Streamlining vector support and PostgreSQL extension documentation #58899
Unanswered
WoutervdBrink
asked this question in
General
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
I just created a PR that adds information about requiring the
pgvectorextension to be loaded throughSchema::ensureVectorExtensionExists()when using PostgreSQL.So far, the
Schema::ensureVectorExtensionExists()method is currently mentioned in the AI SDK docs (here) and the vector search docs (here. The underlying, but perhaps quite usefulSchema::ensureExtensionExists(string $name, ?string $schema = null)method is not documented.Furthermore, the documentation currently states that only PostgreSQL supports vector search. While this is indeed currently the case (see here), MySQL 9.3 added support for vector types and vector functions, so support could be added there as well.
I think it would be useful to add information about the
ensureVectorExtensionExists()and/orensureExtensionExistsmethods. I would suggest a separate first-level heading inmigrations.md, between 'Indexes' and 'Events', but I'm not too sure about that, which is why I'm opening this discussion rather than immediately submitting a PR.Furthermore, I would like to know whether it would be considered useful to add vector support for MySQL servers (through the new column type and functions). Such a change would impact the documentation, as it adds a choice between MySQL and PostgreSQL rather than a requirement for PostgreSQL+pgvector.
Beta Was this translation helpful? Give feedback.
All reactions