Re: [HACKERS] patternsel() and histogram_selectivity() and the hard cutoff of 100
Hi Greg, So I had a thought about how to soften the controversial hard cutoff of 100 for the use of the histogram selectivity. Instead of switching 100% one way or the other between the two heuristics why not calculate both and combine them. The larger the sample size from the histogram the more we can weight the histogram calculation. The smaller the histogram size the more we weight the heuristic. My first thought was to scale it linearly so we use 10% of the histogram sample + 90% of the heuristic for default statistic sizes of 10 samples. That degenerates to the status quo for 100 samples and up. Incidentally I hacked up a patch to do this: Sounds sensible to me, at least much more than a hardcoded magic number a few people know about... Cheers -- Matteo Beccati Openads - http://www.openads.org ---(end of broadcast)--- TIP 3: Have you checked our extensive FAQ? http://www.postgresql.org/docs/faq
Re: [HACKERS] Patch review
Magnus Hagander wrote: It would also be better to be able to off-load it to more than one person. For example, I would like to be able to get into the unapplied patches list and remove the email about events on 8.3RC1. First of all, it's not a patch, but it's listed under it. But more importantly, it has been fixed and should just be removed. So I now have to email you to ask you to remove it, and then you have to do the actual work, which means double work. True. Those web pages are emails pulled from the stream of emails that I think are worthy of discussion during 8.4 development. If we assume we want to continue communicating via email I need a way to pull items out and collect them, and unfortunately right now it is hard for others to help in that. I have a few ideas. First I could easily create an email address that would allow others to _add_ emails to the web page (via bounce) but that doesn't solve the issue of allowing people to comment on and delete items. Do most email readers support bouncing emails to another address, so the to/from fields are not modified? (Seems there is a Thunderbird extension to do it, http://blog.mecworks.com/articles/2005/04/20/bouncing-mail-in-thunderbird/.) There is no reason I have to host the list here. I can _bounce_ emails to any address. Is there a service we can use that allows emails to be accepted and displayed on a web site and that allows deletions and comments, and has stable URLs for every email message? Is there software I can install on my server to do this? If we want to communicate via a web interface, I would still need a way to collect specific messages (not always entire threads). -- Bruce Momjian [EMAIL PROTECTED]http://momjian.us EnterpriseDB http://postgres.enterprisedb.com + If your life is a hard drive, Christ can be your backup. + ---(end of broadcast)--- TIP 2: Don't 'kill -9' the postmaster
Re: [HACKERS] Patch review
Bruce Momjian wrote: Bruce Momjian wrote: Bruce Momjian wrote: Tom Lane wrote: Bruce Momjian [EMAIL PROTECTED] writes: I have added message-id's to both patches web sites. The message id appears next to the author in the thread listing, and at the top of the message page. That's an improvement, but it doesn't solve the other fundamental problem, which is the lack of any way to annotate the list. Agreed, and it doesn't allow people to delete items either. Hold, I think I can add annotations if that's what people want; you can see an example in my blog: http://momjian.us/main/blogs/blog.html OK, comments added, and they are based on message-id, so they will not change over time. I am using JS-Kit, http://js-kit.com/comments/, article http://blog.wired.com/monkeybites/2006/11/jskit_add_comme.html. Seems the comments slowed down the page load so I now have the thread broken down into 50 emails per page. I got a message from someone saying I was trying too hard to avoid using a tracker. I think I was clear what functionality I needed and no one had any ready-made solutions. Let me give a practical example of what I need. Suppose we were using a web-based discussion forum, rather than email. Assume it is something like Slashdot, where you have article titles and comments. For the patches lists I need to take sometimes entire threads, sometimes groups of comments, and store them in a format so people can review them as a digest. And I want to allow comments on these items, and ideally allow multiple people to delete them. If I link to a comment URL, how do people know if they should look at that comment or all comments below it? If I had omment URLs, how would I present those in a threaded way? A flat list of URL would be too crude and lack titles. So, ultimately I would need a threaded way to present the URLs, which is pretty much what mhonarc does now. Comments would certainly be easier because you could just add comments to the discussion threads. Deleting URLs would be the same as mhonarc, except the URLs would be static. So, if we did have a tracker, how would it be different? Comments would be more integrated but I am unclear how the patches_hold queue would be different. Basically what I do now is to take the email stream and chop pieces out of it for later review. I see a tracker making some of that process easier and more distributed/shared, but some of it harder and more complex. -- Bruce Momjian [EMAIL PROTECTED]http://momjian.us EnterpriseDB http://postgres.enterprisedb.com + If your life is a hard drive, Christ can be your backup. + ---(end of broadcast)--- TIP 6: explain analyze is your friend
Re: [HACKERS] Patch review
Brendan Jurd wrote: On Feb 13, 2008 10:45 AM, Bruce Momjian [EMAIL PROTECTED] wrote: For the patches lists I need to take sometimes entire threads, sometimes groups of comments, and store them in a format so people can review them as a digest. And I want to allow comments on these items, and ideally allow multiple people to delete them. Hi Bruce, If we were using a tracker, why would you need to produce this digest at all? Why would you not simply refer people to, e.g., the tracker's report of all outstanding tickets in 8.4? Uh, well, it is kind of hard because it assumes discussions are linear, meaning an item has only one ticket, and the ticket is clear on what to do. In practice, that is not usually the case. -- Bruce Momjian [EMAIL PROTECTED]http://momjian.us EnterpriseDB http://postgres.enterprisedb.com + If your life is a hard drive, Christ can be your backup. + ---(end of broadcast)--- TIP 3: Have you checked our extensive FAQ? http://www.postgresql.org/docs/faq
Re: [HACKERS] Patch review
On Feb 13, 2008 10:45 AM, Bruce Momjian [EMAIL PROTECTED] wrote: For the patches lists I need to take sometimes entire threads, sometimes groups of comments, and store them in a format so people can review them as a digest. And I want to allow comments on these items, and ideally allow multiple people to delete them. Hi Bruce, If we were using a tracker, why would you need to produce this digest at all? Why would you not simply refer people to, e.g., the tracker's report of all outstanding tickets in 8.4? Regards, BJ ---(end of broadcast)--- TIP 1: if posting/reading through Usenet, please send an appropriate subscribe-nomail command to [EMAIL PROTECTED] so that your message can get through to the mailing list cleanly
Re: [HACKERS] Patch review
Gregory Stark wrote: Bruce Momjian [EMAIL PROTECTED] writes: Suppose we were using a web-based discussion forum, rather than email. That would be crazy, why would I suppose such a thing? For the patches lists I need to take sometimes entire threads, sometimes groups of comments, and store them in a format so people can review them as a digest. Why do you need to do any such thing? What does any of that have to do with a patches queue? TODO items and patches are often in the middle of threads. If I link to a comment URL, how do people know if they should look at that comment or all comments below it? They should look at whatever they want to. I usually have to back up several messages to understand the context and then follow several messages later. You can't possibly know how much context people will need to understand. You can't try to control people to that level of detail. If I had omment URLs, how would I present those in a threaded way? So, if we did have a tracker, how would it be different? Comments would be more integrated but I am unclear how the patches_hold queue would be different. A patches queue is just a list of patches with their current status. Not a replacement for our mailing lists. You're trying to solve the wrong problem. The current status for a patch is just something like waiting for feedback on questions from message [link] or waiting for new version addressing issues raised in review [link]. That's it. The critical information we need are: What's the most recent version of the patch? what is it blocking on? and who is it blocking on? The discussion was mostly related to the 8.3 patches_hold queue where people wanted help processing it. -- Bruce Momjian [EMAIL PROTECTED]http://momjian.us EnterpriseDB http://postgres.enterprisedb.com + If your life is a hard drive, Christ can be your backup. + ---(end of broadcast)--- TIP 4: Have you searched our list archives? http://archives.postgresql.org
Re: [HACKERS] Patch review
Bruce Momjian [EMAIL PROTECTED] writes: Suppose we were using a web-based discussion forum, rather than email. That would be crazy, why would I suppose such a thing? For the patches lists I need to take sometimes entire threads, sometimes groups of comments, and store them in a format so people can review them as a digest. Why do you need to do any such thing? What does any of that have to do with a patches queue? If I link to a comment URL, how do people know if they should look at that comment or all comments below it? They should look at whatever they want to. I usually have to back up several messages to understand the context and then follow several messages later. You can't possibly know how much context people will need to understand. You can't try to control people to that level of detail. If I had omment URLs, how would I present those in a threaded way? So, if we did have a tracker, how would it be different? Comments would be more integrated but I am unclear how the patches_hold queue would be different. A patches queue is just a list of patches with their current status. Not a replacement for our mailing lists. You're trying to solve the wrong problem. The current status for a patch is just something like waiting for feedback on questions from message [link] or waiting for new version addressing issues raised in review [link]. That's it. The critical information we need are: What's the most recent version of the patch? what is it blocking on? and who is it blocking on? -- Gregory Stark EnterpriseDB http://www.enterprisedb.com Ask me about EnterpriseDB's Slony Replication support! ---(end of broadcast)--- TIP 6: explain analyze is your friend
[HACKERS] REL8_3_STABLE branch created ...
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Just set down the branch so that movement can happen towards 8.4.0 ... let me know if there are any problems ... - Marc G. Fournier Hub.Org Networking Services (http://www.hub.org) Email . [EMAIL PROTECTED] MSN . [EMAIL PROTECTED] Yahoo . yscrappy Skype: hub.orgICQ . 7615664 -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.4 (FreeBSD) iD8DBQFHsmG24QvfyHIvDvMRAuxJAKCy9OuplihNOTivwd08va7rAJxKMwCeLl1Z 0n8apgMIsvXC6d6Q8oJVE2o= =Bi3l -END PGP SIGNATURE- ---(end of broadcast)--- TIP 7: You can help support the PostgreSQL project by donating at http://www.postgresql.org/about/donate
[HACKERS] Locale vs LIKE
Given the improvements in 8.3 listed in the release notes: - Improve efficiency of LIKE/ILIKE, especially for multi-byte character sets like UTF-8 (Andrew, Itagaki Takahiro) Does this still hold: http://www.postgresql.org/docs/8.3/interactive/locale.html The drawback of using locales other than C or POSIX in PostgreSQL is its performance impact. It slows character handling and prevents ordinary indexes from being used by LIKE. For this reason use locales only if you actually need them. i.e. Do I still have to either initdb --locale=C or explicitly use text_pattern_ops? (Queries include predicates of the form [indexed text expression] like ABC%) Stephen Denne Disclaimer: At the Datamail Group we value team commitment, respect, achievement, customer focus, and courage. This email with any attachments is confidential and may be subject to legal privilege. If it is not intended for you please advise by reply immediately, destroy it and do not copy, disclose or use it in any way. __ This email has been scanned by the DMZGlobal Business Quality Electronic Messaging Suite. Please see http://www.dmzglobal.com/services/bqem.htm for details. __ ---(end of broadcast)--- TIP 5: don't forget to increase your free space map settings
[HACKERS] Branched for 8.4
We have branced CVS for 8.4. The 8.3 CVS tag is REL8_3_STABLE. I have started tracking our development status at the top of my Postgres blog page. -- Bruce Momjian [EMAIL PROTECTED]http://momjian.us EnterpriseDB http://postgres.enterprisedb.com + If your life is a hard drive, Christ can be your backup. + ---(end of broadcast)--- TIP 7: You can help support the PostgreSQL project by donating at http://www.postgresql.org/about/donate
Re: [HACKERS] Patch review
Bruce Momjian [EMAIL PROTECTED] writes: Gregory Stark wrote: Bruce Momjian [EMAIL PROTECTED] writes: For the patches lists I need to take sometimes entire threads, sometimes groups of comments, and store them in a format so people can review them as a digest. Why do you need to do any such thing? What does any of that have to do with a patches queue? TODO items and patches are often in the middle of threads. No, messages which propose the items and patches are in the middle of the thread. What we need is a list of the actual items. We already have an archive of the messages. Looking at your held queue, how many patches are in there? Who are the authors? Which ones are waiting for feedback before they can continue? The critical information we need are: What's the most recent version of the patch? what is it blocking on? and who is it blocking on? The discussion was mostly related to the 8.3 patches_hold queue where people wanted help processing it. Yes, the information I listed is the what we need to do that. -- Gregory Stark EnterpriseDB http://www.enterprisedb.com Ask me about EnterpriseDB's PostGIS support! ---(end of broadcast)--- TIP 4: Have you searched our list archives? http://archives.postgresql.org
Re: [HACKERS] NLS on MSVC strikes back!
Hi Alvaro-san. Yes, However, It is not in good condition at the reason a message catalog still is not enough. Then, I have tried these adjustments with the problem of a locale. Still, it is not much time. But, My condition is not good so that I'm may be influenza. :-( Regards, Hiroshi Saito - Original Message - From: Alvaro Herrera [EMAIL PROTECTED] Hiroshi Saito wrote: Hi --Document.-- Table 9-22. TM prefix translation mode (print localized day and month names based on lc_messages) -- http://winpg.jp/~saito/pg_work/NLS_TO_CHAR_JP.png But, Although Japanese is out of condition Does the fundamental specification change? Humm, I cannot read Japanese so as far as I can make, this is working fine? -- Alvaro Herrerahttp://www.CommandPrompt.com/ PostgreSQL Replication, Consulting, Custom Development, 24x7 support ---(end of broadcast)--- TIP 7: You can help support the PostgreSQL project by donating at http://www.postgresql.org/about/donate
Re: [HACKERS] NLS on MSVC strikes back!
From: Alvaro Herrera [EMAIL PROTECTED] Dave Page wrote: On Feb 12, 2008 4:44 PM, Alvaro Herrera [EMAIL PROTECTED] wrote: What this really means is that locale support is completely broken in the MSVC build, i.e. you cannot get localized strings at all (not just to_char()). Is this correct? If so, this is a serious problem. Not judging by Hiroshi's screenshot. I don't speak Japanese either, but it certainly didn't say 'tuesday' after he adjusted LC_MESSAGES. But it has the string mingw in the path, so I suspect it's not MSVC ... Hiroshi-san, can you confirm? Yeah, However, I go to a hospital from now sorry. I will check it, after returning. Regards, Hiroshi Saito ---(end of broadcast)--- TIP 3: Have you checked our extensive FAQ? http://www.postgresql.org/docs/faq
Re: [HACKERS] REL8_3_STABLE branch created ...
Marc G. Fournier [EMAIL PROTECTED] writes: Just set down the branch so that movement can happen towards 8.4.0 ... let me know if there are any problems ... Looks good from here --- I stamped HEAD as 8.4devel. Let the games begin! regards, tom lane ---(end of broadcast)--- TIP 5: don't forget to increase your free space map settings
Re: [HACKERS] NLS on MSVC strikes back!
What this really means is that locale support is completely broken in the MSVC build, i.e. you cannot get localized strings at all (not just to_char()). Is this correct? If so, this is a serious problem. The way one could confirm this is by: 1. rename share/locale/de to share/locale/German_Germany 2. start a dosbox 3. before starting PG, set an env variable SET LANGUAGE=German_Germany in that dosbox, which will force all child threads to have that locale. 3. start PG form that dosbox and SET LC_MESSAGES to es_ES or Spanish_Spain or es_ES.utf8 or anything else...; select to_char(now(),'TMDay TMMonth '); 4. The result will be in German and not Spanish as one expects. If handling locale in to_char will remain using gettext then I can start working on a patch to fix this. Regards, Gevik Babakhani PostgreSQL NL http://www.postgresql.nl TrueSoftware BV http://www.truesoftware.nl ---(end of broadcast)--- TIP 1: if posting/reading through Usenet, please send an appropriate subscribe-nomail command to [EMAIL PROTECTED] so that your message can get through to the mailing list cleanly
Re: [HACKERS] Patch review
Bruce Momjian wrote: Bruce Momjian wrote: Tom Lane wrote: Bruce Momjian [EMAIL PROTECTED] writes: I have added message-id's to both patches web sites. The message id appears next to the author in the thread listing, and at the top of the message page. That's an improvement, but it doesn't solve the other fundamental problem, which is the lack of any way to annotate the list. Agreed, and it doesn't allow people to delete items either. Hold, I think I can add annotations if that's what people want; you can see an example in my blog: http://momjian.us/main/blogs/blog.html OK, comments added, and they are based on message-id, so they will not change over time. I am using JS-Kit, http://js-kit.com/comments/, article http://blog.wired.com/monkeybites/2006/11/jskit_add_comme.html. Seems the comments slowed down the page load so I now have the thread broken down into 50 emails per page. -- Bruce Momjian [EMAIL PROTECTED]http://momjian.us EnterpriseDB http://postgres.enterprisedb.com + If your life is a hard drive, Christ can be your backup. + ---(end of broadcast)--- TIP 7: You can help support the PostgreSQL project by donating at http://www.postgresql.org/about/donate
Re: [HACKERS] Patch review
Bruce Momjian wrote: Tom Lane wrote: Bruce Momjian [EMAIL PROTECTED] writes: I have added message-id's to both patches web sites. The message id appears next to the author in the thread listing, and at the top of the message page. That's an improvement, but it doesn't solve the other fundamental problem, which is the lack of any way to annotate the list. Agreed, and it doesn't allow people to delete items either. Hold, I think I can add annotations if that's what people want; you can see an example in my blog: http://momjian.us/main/blogs/blog.html OK, comments added, and they are based on message-id, so they will not change over time. I am using JS-Kit, http://js-kit.com/comments/, article http://blog.wired.com/monkeybites/2006/11/jskit_add_comme.html. -- Bruce Momjian [EMAIL PROTECTED]http://momjian.us EnterpriseDB http://postgres.enterprisedb.com + If your life is a hard drive, Christ can be your backup. + ---(end of broadcast)--- TIP 5: don't forget to increase your free space map settings
Re: [HACKERS] NLS on MSVC strikes back!
Alvaro Herrera wrote: Gevik Babakhani wrote: 4. Locale names are different in MS Windows. I created a C app to test gettext on Windows. setting LC_MESSAGES to Spanisg_Spain and German_Germany works but es_ES and de_DE do not :( SET LC_MESSAGES to '' has no effect because: A. gettext compiled/linked in MSVC looks for the locale of the current thread and NOT the LC_MESSAGES,LANGIAGE,LANG... environment variables. See gettext's sources B. Given current thread's locale Spanish_Spain, gettext fails to find a directory called Spanish_Spain in share/locale. As result English names are returned. What this really means is that locale support is completely broken in the MSVC build, i.e. you cannot get localized strings at all (not just to_char()). Is this correct? If so, this is a serious problem. I've certainly managed to get output in Swedish more than once... Annoying as hell because it's the default if you build with NLS and has your English language Windows configured with Swedish time/date format :) //Magnus ---(end of broadcast)--- TIP 4: Have you searched our list archives? http://archives.postgresql.org
Re: [HACKERS] Patch review
Tom Lane wrote: Bruce Momjian [EMAIL PROTECTED] writes: I have added message-id's to both patches web sites. The message id appears next to the author in the thread listing, and at the top of the message page. That's an improvement, but it doesn't solve the other fundamental problem, which is the lack of any way to annotate the list. Agreed, and it doesn't allow people to delete items either. Hold, I think I can add annotations if that's what people want; you can see an example in my blog: http://momjian.us/main/blogs/blog.html -- Bruce Momjian [EMAIL PROTECTED]http://momjian.us EnterpriseDB http://postgres.enterprisedb.com + If your life is a hard drive, Christ can be your backup. + ---(end of broadcast)--- TIP 5: don't forget to increase your free space map settings
Re: [HACKERS] Patch review
Joshua D. Drake wrote: What the heck is a steam hammer? :P http://en.wikipedia.org/wiki/Steam_hammer The same people went on to invent the steam television ... cheers andrew ---(end of broadcast)--- TIP 2: Don't 'kill -9' the postmaster
Re: [HACKERS] Patch review
Stefan Kaltenbrunner wrote: Stefan Kaltenbrunner wrote: Tom Lane wrote: [...] Personally I'd be happier with an editable wiki page consisting of links to the original messages in the mail list archives, plus free-format annotations (such as status). This should be trivial to set up and reasonably flexible. With experience we might find we need something fancier, but let's not overdesign our solution at the start. I will see if I can come up with a proposal on the developer wiki for a list that looks like that tomorrow. I have simply converted the old 8.3 patch status page for now by removing all the applied and rejected patches: http://developer.postgresql.org/index.php/Todo:PatchStatus I'm slowly adding some more stuff from the patches_hold list there but given the way that list is structured this is a fairly difficult and tendious task(a lot there is only discussion/design sketches and in effect one has to check every single mail in there to see if should be added) :-( I have added message-id's to both patches web sites. The message id appears next to the author in the thread listing, and at the top of the message page. Unfortunately you can't search by message id in our archives, but if that is fixed, I can add HTML to get that to work too. I was hoping someone had a completed solution, but I guess not. -- Bruce Momjian [EMAIL PROTECTED]http://momjian.us EnterpriseDB http://postgres.enterprisedb.com + If your life is a hard drive, Christ can be your backup. + ---(end of broadcast)--- TIP 9: In versions below 8.0, the planner will ignore your desire to choose an index scan if your joining column's datatypes do not match
Re: [HACKERS] Patch review
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Mon, 11 Feb 2008 17:09:45 -0600 Decibel! [EMAIL PROTECTED] wrote: On Feb 9, 2008, at 1:08 AM, Tom Lane wrote: Let's not swat flies with steam hammers. What the heck is a steam hammer? :P http://en.wikipedia.org/wiki/Steam_hammer Joshua D. Drake - -- The PostgreSQL Company since 1997: http://www.commandprompt.com/ PostgreSQL Community Conference: http://www.postgresqlconference.org/ Donate to the PostgreSQL Project: http://www.postgresql.org/about/donate PostgreSQL SPI Liaison | SPI Director | PostgreSQL political pundit -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQFHsdBfATb/zqfZUUQRAnkwAKCo5TaaIOWHVYFWCW/i5s0mHd5vDQCfUw1P 0sh/PuDHAbvQQITVuQbeT3s= =EUBM -END PGP SIGNATURE- ---(end of broadcast)--- TIP 7: You can help support the PostgreSQL project by donating at http://www.postgresql.org/about/donate
Re: [HACKERS] Patch review
Bruce Momjian [EMAIL PROTECTED] writes: I have added message-id's to both patches web sites. The message id appears next to the author in the thread listing, and at the top of the message page. That's an improvement, but it doesn't solve the other fundamental problem, which is the lack of any way to annotate the list. regards, tom lane ---(end of broadcast)--- TIP 6: explain analyze is your friend
Re: [HACKERS] NLS on MSVC strikes back!
On Feb 12, 2008 4:44 PM, Alvaro Herrera [EMAIL PROTECTED] wrote: What this really means is that locale support is completely broken in the MSVC build, i.e. you cannot get localized strings at all (not just to_char()). Is this correct? If so, this is a serious problem. Not judging by Hiroshi's screenshot. I don't speak Japanese either, but it certainly didn't say 'tuesday' after he adjusted LC_MESSAGES. -- Dave Page EnterpriseDB UK: http://www.enterprisedb.com The Oracle-compatible database company ---(end of broadcast)--- TIP 1: if posting/reading through Usenet, please send an appropriate subscribe-nomail command to [EMAIL PROTECTED] so that your message can get through to the mailing list cleanly
Re: [HACKERS] NLS on MSVC strikes back!
Gevik Babakhani wrote: 4. Locale names are different in MS Windows. I created a C app to test gettext on Windows. setting LC_MESSAGES to Spanisg_Spain and German_Germany works but es_ES and de_DE do not :( SET LC_MESSAGES to '' has no effect because: A. gettext compiled/linked in MSVC looks for the locale of the current thread and NOT the LC_MESSAGES,LANGIAGE,LANG... environment variables. See gettext's sources B. Given current thread's locale Spanish_Spain, gettext fails to find a directory called Spanish_Spain in share/locale. As result English names are returned. What this really means is that locale support is completely broken in the MSVC build, i.e. you cannot get localized strings at all (not just to_char()). Is this correct? If so, this is a serious problem. -- Alvaro Herrerahttp://www.CommandPrompt.com/ The PostgreSQL Company - Command Prompt, Inc. ---(end of broadcast)--- TIP 5: don't forget to increase your free space map settings
Re: [HACKERS] NLS on MSVC strikes back!
Andrew Dunstan [EMAIL PROTECTED] writes: On Nov 24 2006 this was committed dor src/backend/utils/adt/formatting.c: r 1.114: Fix to_char() locale handling to honor LC_TIME, not LC_MESSAGES. A few hours later it was supposedly reverted: r 1.115: Revert (too late in beta): Fix to_char() locale handling to honor LC_TIME, not LC_MESSAGES. Apparently this was never returned to (it should probably have been put on the patches hold queue). to_char's month/day name localization is implemented with gettext() not strftime(), which is why it depends on LC_MESSAGES not LC_TIME. I seem to recall that we didn't like the side-effects of the patch you are mentioning, and so it ended up being rejected outright. regards, tom lane ---(end of broadcast)--- TIP 5: don't forget to increase your free space map settings
Re: [HACKERS] NLS on MSVC strikes back!
Andrew Dunstan wrote: Alvaro Herrera wrote: Gevik Babakhani wrote: I have compiled 8.3 (CVS HEAD) with NLS on a XP box. But it seems that NLS support is broke. Could someone please confirm this. What I can confirm is that lc_messages is supposed to work for to_char, because it shows the localized output for me as I change lc_messages; and it doesn't if I change lc_time. FSVO supposed to. The CVS history is somewhat murky. On Nov 24 2006 this was committed dor src/backend/utils/adt/formatting.c: r 1.114: Fix to_char() locale handling to honor LC_TIME, not LC_MESSAGES. A few hours later it was supposedly reverted: r 1.115: Revert (too late in beta): Fix to_char() locale handling to honor LC_TIME, not LC_MESSAGES. Apparently this was never returned to (it should probably have been put on the patches hold queue). Added to TODO list: o Use LC_TIME for localized weekday/month names, rather than LC_MESSAGES -- Bruce Momjian [EMAIL PROTECTED]http://momjian.us EnterpriseDB http://postgres.enterprisedb.com + If your life is a hard drive, Christ can be your backup. + ---(end of broadcast)--- TIP 3: Have you checked our extensive FAQ? http://www.postgresql.org/docs/faq
Re: [HACKERS] NLS on MSVC strikes back!
Hi --Document.-- Table 9-22. TM prefix translation mode (print localized day and month names based on lc_messages) -- http://winpg.jp/~saito/pg_work/NLS_TO_CHAR_JP.png But, Although Japanese is out of condition Does the fundamental specification change? Regards, Hiroshi Saito - Original Message - From: Alvaro Herrera [EMAIL PROTECTED] Gevik Babakhani wrote: I have compiled 8.3 (CVS HEAD) with NLS on a XP box. But it seems that NLS support is broke. Could someone please confirm this. What I can confirm is that lc_messages is supposed to work for to_char, because it shows the localized output for me as I change lc_messages; and it doesn't if I change lc_time. In any case, Gevik=# show lc_messages ; lc_messages English_United States.1252 (1 row) Gevik=# Gevik=# set lc_messages TO 'de_DE.utf8'; SET I wonder if this is really valid. Shouldn't you be using a locale name like 'German_Germany.65001' or some such? I thought Windows did not recognize the de_DE.utf8 form of names. Perhaps it failed to raise an error here? It does for me: alvherre=# set lc_time to 'fr_CA.utf8'; SET alvherre=# set lc_time to 'de_DE.utf8'; ERROR: valor no válido para el parámetro «lc_time»: «de_DE.utf8» Suffice to say I don't have the de_DE locale installed. -- Alvaro Herrerahttp://www.CommandPrompt.com/ PostgreSQL Replication, Consulting, Custom Development, 24x7 support ---(end of broadcast)--- TIP 1: if posting/reading through Usenet, please send an appropriate subscribe-nomail command to [EMAIL PROTECTED] so that your message can get through to the mailing list cleanly ---(end of broadcast)--- TIP 6: explain analyze is your friend
[HACKERS] postmaster in a tight loop
Folks, I encountered a situation on Sunday night where the postmaster was in a tight loop. That's the conclusion we reached, but have no real proof. I also have no idea how to reproduce this situation. This post is just an FYI in case it helps. The laptop was running hot so I looked around and found pgsql to be the cause. I decided to shutdown the postmaster, but it would not shutdown: # /usr/local/etc/rc.d/postgresql stop pg_ctl: server does not shut down After that, I started looking closer. For starters, here is what the system looked like: $ date Sun Feb 10 19:38:10 EST 2008 $ ps auwx | grep pgsql pgsql 1172 90.0 0.4 44636 3088 ?? Rs 10:45AM 21:55.16 / usr/local/bin/postgres -D /usr/local/pgsql/data pgsql 1183 0.0 0.4 44652 3392 ?? Ss 10:45AM 0:06.77 postgres: writer process(postgres) pgsql 1184 0.0 0.4 44652 3176 ?? Ss 10:45AM 0:04.42 postgres: wal writer process(postgres) pgsql 1185 0.0 0.4 44884 3336 ?? Ss 10:45AM 0:03.35 postgres: autovacuum launcher process(postgres) pgsql 1186 0.0 0.4 8588 3060 ?? Ss 10:45AM 0:03.77 postgres: stats collector process(postgres) A little bit from top: last pid: 89359; load averages: 3.10, 2.84, 2.30 up 0+08:53:20 19:37:54 84 processes: 5 running, 79 sleeping CPU states: 48.3% user, 0.8% nice, 50.8% system, 0.0% interrupt, 0.0% idle Mem: 306M Active, 235M Inact, 104M Wired, 27M Cache, 85M Buf, 68M Free Swap: 512M Total, 512M Free PID USERNAMETHR PRI NICE SIZERES STATETIME WCPU COMMAND 1172 pgsql 1 1300 44636K 3088K RUN 22:18 77.34% postgres 1387 root 1 1000 85936K 62032K RUN 10:48 4.49% Xorg 89357 dan 1 80 1860K 1424K wait 0:00 1.00% sh 46507 dan 1 960 15636K 9900K select 1:52 0.05% npviewer.bin 88563 dan 1 960 28856K 21128K RUN 0:01 0.05% kdeinit $ uname -a FreeBSD laptop.unixathome.org 6.3-PRERELEASE FreeBSD 6.3-PRERELEASE #2: Wed Nov 7 10:54:48 EST 2007 [EMAIL PROTECTED]:/usr/obj/usr/src/sys/ PCBSD i386 Running PostgreSQL 8.3.0 Looking at ktrace output, I saw a lot of this: 1172 postgres CALL kse_release(0xbfbfd500) 1172 postgres RET kse_release -1 errno 22 Invalid argument The ldd output for postgres is: [EMAIL PROTECTED]:/usr/home/dan] $ ldd /usr/local/bin/postgres /usr/local/bin/postgres: libintl.so.8 = /usr/local/lib/libintl.so.8 (0x2835a000) libxml2.so.5 = /usr/local/lib/libxml2.so.5 (0x28363000) libssl.so.4 = /usr/lib/libssl.so.4 (0x2846f000) libcrypto.so.4 = /lib/libcrypto.so.4 (0x2849d000) libcrypt.so.3 = /lib/libcrypt.so.3 (0x2859) libm.so.4 = /lib/libm.so.4 (0x285a8000) libpthread.so.2 = /lib/libpthread.so.2 (0x285be000) libc.so.6 = /lib/libc.so.6 (0x285e3000) libiconv.so.3 = /usr/local/lib/libiconv.so.3 (0x286c9000) libz.so.3 = /lib/libz.so.3 (0x287b6000) [EMAIL PROTECTED]:/usr/home/dan] $ The server was compiled with these options: [EMAIL PROTECTED]:~] $ less /var/db/ports/postgresql83/options # This file is auto-generated by 'make config'. # No user-servicable parts inside! # Options for postgresql-client-8.3.0 _OPTIONS_READ=postgresql-client-8.3.0 WITH_NLS=true WITHOUT_PAM=true WITHOUT_LDAP=true WITHOUT_MIT_KRB5=true WITHOUT_HEIMDAL_KRB5=true WITHOUT_OPTIMIZED_CFLAGS=true WITH_XML=true WITH_TZDATA=true WITHOUT_DEBUG=true WITHOUT_INTDATE=true [EMAIL PROTECTED]:~] $ -- Dan Langille -- http://www.langille.org/ [EMAIL PROTECTED] ---(end of broadcast)--- TIP 6: explain analyze is your friend
Re: [HACKERS] NLS on MSVC strikes back!
Alvaro Herrera wrote: Gevik Babakhani wrote: I have compiled 8.3 (CVS HEAD) with NLS on a XP box. But it seems that NLS support is broke. Could someone please confirm this. What I can confirm is that lc_messages is supposed to work for to_char, because it shows the localized output for me as I change lc_messages; and it doesn't if I change lc_time. FSVO supposed to. The CVS history is somewhat murky. On Nov 24 2006 this was committed dor src/backend/utils/adt/formatting.c: r 1.114: Fix to_char() locale handling to honor LC_TIME, not LC_MESSAGES. A few hours later it was supposedly reverted: r 1.115: Revert (too late in beta): Fix to_char() locale handling to honor LC_TIME, not LC_MESSAGES. Apparently this was never returned to (it should probably have been put on the patches hold queue). I can't see offhand what else might have caused the behaviour change. cheers andrew ---(end of broadcast)--- TIP 9: In versions below 8.0, the planner will ignore your desire to choose an index scan if your joining column's datatypes do not match
Re: [HACKERS] NLS on MSVC strikes back!
Hiroshi Saito wrote: Hi --Document.-- Table 9-22. TM prefix translation mode (print localized day and month names based on lc_messages) -- http://winpg.jp/~saito/pg_work/NLS_TO_CHAR_JP.png But, Although Japanese is out of condition Does the fundamental specification change? Humm, I cannot read Japanese so as far as I can make, this is working fine? -- Alvaro Herrerahttp://www.CommandPrompt.com/ PostgreSQL Replication, Consulting, Custom Development, 24x7 support ---(end of broadcast)--- TIP 2: Don't 'kill -9' the postmaster
Re: [HACKERS] postmaster in a tight loop
Alvaro Herrera [EMAIL PROTECTED] writes: I find it a bit worrying that the postmaster is calling that syscall at all. Yeah. Misguided thread-aware library perhaps? Next time please try to get a stack trace. regards, tom lane ---(end of broadcast)--- TIP 3: Have you checked our extensive FAQ? http://www.postgresql.org/docs/faq
Re: [HACKERS] postmaster in a tight loop
Dan Langille wrote: Looking at ktrace output, I saw a lot of this: 1172 postgres CALL kse_release(0xbfbfd500) 1172 postgres RET kse_release -1 errno 22 Invalid argument Humm, kse_release seems related to multithreading. Or so says http://nixdoc.net/man-pages/FreeBSD/kse_release.2.html I find it a bit worrying that the postmaster is calling that syscall at all. -- Alvaro Herrerahttp://www.CommandPrompt.com/ PostgreSQL Replication, Consulting, Custom Development, 24x7 support ---(end of broadcast)--- TIP 6: explain analyze is your friend
Re: [HACKERS] NLS on MSVC strikes back!
Gevik Babakhani wrote: I have compiled 8.3 (CVS HEAD) with NLS on a XP box. But it seems that NLS support is broke. Could someone please confirm this. What I can confirm is that lc_messages is supposed to work for to_char, because it shows the localized output for me as I change lc_messages; and it doesn't if I change lc_time. In any case, Gevik=# show lc_messages ; lc_messages English_United States.1252 (1 row) Gevik=# Gevik=# set lc_messages TO 'de_DE.utf8'; SET I wonder if this is really valid. Shouldn't you be using a locale name like 'German_Germany.65001' or some such? I thought Windows did not recognize the de_DE.utf8 form of names. Perhaps it failed to raise an error here? It does for me: alvherre=# set lc_time to 'fr_CA.utf8'; SET alvherre=# set lc_time to 'de_DE.utf8'; ERROR: valor no válido para el parámetro «lc_time»: «de_DE.utf8» Suffice to say I don't have the de_DE locale installed. -- Alvaro Herrerahttp://www.CommandPrompt.com/ PostgreSQL Replication, Consulting, Custom Development, 24x7 support ---(end of broadcast)--- TIP 1: if posting/reading through Usenet, please send an appropriate subscribe-nomail command to [EMAIL PROTECTED] so that your message can get through to the mailing list cleanly
Re: [HACKERS] NLS on MSVC strikes back!
Right, I know that. But I didn't see you trying lc_time as Andrew suggested. Did that too, but no luck :( ---(end of broadcast)--- TIP 3: Have you checked our extensive FAQ? http://www.postgresql.org/docs/faq
Re: [HACKERS] NLS on MSVC strikes back!
Well, that should be considered a bug, not a feature. Perhaps it was fixed in 8.3. This is 8.3 I am testing with. ---(end of broadcast)--- TIP 5: don't forget to increase your free space map settings
Re: [HACKERS] NLS on MSVC strikes back!
Gevik Babakhani wrote: Should that not be lc_time you are setting? lc_messages is for, uh, messages. No. The same thing works on 8.2.6 Well, that should be considered a bug, not a feature. Perhaps it was fixed in 8.3. -- Alvaro Herrerahttp://www.CommandPrompt.com/ The PostgreSQL Company - Command Prompt, Inc. ---(end of broadcast)--- TIP 6: explain analyze is your friend
Re: [HACKERS] NLS on MSVC strikes back!
Gevik Babakhani wrote: Well, that should be considered a bug, not a feature. Perhaps it was fixed in 8.3. This is 8.3 I am testing with. You are missing Alvaro's point. He is saying that the behaviour you relied on in 8.2.6 was a bug, and possibly you can no longer rely on the buggy behaviour in 8.3. cheers andrew ---(end of broadcast)--- TIP 7: You can help support the PostgreSQL project by donating at http://www.postgresql.org/about/donate
Re: [HACKERS] Patch review
Martijn van Oosterhout wrote: On Tue, Feb 12, 2008 at 05:49:12AM -0500, Bruce Momjian wrote: There is no reason I have to host the list here. I can _bounce_ emails to any address. Is there a service we can use that allows emails to be accepted and displayed on a web site and that allows deletions and comments, and has stable URLs for every email message? Is there software I can install on my server to do this? debbugs? *duck* debbugs is nice but it is more or less developed purely for debian with rarely any regular releases afaik ? Stefan ---(end of broadcast)--- TIP 6: explain analyze is your friend
Re: [HACKERS] Patch review
On Tue, Feb 12, 2008 at 05:49:12AM -0500, Bruce Momjian wrote: There is no reason I have to host the list here. I can _bounce_ emails to any address. Is there a service we can use that allows emails to be accepted and displayed on a web site and that allows deletions and comments, and has stable URLs for every email message? Is there software I can install on my server to do this? debbugs? *duck* Personally I like trac for its simplicity. You can add comments and post patches and track status. Which seems more than enough for what we do. Have a nice day, -- Martijn van Oosterhout [EMAIL PROTECTED] http://svana.org/kleptog/ Those who make peaceful revolution impossible will make violent revolution inevitable. -- John F Kennedy signature.asc Description: Digital signature
Re: [HACKERS] Patch review
Bruce Momjian [EMAIL PROTECTED] writes: If we want to communicate via a web interface, I would still need a way to collect specific messages (not always entire threads). You're talking about collecting messages. Everyone else wants to collect patches or proposed changes. There's not a one-to-one mapping between the two. One example of what we would want is a table in a wiki where it was easy to insert a reference to the most current patch, a few pointers to messages where it was discussed, and the current status. Or something like the review-board dashboard which lists the patches under consideration with their current status. Ideally with pgsql-patches gatewayed to create new patches in the list and comments gatewayed back to the list. I'm sure we want to communicate by email. We also want to have an interface we can all use to maintain a queue of patches under consideration. They're two separate, but related, things. -- Gregory Stark EnterpriseDB http://www.enterprisedb.com Ask me about EnterpriseDB's On-Demand Production Tuning ---(end of broadcast)--- TIP 9: In versions below 8.0, the planner will ignore your desire to choose an index scan if your joining column's datatypes do not match
Re: [HACKERS] Patch review
Gregory Stark wrote: Bruce Momjian [EMAIL PROTECTED] writes: If we want to communicate via a web interface, I would still need a way to collect specific messages (not always entire threads). You're talking about collecting messages. Everyone else wants to collect patches or proposed changes. There's not a one-to-one mapping between the two. One example of what we would want is a table in a wiki where it was easy to insert a reference to the most current patch, a few pointers to messages where it was discussed, and the current status. yeah but generating that list is not so easy - in a perfect world that short overview list would be generated by whatever tool we use. Or something like the review-board dashboard which lists the patches under consideration with their current status. Ideally with pgsql-patches gatewayed to create new patches in the list and comments gatewayed back to the list. most modern trackers can do that more or less (like BZ can track conversations provided it has some way to infer what mail might be related to what bug/patch/featurerequest). I'm trying to hack that feature up for the demo install to simply do that transparently for -bugs so that we can see this working for the simple case of -bugs without having to do any chances ... I'm sure we want to communicate by email. We also want to have an interface we can all use to maintain a queue of patches under consideration. They're two separate, but related, things. yes well but that interface could very well be a more featureful tracker that we use to generate that list ... Stefan ---(end of broadcast)--- TIP 4: Have you searched our list archives? http://archives.postgresql.org
[HACKERS] NLS on MSVC strikes back!
I have compiled 8.3 (CVS HEAD) with NLS on a XP box. But it seems that NLS support is broke. Could someone please confirm this. My test: Gevik=# Gevik=# show lc_messages ; lc_messages English_United States.1252 (1 row) Gevik=# Gevik=# set lc_messages TO 'de_DE.utf8'; SET Gevik=# show lc_messages ; lc_messages - de_DE.utf8 (1 row) Gevik=# select to_char(now(), 'TMDay, DD TMMonth '); to_char --- Tuesday, 12 February 2008 (1 row) Gevik=# Gevik=# set lc_messages TO 'es_ES.utf8'; SET Gevik=# show lc_messages ; lc_messages - es_ES.utf8 (1 row) Gevik=# select to_char(now(), 'TMDay, DD TMMonth '); to_char --- Tuesday, 12 February 2008 (1 row) ---(end of broadcast)--- TIP 6: explain analyze is your friend
Re: [HACKERS] NLS on MSVC strikes back!
Gevik Babakhani wrote: I have compiled 8.3 (CVS HEAD) with NLS on a XP box. But it seems that NLS support is broke. Could someone please confirm this. My test: Gevik=# Gevik=# set lc_messages TO 'de_DE.utf8'; SET Gevik=# show lc_messages ; lc_messages - de_DE.utf8 (1 row) Gevik=# select to_char(now(), 'TMDay, DD TMMonth '); to_char --- Tuesday, 12 February 2008 (1 row) Should that not be lc_time you are setting? lc_messages is for, uh, messages. cheers andrew ---(end of broadcast)--- TIP 2: Don't 'kill -9' the postmaster
Re: [HACKERS] NLS on MSVC strikes back!
Should that not be lc_time you are setting? lc_messages is for, uh, messages. No. The same thing works on 8.2.6 ---(end of broadcast)--- TIP 6: explain analyze is your friend
Re: [HACKERS] NLS on MSVC strikes back!
Gevik Babakhani wrote: Well, that should be considered a bug, not a feature. Perhaps it was fixed in 8.3. This is 8.3 I am testing with. Right, I know that. But I didn't see you trying lc_time as Andrew suggested. -- Alvaro Herrerahttp://www.CommandPrompt.com/ The PostgreSQL Company - Command Prompt, Inc. ---(end of broadcast)--- TIP 6: explain analyze is your friend
Re: [HACKERS] NLS on MSVC strikes back!
to_char's month/day name localization is implemented with gettext() not strftime(), which is why it depends on LC_MESSAGES not LC_TIME. I seem to recall that we didn't like the side-effects of the patch you are mentioning, and so it ended up being rejected outright. Correct. I have been looking for the cause of this problem and these are my finding: 1. The platforms discussed and tested here are MS Windows XP and 2003 using Microsoft VC++ 2005 and NOT MINGW. 2. As Tom described above, to_char is implemented with gettext which depends on LC_MESSAGES. I guess this is okay. 3. Changing LC_MESSAGES is done in pg_locale.c:94:pg_perm_setlocale by setting LC_MESSAGES environment variable. This works for MINGW (tested by installing PG 8.2.6), but it does not work for MSVC++ due different locale handling of gettext. Please see gettext sources:1087:localenames.c 4. Locale names are different in MS Windows. I created a C app to test gettext on Windows. setting LC_MESSAGES to Spanisg_Spain and German_Germany works but es_ES and de_DE do not :( SET LC_MESSAGES to '' has no effect because: A. gettext compiled/linked in MSVC looks for the locale of the current thread and NOT the LC_MESSAGES,LANGIAGE,LANG... environment variables. See gettext's sources B. Given current thread's locale Spanish_Spain, gettext fails to find a directory called Spanish_Spain in share/locale. As result English names are returned. A possible solutions: - Keep to_char/LC_MESSAGES handling and create a fix for B. Because we do not want to change/maintain our own version of Gettext this would also involve creating a different directory/name structure for Windows. For example share\locale\de would be share\locale\German_Germany. Any thoughts, Regards, Gevik Babakhani PostgreSQL NL http://www.postgresql.nl TrueSoftware BV http://www.truesoftware.nl ---(end of broadcast)--- TIP 5: don't forget to increase your free space map settings
Re: [HACKERS] Locale vs LIKE
Stephen Denne [EMAIL PROTECTED] writes: i.e. Do I still have to either initdb --locale=C or explicitly use text_pattern_ops? yes, if you want an index to be used -- Gregory Stark EnterpriseDB http://www.enterprisedb.com Ask me about EnterpriseDB's PostGIS support! ---(end of broadcast)--- TIP 2: Don't 'kill -9' the postmaster
