Log In - Home Page

CommitFest 2011-06 (Closed)

New Patch - Activity Log - CommitFest Topics

The most recent three comments for each patch will be displayed below. To view all the comments for a particular patch, or to add a comment or make other changes, click on the patch name.

Status Summary. Committed: 29, Returned with Feedback: 29, Rejected: 1. Total: 59.

Pending Patches

No patches.

Committed Patches

Patch Name Status Author Reviewers Date Closed
Binary in/out for void
Patch by rsmogura on 2011-02-21: Initial version.
Committed Radosław Smogura Nobody 2011-02-22
PL/Python tracebacks
Patch by wulczer on 2011-03-03: Tracebacks now in errcontext.
Patch by tgl on 2011-03-07: Patch split into 3 parts that address slightly distinct issues.
Comment by wulczer on 2011-03-20: New version with the remaining part revamped.
Committed Jan Urbański Alex Hunsaker 2011-04-06
missing tab completions for COMMENT ON
Patch by schmiddy on 2011-05-29: Initial version.
Review by rhaas on 2011-06-12: Committed and back-patched to 9.1 (with some changes)
Committed Josh Kupershmidt Robert Haas 2011-06-12
Use symbolic constants when indexing arrays instead of i++
Patch by rhpostgres on 2011-06-14: Initial version.
Review by fgp on 2011-06-16: Patch looks good and passes "make check". Marking as Ready for Committer.
Committed Richard Hopkins Florian Pflug 2011-06-16
pg_hba.conf field-specific keywords
Patch by direvus on 2011-06-17: v2 of patch, updated to HEAD.
Patch by alvherre on 2011-06-21: v3
Comment by okbobcz on 2011-06-27: processing of load_ident is different than load_hba
Committed Brendan Jurd Pavel Stehule, Álvaro Herrera 2011-06-20
Eliding no-op varchar length coercions
Patch by nmisch on 2011-06-11: v2
Review by rhaas on 2011-06-19: Could use a few more comments, but otherwise looks good to me.
Patch by nmisch on 2011-06-19: v3
Committed Noah Misch Robert Haas, Alexey Klyukin 2011-06-22
deadlock_timeout at PGC_SUSET
Patch by nmisch on 2011-06-11: Initial version.
Review by hanada on 2011-06-17: Seems fine, marked as "Ready for committer"
Committed Noah Misch Shigeru Hanada 2011-06-22
smallserial / serial2 support
Patch by mikepultz on 2011-06-08: Updated patch including documentation
Review by brar on 2011-06-08: Everything seems fine so far. We probably need some more agreement on whether this is useful/necessary and a quick re-review from a more experienced person than me.
Committed Mike Pultz Brar Piening 2011-06-22
crash-safe visibility map
Patch by nmisch on 2011-06-19: v6 from rhaas
Review by nmisch on 2011-06-19: patch has some bogus hunks, but otherwise ready
Patch by nmisch on 2011-06-20: v7 from rhaas - extra hunks removed
Committed Robert Haas Noah Misch 2011-06-22
heap_hot_search_buffer refactoring
Patch by rhaas on 2011-06-06: Initial version.
Review by jdavis on 2011-06-25: Bug
Review by jdavis on 2011-06-25: Review
Committed Robert Haas based on a patch by Heikki Linnakanagas Jeff Davis 2011-06-27
use less space in xl_xact_commit
Patch by leonardof on 2011-05-19: Initial version.
Patch by leonardof on 2011-05-25: Fixed some English mistakes
Patch by leonardof on 2011-06-15: New patch, based on Simon Riggs suggestion
Committed Leonardo Francalanci Simon Riggs 2011-06-28
DROP Reworks Part.0 - missing_ok support on get_object_address
Patch by kaigai on 2011-06-19: Patch revised. (v3)
Review by rhaas on 2011-06-22: Gripes about locking.
Patch by kaigai on 2011-06-23: Patch revised. (v4)
Committed KaiGai Kohei Robert Haas 2011-06-28
CHECK constraints with NOT VALID
Patch by alvherre on 2011-06-01: Initial version.
Patch by alvherre on 2011-06-16: Updated patch
Comment by jcasanov on 2011-06-18: Reviewed by Thom Brown and Jaime Casanova. Currently the patch compiles and works as expected... besides a few comments about the docs everything looks ok.
Committed Álvaro Herrera Jaime Casanova 2011-06-30
psql include files using relative paths
Review by schmiddy on 2011-05-28: Review of the third version.
Patch by singh.gurjeet on 2011-06-05: Newer version of the patch submitted, with review comments incorporated.
Patch by singh.gurjeet on 2011-06-06: Minor C code comment improvements.
Committed Gurjeet Singh Josh Kupershmidt 2011-07-06
standardize queries in describe.c
Patch by schmiddy on 2011-05-13: Initial version.
Review by merlinm on 2011-06-14: Brought patch up to HEAD. Looks good -- can commit as-is, but author should justify putting semi-colons in each query vs the print function. Marked waiting on author.
Comment by merlinm on 2011-06-16: Given lack of comment, decided to push patch as is...
Committed Josh Kupershmidt Merlin Moncure 2011-07-06
Latch implementation: wake on postmaster death, reduce archiver wakeups
Patch by sternocera on 2011-06-22: New patch that addresses Fujii's second, more recent set of concerns
Patch by sternocera on 2011-06-25: Patch that addresses Fujii's third and most recent set of concerns.
Comment by MasaoFujii on 2011-06-30: Further comments
Committed Peter Geoghegan Heikki Linnakangas, Fujii Masao 2011-07-08
Make relation_openrv atomic wrt DDL
Review by rhaas on 2011-07-06: Review.
Patch by rhaas on 2011-07-07: Hack, hack.
Review by nmisch on 2011-07-07: rhaas version seems sound; cosmetic points
Committed Noah Misch Greg Stark, Robert Haas 2011-07-09
Move childoffnum to child in GISTInsertStack
Patch by smagen on 2011-05-24: Initial version.
Comment by aglio on 2011-07-14: Patch needs further revision to be safe per discussion. Deferring.
Committed Alexander Korotkov Jeff Janes, Heikki Linnakangas 2011-07-14
Add ability to constrain backend temporary file space
Patch by c2main on 2011-06-15: updated patch from Mark
Review by c2main on 2011-06-20: review (except performance review) + updated patch with a fix and cosmetic changes.
Patch by aglio on 2011-06-27: Updated patch version
Committed Mark Kirkwood Tatsuo Ishii 2011-07-17
reduce overhead of frequent table locks
Patch by rhaas on 2011-06-12: v3. Fixed GetLockConflicts(). Other minor fixes. Discussion of remaining issues.
Patch by rhaas on 2011-06-27: v4. Fixed pg_locks, removed debugging code, a bit of renaming.
Review by jdavis on 2011-07-12: Looks good!
Committed Robert Haas Jeff Davis 2011-07-18
Avoid index rebuilds for no-rewrite ALTER TABLE ALTER TYPE
Review by rhaas on 2011-07-06: doesn't seem to compile?
Patch by nmisch on 2011-07-06: v4
Review by rhaas on 2011-07-07: Looks good, with minor caveats. Marking Ready for Committer.
Committed Noah Misch Robert Haas 2011-07-18
enhanced diagnostics
Patch by okbobcz on 2011-07-07: fixed issues, merge with Hanada's patch
Review by theory on 2011-07-10: Looks good, just tweaked the documentation in a new patch.
Patch by okbobcz on 2011-07-15: final patch
Committed Pavel Stehule David Wheeler 2011-07-18
savepoint commit performance
Patch by rhaas on 2011-02-24: Initial version.
Review by heikki on 2011-06-06: Needs some fixing
Committed Simon Riggs Heikki Linnakangas 2011-07-19
Cascade Replication
Patch by MasaoFujii on 2011-07-05: Updated version; simplify the patch to focus on the main problem.
Comment by MasaoFujii on 2011-07-11: Simon's another comments.
Patch by MasaoFujii on 2011-07-11: Updated version; addressed all the issues raised by Simon.
Committed Fujii Masao Simon Riggs 2011-07-19
security label of shared object
Patch by kaigai on 2011-07-02: pg_shseclabel.provider was re-defined as NameData, instead of Text
Patch by kaigai on 2011-07-06: Toast table & index removed (v5)
Comment by rhaas on 2011-07-20: committed... mostly
Committed KaiGai Kohei Joe Conway, Robert Haas 2011-07-20
XML error handling improvement to fix XPATH bug
Patch by fgp on 2011-06-22: Made to work libxml 2.6.23 and 2.7.8 and fixes all but two of Noah's complains. These two are still being discussed.
Review by nmisch on 2011-06-24: New version achieves goals. Further discussion, minor code notes
Patch by nmisch on 2011-06-27: Cosmetic revisions; looks good
Committed Florian Pflug Noah Misch 2011-07-20
Bugfix for XPATH() if expression returns a scalar value
Patch by fgp on 2011-06-09: Rebased onto HEAD
Review by fgp on 2011-06-29: Radoslaw thinks the patch is fine, expect for the escaping of the return values, even though that will create values of type XML which are not well-formed.
Review by rsmogura on 2011-07-14: I like scalars, I don't like escaping, according to others escaping is ok. My objections to escaping are as follow:
- possibility of double escaping
- problems with de-escaping this on client side
- this patch doens't resolves all mentioned problems about inserting wrong XMLs, if we are going in this mentioned spirit the result of text node must not only be escaped, but wrapped by any kind of root element
- backward compatibility problems
- when I ask for text() I want to extract text not a XML, if I want to get node I will ask /* not a text(/*)
- if user recreates XML which is serialized as text in some node, after patch he will insert improper XMLs, because he will get escaped text - no root element.
Committed Florian Pflug Radosław Smogura 2011-07-21
Bugfix for XPATH() if text or attribute nodes are selected
Review by fgp on 2011-06-29: Radoslaw believes that the current behaviour is per design, even though it creates instances of type XML which are not well-formed.
Review by rsmogura on 2011-07-14: Peter Eisentraut is up for this so I bump it up. My objections was:
- possibility of double escaping
- problems with de-escaping this on client side
- this patch doens't resolves all mentioned problems about inserting wrong XMLs, if we are going in this mentioned spirit the result of text node must not only be escaped, but wrapped by any kind of root element
- backward compatibility problems
- if user recreates XML which is serialized as text in some node, after patch he will insert improper XMLs, because he will get escaped text - no root element.
- when I ask for text() I want to extract text not a XML, if I want to get node I will ask /* not a text(/*)
Comment by aglio on 2011-07-14: Ready to go, per Radoslaw.
Committed Florian Pflug Radosław Smogura 2011-07-21
Bugfix for libpq's PQsetvalue
Patch by pasha_golub on 2011-07-18: Initial version.
Comment by pasha_golub on 2011-07-18: Merlin Moncure wrote about this bug and about my proposal about adding PQdeleteTupple
Comment by merlinm on 2011-07-18: this bug goes back to 9.0 (calling PQsetvalue will crash the client if you call it on a result returned from a query)
Committed Andrew Chernow Robert Haas 2011-07-21

Returned with Feedback

Patch Name Status Author Reviewers Date Closed
Expanding "a like any(array[1,2...])" into "a like 1 or a like 2"
Patch by chetan on 2011-06-02: Initial version.
Review by heikki on 2011-06-06: Has issues with multiple evaluation
Returned with Feedback Chetan Suttraway Heikki Linnakangas 2011-06-06
Prefered Types
Review by rhaas on 2011-06-06: Some concerns from Tom Lane.
Review by rhaas on 2011-06-06: Some concerns from Alvaro Herrera.
Comment by gatt on 2011-06-13: Please tell me, what must I do?
I see what you don`t know why. I can tell what i use this feature now, because i need call function without overloads use assign cast rules. First my step was using implicit casts... after it i have problems with too many implicit casts without priority and i try resolve this problem. that`s all.
The best will be calling single functions(without overloads) use assign cast rules. i don`t try to do this.
PS
Using casts priority you will get more complicated system and no more, it will be more complicated than typeisprefered.
PPS
I can change some problems in patch, but i think problem not in patch, problem in heads... As long as you cant choose best way to resolve this problems class i can just wait... May be this problem class has no one solution, as you want think..?

PPPS
After few days thinking...
1. Overload functions must marked as OVERLOAD (or for commpatibility for functions without overload like SINGLE)
2. Functions without overloading must use Assign cast rules
3. Overload functions can work like now (but more TRUEst solution will be type equal)
Returned with Feedback Zotov <zotov@oe-it.ru> Nobody 2011-06-06
Binary in/out for aclitem
Patch by rsmogura on 2011-02-23: Initial version.
Review by rhaas on 2011-06-06: Patch needs to add binary in/out for all standard types, not just fix one at a time.
Returned with Feedback Radosław Smogura Nobody 2011-06-06
Buffer Cache Hibernation
Comment by c2main on 2011-05-31: the patch need to be send by mail to postgresql-hackers mailling list
Patch by iwasaki on 2011-06-05: This is the final version from me.
Review by gsmith on 2011-06-07: Suggestions for GUC renaming, code adjustment, making file access less tied to cache internals. Downstream comments suggest an extension may be a more appropriate way to code this too.
Returned with Feedback Mitsuru IWASAKI Greg Smith 2011-06-09
TG_DEPTH
Comment by kgrittn on 2011-01-28: Initial discussion.
Review by fgp on 2011-06-10: Reviewed patch and rebased onto HEAD. Depth tracking should IMHO be decoupled from subtransaction enter/exit.
Comment by kgrittn on 2011-06-10: Flagging RwF to leave time for SSI. Will pick up from Florian's review.
Returned with Feedback Kevin Grittner Florian Pflug 2011-06-10
triggered_change_notification
Patch by kgrittn on 2011-02-22: Initial version.
Patch by kgrittn on 2011-02-23: v2 - use syscache; fix error messages
Comment by kgrittn on 2011-06-10: Marking RwF to free time for any SSI work. Will resubmit as an extension in a later CF.
Returned with Feedback Kevin Grittner Dimitri Fontaine 2011-06-10
POSIX shared memory
Patch by agentm on 2011-04-12: Initial version.
Review by heikki on 2011-06-16: Bitrotted. Locking changes need to be splitted out
Returned with Feedback agentm Heikki Linnakangas 2011-06-16
Fix nested CASE-WHEN
Patch by heikki on 2011-06-08: Initial version.
Comment by okbobcz on 2011-06-16: there is bug when nested CASE statement was used

ERROR: invalid expression parameter reference (1 levels up, while
stack is only 1 elements deep)
Comment by okbobcz on 2011-06-27: This patch was taken back
Returned with Feedback Heikki Linnakangas Pavel Stehule 2011-06-27
hint bit i/o reduction cache
Patch by merlinm on 2011-05-13: Initial version.
Comment by merlinm on 2011-05-13: The original idea for the cache. Note Tom Lane's objection downthread.
Review by merlinm on 2011-07-05: Review/commentary by rhaas. In short: approach has potential but needs work.
Returned with Feedback Merlin Moncure Simon Riggs, Robert Haas 2011-07-05
add AS EXPLICIT to CREATE CAST
Patch by fabien on 2011-05-27: Initial version.
Comment by direvus on 2011-06-17: Patch doesn't apply, plus some commentary.
Returned with Feedback Fabien Brendan Jurd 2011-07-05
Cache estimates & cache access cost - WIP
Patch by gsmith on 2011-05-26: Initial version.
Patch by c2main on 2011-06-14: Updated patch, include GUC, pg_class and ANALYZE OSCACHE changes, not the plugin.
Review by rhaas on 2011-06-14: objections
Returned with Feedback Cédric Villemain Greg Smith 2011-07-06
Parameterized aggregate subquery
Patch by umitanuki on 2011-06-17: Revised with a few editorial changes. Nothing new.
Review by yhavinga on 2011-06-29: Some comments about the patch and a test that resulted in the error: plan should not reference subplan's variable
Patch by umitanuki on 2011-07-02: Fix error, and improve comments.
Returned with Feedback Hitoshi Harada Yeb Havinga 2011-07-08
Time-Delayed Standbys
Patch by rhaas on 2011-06-29: v2.
Comment by MasaoFujii on 2011-06-30: Another comments.
Comment by rhaas on 2011-07-09: needs more work
Returned with Feedback Robert Haas Simon Riggs, Fujii Masao 2011-07-09
New Timeline for PITR
Patch by dfetter on 2011-06-15: Initial version.
Comment by jcasanov on 2011-06-27: After some comments from Josh, Heikki, Robert, Tom and others. Author decided to make a new version
Returned with Feedback David Fetter Jaime Casanova 2011-07-09
add GiST support for BOX @> POINT queries
Comment by rhaas on 2011-06-14: Ping.
Patch by umitanuki on 2011-06-20: Ask about adding new function.
Comment by umitanuki on 2011-06-20: Additional comment. Old design that wants to modify original function would be ok. Why not considering circle and polygon.
Returned with Feedback Andrew Tipton Hitoshi Harada 2011-07-09
Core Extensions relocation
Patch by gsmith on 2011-06-09: Initial version.
Review by gsmith on 2011-06-16: Need to refresh makefiles
Comment by rhaas on 2011-07-05: Ping.
Returned with Feedback Greg Smith Josh Berkus 2011-07-09
DROP Reworks Part.1 - Consolidate routines to handle DropStmt
Review by rhaas on 2011-07-06: Some suggested changes.
Patch by kaigai on 2011-07-08: Revised patch according to the suggestion (v3)
Review by rhaas on 2011-07-09: Have a plan, but not enough time to implement it in this CF.
Returned with Feedback KaiGai Kohei Robert Haas 2011-07-09
Fast GiST Index Build (WIP)
Patch by smagen on 2011-06-21: New version of patch
Patch by smagen on 2011-07-01: New version of patch. New options, comments, bug fixes.
Comment by rhaas on 2011-07-09: still WIP, marking Returned with Feedback
Returned with Feedback Alexander Korotkov Heikki Linnakangas 2011-07-09
Fix leaky-views, part1
Review by nmisch on 2011-07-07: Needs more future-proofing
Patch by nmisch on 2011-07-10: v6
Review by nmisch on 2011-07-10: Ready for committer as far as this part goes, but shall we return with feedback and merge into the part 2 patch?
Returned with Feedback KaiGai Kohei Noah Misch 2011-07-10
Fix leaky-views, part2
Patch by kaigai on 2011-06-06: Initial version.
Patch by kaigai on 2011-07-03: Syntax support got separated. Also see the part-1 thread.
Review by nmisch on 2011-07-07: Criteria for identifying leakproof functions must change
Returned with Feedback KaiGai Kohei Noah Misch 2011-07-11
PITR using backup history file
Patch by Jun Ishizuka on 2011-07-08: v3 patch
Review by ssinger on 2011-07-10: Review of updated patch.
A few questions for the author.
Comment by Jun Ishizuka on 2011-07-12: Change "Returned with feedback"
Returned with Feedback Jun Ishizuka Steve Singer, Fujii Masao 2011-07-12
Allow pg_archivecleanup to ignore extensions
Patch by gsmith on 2011-04-07: Initial version.
Comment by aglio on 2011-07-14: Simon claims to have reviewed the patch, but I cannot find any review in the archives. No response from patch authors. Returing.
Returned with Feedback Jaime Casanova, Greg Smith Simon Riggs 2011-07-14
Add json contrib module
Comment by aglio on 2011-07-07: Joey will provide updated patch soon.
Comment by aglio on 2011-07-14: Still waiting for new patch version.
Patch by joeyadams on 2011-07-15: Addressed issues discussed in review, but have questions about Unicode handling.
Returned with Feedback Joey Adams Bernd Helmle 2011-07-14
csvlog_fields/csvlog_header
Patch by sfrost on 2011-02-15: Initial version.
Review by alexhunsaker on 2011-07-09: new patch, still a few questions
Returned with Feedback Stephen Frost Alex Hunsaker 2011-07-15
enhanced-errors
Patch by ssinger on 2011-06-21: Updated patch sent in by Pavel
Comment by ssinger on 2011-06-21: The updated patch addresses the issues from the review.
Review by tgl on 2011-07-15: I think this needs to hew a lot more closely to the SQL standard.
Returned with Feedback Pavel Stěhule Steve Singer 2011-07-15
libpq SSL with non-blocking sockets (WIP)
Patch by ssinger on 2011-07-03: Updated patch from Martin
Review by ssinger on 2011-07-03: This version of the patch looks good
Review by tgl on 2011-07-15: Still needs considerable work: doesn't handle low-memory situations sanely, and covers only one of three places where we are at risk.
Returned with Feedback Martin Pihlak Steve Singer 2011-07-15
pg_comments system view
Patch by kaigai on 2011-06-19: The v5 patch
Patch by schmiddy on 2011-07-08: v6 patch, included is_system column to pg_comments
Review by rhaas on 2011-07-18: is_system columns seems too psql-specific
Returned with Feedback Robert Haas, Josh Kupershmidt KaiGai Kohei 2011-07-18
Show all errors in postgresql.conf instead of only the first one
Comment by tgl on 2011-07-16: I think we should take another look at what behavior we actually want overall.
Patch by tgl on 2011-07-19: This is Alvaro's V3, which for some reason he didn't add to the CF entry.
Comment by alvherre on 2011-07-19: Alexey is going to rework per Tom's comments above. I'm going to mark it as returned with feedback.
Returned with Feedback Alexey Klyukin Florian Pflug, Alvaro Herrera 2011-07-19
sepgsql - userspace access vector cache
Patch by kaigai on 2011-07-14: The latest version (v4)
Review by yhavinga on 2011-07-20: Some minor remarks, nothing serious, great improvement in performance.
Review by rhaas on 2011-07-20: Syscache memory usage and initialization cost seem problematic; still need to review UAVC portion of patch.
Returned with Feedback KaiGai Kohei Yeb Havinga 2011-07-20

Rejected Patches

Patch Name Status Author Reviewers Date Closed
Allow \dd to show constraint comments
Patch by schmiddy on 2011-05-18: Initial version.
Review by merlinm on 2011-06-29: looks good.
Review by rhaas on 2011-07-08: Propose we reject this.
Rejected Josh Kupershmidt Merlin Moncure 2011-07-08