Horton

Re: review: CHECK FUNCTION statement

Albe Laurenz <laurenz.albe@wien.gv.at>

View thread
From: "Albe Laurenz" <laurenz.albe@wien.gv.at>
To: "Pavel Stehule *EXTERN*" <pavel.stehule@gmail.com>
Cc: "Tom Lane *EXTERN*" <tgl@sss.pgh.pa.us>, "PostgreSQL Hackers" <pgsql-hackers@postgresql.org>
Date: 2011-12-09T11:27:36Z
Lists: pgsql-hackers

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →

  1. Add const qualifiers to node inspection functions

Pavel Stehule wrote:
> updated version
>
> changes:
>
> * CHECK FUNCTION ALL; is enabled - in this case functions from
> pg_catalog schema are ignored
>
> I looked on parser, and I didn't other changes there - IN SCHEMA, FOR
> ROLE are used more time there, so our usage will be consistent

> a small addition
> 
> * don't check SQL functions - are checked well now
> * don't check functions from information_schema too

One hunk in the patch fails due to conflict with
commit d5f23af6bfbc454e86dd16e5c7a0bfc0cf6189d0
(Peter Eisentraut's const patch).

There are also compiler warnings about discarded const
qualifiers in backend/nodes/copyfuncs.c,
backend/nodes/equalfuncs.c and backend/parser/gram.y.

There is a bug when ALL IN SCHEMA or ALL IN LANGUAGE
is used:

test=> CHECK FUNCTION ALL IN LANGUAGE plpgsql;
ERROR:  language "language" does not exist
test=> CHECK FUNCTION ALL IN SCHEMA laurenz;
ERROR:  schema "schema" does not exist

Something gets mixed up here.

I like the idea that CHECK FUNCTION ALL without additional
clauses works and ignores pg_catalog and information_schema!

I'm working on some documentation, but it won't be final before
the functionality is agreed upon.

Yours,
Laurenz Albe