dima/supabase
1
0 Fork 0
mirror of https://github.com/supabase/supabase.git synced 2026-10-06 09:55:06 +03:00
Code Issues Packages Projects Releases Wiki Activity
38,866 Commits 1,781 Branches 62 Tags
ab9ccf00a3420f7aa12e552d578e1118c19ccdb5
Commit Graph
3 Commits
Author SHA1 Message Date
Charis ec1c889349 feat(studio): logs SQL brands + execution data layer (#48301) 
## I have read the
[CONTRIBUTING.md](https://github.com/supabase/supabase/blob/master/CONTRIBUTING.md)
file.

YES

## What kind of change does this PR introduce?

Feature (data layer only — PR 1 of the SQL-editor query-source stack;
nothing user-visible yet, no consumers).

## What is the current behavior?

The Studio SQL editor only runs queries against Postgres. There is no
type-safe brand for user-authored logs SQL and no
execution/normalization layer for running SQL against the logs/analytics
(ClickHouse) backend.

## What is the new behavior?

Pure additions, no behavior change:

- `data/logs/safe-analytics-sql.ts` — adds distinct untrusted/safe
brands for user-authored logs SQL (`UntrustedLogSqlFragment`,
`untrustedLogSql`, `acceptUntrustedLogsSql`), mirroring pg-meta's
`UntrustedSqlFragment` but kept intentionally disjoint so Postgres and
logs SQL can never cross boundaries.
- `data/logs/execute-logs-sql-mutation.ts` (new) — `executeLogsSql`
wraps `executeAnalyticsSql`, attaches the resolved time range as request
params (`iso_timestamp_start/end`, never spliced into SQL), and
normalizes to `{ rows, error? }`; `mapLogsError` normalizes the
analytics backend's structured 200-body error into the `{ message }`
shape the result pane reads; `useExecuteLogsSqlMutation` collapses
transport and 200-body errors into React Query's single `onError` path.
- Unit tests for `mapLogsError`, the brands (including compile-time
disjointness vs pg-meta brands), and safe composition.

Verification: `pnpm test:studio` (new suites, 26 passed), `pnpm
typecheck`, `lint:ratchet` (no new warnings), and Prettier all pass.

## Additional context

<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit

* **New Features**
* Added the ability to run user-authored logs SQL with resolved
start/end timestamps.
* Normalized query error handling so failures surface a clear message
(including sensible fallbacks) and integrates with mutation error flows
(with a default error toast when not customized).
* Introduced safety branding for logs SQL fragments, including promotion
to runnable safe SQL.
* **Tests**
* Added tests covering error normalization across multiple
malformed/empty error shapes.
* Added tests ensuring logs SQL branding preserves/accepts only the
intended types and rejects unsafe inputs.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
 2026-07-28 10:26:30 -04:00
Charis a7d51cdf52 feat(logs): brand legacy analytics SQL stack with SafeLogSqlFragment (#46351) 
## I have read the
[CONTRIBUTING.md](https://github.com/supabase/supabase/blob/master/CONTRIBUTING.md)
file.

YES

## What kind of change does this PR introduce?

Refactor / type safety improvement

## What is the current behavior?

The legacy log query stack (`genDefaultQuery`, `genCountQuery`,
`genChartQuery`, `genWhereStatement`, `useLogsPreview`, `useSingleLog`)
builds SQL from raw strings with no type-level guarantee that values are
safely interpolated. Identifier helpers (`bqIdent`, `bqDottedIdent`,
`clickhouseIdent`, `clickhouseDottedIdent`) are duplicated across
BigQuery and ClickHouse variants, and `bqDottedIdent` wraps the entire
dotted path in one backtick pair (`` `request.pathname` ``), which
BigQuery treats as a literal column name rather than a UNNEST alias
field — causing runtime query failures on dotted filter keys.

## What is the new behavior?

- All gen functions return `SafeLogSqlFragment` and all callers route
through `executeAnalyticsSql`, enforcing compile-time SQL provenance
tracking across the legacy stack.
- `bqIdent` / `bqDottedIdent` / `clickhouseIdent` /
`clickhouseDottedIdent` are replaced by a single `quotedIdent` function
that backtick-quotes each segment individually (e.g. ``
`request`.`pathname` ``). ClickHouse natively accepts backticks, so one
function serves both engines and the dotted-path quoting bug is fixed.
- `SQL_FILTER_TEMPLATES` entries are converted to `SafeLogSqlFragment`
(static via `safeSql`, dynamic via `safeSql` + `analyticsLiteral`).
- `buildWhereClauses` is extracted as a private helper returning
`SafeLogSqlFragment[]` so the pg_cron path can merge clauses without
unsafe slice-and-cast.

## Additional context

<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit

* **Refactor**
* Logs query generation migrated to safer, engine-agnostic SQL
fragments, typed filter templates, and unified identifier quoting for
stronger injection protection and more consistent queries.
* Logs preview and single-log retrieval now execute analytics SQL
end-to-end using the unified executor.

* **New Features**
* Analytics SQL executor can call the backend via GET or POST and
accepts method selection.

* **Tests**
* Updated tests to validate unified identifier quoting and safe-SQL
helper behavior.

<!-- review_stack_entry_start -->

[![Review Change
Stack](https://storage.googleapis.com/coderabbit_public_assets/review-stack-in-coderabbit-ui.svg)](https://app.coderabbit.ai/change-stack/supabase/supabase/pull/46351?utm_source=github_walkthrough&utm_medium=github&utm_campaign=change_stack)

<!-- review_stack_entry_end -->
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
 2026-05-26 15:20:54 -04:00
Charis d117e70f6c feat: add safe SQL execution for analytics queries (BigQuery/ClickHouse) (#46287) 
## I have read the
[CONTRIBUTING.md](https://github.com/supabase/supabase/blob/master/CONTRIBUTING.md)
file.

YES

## What kind of change does this PR introduce?

Feature - Security infrastructure

## What is the current behavior?

Analytics queries (BigQuery for legacy cloud, ClickHouse for self-hosted
OTEL) lack a compile-time safety model to prevent SQL injection from
untrusted input sources like URL parameters, UI inputs, or LLM output.

## What is the new behavior?

Implement a security model with a branded type `SafeLogSqlFragment` that
ensures all SQL fragments originate from either static code or
sanitization helpers. This includes:

- `analyticsLiteral()` for escaping string/number/boolean values
- `bqIdent()` and `clickhouseIdent()` for quoting identifiers with
engine-specific syntax
- `safeSql` template tag for composing fragments safely
- `executeAnalyticsSql()` wire boundary that rejects plain strings at
compile time

The pattern prevents cross-engine confusion by keeping
`SafeLogSqlFragment` (analytics) distinct from pg-meta's
`SafeSqlFragment` (Postgres).

## Additional context

<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->

## Summary by CodeRabbit

* **New Features**
* Introduced analytics SQL execution capabilities with built-in safety
validation for queries.
* Enhanced query robustness through keyword and identifier validation
mechanisms.
  * Improved error handling and reporting for analytics operations.

* **Tests**
* Added comprehensive test suite for analytics SQL safety and validation
utilities.

<!-- review_stack_entry_start -->

[![Review Change
Stack](https://storage.googleapis.com/coderabbit_public_assets/review-stack-in-coderabbit-ui.svg)](https://app.coderabbit.ai/change-stack/supabase/supabase/pull/46287?utm_source=github_walkthrough&utm_medium=github&utm_campaign=change_stack)

<!-- review_stack_entry_end -->

<!-- end of auto-generated comment: release notes by coderabbit.ai -->
 2026-05-25 08:40:18 -04:00