Обсуждение: Warning about the 8.4 release
I am now warning that we have an unusually large number of open items that must be either completed or moved to the TODO list before 8.4 can be released. Everyone knows about the commit fest wiki items, but I am tracking 291 threads that need some type of attention; if only 50% of them are significant for 8.4, that still leaves +100 items that should be completed in the next month. You can see the full list here: http://momjian.us/cgi-bin/pgpatches If we just move them all to the TODO list, the TODO list will be significantly larger for 8.4 than in previous releases, and fixing items only gets harder as we get farther away from the email discussion that prompted the item (and we risk a MySQL 5.1 debacle). I have already approached developers to get help in completing these items, but got little assistance. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://enterprisedb.com + If your life is a hard drive, Christ can be your backup. +
Bruce Momjian <bruce@momjian.us> writes: > Everyone knows about the commit fest wiki items, but I am tracking 291 > threads that need some type of attention; if only 50% of them are > significant for 8.4, that still leaves +100 items that should be > completed in the next month. You can see the full list here: > http://momjian.us/cgi-bin/pgpatches > If we just move them all to the TODO list, the TODO list will be > significantly larger for 8.4 than in previous releases, ... and will be mostly junk. 90% of what you've got there is either already dealt with or not a bug, and certainly not new bugs in 8.4. regards, tom lane
Tom Lane wrote: > Bruce Momjian <bruce@momjian.us> writes: > > Everyone knows about the commit fest wiki items, but I am tracking 291 > > threads that need some type of attention; if only 50% of them are > > significant for 8.4, that still leaves +100 items that should be > > completed in the next month. You can see the full list here: > > > http://momjian.us/cgi-bin/pgpatches > > > If we just move them all to the TODO list, the TODO list will be > > significantly larger for 8.4 than in previous releases, > > ... and will be mostly junk. 90% of what you've got there is either > already dealt with or not a bug, and certainly not new bugs in 8.4. You are right that 90% are either not new or not bugs, but they are things that need correction, and no one benefits by just pushing them off to a later release. As I said, if even 50% are things that need work, they should be completed for 8.4 rather than added to the TODO list. These are all things that have been reported during the 8.4 release cycle. Are you saying we no longer care about fixing small issues? I think the project quality will suffer. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://enterprisedb.com + If your life is a hard drive, Christ can be your backup. +
Bruce Momjian wrote: > Everyone knows about the commit fest wiki items, but I am tracking 291 > threads that need some type of attention; if only 50% of them are > significant for 8.4, that still leaves +100 items that should be > completed in the next month. You can see the full list here: > > http://momjian.us/cgi-bin/pgpatches I notice that some of the items in this commitfest are in your queue. Some things are duplicated; for example parallel restore appear more than once (I count six times). Column privileges is also several times. Also some patches have already been applied; for example: Item 14, "[HACKERS] Another refactoring proposal: move stuff into nodes/nodeFuncs.[ch]" was applied. Item 57, "Re: [HACKERS] autovacuum and TOAST tables" was applied. Item 78, bug 4495 is not a memory leak Item 79, bug 4496 is not a leak either Item 86 was rejected Item 103 (plpython) is a commitfest item that was returned to author I'm not sure why is item 106 (heap_formtuple) listed? Item 118 (pgdump roles, Ibrar Ahmed) is applied Item 130 (PD_PAGE_FULL) is bogus; drop it I'm not sure why you have planetpostgresql.org items? Item 155 (toast by chunk end) can be dropped from this list; it's an unneeded patch. Item 162 is an open item??? "cool hacks with recursive queries" Item 166, "designated initializers" can be dropped (part of reloptions patch that's already committed) Item 184 "working in the US" is a pgsql-core email, not even in the archives ... Item 197, "links to cvsweb from archive" can be dropped; problem was fixed. [checks] Uh, actually it is broken again. I guess Marc upgraded the web server. Item 242, "generic reloptions improvement", is applied Item 272, "protein database", is here because ...? Item 286 (32/64 bits) was already fixed by yourself? -- Alvaro Herrera http://www.CommandPrompt.com/ PostgreSQL Replication, Consulting, Custom Development, 24x7 support
I realize the list is incomplete and inaccurate. My point is if only 50% need work, we still have lots of work to do. I will update my mailbox and post a new version soon. --------------------------------------------------------------------------- Alvaro Herrera wrote: > Bruce Momjian wrote: > > > Everyone knows about the commit fest wiki items, but I am tracking 291 > > threads that need some type of attention; if only 50% of them are > > significant for 8.4, that still leaves +100 items that should be > > completed in the next month. You can see the full list here: > > > > http://momjian.us/cgi-bin/pgpatches > > I notice that some of the items in this commitfest are in your queue. > Some things are duplicated; for example parallel restore appear more > than once (I count six times). Column privileges is also several > times. > > Also some patches have already been applied; for example: > > Item 14, "[HACKERS] Another refactoring proposal: move stuff into > nodes/nodeFuncs.[ch]" was applied. > > Item 57, "Re: [HACKERS] autovacuum and TOAST tables" was applied. > > Item 78, bug 4495 is not a memory leak > > Item 79, bug 4496 is not a leak either > > Item 86 was rejected > > Item 103 (plpython) is a commitfest item that was returned to author > > I'm not sure why is item 106 (heap_formtuple) listed? > > Item 118 (pgdump roles, Ibrar Ahmed) is applied > > Item 130 (PD_PAGE_FULL) is bogus; drop it > > I'm not sure why you have planetpostgresql.org items? > > Item 155 (toast by chunk end) can be dropped from this list; it's an unneeded patch. > > Item 162 is an open item??? "cool hacks with recursive queries" > > Item 166, "designated initializers" can be dropped (part of reloptions > patch that's already committed) > > Item 184 "working in the US" is a pgsql-core email, not even in the > archives ... > > Item 197, "links to cvsweb from archive" can be dropped; problem was > fixed. [checks] Uh, actually it is broken again. I guess Marc upgraded > the web server. > > Item 242, "generic reloptions improvement", is applied > > Item 272, "protein database", is here because ...? > > Item 286 (32/64 bits) was already fixed by yourself? > > -- > Alvaro Herrera http://www.CommandPrompt.com/ > PostgreSQL Replication, Consulting, Custom Development, 24x7 support -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://enterprisedb.com + If your life is a hard drive, Christ can be your backup. +
On Tue, Jan 6, 2009 at 3:21 PM, Bruce Momjian <bruce@momjian.us> wrote: > I am now warning that we have an unusually large number of open items > that must be either completed or moved to the TODO list before 8.4 can > be released. > > Everyone knows about the commit fest wiki items, but I am tracking 291 > threads that need some type of attention; if only 50% of them are > significant for 8.4, that still leaves +100 items that should be > completed in the next month. You can see the full list here: 243 seems like a priority for release :-p 253 was a website issue that's been fixed. 158 was fixed this morning by pure chance. 170 is a private message on -core And the one bug (which I can't find now) that I've logged recently is missing... -- Dave Page EnterpriseDB UK: http://www.enterprisedb.com
Dave Page wrote: > On Tue, Jan 6, 2009 at 3:21 PM, Bruce Momjian <bruce@momjian.us> wrote: > > I am now warning that we have an unusually large number of open items > > that must be either completed or moved to the TODO list before 8.4 can > > be released. > > > > Everyone knows about the commit fest wiki items, but I am tracking 291 > > threads that need some type of attention; if only 50% of them are > > significant for 8.4, that still leaves +100 items that should be > > completed in the next month. You can see the full list here: > > 243 seems like a priority for release :-p > 253 was a website issue that's been fixed. > 158 was fixed this morning by pure chance. Can you give subject lines on this? I didn't mean for people to actually start working on items, but rather just to see the scope of the problem. I am making updates based on Alvaro's comments now. > 170 is a private message on -core OK removed. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://enterprisedb.com + If your life is a hard drive, Christ can be your backup. +
On Tue, 2009-01-06 at 10:21 -0500, Bruce Momjian wrote: > I have already approached developers to get help in completing these > items, but got little assistance. If you can send me the list that you think applies to me, I'll work on it. I don't want to spend the time to read every entry if you already have. Thanks. -- Simon Riggs www.2ndQuadrant.comPostgreSQL Training, Services and Support
Alvaro Herrera wrote: > Bruce Momjian wrote: > > > Everyone knows about the commit fest wiki items, but I am tracking 291 > > threads that need some type of attention; if only 50% of them are > > significant for 8.4, that still leaves +100 items that should be > > completed in the next month. You can see the full list here: > > > > http://momjian.us/cgi-bin/pgpatches > > I notice that some of the items in this commitfest are in your queue. > Some things are duplicated; for example parallel restore appear more > than once (I count six times). Column privileges is also several > times. > > Also some patches have already been applied; for example: > > Item 14, "[HACKERS] Another refactoring proposal: move stuff into > nodes/nodeFuncs.[ch]" was applied. OK, removed. > Item 57, "Re: [HACKERS] autovacuum and TOAST tables" was applied. Removed. > Item 78, bug 4495 is not a memory leak > > Item 79, bug 4496 is not a leak either > > Item 86 was rejected OK, on above. > Item 103 (plpython) is a commitfest item that was returned to author OK, I will grab that from the commit wiki and add to the TODO list later. Below all removed: > I'm not sure why is item 106 (heap_formtuple) listed? > > Item 118 (pgdump roles, Ibrar Ahmed) is applied > > Item 130 (PD_PAGE_FULL) is bogus; drop it > > I'm not sure why you have planetpostgresql.org items? > > Item 155 (toast by chunk end) can be dropped from this list; it's an unneeded patch. > > Item 162 is an open item??? "cool hacks with recursive queries" > > Item 166, "designated initializers" can be dropped (part of reloptions > patch that's already committed) > > Item 184 "working in the US" is a pgsql-core email, not even in the > archives ... > > Item 197, "links to cvsweb from archive" can be dropped; problem was > fixed. [checks] Uh, actually it is broken again. I guess Marc upgraded > the web server. > > Item 242, "generic reloptions improvement", is applied > > Item 272, "protein database", is here because ...? > Item 286 (32/64 bits) was already fixed by yourself? This needs Magnus to do the MSVC build. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://enterprisedb.com + If your life is a hard drive, Christ can be your backup. +
Dave Page wrote: > On Tue, Jan 6, 2009 at 4:12 PM, Bruce Momjian <bruce@momjian.us> wrote: > > Dave Page wrote: > > >> 243 seems like a priority for release :-p > > [GENERAL] happy holidays, christmas etc., Removed. > >> 253 was a website issue that's been fixed. > > [pgsql-www] Re: [pgsql-advocacy] Problem with "File Browser" link on > downloadspage Great. > >> 158 was fixed this morning by pure chance. > > Re: [BUGS] BUG #4538: shared memory Good. > > Can you give subject lines on this? > > Yikes, I didn't spot that the numbers could change. Seems to make them > kinda useless. Yep. > > I didn't mean for people to > > actually start working on items, but rather just to see the scope of the > > problem. > > Well, I don't think we are seeing the scope of the problem, given that > we've had 20+ false non-problems spotted within a few minutes of your > list being posted. Once it's cleaned up we'll see the scope of the > problem (not that I believe we have one - I think we have what is > shaping up to be a superb release, if a touch later than originally > planned). > > BTW, the missing bug is > http://archives.postgresql.org/pgsql-bugs/2008-11/msg00106.php I have added that to the 8.4 open items wiki. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://enterprisedb.com + If your life is a hard drive, Christ can be your backup. +
On Tue, Jan 6, 2009 at 4:12 PM, Bruce Momjian <bruce@momjian.us> wrote: > Dave Page wrote: >> 243 seems like a priority for release :-p [GENERAL] happy holidays, christmas etc., >> 253 was a website issue that's been fixed. [pgsql-www] Re: [pgsql-advocacy] Problem with "File Browser" link on downloadspage >> 158 was fixed this morning by pure chance. Re: [BUGS] BUG #4538: shared memory > Can you give subject lines on this? Yikes, I didn't spot that the numbers could change. Seems to make them kinda useless. > I didn't mean for people to > actually start working on items, but rather just to see the scope of the > problem. Well, I don't think we are seeing the scope of the problem, given that we've had 20+ false non-problems spotted within a few minutes of your list being posted. Once it's cleaned up we'll see the scope of the problem (not that I believe we have one - I think we have what is shaping up to be a superb release, if a touch later than originally planned). BTW, the missing bug is http://archives.postgresql.org/pgsql-bugs/2008-11/msg00106.php -- Dave Page EnterpriseDB UK: http://www.enterprisedb.com
> Can you give subject lines on this? I didn't mean for people to > actually start working on items, but rather just to see the scope of the > problem. Hmm, well, when you are ready for people to start working on items, I might be able to work on some items, if there are things that a non-core, non-committer community member such as myself can help with. ...Robert
Bruce Momjian wrote:
> I am now warning that we have an unusually large number of open items
> that must be either completed or moved to the TODO list before 8.4 can
> be released.
>
> Everyone knows about the commit fest wiki items, but I am tracking 291
> threads that need some type of attention; if only 50% of them are
> significant for 8.4, that still leaves +100 items that should be
> completed in the next month. You can see the full list here:
OK, based on excellent feedback I have updated the list of open issues
(down from 291 to 261):
http://momjian.us/cgi-bin/pgpatches
If people have further updates please, please send them (with subject
headings please). ;-)
I am excited people looked at the list and were able to tell me some of
them are closed.
-- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB
http://enterprisedb.com
+ If your life is a hard drive, Christ can be your backup. +
Robert Haas wrote: > > Can you give subject lines on this? I didn't mean for people to > > actually start working on items, but rather just to see the scope of the > > problem. > > Hmm, well, when you are ready for people to start working on items, I > might be able to work on some items, if there are things that a > non-core, non-committer community member such as myself can help with. Yep, many are just documentation updates or things where you can send in patches. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://enterprisedb.com + If your life is a hard drive, Christ can be your backup. +
On Tue, 2009-01-06 at 11:30 -0500, Bruce Momjian wrote: > Simon Riggs wrote: > > > > On Tue, 2009-01-06 at 10:21 -0500, Bruce Momjian wrote: > > > > > I have already approached developers to get help in completing these > > > items, but got little assistance. > > > > If you can send me the list that you think applies to me, I'll work on > > it. I don't want to spend the time to read every entry if you already > > have. Thanks. > > Sure, I think all your stuff is on the commit fest page so it is already > being dealt with. I've looked at 65, 65, 67, 69, 104, 109, 112, 145, 163, 175, 178, 225, 226, 233, 276 and 285 to extract any additional points not already noted on the Hot Standby wiki. Nothing burning, just couple of minor issues picked up. 112 is a duplicate of 178. Both are unrelated directly to HS, though it needs attention for 8.4. Thanks, -- Simon Riggs www.2ndQuadrant.comPostgreSQL Training, Services and Support
It seems like it would be helpful if you made a pass through it yourself just looking for duplicates and commitfest items, since it's probably just as fast for you to find them and fix them as it is for us to tell you about them - maybe faster, since the links on this page don't seem to work very well. But... - you have column-level privileges in there four times, it's also in the commitfest - updatable views is in the commitfest, and is the same item as WIP: Automatic view update rules - WIP: Hash Join-Filter Pruning using Bloom Filters is in the commitfest - SQL/MED compatible connection manager is in the commitfest and in fact comitted - optimizing copy is on the commitfest page (under "optimizing copy with memchr") - htup and bufpage API clean up is on the commitfest page - HeapTuple version extension + code cleanup is on the commitfest page - Synchronous replication patch v1 and synchronous replication patch v2 are earlier versions of the patch that is now one of the major items left to resolve for this commitfest - updated hash functions for postgresql v1 is in on the commitfest page - WIP: Page space reservation (pgupgrade) is an idea that was rejected, IIRC. pg_upgrade project status is more of the same thing. there are several more pg_upgrade related items on here as well, most of which are probably unnecessary. - Infrastructure changes for recovery (v8) is on the commitfest page - parallel restore is on here several times under various times: parallel pg_restore - WIP patch, parallel pg_restore, parallel pg_restore design issues. ...Robert
On Tuesday 06 January 2009 18:49:00 Bruce Momjian wrote: > If people have further updates please, please send them (with subject > headings please). Most of the entries are quite frankly junk, either already committed, already rejected, patches not under consideration, irrelevant discussions, or completely silly, e.g., 261. Warning about the 8.4 release
Peter Eisentraut wrote: > On Tuesday 06 January 2009 18:49:00 Bruce Momjian wrote: > > If people have further updates please, please send them (with subject > > headings please). > > Most of the entries are quite frankly junk, either already committed, already > rejected, patches not under consideration, irrelevant discussions, or > completely silly, e.g., > > 261. Warning about the 8.4 release That was the same reaction Tom had. Again, many might be junk, but is it 100% junk. What about: 8.4 - psql output for \lOverriding Kerberos parametersHAVE_FSEEKO for WIN32stat() vs cygwinMemory mess introduced by recentfuncapi.c patchThe suppress_redundant_updates_trigger() works incorrectlySo what's an "empty" array anyway? I just picked those at random. And again, though they are not required for the release, they appeared during 8.4 development and should be fixed now, if possible. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://enterprisedb.com + If your life is a hard drive, Christ can be your backup. +
Bruce Momjian <bruce@momjian.us> writes:
> Peter Eisentraut wrote:
>> Most of the entries are quite frankly junk,
> That was the same reaction Tom had. Again, many might be junk, but is
> it 100% junk. What about:
> 8.4 - psql output for \l
Done (and this is on the commitfest page anyway)
> Overriding Kerberos parameters
Just submitted 4 days ago, and anyway Magnus is quite capable of committing it for himself
> HAVE_FSEEKO for WIN32
Proposes moving win32-only code into port.h, which is 100% wrong IMHO
> stat() vs cygwin
According to the thread, no one but you thinks there's a problem.
> Memory mess introduced by recent funcapi.c patch
Fixed
> The suppress_redundant_updates_trigger() works incorrectly
Fixed
> So what's an "empty" array anyway?
Blue-sky discussion about a fundamental behavior change that should certainly not be put into 8.4 in a rush.
As-is, this list is completely unhelpful. It looks like you've dumped
all your unread mail onto this page and asked the rest of us to sort it
for you. I'm sorry, but I've got other things to do.
regards, tom lane
Tom Lane wrote: > Bruce Momjian <bruce@momjian.us> writes: > > Peter Eisentraut wrote: > >> Most of the entries are quite frankly junk, > > > That was the same reaction Tom had. Again, many might be junk, but is > > it 100% junk. What about: > > > 8.4 - psql output for \l > > Done (and this is on the commitfest page anyway) Did you respond to this comment? http://archives.postgresql.org/pgsql-hackers/2009-01/msg00154.php > > HAVE_FSEEKO for WIN32 > > Proposes moving win32-only code into port.h, > which is 100% wrong IMHO The problem is that we override fseeko only in pg_dump, while it should be done globally for Win32; that is what Andrew was asking about. Am I missing something? > > stat() vs cygwin > > According to the thread, no one but you thinks there's a > problem. Well, I offered to document the problem, and have done that now. > > Memory mess introduced by recent funcapi.c patch > > Fixed OK. > > The suppress_redundant_updates_trigger() works incorrectly > > Fixed OK. > > So what's an "empty" array anyway? > > Blue-sky discussion about a fundamental behavior change > that should certainly not be put into 8.4 in a rush. Added to TODO. > As-is, this list is completely unhelpful. It looks like you've dumped > all your unread mail onto this page and asked the rest of us to sort it > for you. I'm sorry, but I've got other things to do. That explains why people were confused. I never intended for the list to be scanned --- I only wanted to give people and idea of the _volume_ I am looking at; I will approach people individually to close them, though the comments I did get were very helpful. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://enterprisedb.com + If your life is a hard drive, Christ can be your backup. +
On Tue, Jan 6, 2009 at 12:38 PM, Robert Haas <robertmhaas@gmail.com> wrote:
I'm pulling this patch and resubmitting for 8.5.
--
Jonah H. Harris, Senior DBA
myYearbook.com
- WIP: Hash Join-Filter Pruning using Bloom Filters is in the commitfest
I'm pulling this patch and resubmitting for 8.5.
--
Jonah H. Harris, Senior DBA
myYearbook.com
Alvaro Herrera wrote: > Bruce Momjian wrote: > >> Everyone knows about the commit fest wiki items, but I am tracking 291 >> threads that need some type of attention; if only 50% of them are >> significant for 8.4, that still leaves +100 items that should be >> completed in the next month. You can see the full list here: >> >> http://momjian.us/cgi-bin/pgpatches > > I notice that some of the items in this commitfest are in your queue. > Some things are duplicated; for example parallel restore appear more > than once (I count six times). Column privileges is also several > times. * Updates of SE-PostgreSQL 8.4devel patches (r1389) * Updates of SE-PostgreSQL 8.4devel patches (r1386) The two items are also duplicated, and now on CommitFest wikipage. I submit my messages with diffent subject line for each revision, whenever the patch set is updated. Is it unconfortable manner? Thanks, -- OSS Platform Development Division, NEC KaiGai Kohei <kaigai@ak.jp.nec.com>
On Wed, Jan 7, 2009 at 1:23 AM, Bruce Momjian <bruce@momjian.us> wrote: > Tom Lane wrote: > >> As-is, this list is completely unhelpful. It looks like you've dumped >> all your unread mail onto this page and asked the rest of us to sort it >> for you. I'm sorry, but I've got other things to do. > > That explains why people were confused. I never intended for the list > to be scanned --- I only wanted to give people and idea of the _volume_ > I am looking at; I will approach people individually to close them, > though the comments I did get were very helpful. > Should we dump your list to a separate Wiki page so that people can directly edit/comment/remove the items which they are sure about ? Thanks, Pavan -- Pavan Deolasee EnterpriseDB http://www.enterprisedb.com
Robert Haas píše v út 06. 01. 2009 v 12:38 -0500:
> - WIP: Page space reservation (pgupgrade) is an idea that was
> rejected, IIRC. pg_upgrade project status is more of the same thing.
> there are several more pg_upgrade related items on here as well, most
> of which are probably unnecessary.
space reservation is not rejected idea. It is necessary for online page
conversion. There were some general comments, without any conclusion.
There is list of invalid/rejected/outdate items
29 Prototype: In-place upgrade v02
111 WIP: Page space reservation (pgupgrade)
112 pg_upgrade project status
133 pg_upgrade: How to deal with toast -> postponed to next release
143 In-place upgrade
Valid items are:
174 Space reservation (pgupgrade)
181: WIP: pre-upgrade page reservation
Zdenek
Zdenek Kotala wrote: > > Robert Haas p??e v ?t 06. 01. 2009 v 12:38 -0500: > > > - WIP: Page space reservation (pgupgrade) is an idea that was > > rejected, IIRC. pg_upgrade project status is more of the same thing. > > there are several more pg_upgrade related items on here as well, most > > of which are probably unnecessary. > > space reservation is not rejected idea. It is necessary for online page > conversion. There were some general comments, without any conclusion. > > There is list of invalid/rejected/outdate items > > 29 Prototype: In-place upgrade v02 > 111 WIP: Page space reservation (pgupgrade) > 112 pg_upgrade project status > 133 pg_upgrade: How to deal with toast -> postponed to next release > 143 In-place upgrade Thanks for the updates. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://enterprisedb.com + If your life is a hard drive, Christ can be your backup. +
Pavan Deolasee wrote: > On Wed, Jan 7, 2009 at 1:23 AM, Bruce Momjian <bruce@momjian.us> wrote: > > Tom Lane wrote: > > > > >> As-is, this list is completely unhelpful. It looks like you've dumped > >> all your unread mail onto this page and asked the rest of us to sort it > >> for you. I'm sorry, but I've got other things to do. > > > > That explains why people were confused. I never intended for the list > > to be scanned --- I only wanted to give people and idea of the _volume_ > > I am looking at; I will approach people individually to close them, > > though the comments I did get were very helpful. > > > > Should we dump your list to a separate Wiki page so that people can > directly edit/comment/remove the items which they are sure about ? Not yet --- I need to work for a week or two to trim the list down to things that actually still need attention. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://enterprisedb.com + If your life is a hard drive, Christ can be your backup. +
Bruce Momjian wrote:
> I am now warning that we have an unusually large number of open items
> that must be either completed or moved to the TODO list before 8.4 can
> be released.
More threads to remove:
* [BUGS] BUG #4553: HOLD cursors not materializing results fully
was fixed here:
revision 1.77
date: 2008-12-01 14:06:21 -0300; author: tgl; state: Exp; lines: +7 -3;
Ensure that the contents of a holdable cursor don't depend on out-of-line
toasted values, since those could get dropped once the cursor's transaction
is over. Per bug #4553 from Andrew Gierth.
Back-patch as far as 8.1. The bug actually exists back to 7.4 when holdable
cursors were introduced, but this patch won't work before 8.1 without
significant adjustments. Given the lack of field complaints, it doesn't seem
worth the work (and risk of introducing new bugs) to try to make a patch for
the older branches.
* [COMMITTERS] pgsql: Have psql's \d+ print reloptions.
the problematic comment is just not true; you see index reloptions along
the table description
* [HACKERS] pg_type.h regression?,
was fixed here:
revision 1.200
date: 2008-09-24 23:28:56 -0400; author: tgl; state: Exp; lines: +2 -2;
Establish the rule that array types should have the same typdelim as their
element types. Since the backend doesn't actually pay attention to the array
type's delimiter, this has no functional effect, but it seems better for the
catalog entries to be consistent. Per gripe from Greg Mullane and subsequent
discussion.
* Re: SQL:2008 LIMIT/OFFSET
is a committed patch here, with the comment from Tom about position already
handled:
revision 2.628
date: 2008-10-22 08:00:34 -0300; author: petere; state: Exp; lines: +42 -2;
SQL:2008 alternative syntax for LIMIT/OFFSET:
OFFSET num {ROW|ROWS} FETCH {FIRST|NEXT} [num] {ROW|ROWS} ONLY
The vast majority of the stuff seems to be patches that are in the
commitfest queue ...
--
Alvaro Herrera http://www.CommandPrompt.com/
PostgreSQL Replication, Consulting, Custom Development, 24x7 support