf("expected no error for aggregations with raw request type, got: %v", err)
}
@@ -1230,7 +1230,7 @@ func TestNonAggregationFieldsSkipped(t *testing.T) {
{Expression: "count()"},
},
}
err := query.Validate(RequestTypeRawStream)
err := query.Validate(getValidationOptions(false)...)
if err != nil {
t.Errorf("expected no error for aggregations with raw_stream request type, got: %v", err)
}
@@ -1248,12 +1248,12 @@ func TestNonAggregationFieldsSkipped(t *testing.T) {
},
}
// Should error for raw (selectFields are validated)
err := query.Validate(RequestTypeRaw)
err := query.Validate(getValidationOptions(false)...)
if err == nil {
t.Errorf("expected error for isRoot in selectFields with raw request type")
}
// Should pass for timeseries (selectFields skipped)
err = query.Validate(RequestTypeTimeSeries)
err = query.Validate(getValidationOptions(true)...)
if err != nil {
t.Errorf("expected no error for isRoot in selectFields with timeseries request type, got: %v", err)
}

View File

@@ -0,0 +1,634 @@
import csv
import io
import json
from datetime import datetime, timedelta, timezone
from http import HTTPStatus
from typing import Callable, List
from urllib.parse import urlencode
import requests
from fixtures import types
from fixtures.auth import USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD
from fixtures.logs import Logs
def test_export_logs_csv(
signoz: types.SigNoz,
create_user_admin: None, # pylint: disable=unused-argument
get_token: Callable[[str, str], str],
insert_logs: Callable[[List[Logs]], None],
) -> None:
"""
Setup:
Insert 3 logs with different severity levels and attributes.
Tests:
1. Export logs as CSV format
2. Verify CSV structure and content
3. Validate headers are present
4. Check log data is correctly formatted
"""
now = datetime.now(tz=timezone.utc).replace(second=0, microsecond=0)
insert_logs(
[
Logs(
timestamp=now - timedelta(seconds=10),
body="Application started successfully",
severity_text="INFO",
resources={
"service.name": "api-service",
"deployment.environment": "production",
"host.name": "server-01",
},
attributes={
"http.method": "GET",
"http.status_code": 200,
"user.id": "user123",
},
),
Logs(
timestamp=now - timedelta(seconds=8),
body="Connection to database failed",
severity_text="ERROR",
resources={
"service.name": "api-service",
"deployment.environment": "production",
"host.name": "server-01",
},
attributes={
"error.type": "ConnectionError",
"db.name": "production_db",
},
),
Logs(
timestamp=now - timedelta(seconds=5),
body="Request processed",
severity_text="DEBUG",
resources={
"service.name": "worker-service",
"deployment.environment": "production",
"host.name": "server-02",
},
attributes={
"request.id": "req-456",
"duration_ms": 150.5,
},
),
]
)
token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)
# Calculate timestamps in nanoseconds
start_ns = int((now - timedelta(minutes=5)).timestamp() * 1e9)
end_ns = int(now.timestamp() * 1e9)
params = {
"start": start_ns,
"end": end_ns,
}
# Export logs as CSV (default format, no source needed)
response = requests.get(
signoz.self.host_configs["8080"].get(
f"/api/v1/export_raw_data?{urlencode(params)}"
),
timeout=30,
headers={
"authorization": f"Bearer {token}",
},
)
assert response.status_code == HTTPStatus.OK
assert response.headers["Content-Type"] == "text/csv"
assert "attachment" in response.headers.get("Content-Disposition", "")
# Parse CSV content
csv_content = response.text
csv_reader = csv.DictReader(io.StringIO(csv_content))
rows = list(csv_reader)
assert len(rows) == 3, f"Expected 3 rows, got {len(rows)}"
# Verify log bodies are present in the exported data
bodies = [row.get("body") for row in rows]
assert "Application started successfully" in bodies
assert "Connection to database failed" in bodies
assert "Request processed" in bodies
# Verify severity levels
severities = [row.get("severity_text") for row in rows]
assert "INFO" in severities
assert "ERROR" in severities
assert "DEBUG" in severities
def test_export_logs_jsonl(
signoz: types.SigNoz,
create_user_admin: None, # pylint: disable=unused-argument
get_token: Callable[[str, str], str],
insert_logs: Callable[[List[Logs]], None],
) -> None:
"""
Setup:
Insert 2 logs with different attributes.
Tests:
1. Export logs as JSONL format
2. Verify JSONL structure and content
3. Check each line is valid JSON
4. Validate log data is correctly formatted
"""
now = datetime.now(tz=timezone.utc).replace(second=0, microsecond=0)
insert_logs(
[
Logs(
timestamp=now - timedelta(seconds=10),
body="User logged in",
severity_text="INFO",
resources={
"service.name": "auth-service",
"deployment.environment": "staging",
},
attributes={
"user.email": "test@example.com",
"session.id": "sess-789",
},
),
Logs(
timestamp=now - timedelta(seconds=5),
body="Payment processed successfully",
severity_text="INFO",
resources={
"service.name": "payment-service",
"deployment.environment": "staging",
},
attributes={
"transaction.id": "txn-123",
"amount": 99.99,
},
),
]
)
token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)
# Calculate timestamps in nanoseconds
start_ns = int((now - timedelta(minutes=5)).timestamp() * 1e9)
end_ns = int(now.timestamp() * 1e9)
params = {
"start": start_ns,
"end": end_ns,
"format": "jsonl",
"source": "logs",
}
# Export logs as JSONL
response = requests.get(
signoz.self.host_configs["8080"].get(
f"/api/v1/export_raw_data?{urlencode(params)}"
),
timeout=10,
headers={
"authorization": f"Bearer {token}",
},
)
assert response.status_code == HTTPStatus.OK
assert response.headers["Content-Type"] == "application/x-ndjson"
assert "attachment" in response.headers.get("Content-Disposition", "")
# Parse JSONL content
jsonl_lines = response.text.strip().split("\n")
assert len(jsonl_lines) == 2, f"Expected 2 lines, got {len(jsonl_lines)}"
# Verify each line is valid JSON
json_objects = []
for line in jsonl_lines:
obj = json.loads(line)
json_objects.append(obj)
assert "id" in obj
assert "timestamp" in obj
assert "body" in obj
assert "severity_text" in obj
# Verify log bodies
bodies = [obj.get("body") for obj in json_objects]
assert "User logged in" in bodies
assert "Payment processed successfully" in bodies
def test_export_logs_with_filter(
signoz: types.SigNoz,
create_user_admin: None, # pylint: disable=unused-argument
get_token: Callable[[str, str], str],
insert_logs: Callable[[List[Logs]], None],
) -> None:
"""
Setup:
Insert logs with different severity levels.
Tests:
1. Export logs with filter applied
2. Verify only filtered logs are returned
"""
now = datetime.now(tz=timezone.utc).replace(second=0, microsecond=0)
insert_logs(
[
Logs(
timestamp=now - timedelta(seconds=10),
body="Info message",
severity_text="INFO",
resources={
"service.name": "test-service",
},
attributes={},
),
Logs(
timestamp=now - timedelta(seconds=8),
body="Error message",
severity_text="ERROR",
resources={
"service.name": "test-service",
},
attributes={},
),
Logs(
timestamp=now - timedelta(seconds=5),
body="Another error message",
severity_text="ERROR",
resources={
"service.name": "test-service",
},
attributes={},
),
]
)
token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)
# Calculate timestamps in nanoseconds
start_ns = int((now - timedelta(minutes=5)).timestamp() * 1e9)
end_ns = int(now.timestamp() * 1e9)
params = {
"start": start_ns,
"end": end_ns,
"format": "jsonl",
"source": "logs",
"filter": "severity_text = 'ERROR'",
}
# Export logs with filter
response = requests.get(
signoz.self.host_configs["8080"].get(
f"/api/v1/export_raw_data?{urlencode(params)}"
),
timeout=10,
headers={
"authorization": f"Bearer {token}",
},
)
assert response.status_code == HTTPStatus.OK
assert response.headers["Content-Type"] == "application/x-ndjson"
# Parse JSONL content
jsonl_lines = response.text.strip().split("\n")
assert len(jsonl_lines) == 2, f"Expected 2 lines (filtered), got {len(jsonl_lines)}"
# Verify only ERROR logs are returned
for line in jsonl_lines:
obj = json.loads(line)
assert obj["severity_text"] == "ERROR"
assert "error message" in obj["body"].lower()
def test_export_logs_with_limit(
signoz: types.SigNoz,
create_user_admin: None, # pylint: disable=unused-argument
get_token: Callable[[str, str], str],
insert_logs: Callable[[List[Logs]], None],
) -> None:
"""
Setup:
Insert 5 logs.
Tests:
1. Export logs with limit applied
2. Verify only limited number of logs are returned
"""
now = datetime.now(tz=timezone.utc).replace(second=0, microsecond=0)
logs = []
for i in range(5):
logs.append(
Logs(
timestamp=now - timedelta(seconds=i),
body=f"Log message {i}",
severity_text="INFO",
resources={
"service.name": "test-service",
},
attributes={
"index": i,
},
)
)
insert_logs(logs)
token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)
# Calculate timestamps in nanoseconds
start_ns = int((now - timedelta(minutes=5)).timestamp() * 1e9)
end_ns = int(now.timestamp() * 1e9)
params = {
"start": start_ns,
"end": end_ns,
"format": "csv",
"source": "logs",
"limit": 3,
}
# Export logs with limit
response = requests.get(
signoz.self.host_configs["8080"].get(
f"/api/v1/export_raw_data?{urlencode(params)}"
),
timeout=10,
headers={
"authorization": f"Bearer {token}",
},
)
assert response.status_code == HTTPStatus.OK
assert response.headers["Content-Type"] == "text/csv"
# Parse CSV content
csv_content = response.text
csv_reader = csv.DictReader(io.StringIO(csv_content))
rows = list(csv_reader)
assert len(rows) == 3, f"Expected 3 rows (limited), got {len(rows)}"
def test_export_logs_with_columns(
signoz: types.SigNoz,
create_user_admin: None, # pylint: disable=unused-argument
get_token: Callable[[str, str], str],
insert_logs: Callable[[List[Logs]], None],
) -> None:
"""
Setup:
Insert logs with various attributes.
Tests:
1. Export logs with specific columns
2. Verify only specified columns are returned
"""
now = datetime.now(tz=timezone.utc).replace(second=0, microsecond=0)
insert_logs(
[
Logs(
timestamp=now - timedelta(seconds=10),
body="Test log message",
severity_text="INFO",
resources={
"service.name": "test-service",
"deployment.environment": "production",
},
attributes={
"http.method": "GET",
"http.status_code": 200,
},
),
]
)
token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)
# Calculate timestamps in nanoseconds
start_ns = int((now - timedelta(minutes=5)).timestamp() * 1e9)
end_ns = int(now.timestamp() * 1e9)
# Request only specific columns
params = {
"start": start_ns,
"end": end_ns,
"format": "csv",
"source": "logs",
"columns": ["timestamp", "severity_text", "body"],
}
# Export logs with specific columns
response = requests.get(
signoz.self.host_configs["8080"].get(
f"/api/v1/export_raw_data?{urlencode(params, doseq=True)}"
),
timeout=10,
headers={
"authorization": f"Bearer {token}",
},
)
assert response.status_code == HTTPStatus.OK
assert response.headers["Content-Type"] == "text/csv"
# Parse CSV content
csv_content = response.text
csv_reader = csv.DictReader(io.StringIO(csv_content))
rows = list(csv_reader)
assert len(rows) == 1
# Verify the specified columns are present
row = rows[0]
assert "timestamp" in row
assert "severity_text" in row
assert "body" in row
assert row["severity_text"] == "INFO"
assert row["body"] == "Test log message"
def test_export_logs_with_order_by(
signoz: types.SigNoz,
create_user_admin: None, # pylint: disable=unused-argument
get_token: Callable[[str, str], str],
insert_logs: Callable[[List[Logs]], None],
) -> None:
"""
Setup:
Insert logs at different timestamps.
Tests:
1. Export logs with ascending timestamp order
2. Verify logs are returned in correct order
"""
now = datetime.now(tz=timezone.utc).replace(second=0, microsecond=0)
insert_logs(
[
Logs(
timestamp=now - timedelta(seconds=10),
body="First log",
severity_text="INFO",
resources={
"service.name": "test-service",
},
attributes={},
),
Logs(
timestamp=now - timedelta(seconds=5),
body="Second log",
severity_text="INFO",
resources={
"service.name": "test-service",
},
attributes={},
),
Logs(
timestamp=now - timedelta(seconds=1),
body="Third log",
severity_text="INFO",
resources={
"service.name": "test-service",
},
attributes={},
),
]
)
token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)
# Calculate timestamps in nanoseconds
start_ns = int((now - timedelta(minutes=5)).timestamp() * 1e9)
end_ns = int(now.timestamp() * 1e9)
params = {
"start": start_ns,
"end": end_ns,
"format": "jsonl",
"source": "logs",
"order_by": "timestamp:asc",
}
# Export logs with ascending order
response = requests.get(
signoz.self.host_configs["8080"].get(
f"/api/v1/export_raw_data?{urlencode(params)}"
),
timeout=10,
headers={
"authorization": f"Bearer {token}",
},
)
assert response.status_code == HTTPStatus.OK
assert response.headers["Content-Type"] == "application/x-ndjson"
# Parse JSONL content
jsonl_lines = response.text.strip().split("\n")
assert len(jsonl_lines) == 3
# Verify order - first log should be "First log" (oldest)
json_objects = [json.loads(line) for line in jsonl_lines]
assert json_objects[0]["body"] == "First log"
assert json_objects[1]["body"] == "Second log"
assert json_objects[2]["body"] == "Third log"
def test_export_logs_with_complex_filter(
signoz: types.SigNoz,
create_user_admin: None, # pylint: disable=unused-argument
get_token: Callable[[str, str], str],
insert_logs: Callable[[List[Logs]], None],
) -> None:
"""
Setup:
Insert logs with various service names and severity levels.
Tests:
1. Export logs with complex filter (multiple conditions)
2. Verify only logs matching all conditions are returned
"""
now = datetime.now(tz=timezone.utc).replace(second=0, microsecond=0)
insert_logs(
[
Logs(
timestamp=now - timedelta(seconds=10),
body="API error occurred",
severity_text="ERROR",
resources={
"service.name": "api-service",
},
attributes={},
),
Logs(
timestamp=now - timedelta(seconds=8),
body="Worker info message",
severity_text="INFO",
resources={
"service.name": "worker-service",
},
attributes={},
),
Logs(
timestamp=now - timedelta(seconds=5),
body="API info message",
severity_text="INFO",
resources={
"service.name": "api-service",
},
attributes={},
),
]
)
token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)
# Calculate timestamps in nanoseconds
start_ns = int((now - timedelta(minutes=5)).timestamp() * 1e9)
end_ns = int(now.timestamp() * 1e9)
# Filter for api-service AND ERROR severity
params = {
"start": start_ns,
"end": end_ns,
"format": "jsonl",
"source": "logs",
"filter": "service.name = 'api-service' AND severity_text = 'ERROR'",
}
# Export logs with complex filter
response = requests.get(
signoz.self.host_configs["8080"].get(
f"/api/v1/export_raw_data?{urlencode(params)}"
),
timeout=10,
headers={
"authorization": f"Bearer {token}",
},
)
assert response.status_code == HTTPStatus.OK
assert response.headers["Content-Type"] == "application/x-ndjson"
# Parse JSONL content
jsonl_lines = response.text.strip().split("\n")
assert (
len(jsonl_lines) == 1
), f"Expected 1 line (complex filter), got {len(jsonl_lines)}"
# Verify the filtered log
filtered_obj = json.loads(jsonl_lines[0])
assert filtered_obj["body"] == "API error occurred"
assert filtered_obj["severity_text"] == "ERROR"

View File

@@ -0,0 +1,782 @@
import csv
import io
import json
from datetime import datetime, timedelta, timezone
from http import HTTPStatus
from typing import Callable, List
from urllib.parse import urlencode
import requests
from fixtures import types
from fixtures.auth import USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD
from fixtures.traces import TraceIdGenerator, Traces, TracesKind, TracesStatusCode
def test_export_traces_csv(
signoz: types.SigNoz,
create_user_admin: None, # pylint: disable=unused-argument
get_token: Callable[[str, str], str],
insert_traces: Callable[[List[Traces]], None],
) -> None:
"""
Setup:
Insert 3 traces with different attributes.
Tests:
1. Export traces as CSV format
2. Verify CSV structure and content
3. Validate headers are present
4. Check trace data is correctly formatted
"""
http_service_trace_id = TraceIdGenerator.trace_id()
http_service_span_id = TraceIdGenerator.span_id()
http_service_db_span_id = TraceIdGenerator.span_id()
topic_service_trace_id = TraceIdGenerator.trace_id()
topic_service_span_id = TraceIdGenerator.span_id()
now = datetime.now(tz=timezone.utc).replace(second=0, microsecond=0)
insert_traces(
[
Traces(
timestamp=now - timedelta(seconds=4),
duration=timedelta(seconds=3),
trace_id=http_service_trace_id,
span_id=http_service_span_id,
parent_span_id="",
name="POST /integration",
kind=TracesKind.SPAN_KIND_SERVER,
status_code=TracesStatusCode.STATUS_CODE_OK,
status_message="",
resources={
"deployment.environment": "production",
"service.name": "http-service",
"os.type": "linux",
"host.name": "linux-000",
},
attributes={
"net.transport": "IP.TCP",
"http.scheme": "http",
"http.user_agent": "Integration Test",
"http.request.method": "POST",
"http.response.status_code": "200",
},
),
Traces(
timestamp=now - timedelta(seconds=3.5),
duration=timedelta(seconds=0.5),
trace_id=http_service_trace_id,
span_id=http_service_db_span_id,
parent_span_id=http_service_span_id,
name="SELECT",
kind=TracesKind.SPAN_KIND_CLIENT,
status_code=TracesStatusCode.STATUS_CODE_OK,
status_message="",
resources={
"deployment.environment": "production",
"service.name": "http-service",
"os.type": "linux",
"host.name": "linux-000",
},
attributes={
"db.name": "integration",
"db.operation": "SELECT",
"db.statement": "SELECT * FROM integration",
},
),
Traces(
timestamp=now - timedelta(seconds=1),
duration=timedelta(seconds=2),
trace_id=topic_service_trace_id,
span_id=topic_service_span_id,
parent_span_id="",
name="topic publish",
kind=TracesKind.SPAN_KIND_PRODUCER,
status_code=TracesStatusCode.STATUS_CODE_OK,
status_message="",
resources={
"deployment.environment": "production",
"service.name": "topic-service",
"os.type": "linux",
"host.name": "linux-001",
},
attributes={
"message.type": "SENT",
"messaging.operation": "publish",
"messaging.message.id": "001",
},
),
]
)
token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)
# Calculate timestamps in nanoseconds
start_ns = int((now - timedelta(minutes=5)).timestamp() * 1e9)
end_ns = int(now.timestamp() * 1e9)
params = {
"start": start_ns,
"end": end_ns,
"source": "traces",
"limit": 1000,
}
# Export traces as CSV (GET for simple queries)
response = requests.get(
signoz.self.host_configs["8080"].get(
f"/api/v1/export_raw_data?{urlencode(params)}"
),
timeout=30,
headers={
"authorization": f"Bearer {token}",
},
)
assert response.status_code == HTTPStatus.OK
assert response.headers["Content-Type"] == "text/csv"
assert "attachment" in response.headers.get("Content-Disposition", "")
# Parse CSV content
csv_content = response.text
csv_reader = csv.DictReader(io.StringIO(csv_content))
rows = list(csv_reader)
assert len(rows) == 3, f"Expected 3 rows, got {len(rows)}"
# Verify trace IDs are present in the exported data
trace_ids = [row.get("trace_id") for row in rows]
assert http_service_trace_id in trace_ids
assert topic_service_trace_id in trace_ids
# Verify span names are present
span_names = [row.get("name") for row in rows]
assert "POST /integration" in span_names
assert "SELECT" in span_names
assert "topic publish" in span_names
def test_export_traces_jsonl(
signoz: types.SigNoz,
create_user_admin: None, # pylint: disable=unused-argument
get_token: Callable[[str, str], str],
insert_traces: Callable[[List[Traces]], None],
) -> None:
"""
Setup:
Insert 2 traces with different attributes.
Tests:
1. Export traces as JSONL format
2. Verify JSONL structure and content
3. Check each line is valid JSON
4. Validate trace data is correctly formatted
"""
http_service_trace_id = TraceIdGenerator.trace_id()
http_service_span_id = TraceIdGenerator.span_id()
topic_service_trace_id = TraceIdGenerator.trace_id()
topic_service_span_id = TraceIdGenerator.span_id()
now = datetime.now(tz=timezone.utc).replace(second=0, microsecond=0)
insert_traces(
[
Traces(
timestamp=now - timedelta(seconds=4),
duration=timedelta(seconds=3),
trace_id=http_service_trace_id,
span_id=http_service_span_id,
parent_span_id="",
name="POST /api/test",
kind=TracesKind.SPAN_KIND_SERVER,
status_code=TracesStatusCode.STATUS_CODE_OK,
status_message="",
resources={
"service.name": "api-service",
"deployment.environment": "staging",
},
attributes={
"http.request.method": "POST",
"http.response.status_code": "201",
},
),
Traces(
timestamp=now - timedelta(seconds=2),
duration=timedelta(seconds=1),
trace_id=topic_service_trace_id,
span_id=topic_service_span_id,
parent_span_id="",
name="queue.process",
kind=TracesKind.SPAN_KIND_CONSUMER,
status_code=TracesStatusCode.STATUS_CODE_OK,
status_message="",
resources={
"service.name": "queue-service",
"deployment.environment": "staging",
},
attributes={
"messaging.operation": "process",
"messaging.system": "rabbitmq",
},
),
]
)
token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)
# Calculate timestamps in nanoseconds
start_ns = int((now - timedelta(minutes=5)).timestamp() * 1e9)
end_ns = int(now.timestamp() * 1e9)
params = {
"start": start_ns,
"end": end_ns,
"format": "jsonl",
"source": "traces",
"limit": 1000,
}
# Export traces as JSONL (GET for simple queries)
response = requests.get(
signoz.self.host_configs["8080"].get(
f"/api/v1/export_raw_data?{urlencode(params)}"
),
timeout=10,
headers={
"authorization": f"Bearer {token}",
},
)
assert response.status_code == HTTPStatus.OK
assert response.headers["Content-Type"] == "application/x-ndjson"
assert "attachment" in response.headers.get("Content-Disposition", "")
# Parse JSONL content
jsonl_lines = response.text.strip().split("\n")
assert len(jsonl_lines) == 2, f"Expected 2 lines, got {len(jsonl_lines)}"
# Verify each line is valid JSON
json_objects = []
for line in jsonl_lines:
obj = json.loads(line)
json_objects.append(obj)
assert "trace_id" in obj
assert "span_id" in obj
assert "name" in obj
# Verify trace IDs are present
trace_ids = [obj.get("trace_id") for obj in json_objects]
assert http_service_trace_id in trace_ids
assert topic_service_trace_id in trace_ids
# Verify span names are present
span_names = [obj.get("name") for obj in json_objects]
assert "POST /api/test" in span_names
assert "queue.process" in span_names
def test_export_traces_with_filter(
signoz: types.SigNoz,
create_user_admin: None, # pylint: disable=unused-argument
get_token: Callable[[str, str], str],
insert_traces: Callable[[List[Traces]], None],
) -> None:
"""
Setup:
Insert traces with different service names.
Tests:
1. Export traces with filter applied
2. Verify only filtered traces are returned
"""
service_a_trace_id = TraceIdGenerator.trace_id()
service_a_span_id = TraceIdGenerator.span_id()
service_b_trace_id = TraceIdGenerator.trace_id()
service_b_span_id = TraceIdGenerator.span_id()
now = datetime.now(tz=timezone.utc).replace(second=0, microsecond=0)
insert_traces(
[
Traces(
timestamp=now - timedelta(seconds=4),
duration=timedelta(seconds=1),
trace_id=service_a_trace_id,
span_id=service_a_span_id,
parent_span_id="",
name="operation-a",
kind=TracesKind.SPAN_KIND_SERVER,
status_code=TracesStatusCode.STATUS_CODE_OK,
status_message="",
resources={
"service.name": "service-a",
},
attributes={},
),
Traces(
timestamp=now - timedelta(seconds=2),
duration=timedelta(seconds=1),
trace_id=service_b_trace_id,
span_id=service_b_span_id,
parent_span_id="",
name="operation-b",
kind=TracesKind.SPAN_KIND_SERVER,
status_code=TracesStatusCode.STATUS_CODE_OK,
status_message="",
resources={
"service.name": "service-b",
},
attributes={},
),
]
)
token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)
# Calculate timestamps in nanoseconds
start_ns = int((now - timedelta(minutes=5)).timestamp() * 1e9)
end_ns = int(now.timestamp() * 1e9)
params = {
"start": start_ns,
"end": end_ns,
"format": "jsonl",
"source": "traces",
"limit": 1000,
"filter": "service.name = 'service-a'",
}
# Export traces with filter (GET supports filter param)
response = requests.get(
signoz.self.host_configs["8080"].get(
f"/api/v1/export_raw_data?{urlencode(params)}"
),
timeout=10,
headers={
"authorization": f"Bearer {token}",
},
)
assert response.status_code == HTTPStatus.OK
assert response.headers["Content-Type"] == "application/x-ndjson"
# Parse JSONL content
jsonl_lines = response.text.strip().split("\n")
assert len(jsonl_lines) == 1, f"Expected 1 line (filtered), got {len(jsonl_lines)}"
# Verify the filtered trace
filtered_obj = json.loads(jsonl_lines[0])
assert filtered_obj["trace_id"] == service_a_trace_id
assert filtered_obj["name"] == "operation-a"
def test_export_traces_with_limit(
signoz: types.SigNoz,
create_user_admin: None, # pylint: disable=unused-argument
get_token: Callable[[str, str], str],
insert_traces: Callable[[List[Traces]], None],
) -> None:
"""
Setup:
Insert 5 traces.
Tests:
1. Export traces with limit applied
2. Verify only limited number of traces are returned
"""
now = datetime.now(tz=timezone.utc).replace(second=0, microsecond=0)
traces = []
for i in range(5):
traces.append(
Traces(
timestamp=now - timedelta(seconds=i),
duration=timedelta(seconds=1),
trace_id=TraceIdGenerator.trace_id(),
span_id=TraceIdGenerator.span_id(),
parent_span_id="",
name=f"operation-{i}",
kind=TracesKind.SPAN_KIND_SERVER,
status_code=TracesStatusCode.STATUS_CODE_OK,
status_message="",
resources={
"service.name": "test-service",
},
attributes={},
)
)
insert_traces(traces)
token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)
# Calculate timestamps in nanoseconds
start_ns = int((now - timedelta(minutes=5)).timestamp() * 1e9)
end_ns = int(now.timestamp() * 1e9)
params = {
"start": start_ns,
"end": end_ns,
"format": "csv",
"source": "traces",
"limit": 3,
}
# Export traces with limit (GET supports limit param)
response = requests.get(
signoz.self.host_configs["8080"].get(
f"/api/v1/export_raw_data?{urlencode(params)}"
),
timeout=10,
headers={
"authorization": f"Bearer {token}",
},
)
assert response.status_code == HTTPStatus.OK
assert response.headers["Content-Type"] == "text/csv"
# Parse CSV content
csv_content = response.text
csv_reader = csv.DictReader(io.StringIO(csv_content))
rows = list(csv_reader)
assert len(rows) == 3, f"Expected 3 rows (limited), got {len(rows)}"
def test_export_traces_multiple_queries_rejected(
signoz: types.SigNoz,
create_user_admin: None, # pylint: disable=unused-argument
get_token: Callable[[str, str], str],
) -> None:
"""
Tests:
1. POST with multiple builder queries but no trace operator is rejected
2. Verify 400 error is returned
"""
now = datetime.now(tz=timezone.utc).replace(second=0, microsecond=0)
start_ns = int((now - timedelta(minutes=5)).timestamp() * 1e9)
end_ns = int(now.timestamp() * 1e9)
token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)
body = {
"start": start_ns,
"end": end_ns,
"compositeQuery": {
"queries": [
{
"type": "builder_query",
"spec": {
"signal": "traces",
"name": "A",
"limit": 1000,
"filter": {"expression": "service.name = 'service-a'"},
},
},
{
"type": "builder_query",
"spec": {
"signal": "traces",
"name": "B",
"limit": 1000,
"filter": {"expression": "service.name = 'service-b'"},
},
},
]
},
}
url = signoz.self.host_configs["8080"].get("/api/v1/export_raw_data?format=jsonl")
response = requests.post(
url,
json=body,
timeout=10,
headers={
"authorization": f"Bearer {token}",
"Content-Type": "application/json",
},
)
assert response.status_code == HTTPStatus.BAD_REQUEST
def test_export_traces_with_composite_query_trace_operator(
signoz: types.SigNoz,
create_user_admin: None, # pylint: disable=unused-argument
get_token: Callable[[str, str], str],
insert_traces: Callable[[List[Traces]], None],
) -> None:
"""
Setup:
Insert multiple traces with parent-child relationships.
Tests:
1. Export traces using trace operator in composite query (POST)
2. Verify trace operator query works correctly
"""
parent_trace_id = TraceIdGenerator.trace_id()
parent_span_id = TraceIdGenerator.span_id()
child_span_id_1 = TraceIdGenerator.span_id()
child_span_id_2 = TraceIdGenerator.span_id()
now = datetime.now(tz=timezone.utc).replace(second=0, microsecond=0)
insert_traces(
[
Traces(
timestamp=now - timedelta(seconds=10),
duration=timedelta(seconds=5),
trace_id=parent_trace_id,
span_id=parent_span_id,
parent_span_id="",
name="parent-operation",
kind=TracesKind.SPAN_KIND_SERVER,
status_code=TracesStatusCode.STATUS_CODE_OK,
status_message="",
resources={
"service.name": "parent-service",
},
attributes={
"operation.type": "parent",
},
),
Traces(
timestamp=now - timedelta(seconds=9),
duration=timedelta(seconds=2),
trace_id=parent_trace_id,
span_id=child_span_id_1,
parent_span_id=parent_span_id,
name="child-operation-1",
kind=TracesKind.SPAN_KIND_INTERNAL,
status_code=TracesStatusCode.STATUS_CODE_OK,
status_message="",
resources={
"service.name": "parent-service",
},
attributes={
"operation.type": "child",
},
),
Traces(
timestamp=now - timedelta(seconds=7),
duration=timedelta(seconds=1),
trace_id=parent_trace_id,
span_id=child_span_id_2,
parent_span_id=parent_span_id,
name="child-operation-2",
kind=TracesKind.SPAN_KIND_INTERNAL,
status_code=TracesStatusCode.STATUS_CODE_OK,
status_message="",
resources={
"service.name": "parent-service",
},
attributes={
"operation.type": "child",
},
),
]
)
token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)
# Calculate timestamps in nanoseconds
start_ns = int((now - timedelta(minutes=5)).timestamp() * 1e9)
end_ns = int(now.timestamp() * 1e9)
# A: spans with operation.type = 'parent'
query_a = {
"type": "builder_query",
"spec": {
"signal": "traces",
"name": "A",
"limit": 1000,
"filter": {"expression": "operation.type = 'parent'"},
},
}
# B: spans with operation.type = 'child'
query_b = {
"type": "builder_query",
"spec": {
"signal": "traces",
"name": "B",
"limit": 1000,
"filter": {"expression": "operation.type = 'child'"},
},
}
# Trace operator: find traces where A has a direct descendant B
query_c = {
"type": "builder_trace_operator",
"spec": {
"name": "C",
"expression": "A => B",
"returnSpansFrom": "A",
"limit": 1000,
"order": [{"key": {"name": "timestamp"}, "direction": "desc"}],
},
}
body = {
"start": start_ns,
"end": end_ns,
"requestType": "raw",
"compositeQuery": {
"queries": [query_a, query_b, query_c],
},
}
url = signoz.self.host_configs["8080"].get("/api/v1/export_raw_data?format=jsonl")
response = requests.post(
url,
json=body,
timeout=10,
headers={
"authorization": f"Bearer {token}",
"Content-Type": "application/json",
},
)
print(response.text)
assert response.status_code == HTTPStatus.OK
assert response.headers["Content-Type"] == "application/x-ndjson"
# Parse JSONL content
jsonl_lines = response.text.strip().split("\n")
assert len(jsonl_lines) == 1, f"Expected at least 1 line, got {len(jsonl_lines)}"
# Verify all returned spans belong to the matched trace
json_objects = [json.loads(line) for line in jsonl_lines]
trace_ids = [obj.get("trace_id") for obj in json_objects]
assert all(tid == parent_trace_id for tid in trace_ids)
# Verify the parent span (returnSpansFrom = "A") is present
span_names = [obj.get("name") for obj in json_objects]
assert "parent-operation" in span_names
def test_export_traces_with_select_fields(
signoz: types.SigNoz,
create_user_admin: None, # pylint: disable=unused-argument
get_token: Callable[[str, str], str],
insert_traces: Callable[[List[Traces]], None],
) -> None:
"""
Setup:
Insert traces with various attributes.
Tests:
1. Export traces with specific select fields via POST
2. Verify only specified fields are returned in the output
"""
trace_id = TraceIdGenerator.trace_id()
span_id = TraceIdGenerator.span_id()
now = datetime.now(tz=timezone.utc).replace(second=0, microsecond=0)
insert_traces(
[
Traces(
timestamp=now - timedelta(seconds=10),
duration=timedelta(seconds=2),
trace_id=trace_id,
span_id=span_id,
parent_span_id="",
name="test-operation",
kind=TracesKind.SPAN_KIND_SERVER,
status_code=TracesStatusCode.STATUS_CODE_OK,
status_message="",
resources={
"service.name": "test-service",
"deployment.environment": "production",
"host.name": "server-01",
},
attributes={
"http.method": "POST",
"http.status_code": "201",
"user.id": "user123",
},
),
]
)
token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)
# Calculate timestamps in nanoseconds
start_ns = int((now - timedelta(minutes=5)).timestamp() * 1e9)
end_ns = int(now.timestamp() * 1e9)
body = {
"start": start_ns,
"end": end_ns,
"requestType": "raw",
"compositeQuery": {
"queries": [
{
"type": "builder_query",
"spec": {
"signal": "traces",
"name": "A",
"limit": 1000,
"selectFields": [
{
"name": "trace_id",
"fieldDataType": "string",
"fieldContext": "span",
"signal": "traces",
},
{
"name": "span_id",
"fieldDataType": "string",
"fieldContext": "span",
"signal": "traces",
},
{
"name": "name",
"fieldDataType": "string",
"fieldContext": "span",
"signal": "traces",
},
{
"name": "service.name",
"fieldDataType": "string",
"fieldContext": "resource",
"signal": "traces",
},
],
},
}
]
},
}
url = signoz.self.host_configs["8080"].get("/api/v1/export_raw_data?format=jsonl")
response = requests.post(
url,
json=body,
timeout=10,
headers={
"authorization": f"Bearer {token}",
"Content-Type": "application/json",
},
)
assert response.status_code == HTTPStatus.OK
assert response.headers["Content-Type"] == "application/x-ndjson"
# Parse JSONL content
jsonl_lines = response.text.strip().split("\n")
assert len(jsonl_lines) == 1
# Verify the selected fields are present
result = json.loads(jsonl_lines[0])
assert "trace_id" in result
assert "span_id" in result
assert "name" in result
# Verify values
assert result["trace_id"] == trace_id
assert result["span_id"] == span_id
assert result["name"] == "test-operation"