Tuesday, 2009-04-07

../irclogs/#mantishelp.2009-04-07.log
--- scribe started ---00:00
mantisbotNew bug: Bug 10296 - nobswolf - open - new05:36
mantisbotNew bug: single project is not selected - http://www.mantisbt.org/bugs/view.php?id=1029605:36
mantisbotNew bug: Bug 10297 - nobswolf - open - new08:37
mantisbotNew bug: allow a reference from a custom column - http://www.mantisbt.org/bugs/view.php?id=1029708:37
mantisbotNew bug: Bug 10298 - jan jindra - open - new10:27
mantisbotNew bug: erewre - http://www.mantisbt.org/bugs/view.php?id=1029810:27
CIA-18Mantisbt: paul * r7a111a586dc5 /core/ (error_api.php gpc_api.php lang_api.php obsolete.php): PHPDoc updates10:40
CIA-18Mantisbt: paul * r96cb7ba31225 /graphs/ (8 files): db_get_table should be used to retrieve table names10:40
paulrhi kirill10:57
[KK]Kirillpaulr: Hi11:09
[KK]KirillMy stupid work take more time...11:10
[KK]KirillI see John work with source-integrator, and append new events in mantisbt11:10
[KK]Kirillbut my wish not append11:10
paulrhmm?11:12
[KK]KirillI want add event to work with uploaded files11:13
[KK]Kirillfor example compress file to zip and put in issue archived files, not original11:14
[KK]Kirillor secont example - change filename from national character to latin11:15
[KK]Kirilldid you understand what I want?11:15
paulrhmm11:15
paulras plugin events?11:16
[KK]Kirillyes11:16
[KK]KirillI want plugin event for work with attached files11:16
[KK]Kirillbefore they attached11:16
[KK]Kirillor/and after attached11:17
paulrsounds nasty11:17
paulri.e. I can see that breaking stuff11:17
[KK]Kirillpaulr: sorry, I don't understand11:19
paulr[12:14:32] <[KK]Kirill> for example compress file to zip and put in issue archived files, not original11:19
[KK]Kirillyes11:19
[KK]Kirillexample from real life11:20
[KK]KirillI have mantis for my company to support other it-personal11:20
[KK]Kirillthey upload files in national characters11:21
[KK]Kirillinissue they show nice11:21
[KK]Kirillcorrectly11:21
[KK]Kirillbut when I try download in different browser I take break filename11:22
[KK]Kirillin unicode-codes11:22
paulris that a bug in filedownload though?11:22
[KK]Kirillmay be.11:23
[KK]KirillI can understand why11:23
[KK]Kirillmay be in web-server misconfig11:23
[KK]Kirillmay be browser or mantisbt11:23
paulradding a plugin to zip files could be interesting ( although I *think* it would probably result in hassle)11:23
[KK]Kirill:)11:24
[KK]Kirillok. it's will be nice to work with files11:24
paulradding a plugin to change filename's strikes me as something that will probably cause some level of grief11:24
[KK]Kirillpaulr: change name to unique filenames11:25
paulri.e. if someone attaches a file atm in chinese, i'd expect them to get the file back with a chinese filename11:25
[KK]KirillI want store original filename in plugin-table11:25
paulrdont we already store the original filename in the file_table?11:26
[KK]Kirillin mantis_file_table store changed filenames, in plugin file table store original. In all issues show original file names11:29
[KK]Kirillbut when download plugin put changed filename11:29
[KK]Kirillsorry, not put - give11:30
[KK]Kirilland downloaded filename with changed filename11:30
paulrlo daryn13:20
darynmornin13:20
daryner...afternoon i guess13:21
paulrdaryn: i'm so bored13:44
[KK]Kirillwho?13:45
[KK]KirillI?13:45
paulr<<13:46
[KK]Kirill:)13:46
paulrnuclear_eclipse: ./13:49
paulrdaryn: wherd you get to with cddef classes?14:51
darynmmm15:00
darynbeen awhile15:00
* paulr is playing with cc15:00
daryni think the def classes near working order but i was trying to figure out how to do the cf types15:01
paulrcheck out http://www.mantisforge.org:8080/cruisecontrol/buildresults/mantisbt15:01
darynsweet15:01
paulrit's kinda broken atm though15:02
darynwe use hudson in house15:02
darynnot for mantis though15:02
paulrhudson?15:02
darynanother automated build system15:02
daryncontinuous integration15:03
* paulr is currently working on php codesnifer rules15:03
paulrso nows a good time to tell me if we dont like anything15:04
daryni split the cf classes, api changes, and page changes into separate branches. was trying to make sure existing functionality stays the same15:05
paulr 2977 | ERROR | Functions must not contain multiple empty lines in a row; found15:06
paulr      |       | 2 empty lines15:06
* paulr slaps daryn ;p15:06
darynbasically i'm implementing the api's into classes while mostly leaving the custom_field_api file alone.15:06
daryn:P15:06
darynit's not done15:06
darynso...leave me alone15:06
paulrthat was in filter api15:06
paulr;p15:06
darynthat's not done either15:07
darynrebooting...back momentarily15:08
nuclear_eclipsepaulr: hi15:18
paulrlo15:18
paulr?>15:21
darynpaulr: I thought we weren't supposed to use closing php tag anymore...15:22
nuclear_eclipsedaryn: you're correct15:23
paulrreally?!15:23
nuclear_eclipseright, the whole point of leaving off the closing tags is to prevent unintentional whitespace output15:24
paulrthat's quite interesting you mention this15:24
paulr:(15:25
CIA-18Mantisbt: paul * rb4cf7972d817 / (28 files in 3 dirs): Remove remaining closing tags15:26
CIA-18Mantisbt: paul * r5cf7970fb208 / (197 files in 8 dirs): Consistency update15:26
darynheh...i was just trying to be funny15:28
daryn[10:21] <paulr> ?>15:28
paulroh so you didn't mean to do it? ;p15:32
darynno, but it's fine by me.15:35
nuclear_eclipsepaulr: bug 9856!15:39
mantisbotBug 9856 - jreese - open - assigned15:39
mantisbotImplement method for tracking changes to Description, Steps to Reproduce, and Additional Info - http://www.mantisbt.org/bugs/view.php?id=985615:39
paulraye15:39
* paulr needs a break soon so will look then :)15:40
paulrjohn: is 9856 sync'd with trunk ?15:42
nuclear_eclipsecheck git.mantisforge.org/w/mantisbt/jreese.git - that one is15:42
paulrlooks to be out of date ;p15:43
paulr45 hours agoPaul Richards15:43
* paulr grins15:43
paulrright lets see15:43
paulrchanges for this are only in 12 files?15:44
paulraka, I can just review http://git.mantisforge.org/w/mantisbt/jreese.git?a=treediff;h=refs/heads/revisions;hp=631dd84ba59f2159dc3f1ec7b6c08e380b01ce99;hb=revisions;hpb=631dd84ba59f2159dc3f1ec7b6c08e380b01ce99 ?15:44
nuclear_eclipsethat's a diff from master to the revisions branch, correct?15:45
paulryes15:46
paulrhow do you undelete a file in git15:50
paulrnm15:52
paulrreset ;/15:52
paulris it FALSE or false ?15:55
nuclear_eclipseis what?15:56
paulrfalse :)15:56
daryni prefer FALSE15:56
nuclear_eclipseit seems that Mantis tends to use null/false/true for most of the codebase...15:56
paulrhttp://pear.php.net/package/PHP_CodeSniffer/docs/latest/PHP_CodeSniffer/Generic_Sniffs_PHP_LowerCaseConstantSniff.html15:57
paulrGeneric_Sniffs_PHP_LowerCaseConstantSniff.15:57
paulrChecks that all uses of true, false and null are lowerrcase.15:57
paulrto run15:57
paulror not to run15:57
nuclear_eclipseI generally wouldn't really consider that a "code smell" though....15:57
paulrGeneric_Sniffs_PHP_UpperCaseConstantSniff.15:58
paulrChecks that all uses of TRUE, FALSE and NULL are uppercase15:58
paulrheh15:58
* paulr tempted to run lowercse so it's consistent15:58
paulrelseif or else if16:03
nuclear_eclipseelse if16:06
darynelseif16:07
darynjust to be contrary16:07
nuclear_eclipse'else if' is easier to read16:09
paulrnuclear_eclipse: where's $p_event in formatting functions?16:10
paulrit's unused?16:10
paulrnuclear_eclipse: do we really want to use ERROR_GENERIC?16:14
paulr(at least, without specifying error_parsm16:15
paulr+       switch( $t_revision['type'] ) {16:15
paulr+       case REV_DESCRIPTION:16:15
paulrand dont we indent that by one16:15
paulr+html_page_top1( bug_format_summary( $t_bug_id, SUMMARY_CAPTION ) );16:17
paulr+html_page_top2();16:17
paulrshould just call html_page_top now16:17
nuclear_eclipse$p_event should still be the first parameter to any hooked function16:17
paulrjust it's not normally used?16:18
nuclear_eclipsepaulr: if there's an ERROR_GENERIC, I'll fix that16:18
paulr $g_bug_revision_drop_threshold = MANAGER;16:18
paulris/should there be a threshold to view previous revisions16:18
nuclear_eclipse don't think there should be, considering it was privy to anyone with access to see the bug in the first place...16:19
paulryou create a revision in bug_create - if so:16:19
paulrshould we not create the first revision when editing for the first time16:19
nuclear_eclipseit does16:19
paulri.e. atm, do you duplicate description on every insert?16:19
nuclear_eclipsebasically "yes", although it's only a "duplicate"  of the current description16:20
paulrso for the 'normal case'16:21
paulri.e. user submits bug, doesn't edit it and we deal with bugnotes16:21
paulrdo we store the bug data twice?16:21
paulrsimilarly, looking at update16:22
paulr+               if ( bug_revision_count( $p_bug_id, REV_DESCRIPTION ) < 1 ) {16:22
paulr+                       $t_revision_id = bug_revision_add( $p_bug_id, $t_current_user, REV_DESCRIPTION, $t_old_data->description );16:22
paulr+               }16:22
paulr+               if ( bug_revision_count( $p_bug_id, REV_DESCRIPTION ) < 1 ) {16:22
paulr+                       $t_revision_id = bug_revision_add( $p_bug_id, $t_current_user, REV_DESCRIPTION, $t_old_data->description );16:22
paulr+               }16:22
paulrthat looks like you double up code?16:22
* paulr is looking at diff at http://git.mantisforge.org/w/mantisbt/jreese.git?a=treediff;h=refs/heads/revisions;hp=631dd84ba59f2159dc3f1ec7b6c08e380b01ce99;hb=revisions;hpb=631dd84ba59f2159dc3f1ec7b6c08e380b01ce9916:22
paulrand thinking the bug_create stuff was supposed to have been deleted?16:23
nuclear_eclipsethat code stores takes care of "grandfathering" in bugs that were created before the revision system was put in place16:23
paulrthe update code?16:24
paulr(can't we always use that and not store a revision in bug_create?16:24
paulrbug_revision_drop is broken for multiple revisions16:26
paulrif ( REV_ANY < $p_type ) {16:27
paulrpart of me thinks we could be using !== so when we end up adding another revision type later16:27
nuclear_eclipseman, it's really been a long time since I put this together...16:28
paulraye16:28
nuclear_eclipsei'm having trouble remembering why I did certain things the way I did...16:28
paulri've been using @return null for functions that dont return anything16:28
paulr+       if ( $p_bugnote_id > 0 ) {16:29
paulr+               $t_query .= ' AND bugnote_id=' . db_param();16:29
paulr+               $t_params[] = $p_bugnote_id;16:29
paulr+       } else {16:29
paulr+               $t_query .= ' AND bugnote_id=0';16:29
paulr+       }16:29
paulrlooks fairly pointless16:29
* paulr wonders what your actually trying to do for bugnote=016:29
paulrshould bug_revision_last return null or array() if empty?16:29
paulryou ad  $t_raw = true; to history_localize_item - I *Thought* we had a seperate function to return *Raw* data, or by raw tdo you more mean 'nolinkified'16:34
* paulr reaches end of file16:34
paulrall i've not done is run it16:34
nuclear_eclipsepaulr: raw = nolinks16:40
nuclear_eclipsebugnote=0 is for revision types not tied to a bugnote (summary, etc)16:43
paulrsure, but that code is the same16:44
paulrif (id > 0) { AND bugnote_id=id } else AND bugnote_id =016:44
nuclear_eclipseagreed it looks silly, it's just a habit trying to handle getting passed bugnote=-1 or some such16:45
paulrahh ok16:45
paulrid=0+id16:45
paulrhm, no16:45
paulrright whatever; /16:45
CIA-18Mantisbt: paul * r9aacdc6443ff / (34 files in 4 dirs): elseif -> else if16:46
CIA-18Mantisbt: paul * r96a1579bcf7e / (23 files in 4 dirs): sizeof->count16:46
* paulr pokes nuclear_eclipse for signs of life17:34
disrupt0wzanyone knows how to make an issue to be automatic daily scheduled?17:43
[KK]Kirillpaulr: ?18:13
[KK]Kirillone question: why need remove last tag http://git.mantisforge.org/w/mantisbt.git?a=commitdiff;h=b4cf7972d817fbcb7d45d038854b7b33dc90f07618:13
[KK]KirillRemove remaining closing tags18:13
nuclear_eclipse[KK]Kirill: the point of leaving off the closing tags is to prevent unintentional whitespace output18:14
[KK]Kirillthanks, John, I understand18:14
paulrnuclear_eclipse: http://www.mantisforge.org:8080/cruisecontrol/buildresults/mantisbt?tab=metrics18:17
nuclear_eclipsefancy18:17
paulrwhere'd we get to with 9856 ?18:17
darynnice paulr18:19
Roahi x)18:21
Enisseohi everyone!19:09
EnisseoI've just finished some work on the graphic design of mantisbt19:10
EnisseoI thought the default design did not help understand and using it19:12
Enisseoso, if you are interested in what I've done, feel free to contact me ;)19:13
paulroh?#19:17
* daryn_away is away: Gone away for now19:24
Enisseo[you can see it on: http://enisseo.net/mantisbt , login: test, password: test, does not work correctly on IE yet, hoping this message will not be seen as spam]19:28
Enisseo[tried to do so without altering too much html code - a few lines modified on the 1.1.6, only by editing the css file]19:30
paulris that css ?19:32
Enisseoyes19:33
paulrany non-css changes?19:33
Enisseono, minor changes to the html code19:33
Enisseolike... 5 .php files, 20 lines19:34
Enisseomainly css classes added (example: <table class="width100"> => <table class="menu width100">)19:34
Enisseoand some "bugs" fixed (wrong colspan number, for example)19:35
Enisseoi'm not sure all pages work fine19:35
nuclear_eclipseEnisseo: problems like incorrect colspans should be reported as a bug on our official tracker please19:36
Enisseoi will ;)19:36
paulrnuclear_eclipse: 9856? :(19:37
nuclear_eclipseotherwise, I find a lot of your design visually distracting compared to the simplicity of the default theme19:37
nuclear_eclipsepaulr: what about it?19:37
paulrwell what we doing about it?19:37
paulrshall I fix?19:37
Enisseonuclear_eclipse: any suggestion?19:37
paulror what :)19:37
nuclear_eclipselast I heard from you, there was no definitive summary of what you thought needed to be changed19:37
paulr1) we should only store a revision on the first change imo19:38
Enisseo[i have to be registered in order to post new issues?]19:38
paulr[yes]19:38
paulrI suspect that should be as simple as just removing 3 lines from bug_create and letting the legacy handler handle it19:39
nuclear_eclipseEnisseo: more subtlety; less contrast; since the arrows don't do anything, they are confusing; the text fields have no top border, so it's difficult to see where they "end"19:39
paulrnuclear_eclipse: you want to ban T_CLOSE_TAG T_INLINE_HTML T_OPEN_TAG sequences right?19:39
paulr(where trim T_INLINE_HTML = '')19:39
nuclear_eclipsebasically, preg_replace_all( '/?>[\s\n]*<?php/', '' )19:40
nuclear_eclipsepaulr: other than deferring revisions from bug_create to only on bug_update, any other changes?19:41
paulrminor whitespace stuff i think19:41
paulri'd need to run it19:41
paulrthe other thing19:41
paulrlocalisation/history19:41
paulriirc, you modified history in one place19:41
paulrI thought there was 2 loops history went into19:41
paulraka a 'get raw data'19:42
paulrand 'get formatted data'19:42
paulrneed to check it's just one19:42
nuclear_eclipseyeah, I handled that appropriately19:42
paulrok19:42
nuclear_eclipseso just the bug_create issue?19:44
paulrthat was the main thing that came out to me whilst looking at the diff19:44
nuclear_eclipseok19:44
paulrif you've got a demo of it running19:44
paulrthat could be nice19:44
nuclear_eclipsesorry, not off hand19:44
nuclear_eclipseI really should set up a public test installation on my server....19:45
paulr(i'm assuming you've been sensible, so more looked at...19:45
nuclear_eclipsepaulr: I just pushed a commit to my pub branch19:46
nuclear_eclipseI gotta run to class - if you can take the time to test the feature or post more feedback as a note on the bug, that'd be excellent, and I'd greatly appreciate it19:47
paulr(i'm right about that aren't i?)19:47
nuclear_eclipsecheers19:47
paulrcya19:47
Enisseonuclear_eclipse: thank you, I do the modifications :) nice suggestions19:50
mantisbotNew bug: Bug 10299 - enisseo - open - new20:15
mantisbotNew bug: Invalid HTML - http://www.mantisbt.org/bugs/view.php?id=1029920:15
* daryn is back.20:30
Enisseo_is there a documentation about the status of an issue?20:35
Enisseo_the "feedback" status, more precisely20:36
paulryay20:57
Enisseo_Can I change the $g_status_enum_string config variable without any risk?21:03
paulrin theory you should be able to21:03
Enisseo_what does this variable mean, by the way?21:04
Enisseo_'10:new,20:feedback,30:acknowledged,40:confirmed,50:assigned,80:resolved,90:closed'21:04
Enisseoenum values, I guess21:07
Roa:>21:22
Roahey :) salut/hola/hello/21:23
Roai need help about mantis x:)21:23
nuclear_eclipse"Just Ask!"21:23
* nuclear_eclipse points to /topic21:23
Roahaha21:24
Roaok ok21:24
Roawell i have problem whit mantis, when configure my smtps (my smtp server) in  mantis21:25
Roa ##Send Email21:25
Roa        $g_phpMailer_method = 2;21:25
Roa        $g_SMTPAuth = true;21:25
Roa        $g_smtp_host  = 'mail.myhost.mx:465';21:25
Roa        $g_smtp_username = 'roa@myhost.com.mx';21:25
Roa        $g_smtp_password = 'qwerty1982';21:26
Roa        $g_smtp_connection_mode = 'ssl';21:26
Roa[B...21:26
Roathe logs in my mail server is.21:26
Roasetting up TLS connection from unknown[189.149.186.60]21:27
RoaApr  7 13:06:04 159395-web1 postfix/smtpd[99433]: SSL_accept error from unknown[189.149.186.60]: -121:27
Roa..21:27
nuclear_eclipsethe port number should not be be included in the host address21:27
nuclear_eclipsesee $g_smtp_port21:28
Roaok21:29
Roaim triying21:30
paulrnuclear_eclipse: wrote a test for ?><>?21:31
Enisseohow can I define a new milestone?21:31
nuclear_eclipseEnisseo: they're versions, and have to be defined when managing a project21:32
Roamm nothing21:32
Roa:S21:32
Roawell they re proble whith phpmailer?21:33
nuclear_eclipseRoa: what version of Mantis are you using?21:33
Roamantis]# pkg_info  | grep mantis21:33
Roamantis-1.1.6        A bug tracking system written in PHP.21:33
Enisseonuclear_eclipse: I've defined a Version (manage/projects/myproject/versions/add version, but it does not appear in the roadmap section :S21:33
slestakanyone using soap + mantis + lighttpd?21:34
nuclear_eclipsehmm, I just realized that I was thinking of 1.2.x; 1.1.x doesn't yet have $g_smtp_port, and works with a much older version of phpmailer....21:34
nuclear_eclipseEnisseo: you need to update bugs to set their target version before anything ever shows up on the roadmap21:34
Roa:>21:34
Roawell.. :S21:35
nuclear_eclipseRoa: also try downloading and installing Mantis directly from us21:35
Road'accord21:35
Enisseonuclear_eclipse: yes, thanks :)21:36
nuclear_eclipsepaulr: poke21:36
Roathe last version  nuclear_eclipse ?21:36
RoaDownload MantisBT 1.2.0a321:36
nuclear_eclipse1.1.6 is the latest official, stable release21:37
paulrnuclear_eclipse: *SIGH*21:37
nuclear_eclipsehi paulr21:37
paulrlo21:37
nuclear_eclipsewhat you sighing about now?21:37
nuclear_eclipseand have you been able to test revisions yet?21:38
paulr<?php print_manage_user_sort_link(  'manage_user_page.php', lang_get( 'username' ), 'username', $c_dir, $c_sort, $c_hide, $c_filter ) ?>21:38
paulr<?php print_sort_icon( $c_dir, $c_sort, 'username' ) ?>21:38
paulrno21:38
paulri've not21:38
paulrwent to play wc321:38
paulrand only just came back21:38
nuclear_eclipsewc3 == warcraft III ?21:39
paulryes21:42
nuclear_eclipseyou play WoW at all?21:43
paulrnope21:43
* nuclear_eclipse plays WoW on a private server w/ no monthly fee21:43
paulrmm21:49
paulrserver seems to have stopped logging errors :)21:49
Roanegative  :>..21:49
Roanuclear_eclipse:21:49
Roa setting up TLS connection from unknown[189.149.186.60]21:50
RoaApr  7 16:47:14 159395-web1 postfix/smtpd[5194]: SSL_accept error from unknown[189.149.186.60]: -121:50
nuclear_eclipseRoa: I guess my only other suggestion is to try the 1.2.x development releases, which use a newer version of PHPMailer... =\21:50
Roaok21:52
paulrnuclear_eclipse: what other rules do we want to enforce?21:53
paulrshall I just work my way through coding standards doc on mantsibt.org21:53
nuclear_eclipsepaulr: after you test 9856 ;)21:54
paulrcan i merge it into my thingie?21:55
paulrthen just sign it off21:55
nuclear_eclipsewhat do you mean?21:55
nuclear_eclipseeg, you want to push 9856 yourself?21:55
paulror more21:55
paulrI cba to reconfigure IIS ;p21:56
Enisseobye, thanks for your help21:56
paulrso if i could import a patchset21:56
paulrin hg you can send changesets about21:56
paulr(did I mention hg before I forget)21:56
nuclear_eclipseoh, use `git format-patch master revisions` iirc to generate patch sets, and then use git-am to import them to a diff repo21:57
paulrcould you generate the file from yours then?21:57
paulrthat should keep history + ownership information right?21:57
nuclear_eclipseyeah21:57
nuclear_eclipsejust a sec21:57
nuclear_eclipsepaulr: http://mantis.pastebin.com/m3259071521:59
nuclear_eclipsesave that as a file, then do `git-am <file>`21:59
paulrstill there john?22:16
nuclear_eclipseyep22:16
paulrlets see if this works then22:17
paulr;022:17
CIA-18Mantisbt: paul * rfbae2c2ea58a / (31 files in 2 dirs): Empty ?> <?php block can be removed.22:17
paulr$ git-am m32590715.txt22:17
paulrsh.exe": git-am: command not found22:17
paulr$ git am m32590715.txt22:18
paulrcat: c:/mantis/git/mantisbt/.git/rebase-apply/next: No such file or directory22:18
paulrprevious rebase directory c:/mantis/git/mantisbt/.git/rebase-apply still exists22:18
paulrbut mbox given.22:18
nuclear_eclipse?!22:18
nuclear_eclipse`git version`22:18
Roanuclear_eclipse: i think the bug is in the phpmailer22:18
RoaxD22:18
paulr1.6.2.msysgit22:18
nuclear_eclipsehmm22:18
nuclear_eclipsepaulr: try doing `git am --abort` or `git rebase --abort`22:19
paulrfatal: git apply: bad git-diff - expected /dev/null on line 3522:21
paulrPatch failed at 0001 Base API and schema changes for tracking revisions to bug t22:21
paulrextarea fields.22:21
paulrhmm22:21
nuclear_eclipseI think your repo is borked?22:23
paulrnah22:23
paulrwindows22:24
nuclear_eclipseset up a vhost on your mantisforge.org server22:24
mantisbotNew bug: Bug 10300 - junior - open - new22:25
mantisbotNew bug: Error in config_defaults_inc.php - http://www.mantisbt.org/bugs/view.php?id=1030022:25
paulrwait, can't i just pull the changes from the reivsions head thing22:27
nuclear_eclipseshould be able to22:27
paulrApplying: Base API and schema changes for tracking revisions to bug textarea fi22:31
paulrlds.22:31
paulrthere we go22:31
paulrthe timestamps are wrong22:38
paulrshould you be able to delete all revisions?22:45
nuclear_eclipseyes, you should be able to delete, but how are the timestamps wrong?22:49
nuclear_eclipsepaulr: *that's* why I was adding revisions in bug_create, so that they had the timestamp when the bug was created, not the timestamp of its first edit22:51
paulrshould bug_revision_last22:53
paulrreturn22:53
paulrnull22:53
paulror22:53
paulrarray()22:53
paulrwhen empty22:53
nuclear_eclipseI don't remember22:53
paulri know what you do i mean22:54
nuclear_eclipsenull means there was no last revision22:54
paulrdoes it make more sense to always return an array22:54
paulrand distinguish no revision with array()22:54
paulror to return two different types22:54
nuclear_eclipsesince the only thing being returned is a row, an empty array would be non-sensical IMO22:54
paulrnuclear_eclipse: http://mantis.pastebin.com/m4ab14a2 are my patches22:58
nuclear_eclipsewhat's the change to filtor_api for?23:06
paulrfix what i broke with my previous commit ;p23:11
nuclear_eclipsepaulr: also, what's the point of putting teh parameter name in the phpdocs?23:11
paulrI thought you needed to23:11
paulrwe do elsewhere23:12
nuclear_eclipse*you* do elsewhere ;)23:12
paulrhmm23:12
nuclear_eclipseI could be wrong, but I don't think you need it; the order of the @params should just match the order of the function arguments23:12
paulrhttp://manual.phpdoc.org/HTMLframesConverter/default/phpDocumentor/tutorial_sample2.pkg.html23:12
paulrthe example there does23:12
paulr@param datatype $paramname description23:13
paulr@param datatype1|datatype2 $paramname description23:13
nuclear_eclipsehmm, well I think that's really stupid and redundant.... =\23:13
* paulr nods23:13
paulrit's fucking annoying23:13
nuclear_eclipsehave you tried generating phpdocs without the param names there?23:14
paulriirc, yes23:14
paulranyway, I added a (optional)timestamp to revision add23:14
paulrwhich I guess could deal with the problem of reivisons23:15
paulrhaving the wrong time23:15
* paulr pokes nuclear_eclipse23:19
nuclear_eclipseyeah23:20
paulrshall i commit my changes (timestamp+phpdoc) and push to trunk ?23:20
nuclear_eclipseI kind of prefer just inserting them directly at bug_create though...23:20
nuclear_eclipseso I assume you didn't find any problems with it in testing?23:22
paulronly my comment of shouly you be able to delete last revision23:25
paulror the currently active reivision etc23:25
nuclear_eclipsehmm23:25
paulr[00:20:29] <nuclear_eclipse> I kind of prefer just inserting them directly at bug_create though...23:25
paulrmy point is well23:25
paulrit's a hack ;p23:25
nuclear_eclipsewhy's that?23:25
paulrif you insert into 'current text table' + 'history table' every bug23:25
paulrwe effectively double the size23:26
nuclear_eclipsedisk space is cheap ;)23:26
paulrmy expectation is v. few bugs change history23:26
nuclear_eclipsetrue23:26
paulrand you've been running this stuff at work haven't you for 6 month?23:27
nuclear_eclipseyep23:27
nuclear_eclipsegranted, we don't yet officially use Mantis as our primary tracker, but it's been on our test installation23:27
paulrbug 985623:28
mantisbotBug 9856 - jreese - open - assigned23:28
mantisbotImplement method for tracking changes to Description, Steps to Reproduce, and Additional Info - http://www.mantisbt.org/bugs/view.php?id=985623:28
nuclear_eclipsepaulr: I'm going to push with your patch to phpdoc/API23:37
paulrok - will it rebase fine at this end /23:38
nuclear_eclipsedunno23:38
paulrheh23:38
paulratm. I have master, base api -john, then my filter api commit23:38
paulr:)23:38
nuclear_eclipseif anything, you can always cherry-pick the filter api commit onto the new master23:39
paulrthere's still some stuff in 9856 we probably need to go through23:39
paulr-html_page_bottom1( __FILE__ );23:39
paulr-23:39
paulr+html_page_bottom( __FILE__ );23:39
paulr-html_page_top1( bug_format_summary( $t_bug_id, SUMMARY_CAPTION ) );23:39
paulr-html_page_top2();23:39
paulr+html_page_top( bug_format_summary( $t_bug_id, SUMMARY_CAPTION ) );23:39
paulrdid you also do thse?23:40
nuclear_eclipseright, should I wait to push then?23:40
paulri'd say no23:40
paulrwhat i mean is23:40
paulrif you look through23:40
nuclear_eclipseI took everything in the patch you pasted, except for the filter_api changes23:40
paulrthe html_page stuff wasn't in the stuff i pasted :)23:40
nuclear_eclipsethen I don't have that23:40
paulr6. Not directly related to this change, but the format of the issue note number used for added/edited is different from the one used in Make Private/Public in the Issue History box.23:41
paulrfrom victor23:41
paulrerm23:41
Roahey nuclear_eclipse is a bug23:41
Roaof phpmailer23:41
Roahttp://webomania.wordpress.com/2007/06/12/getting-phpmailer-to-connect-to-gmails-smtp-server/23:41
nuclear_eclipseRoa: ok23:41
Roa        $g_phpMailer_method = 2;23:42
Roa        $g_SMTPAuth = true;23:42
Roa        $g_smtp_host  = 'ssl://mail.myhost.com:465';23:42
Roa        $g_smtp_username = 'roa@myhost.com';23:42
Roa        $g_smtp_password = 'qwerty';23:42
Roax)23:42
paulrnuclear_eclipse: but unless someone'sreally saying no23:42
paulri'd say push23:42
nuclear_eclipseok23:42
paulrpersonally, think it makes some sense23:42
paulrwe should store changes for reasons you say (malicious)23:42
paulrbug_text/history aren't in the format for that23:43
nuclear_eclipseok, so I'm go for pushing what I have?23:44
nuclear_eclipseI can always let you sort out the nitpicks afterwards, eh? :)23:45
paulrI guess23:45
paulrit's victor/glenn that raised points23:45
paulrbut most of them seemed to be regarding bug_Text<>revision<>history23:45
nuclear_eclipseI meant your nits about page_top|bottom etc23:45
paulrand we are never going to sort *that* mess out23:45
paulr;)23:45
paulrI mean, I wouldn't really want to tell users to run an upgrade script that moved all data out of bug_text into new_bug_text23:46
nuclear_eclipseyeah23:46
paulrwe'd have to do something like23:47
paulrdo one item at a time23:47
nuclear_eclipse*cough* categories! *cough*23:47
nuclear_eclipse:)23:47
paulradd to new table,23:47
paulrfetch from new table23:47
paulrwipe from old table23:47
paulrrepeat for each of 3 columsn23:47
paulrfor 10,000 bugs23:47
nuclear_eclipseyeah23:47
paulrand probably make a user click the mouse every 500 bugs to refresh the page so it does't time out23:48
paulr:)23:48
nuclear_eclipsehehe23:48
nuclear_eclipseok, pushing23:48
nuclear_eclipsewait23:49
nuclear_eclipsegonna do a rebase to add #9856 to the commit messages23:49
nuclear_eclipseI just squashed all the commits together in the end :P23:53
CIA-18Mantisbt: jreese * r678d8a168136 / (12 files in 4 dirs): Issue #9856: Implemented revision history for long textfields and bugnotes.23:53
nuclear_eclipseit's nice having CIA back in the loop :)23:55
paulrnn23:57
paulrthink this just broke my repo though ;/23:57
nuclear_eclipsepaulr: that's why you're supposed to work in disposable branches ;)23:59
paulrnn23:59
nuclear_eclipsegood night23:59

Generated by irclog2html.py