This adds a new log line that shows each schema cache query time individually, only on
`log-level=debug`. Like so:
```
$ PGRST_LOG_LEVEL=debug postgrest-with-pg-17 -f test/spec/fixtures/load.sql postgrest-run
....
10/Apr/2026:21:48:45 -0500: Schema cache queried in 192.2 milliseconds
10/Apr/2026:21:48:45 -0500: tables: 72.027 ms, keydeps: 20.118 ms, rels: 6.189 ms, funcs: 35.010 ms, comprels: 4.319 ms, dreps: 1.614 ms, mhandlers: 7.419 ms, tzones: 43.025 ms
```
This helps debug specific schema cache queries being slow like on
https://github.com/PostgREST/postgrest/issues/4613#issuecomment-4210191065 and
https://github.com/PostgREST/postgrest/issues/3046#issuecomment-3469059948.
It also closes https://github.com/PostgREST/postgrest/issues/3215,
which main motivation was to find out which query is slow.
Implementation details
---------------------
To time each query inside a transaction in pure SQL, we do:
```sql
-- start timer
select set_config('pgrst.tmp_x', clock_timestamp()::text, false);
-- run the query
select <query>
-- end timer
select set_config('pgrst.tmp_x', (clock_timestamp() - current_setting('pgrst.tmp_x', false)::timestamptz)::text, false);
-- .... repeated for every query
-- at the end we capture all the timings with
select extract('milliseconds' from current_setting('pgrst.tmp_x', false)::interval), extract(..;
```
Considerations
--------------
Only added this on `log-level=debug` because while the queries are fast
and the data is valuable, it triples the amount of queries we run during schema cache refresh,
which could be troublesome on slow networks. It's possible to reduce the
amount of queries by starting and stopping timers in one statement, but
this would still double the amount of queries and makes the code messy,
doesn't seem worth it.
Also it would pollute pg_stat_statements, it's only required to debug certain
extreme cases anyway.
To avoid repeated querying of `pg_timezone_names` every time schema
cache is reset, `Prefer: timezone` can be disabled by setting
`db-timezone-enabled = false`.
Signed-off-by: Taimoor Zaeem <taimoorzaeem@gmail.com>
* Calculation of hint message when requested relation is not present in schema cache requires creation of a FuzzySet (to use fuzzy search to find candidate tables). For schemas with many tables it is costly.
This patch introduces dbTablesFuzzyIndex in SchemaCache to memoize the FuzzySet creation.
* Additionally, because of FuzzySet large memory requirements, this patch introduces a limit of 500 relations per schema, above which FuzzySet is not created and hint calculation disabled.
To make schema cache loading wait tests robust it is necessary to provide three separate internal config variables:
* "internal-schema-cache-query-sleep" - introduces delay in schema queries execution
* "internal-schema-cache-load-sleep" - introduces delay between schema queries execution and processing their results
* "internal-schema-cache-relationship-load-sleep" - introduces delay in processing relationship query results
Thanks to these changes it is now possible to test various schema loading scenarios with the right granularity robustly (eg. make sure requests wait for schema loading but not for relationship loading).
Closes https://github.com/PostgREST/postgrest/issues/4245.
It adds a query on schema cache construction, but doesn't add a new
attribute to the SchemaCache type. We only need the information to be
passed to the `initialMediaHandlers` function to build the builtin media
handlers.
Previously, we'd silently report "200 OK" on the root endpoint, but
would never return any endpoints from the schema cache.
Now the schema cache query fails because of the ::regnamespace cast.
The schema cache and OpenAPI output would currently list the first found
enum with the same name instead of the correct type. One other case
where this comes up is when a regular type and an enum type have the
same name. For example in the spec fixtures, we have an enum called
"bit". Every "bit" type, no matter whether it's that enum or the
built-in bit type, will show those enum options in the OpenApi output.
Not adding a test, because OpenAPI is supposed to go away in the future
anyway.
The former depends on th-orphans which does not cross-compile well,
because of template haskell usage.
neat-interpolation is also much better maintained.
This also potentially helps with packaging for Debian/Ubuntu in #2273.
Contravariant.Extras uses Template Haskell, which is hard to
cross-compile. Reducing usage of Template Haskell with the ultimate goal
of solving all cross compilation challenges.
The $subject commit broke internal_schema_cache_sleep for other tests.
This reverts the order change, but keeps the scaling by x1000 to ms and
thus changes other users of this setting to the new scale.
By increasing the delays in this test by factor 400x, postgrest will not
swamp pg with connection retries after the failed schema cache anymore.
This would happen because there is no backoff included after fatal
errors. Once it does, the io tests hang indefinitely in CI.
f9ee1f7e introduced the hoisting of function settings as transaction-scoped
settings. However, this currently doesn't work with memory units, which are
case-sensitive according to the docs [1]. This removes the lowercasing of
values to make them work.
This is not added to the CHANGELOG, because this feature was not released, yet.
[1]: https://www.postgresql.org/docs/current/config-setting.html#CONFIG-SETTING-NAMES-VALUES
Those were left out of the schema dump when the features were introduced, probably
because ByteString doesn't have a toJSON instance. Changing the type to Text solves
this easily.
Resolves#3237
Previously using a generic mimetype handler failed when any kind of select= was given, because
we tried to cast the select-result to the original table type. With this change, this cast is
only applied when select=* is given implicitly or explicitly. This is the only case where this
makes sense, because this guarantees that correct columns are selected in the correct order for
this cast to succeed.
Resolves#3160
