On Fri, Feb 5, 2010 at 06:40, Tim Bunce <Tim(dot)Bunce(at)pobox(dot)com> wrote:
> This is the third update to the fourth of the patches to be split out
> from the former 'plperl feature patch 1'.
> Changes in this patch:
> - Added plperl.on_plperl_init and plperl.on_plperlu_init GUCs
> Both are PGC_SUSET
> SPI functions are not available when the code is run.
> Errors are detected and reported as ereport(ERROR, ...)
> Corresponding documentation and tests for both.
*sniffle* OK I think I agree with everyone else on setting this as
PGC_SUSET until we can either prove it can be USERSET or we can fix it
so we can check permissions on SET properly. It seems if you really
wanted a user to be able to set it you should be able to define a
SECURITY DEFINER function that sets it to a string you pass in or
something. Obviously not part of core postgres...
> - Renamed plperl.on_perl_init to plperl.on_init
*shrug* OK (I think there was still some flack on this var, but I
think its ok-- we can discuss that in a separate thread if people
still disagree :) )
> - Improved state management of select_perl_context()
> An error during interpreter initialization will leave
> the state (interp_state etc) unchanged.
> - The utf8fix code has been greatly simplified.
> - More code comments re PGC_SUSET and no access to SPI functions.
I like the doc changes and think the new comment about %_SHARED being
unsafe is good.
All looks good to me. Ill mark it as "Ready for Committer"
In response to
pgsql-hackers by date
|Next:||From: Tom Lane||Date: 2010-02-05 21:56:22|
|Subject: WIP patch for system-catalog vacuuming via a relation map|
|Previous:||From: Bruce Momjian||Date: 2010-02-05 21:01:52|
|Subject: Re: Confusion over Python drivers|