mirror of
https://github.com/supabase/supabase.git
synced 2026-10-07 02:15:05 +03:00
Makes the Logs Explorer query docs ClickHouse-focused. The Logs Explorer runs on ClickHouse (default since June 2026): a single `logs` table tagged by `source`, with nested fields in a `log_attributes` map. This rewrites `## Querying with the Logs Explorer` to document only that model: - ClickHouse query mechanics (`log_attributes['...']` access, `toInt32OrZero`, `count()`), with example queries and the "don't guess keys" guidance - The old BigQuery `cross join unnest(metadata)` section (timestamp/unnesting/examples) is removed; a short note flags that BigQuery is deprecated - Best practices and the field reference reworded for ClickHouse No dead anchors: the only repo reference to an `#unnesting-arrays` anchor points at `advanced-log-filtering`, which is untouched. Supersedes the earlier two-engine version on this branch. Skill side (agent-skills#112) already points here as the single source for log querying. <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Documentation** * Updated Logs Explorer guides to use ClickHouse as the default query engine. * Added guidance for querying the unified `logs` table and accessing structured fields. * Added examples for filtering errors, SQLSTATE codes, numeric values, and regular expressions. * Documented how to discover available log fields and clarified MCP log-query behavior. * Updated best practices and field references to reflect the current log structure. * **Chores** * Added SQLSTATE and substring terms to the documentation spelling allow-list. <!-- end of auto-generated comment: release notes by coderabbit.ai -->