CF 2009-07: initial reviewing assignments

Initial reviewing assignments are below. Please go to
http://commitfest.postgresql.org/action/commitfest_view/inprogress and
edit the patch you've been assigned, adding your name as a reviewer.
Please do this right away so that it's clear which patches still need
reviewers to be assigned. If you don't want to review the patch that
I've assigned you for some reason, please email me and I'll try to
assign you a different one. Please try to complete your initial
review by end-of-day Saturday (four days from now).

In order to preserve my sanity, it is absolutely critical that you
update http://commitfest.postgresql.org/action/commitfest_view/inprogress
each and every time you review a patch. Go to the patch, set the
comment type to "Review", and enter the message-ID of your post to
-hackers and a brief summary of the main thrust of your review.
Submit. Then edit the patch. If you concluded that the patch is
ready for final review by a committer, set the status of the patch to
"Ready for Committer". If you concluded that the patch requires
changes, set the status to "Waiting on Author". If the patch looks
like it can't reasonably be adjusted in time for this CommitFest, set
the status to "Returned with Feedback" or "Rejected" as seems
appropriate to you. In some cases you will be summarizing the
opinions of others (e.g. committers or other community members who
have weighed in on the topic), not just your own feelings. In other
cases, no one else will weigh in; you'll have to do what you think
best.

If new versions of the patch are posted, the patch author SHOULD add a
"Patch"-type comment and change the patch status back to "Needs
Review". But they might not - in which case, please do it for them.
The most difficult part of managing a CommitFest by far is figuring
out what state all the patches are in, and the more you can keep the
CommitFest up to date, the easier that will be.

I would like to set a general expectation that patches which need more
work should be resubmitted within 3-4 days of the time that you post
your review, and that patches which still need significant work after
being resubmitted twice should be deferred to the next CommitFest. We
haven't discussed these thresholds on pgsql-hackers, so I can't claim
there's any consensus behind them, but I think they're about what is
reasonable. Big, significant patches may need a bit more leeway than
this, and that is fine. Small patches with obvious bugs should be
bounced very aggressively (for example, if the patch that you are
assigned is less than 500 lines and fails to compile, or pass
regression tests, or crashes, or just doesn't work, it's probably fair
to move that to "Returned with Feedback" without further ado. We're
trying to commit the patches that are done now, not the ones that will
be done a month from now; we're trying to provide feedback on the
others.

When you are done reviewing the patch you've been assigned, or at
least far enough along that you don't mind starting on another one,
please reply to THIS list and let me know. I would like to set up a
better system for managing this process so that everyone doesn't need
to spam the list every time they need a new assignment, but if I get
run over by a truck and all of the information is only in my private
email, we'll be hosed - so copy the list when requesting a new
assignment.

Thanks,

...Robert

==SECURITY==

Robert Haas <robertmhaas(at)gmail(dot)com>
- Security Enhanced PostgreSQL

Joshua Tolley <eggyknap(at)gmail(dot)com>
- Security checks on LargeObject

Nikhil Sontakke <nikhil(dot)sontakke(at)enterprisedb(dot)com>
- GRANT ON ALL IN schem

==REPLICATION==

Jaime Casanova <jcasanov(at)systemguards(dot)com(dot)ec>
- Provide support for multiplexing SIGUSR1 signal

Martin Pihlak <martin(dot)pihlak(at)gmail(dot)com>
- Non-blocking communication between a frontend and a backend (pqcomm)

I am not assigning a reviewer for Sync Rep right now because I think
we should review the two patches above (which are prerequisites)
first. However, we will definitely need someone (hopefully more than
one person) to review this later in the CommitFest.

==PERFORMANCE==

Nathan Boley <npboley(at)gmail(dot)com>
- join removal

Dimitri Fontaine <dfontaine(at)hi-media(dot)com>
- ALTER TABLE ... ALTER COLUMN ... SET DISTINCT

Kevin Grittner <Kevin(dot)Grittner(at)wicourts(dot)gov>
- Revise parallel pg_restore's scheduling heuristic

Alex Hunsaker <badalex(at)gmail(dot)com>
- Set PD_ALL_VISIBLE flag in HOT pruning

Dan Colish <dcolish(at)gmail(dot)com>
- Avoid manual shift-and-test logic in AllocSetFreeIndex

Andres Freund <andres(at)anarazel(dot)de>
- remove {join,from}_collapse_limit, add enable_join_ordering

Stephen Frost <sfrost(at)snowman(dot)net>
- Indexam API changes

Josh Berkus <josh(at)agliodbs(dot)com>
- Index-only quals

Abhijit Menon-Sen <ams(at)toroid(dot)org>
- Merge append

==SQL LANGUAGE FEATURES==

Bernd Helmle <bernd(at)oopsware(dot)de>
- Named and mixed notation for PL

Euler Taveira de Oliveira <euler(at)timbira(dot)com>
- Support for EEEE in to_char()

Jeff Davis <pgsql(at)j-davis(dot)com>
- Deferrable Unique constraints

Brendan Jurd <direvus(at)gmail(dot)com>
- Generalized Index Constraints

==CONTRIB MODULES==

David E. Wheeler <david(at)kineticode(dot)com>
- hstore enhancements

==ERROR REPORTING==

Wolfgang Wilhelm <wolfgang20121964(at)yahoo(dot)de>
- Have ParseConfigFile report all parsing errors

Dickson S. Guedes <listas(at)guedesoft(dot)net>
- report key values in duplicate-key errors

==MISCELLANEOUS==

Josh Williams <joshwilliams(at)ij(dot)net>
- Polygons