Wednesday, 2010-03-31

*** Joins: kirillka (~Miranda@global01.vester.ru)00:42
*** Joins: davidinc (~d5374b0d@gateway/web/freenode/x-pkwioyqribifwvoz)01:57
CIA-29Mantisbt: hickseydr master-1.2.x * rb81d23029c03 / (4 files in 2 dirs): Issue #11404: Record dropping of bug revisions in bug history02:16
*** Joins: Cupertino (~Cupez@unaffiliated/cupertino)02:34
*** Joins: giallu (~giallu@fedora/giallu)02:36
*** Joins: wolog (~wolog@wolog.info)03:21
CIA-29Mantisbt: urkle * r4eb5d9aebbe3 /login_page.php: Fix #11415: Check executable bit on admin/ folder, not readable bit04:47
CIA-29Mantisbt: urkle master-1.2.x * r482a493fc42a /login_page.php: Fix #11415: Check executable bit on admin/ folder, not readable bit04:47
*** Quits: wolog (~wolog@wolog.info) (Ping timeout: 240 seconds)05:06
*** Joins: wolog (~wolog@wolog.info)05:07
CIA-29Mantisbt: hickseydr * r9527e0ef9c84 / (71 files in 9 dirs): Issue #11732: Remove built-in source code integration support05:31
paul__dhx_m: last patch06:14
dhx_mpaul__: hey06:14
paul__dont break history06:14
paul__i.e. we need to keep define + history for historical purposes imo06:14
paul__so I wonder about dropping that code entirely...06:15
dhx_mhmmm06:15
dhx_mI'm not sure we can sustain that forever though06:15
paul__i'd be inclined to leave the code in place for now and think about it06:15
paul__i.e. if someone is using cvs://foo links in mantis06:16
paul__and updates to 1.306:16
paul__we need to ensure they'll still work06:16
dhx_mI think they've got bigger problems than cvs:// links if they're still using CVS :p06:16
paul__similarly is probably a bug to have a history id line in db with a constant that does't exist06:16
paul__so to do your patch we'd really need to do a db upgrade to nuke stuff from bug_history table06:17
dhx_mAFAIK it will just drop the display of the history entry06:17
paul__which seems a bad idea :P06:17
dhx_mI was actually thinking of doing that but decided against it (for now) in case people want to downgrade from 1.3.x to 1.2.x06:17
paul__dropping the script/adding new stuff might be a good idea06:17
paul__i.e. the import script06:17
dhx_mwe don't really need history entries for checkins06:17
paul__but i'd be inclined to revert that and leave the code in and check06:18
dhx_mbecause it already adds a bugnote that is timestamped06:18
dhx_mpaul__: while you're here, I was just about to kill the file_send_chunk() function in file_download.php and use readfile() instead06:18
dhx_mbecause PHP's readfile() already does chunking06:19
paul__iirc not06:19
paul__there was reason I moved from readfile to that06:19
dhx_mperhaps that was before we turned off output buffering in file_download.php?06:19
paul__no06:20
paul__we've always turned it off06:20
paul__your missing the issue06:21
paul__issue is if you readfile(2GBfile)06:21
paul__yes php chunks it out06:21
paul__but it reads it all into ram first06:21
dhx_mnot for me...06:21
dhx_mI didn't notice any increase in PHP memory usage when trying with a 50MB file06:22
paul__unless php have changed readfile internals06:22
paul__but it was fixing an isuse ;p06:22
*** Quits: Cupertino (~Cupez@unaffiliated/cupertino) (Quit: I give up...)06:22
dhx_mI imagine it might have been an older version of PHP that was buggy06:22
paul__it was doing it in some version of php5.2 for me06:23
paul__as I always run new php's ;p06:23
dhx_msearch mantisbt.org/bugs for PHP 5.3 and you'll see most of the issues are problems with PHP 5.3 and not MantisBT06:23
paul__anyway, id be inclined to leave that code (for now) at least06:24
paul__and re the checkin changes - maybe revert soe of it06:24
paul__to deal with history06:24
dhx_mI'm trying to hunt down more information at the moment06:24
* paul__ needs to think06:24
paul__bbl06:26
dhx_mok06:26
*** Quits: wolog (~wolog@wolog.info) (Ping timeout: 240 seconds)06:26
CIA-29Mantisbt: hickseydr * rfbfdfe40e07a /file_download.php: Fix #11362: Use readfile instead of custom file_send_chunk06:28
CIA-29Mantisbt: hickseydr master-1.2.x * rf845ed542e27 /file_download.php: Fix #11362: Use readfile instead of custom file_send_chunk06:29
dhx_mI'll revert/fix later if there are problems06:29
dhx_mbut I doubt there will be any... all the blog posts/comments I've seen on the internet about readfile() bugs are from 2006-200706:29
dhx_mand checking PHP's bug tracker shows they were fixed ages ago06:29
* paul__ sighs06:30
paul__i'm pretty sure I can make the need of readfile() return tonight :P06:30
dhx_mdoubt it :p06:30
paul__well, bare in mind06:30
paul__I normally test php trunk06:30
paul__as well as a release06:30
paul__or well, RC's06:30
paul__so if I tested and it was broken in two versions in November06:31
paul__I doubt it's magically fixed now :)06:31
dhx_mwhich version of libc do you use?06:31
paul__windows...06:31
dhx_m"To anyone that's had problems with Readfile() reading large files into memory the problem is not Readfile() itself, it's because you have output buffering on. Just turn off output buffering immediately before the call to Readfile(). Use something like ob_end_flush()."06:32
dhx_mthe 2MB limit error people keep talking about was fixed in 200606:32
paul__in which case, we probably definitely need it :P06:32
dhx_mand I don't see any recent references to readfile() using large amounts of memory06:33
dhx_m... but we turn output buffering off ;)06:33
paul__not if you use zlib.compression_on06:33
paul__which is what MS recommend on windows :P06:33
paul__sas you can't then turn ob off06:33
paul__as it does nothing06:33
paul__I'm pretty sure we need the chunked code :P06:34
dhx_mwe turn off zlib compression explicitly ;)06:34
paul__no06:34
paul__not for IIS06:34
dhx_mif ( ini_get( 'zlib.output_compression' ) && function_exists( 'ini_set' ) ) {06:34
dhx_mini_set( 'zlib.output_compression', false );06:34
dhx_m}06:34
dhx_mno check for IIS06:34
dhx_mIMO the web server should be handling gzip compression, not PHP06:36
paul__anyway, the readfile_chunked was added to fix a bug06:38
paul__so unless there's a reason to remove it06:38
paul__i'd leave it in place for now06:38
paul__as the reason it was added in first place was to fix memory issues06:40
paul__and similarly at some point in futurte would be nice to be able to do resumable file downloads06:40
dhx_mthat's what fpassthru is for :)06:40
paul__no06:41
paul__or well06:41
paul__ok there yes06:41
dhx_mI guess you can't set the upper bound with fpassthru though06:41
paul__aka we'd need a readfile type function to cope with that06:41
paul__nod06:41
paul__so for now, can we leave the chunked stuff in place06:41
paul__until we test properly06:41
dhx_moh06:42
dhx_mI'm just confused as to why so many people had problems with the old function06:44
paul__readfile() or _chunked?06:44
dhx_mchunked06:44
dhx_mI've read through it multiple times and can't work out why it'd ever result in memory exhaustion06:45
paul__there was a potential bug I fixed with return values06:45
paul__readfile() was causing problem in some instances I believe06:46
dhx_mthose bug reports are06:47
dhx_mhttp://www.mantisbt.org/bugs/view.php?id=1013506:47
dhx_m(fixed in July 2009)06:47
dhx_mhttp://www.mantisbt.org/bugs/view.php?id=1076506:47
dhx_mPEBKAC06:48
dhx_m(ie. *VERY* outdated version of PHP)06:48
dhx_mideally we need an attachment_api or something of that nature06:49
dhx_mand to support resuming downloads, etc we'd need to use a chunked function06:50
dhx_mbut for other uses it'd be better to try using XSENDFILE or readfile06:50
*** Joins: wolog (~wolog@wolog.info)07:51
*** Quits: wolog (~wolog@wolog.info) (Ping timeout: 240 seconds)07:56
*** Joins: wolog (~wolog@wolog.info)08:12
*** Joins: mantisbt_53966 (~970d1302@gateway/web/freenode/x-qapgxqumsyuzmefh)08:46
mantisbt_53966anyone online?08:48
mantisbt_53966has nayone tried to do multiple mantis installation on the same machine?08:49
*** Parts: mantisbt_53966 (~970d1302@gateway/web/freenode/x-qapgxqumsyuzmefh)08:52
*** Joins: bthomas (~balbir@82-70-130-246.dsl.in-addr.zen.co.uk)09:02
bthomasHi. I have set $g_file_upload_method = DISK; (as described in manual) then updated project to upload to /var/local/mantis using mantis "manage project". However files are still being uploaded to DATABASE. Am I missing something ?09:07
bthomasmy platform = Debian/Lenny, Mantis Version = 1.1.609:09
*** Quits: davidinc (~d5374b0d@gateway/web/freenode/x-pkwioyqribifwvoz) (Ping timeout: 252 seconds)09:10
CIA-29Mantisbt: hickseydr * r25223c9e4533 / (16 files in 2 dirs): Fix #11530: Support multiple access levels above manage_user_threshold09:26
*** Joins: fanno (~b3g@193.3.95.240)09:31
CIA-29Mantisbt: hickseydr master-1.2.x * r67f43bde8b6e / (16 files in 2 dirs): Fix #11530: Support multiple access levels above manage_user_threshold09:49
*** Joins: daryn (~INTERACT\@rrcs-76-79-4-2.west.biz.rr.com)09:54
*** Quits: fanno (~b3g@193.3.95.240) (Remote host closed the connection)09:54
*** Quits: daryn (~INTERACT\@rrcs-76-79-4-2.west.biz.rr.com) (Quit: daryn)10:02
*** Joins: daryn (~INTERACT\@rrcs-76-79-4-2.west.biz.rr.com)10:02
*** Quits: kirillka (~Miranda@global01.vester.ru) (Quit: kirillka)10:02
*** Quits: wolog (~wolog@wolog.info) (Ping timeout: 240 seconds)10:20
CIA-29Mantisbt: hickseydr * r820b45bea6be /core/html_api.php: Issue #11729: Preselect next highest status in bug_change_status10:22
CIA-29Mantisbt: hickseydr master-1.2.x * r886dccd9a467 /core/html_api.php: Issue #11729: Preselect next highest status in bug_change_status10:22
*** Joins: davidinc (~d5374c13@gateway/web/freenode/x-iscnjvcsfmwiiusn)10:28
*** Quits: davidinc (~d5374c13@gateway/web/freenode/x-iscnjvcsfmwiiusn) (Ping timeout: 252 seconds)10:35
darynpaul__ did we ever decide if it was ok to use a long text type for custom_field_string_table value field?10:56
*** Quits: giallu (~giallu@fedora/giallu) (Read error: Operation timed out)11:14
*** Parts: bthomas (~balbir@82-70-130-246.dsl.in-addr.zen.co.uk)11:51
*** Joins: giallu (~giallu@fedora/giallu)12:00
*** Joins: moto-moi (~hylke@cara.xs4all.nl)12:06
*** Joins: cobexer (~cobexer@91-113-120-76.adsl.highway.telekom.at)12:56
*** Joins: mantisbt_99515 (~d53372d2@gateway/web/freenode/x-vbdqtbsyxoprcgir)12:57
*** Joins: wolog (~wolog@AOrleans-152-1-31-60.w90-21.abo.wanadoo.fr)13:18
*** Joins: siebrand (~beis@sm.xs4all.nl)13:22
*** Joins: mantisbt_76316 (~dcff07a2@gateway/web/freenode/x-ivxqbrgontrrfexa)13:46
mantisbt_76316hshs13:46
*** Quits: mantisbt_76316 (~dcff07a2@gateway/web/freenode/x-ivxqbrgontrrfexa) (Client Quit)13:46
*** cobexer is now known as \cobexer|away14:05
paul__daryn: imo, yes14:44
paul__just need to be *careful* which type14:44
paul__(ref non-mysql dbs)14:45
darynpaul__ i have asmall patch to allow textarea for cf...i just need the schema update. wanna do that since I don't have other db's set up?14:47
*** Joins: AzaToth (~azatoth@wikipedia/AzaToth)15:12
*** Joins: rolfkleef (~rolf@urtica.xs4all.nl)15:32
*** Quits: \cobexer|away (~cobexer@91-113-120-76.adsl.highway.telekom.at) (Remote host closed the connection)17:05
*** Joins: fanno (~Morten@90.184.93.233)17:23
*** Quits: rolfkleef (~rolf@urtica.xs4all.nl) (Ping timeout: 265 seconds)17:26
*** Quits: daryn (~INTERACT\@rrcs-76-79-4-2.west.biz.rr.com) (Quit: daryn)17:38
*** Quits: mantisbt_99515 (~d53372d2@gateway/web/freenode/x-vbdqtbsyxoprcgir) (Quit: Page closed)17:45
siebrandhttp://sourceforge.net/blog/potm-201004/ Mantisbt is SourceForge.net Project of the Month, April 201017:54
moto-moinice :)17:54
*** Quits: fanno (~Morten@90.184.93.233) (Read error: Connection reset by peer)18:14
*** Joins: roy2k (~royk2@wsip-98-172-168-236.sd.sd.cox.net)18:15
roy2kdoes anybody know if someone's tried tackling the concepts behind user pain http://lostgarden.com/2008/05/improving-bug-triage-with-user-pain.html with mantis?18:16
*** Quits: siebrand (~beis@sm.xs4all.nl) ()18:45
*** Quits: moto-moi (~hylke@cara.xs4all.nl) (Quit: Ex-Chat)18:51
*** Quits: scribe9343423 (~scribe934@mantisforge.org) (Remote host closed the connection)20:00
*** Joins: scribe9343423 (~scribe934@mantisforge.org)20:00
*** Quits: roy2k (~royk2@wsip-98-172-168-236.sd.sd.cox.net) (Read error: Connection reset by peer)21:12
*** Quits: AzaToth (~azatoth@wikipedia/AzaToth) (Remote host closed the connection)21:13
*** Joins: squarebox (~d3092347@gateway/web/freenode/x-kemeaqhnpbauipgi)22:01
squarebox@dhx - just have to say you were very very busy yesterday on the bugtracker.22:03
squareboxbtw is there a feature list of stuff you want to add for 1.3?22:04
dhx_msquarebox: hehe thanks :)22:06
dhx_msquarebox: short term aims can be seen in the roadmap... longer term goals can be found at http://sourceforge.net/blog/potm-201004/22:07
squareboxah ok didn't know about the blog22:08
squareboxalso just like to say all your guys hard work is greatly appreciated and seeing mantis's bugtracker updated almost daily is great in term of activity as well as verifying my choice to switch over from Trac22:08
dhx_mwell it's only from today :)22:08
squareboxhaha... well there you go!22:09
dhx_mthanks22:14
dhx_mI'd love to see more contributions though :)22:14
dhx_mbut in comparison to other bug tracking projects I think MantisBT would be at the top (or near the top) in terms of number of contributors22:14
squareboxYeah i've been toying with the idea of trying to learn mantis more so to contribute in my off time as a means to learn PHP22:26
squareboxbut compared to other projects i'd agree that the number of not only the contribitors, but the the fact that they actually maintain their patches for different mantis releases is amazing22:27
dhx_mbackporting patches isn't fun22:28
squareboxmy only comment in that respect is that it seems that the mantis team is slow to incorporate patches into the trunk, but i'm sure you have your reasons22:28
dhx_mthey do generally get lost in the bug tracker22:28
dhx_mhowever the main problem is the quality of patches22:29
squareboxwell, i could def. understand not backporting if that was the big hurdle22:29
squareboxwell it's probably because they are just patches... i mean to say the people creating patches are doing so cause they are being yelled at their bosses to fix it, and so it's more of a band-aid22:30
squareboxso their goal in terms of resovling the issue is a little different that the mantis devs22:32
squareboxthough i always assumed you guys took the patches and fixed them up to include in mantis, as obviously it's hard to expect that the contributor would fix up their code22:32
dhx_mhttp://www.mantisbt.org/bugs/permalink_page.php?url=http%3A%2F%2Fwww.mantisbt.org%2Fbugs%2Fsearch.php%3Fproject_id%3D0%26sticky_issues%3Don%26sortby%3Dlast_updated%26dir%3DDESC%26hide_status_id%3D90%26tag_string%3Dpatch22:33
dhx_mthere aren't too many patches being submitted anyhow22:33
squareboxunforutanely i lack acces to see that page22:34
dhx_moh oops22:34
dhx_mhttp://www.mantisbt.org/bugs/search.php?project_id=0&sticky_issues=on&sortby=last_updated&dir=DESC&hide_status_id=90&tag_string=patch22:34
dhx_mI've never understood why guests aren't allowed to create permanent links...22:34
squareboxon that note, is ther ea specific reason that you don't show the stats/summary for the mantis bugtracker?  it might be a useful marketing tool in the sense of being able to denote how active the bugtracker really is... which in the open source community is a big plus22:35
dhx_mhttp://www.mantisbt.org/bugs/summary_page.php works for me (but I do have a developer account)22:36
dhx_mI agree that it'd be nice to enable graphing support and open it up (as well as the stats) to the public22:36
dhx_mpossibly it's an issue of CPU/memory usage22:37
squareboxyeah access denied and it doesn't show up on the links strip22:38
squareboxwell it's also nice to know there is a dev around in my timezone22:39
dhx_m:)22:39
dhx_mI'd be interested in hearing about the largest MantisBT tracker in the world22:40
dhx_mie. 1mil+ bugs, etc22:40
dhx_mit seems MantisBT is primarily used in closed source development projects22:40
dhx_mwhereas Bugzilla seems to be the predominant bug tracker used by open source projects22:40
squareboxwell you could always ask in your new blog22:41
squareboxas a call out for mantis support to show how good it is22:42
dhx_mI guess the problem is that you can't actually see the MantisBT installation because it's private22:42
squareboxor even in the forum too... thought the forum is kinda a mess22:42
dhx_myeah I don't use it :)22:42
squareboxah22:42
squareboxreading the project of the month thingy... and you really should play up that mantis was created by a japanese guy22:44
squareboxthe japanese goble up anything that was created by a japanese and has become famous22:44
dhx_mlol22:44
squareboxthey are very small community orientated22:44
dhx_mI'm not sure how closely MantisBT 1.2.0 resembles what Kenzaburo had in mind for the project22:45
squareboxdoesn't matter22:46
squareboxto them at least22:46
squareboxit's liek a stamp of approval from a japanese person22:46
dhx_m"made in Japan" :)22:46
squareboxyeah basically22:46
dhx_mthat's quite interesting22:47
squareboxtypically a japanese person in japan isn't really acknowledged until the western world recognizes them... and then they become superstars in japan... like all the japanese baseball players in america...22:48
squareboxwhen they were in japan only the peopel into baseball cared about them22:48
squareboxbut when they went to america, they became famous people and their daily comings and goings is reported on the news here like it was any other famous rock star or something22:49
dhx_mif we ever have a translatable version of mantisbt.org I'll make it follows your tip :)22:49
dhx_m"MantisBT was invented in Japan, by the Japanese, for the Japanese and is used by millions, no wait... trillions of people around the world"22:50
dhx_m:D22:50
squareboxhaha22:53
dhx_mI don't suppose you use Windows and MS Office? I was wondering whether the Word export feature in 1.2.0 works OK without errors popping up in the document about missing CSS files22:53
squareboxthough a good question is if your application is localized, why coudln't you ask the same translators if they would mind translating a localized version of at least the main page and maybe install docs22:54
dhx_mView Issues => Print Reports => Word icon22:54
squareboxyeah i'm in a ms shop company22:54
dhx_mbug 10877 @ http://www.mantisbt.org/bugs/view.php?id=1087722:55
dhx_mthose are typically the sort of patches I don't commit because I can't test them22:55
squareboxyeah i get a similiar error, though i'm on 1.2.0rc2 ... one sec i have a 1.2.0 release env i can test too22:57
dhx_mit'll be the same22:57
squareboxc:\mantis\css\default.css can't find22:57
dhx_mif you get a moment could you provide feedback @ 10877 about whether the patch works?22:57
dhx_mif it does I can commit it :)22:57
squareboxok 1 sec... today is the start of the new fiscal year here and i have absolutely no work anyways22:58
dhx_myeah only if you have time :)23:00
squareboxhmm23:05
squareboxdoesn't seem to change anything for me23:05
squareboxlet me add a note23:05
dhx_mok thanks for testing :)23:06
squareboxwhich is unfortunate, cause this is one piece of functionality that would be nice if it worked23:09
squareboxfeel free to send me an e-mail if there are patches you would like me to test since i'm running in a windows env.  though we aren't using svn integration currently cause of all the bugs we had seen in the tracker as well currently using svn 1.4 with no in house expertise concerning it23:11
dhx_mwell I just dropped the in-built SVN integration23:13
dhx_mbecause it sucks23:13
dhx_mthe SourceIntegration plugin works nicely in its place23:13
squareboxright i saw that and got all excited cause it's a plugin, but then realized i'd have to wait till 1.323:14
dhx_mit works with 1.2.x :)23:14
squareboxOH!23:14
dhx_mall I was saying in the blog post was that you can't use the in-built/old integration system in 1.3.x23:14
squareboxso the bug note is specifcally related to removing the built-in svn support only then23:15
dhx_myou can use either in 1.2.x though23:15
dhx_myep23:15
squareboxah ok... might be something to try23:15
squareboxhold on a sec i might have screwed up applying that patch23:17
squareboxyeah it fixes the default.css stylesheet error23:21
squareboxthough i now get the plugin stylesheet error like he mentions23:21
dhx_maha23:22
dhx_mthanks :)23:22
dhx_mat least I know it's on the right track23:22
CIA-29Mantisbt: hickseydr master-1.2.x * r86fc322dfc41 /view_user_page.php: Fix #11530: Don't expose real name and email of higher users23:24
CIA-29Mantisbt: hickseydr * r7062c67759c8 /view_user_page.php: Fix #11530: Don't expose real name and email of higher users23:24
CIA-29Mantisbt: jonasnockert * r9aff88b637bf /core/html_api.php: Fix #10877: Word export results in missing CSS errors23:30
CIA-29Mantisbt: jonasnockert master-1.2.x * rd1d00c02b944 /core/html_api.php: Fix #10877: Word export results in missing CSS errors23:30
squareboxis CIA the GIT monitor?23:31
dhx_mit's just a bot (from http://cia.vc) that announces new commits23:32
dhx_mbbl23:37
*** Quits: dhx_m (~anonymous@c122-107-157-71.eburwd5.vic.optusnet.com.au) (Quit: Leaving)23:37

Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!