Horton

Re: ExecRTCheckPerms() and many prunable partitions

Amit Langote <amitlangote09@gmail.com>

View thread View on postgr.es
From: Amit Langote <amitlangote09@gmail.com>
To: Alvaro Herrera <alvherre@alvh.no-ip.org>
Cc: Ian Lawrence Barwick <barwick@gmail.com>, Tom Lane <tgl@sss.pgh.pa.us>, David Rowley <dgrowleyml@gmail.com>, Greg Stark <stark@mit.edu>, Julien Rouhaud <rjuju123@gmail.com>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2022-12-02T11:13:28Z
Lists: pgsql-hackers

Attachments

On Fri, Dec 2, 2022 at 7:00 PM Alvaro Herrera <alvherre@alvh.no-ip.org> wrote:
> On 2022-Dec-02, Amit Langote wrote:
> > This sounds like a better idea than adding a new AttrMap, so done this
> > way in the attached 0001.
>
> Thanks for doing that!  I have pushed it, but I renamed
> ri_RootToPartitionMap to ri_RootToChildMap and moved it to another spot
> in ResultRelInfo, which allows to simplify the comments.

Thanks.

> > I've also merged into 0002 the delta patch I had posted earlier to add
> > a copy of RTEPermInfos into the flattened permInfos list instead of
> > adding the Query's copy.
>
> Great.  At this point I have no other comments, except that in both
> parse_relation.c and rewriteManip.c you've chosen to add the new
> functions at the bottom of each file, which is seldom a good choice.
> I think in the case of CombineRangeTables it should be the very first
> function in the file, before all the walker-type stuff; and for
> Add/GetRTEPermissionInfo I would suggest that right below
> addRangeTableEntryForENR might be a decent choice (need to fix the .h
> files to match, of course.)

Okay, I've moved the functions and their .h declarations to the places
you suggest.  While at it, I also uncapitalized Add/Get, because
that's how the nearby functions in the header are named.

Thanks again for the review.  The patch looks much better than it did
3 weeks ago.

-- 
Thanks, Amit Langote
EDB: http://www.enterprisedb.com

Commits

  1. Add a test case for a316a3bc

  2. Correctly set userid of subquery relations' child rels

  3. Fix buggy recursion in flatten_rtes_walker().

  4. Remove some dead code in selfuncs.c

  5. Update outdated comment in ApplyRetrieveRule

  6. Rework query relation permission checking

  7. Generalize ri_RootToPartitionMap to use for non-partition children

  8. Stop accessing checkAsUser via RTE in some cases

  9. Add 'missing_ok' argument to build_attrmap_by_name