Wednesday, 2009-04-22

../irclogs/#mantishelp.2009-04-22.log
--- scribe started ---00:00
truonglong88hi02:54
truonglong88plz tell me the address of bug_page_view02:55
dhx_mhi02:57
dhx_m?02:57
mantisbotNew bug: Bug 10353 - ispolin - open - new03:44
mantisbotNew bug: ??? - http://www.mantisbt.org/bugs/view.php?id=1035303:44
mantisbotNew bug: Bug 10354 - karthisudar - open - new07:00
mantisbotNew bug: test - http://www.mantisbt.org/bugs/view.php?id=1035407:00
mantisbotNew bug: Bug 10355 - karthisudar - open - new07:05
mantisbotNew bug: test - http://www.mantisbt.org/bugs/view.php?id=1035507:05
mantisbotNew bug: Bug 10356 - karthisudar - open - new07:05
mantisbotNew bug: test - http://www.mantisbt.org/bugs/view.php?id=1035607:05
mantisbotNew bug: Bug 10357 - karthisudar - open - new07:05
mantisbotNew bug: test - http://www.mantisbt.org/bugs/view.php?id=1035707:05
dhx_moh dear :)07:07
giallu__grrrr07:18
[KK]KirillJohn, can you update git://git.mantisforge.org/mantisbt/jreese.git to last trank07:31
[KK]Kirill?07:32
dhx_mfor the strings branch?07:40
[KK]Kirillyes07:41
dhx_mis there a merge conflict?07:42
[KK]Kirillnot.. to much diff07:43
dhx_mI ask because I haven't merged stuff before07:44
dhx_mI would have thought that would have merged easily07:44
dhx_mgiven it is 2 patches that don't look like they'd be affected by the latest master head07:44
[KK]Kirillcan you upload patch for last master?07:48
dhx_mI thought with git you normally don't have to merge the master into branches?07:49
[KK]Kirillgood question07:49
[KK]KirillI don't think yet07:49
[KK]Kirillone sec07:50
mantisbotNew bug: Bug 10358 - karthisudar - open - new07:50
mantisbotNew bug: test - http://www.mantisbt.org/bugs/view.php?id=1035807:50
dhx_mis he trying to count to 200 or something?07:51
[KK]KirillBan him07:52
dhx_malso you probably want to be using "git rebase master" in your working branch07:52
dhx_mrather than the merge command07:52
dhx_m... if you need to have the latest master commits in your branch07:53
[KK]KirillI press git commit -a07:53
[KK]Kirillbut not update07:53
[KK]Kirilldiscard changes?07:53
dhx_mfor a lot of small fixes that aren't affected by changes in master since you started the branch, you shouldn't need to merge master into your branch, or rebase your branch07:53
dhx_mso you've merged the strings branch into master?07:54
dhx_mbut haven't pushed yet?07:54
[KK]Kirillnot07:54
[KK]KirillI merge master to strings branch07:54
dhx_mah07:54
dhx_myep discard07:54
[KK]KirillI try coomit strings branch07:54
[KK]Kirillok07:54
dhx_mthen checkout the master07:55
dhx_mand type 'git pull . strings'07:55
dhx_mthat should merge the strings branch into master07:55
dhx_mit should tell you if the automatic merging didn't work07:56
[KK]Kirillone sec07:57
[KK]KirillI commit master to strings :(07:57
dhx_mone of the main benefits of git is that merging branches is 100000000000000000 times easier than with SVN or similar07:57
dhx_mno problem (I think)07:57
[KK]Kirillhow I can revert last commit?07:57
dhx_m"git log" to get the previous commit id (sha1 hash)07:58
dhx_mthen you can "git checkout ID" where ID is the ID of the commit you want to go back to07:58
dhx_mwait maybe this isn't the best way07:58
[KK]KirillIt's normal ?07:59
[KK]Kirill$ git checkout e50944ce028886db539855a0d7d0acbf5dc3a94b07:59
[KK]KirillNote: moving to "e50944ce028886db539855a0d7d0acbf5dc3a94b" which isn't a local branch07:59
[KK]KirillIf you want to create a new branch from this checkout, you may do so07:59
[KK]Kirill(now or later) by using -b with the checkout command again. Example:07:59
[KK]Kirillgit checkout -b <new_branch_name>07:59
[KK]KirillHEAD is now at e50944c... Perf: allow passing in charset to avoid calling lang/config_get in loops. Perf: Replace string_display_line with call to string_html_s07:59
[KK]Kirillpecialchars( string_strip_hrefs in print_project_option list to avoid calling plugin/config loops07:59
dhx_mah07:59
dhx_mdo this:07:59
dhx_mgit reset --hard SHA1_HASH07:59
[KK]KirillHEAD is now at e50944c Perf: allow passing in charset to avoid calling lang/config_get in loops.08:01
dhx_mit'll go back in time and make your working copy exactly the same as what it was like back at revision SHA1_HASH08:01
dhx_mok08:01
dhx_mthat was after the git reset command?08:01
[KK]KirillHEAD is now at e50944c Perf: allow passing in charset to avoid calling lang/config_get in loops.08:02
dhx_msounds OK to me08:02
dhx_mnow you can git pull . strings08:03
[KK]Kirillin msaster branch ?08:04
[KK]Kirillmaster*08:04
dhx_myep08:04
dhx_massuming you have done a git reset08:04
[KK]Kirillcan I copy this branch08:04
dhx_mthe master? or strings?08:04
[KK]KirillI don't want change master branch08:04
[KK]Kirillmaster08:04
dhx_mah ok, just create a new branch from master08:04
dhx_mand merge strings into that branch (I think)08:05
dhx_mbut your master should show "nothing to do" for "git status"08:05
dhx_malso btw don't use git reset if you have already pushed your commits to someone else08:06
dhx_mbecause then they won't be able to trace those commits back to your repo (in your repo, those commits are lost for good... no record)08:06
dhx_mhttp://www.kernel.org/pub/software/scm/git/docs/git-reset.html explains it better08:07
[KK]Kirill$ git pull .strings08:08
[KK]Kirillfatal: '.strings' does not appear to be a git repository08:08
[KK]Kirillfatal: The remote end hung up unexpectedly08:08
dhx_mnah, space needed after the period ;)08:08
dhx_myou're saying "pull the strings branch into my current working branch"08:08
dhx_mso git pull {destination} {source}08:09
[KK]Kirillah.. sorry08:09
[KK]Kirill$ git pull . strings08:09
[KK]KirillFrom .08:09
[KK]Kirill* branch            strings    -> FETCH_HEAD08:09
[KK]KirillAlready up-to-date.08:09
dhx_mdid you "git checkout {branch you want to merge to}"?08:09
[KK]KirillYes. I in branch mastertest08:10
[KK]Kirillwitch I create in master by command git branch mastertest08:10
mantisbotNew bug: Bug 10359 - karthisudar - open - new08:10
mantisbotNew bug: test - http://www.mantisbt.org/bugs/view.php?id=1035908:10
dhx_mdid you use "checkout" after the "branch" command08:11
dhx_mbecause "branch" does not automatically checkout the new branch08:11
dhx_mfrom the error you got, it sounds like you were in the strings branch08:12
[KK]KirillYes. I checkout to this branch08:14
dhx_mhmmm I'll try it myself to see what the problem could be08:14
dhx_mhmmm well I can merge OK with the procedure I outlined08:15
[KK]Kirillone sec08:15
dhx_mmerging my 10330 branch into a new branch I created from the current master08:15
dhx_mI did:08:15
dhx_mgit checkout master08:15
dhx_mgit pull08:15
dhx_mgit branch testmerge08:16
dhx_mgit checkout testmerge08:16
dhx_mgit pull . 1033008:16
dhx_mand that worked OK for me08:16
[KK]Kirill$ git pull . strings08:19
[KK]KirillFrom .08:19
[KK]Kirill* branch            strings    -> FETCH_HEAD08:19
[KK]KirillAuto-merging core/string_api.php08:19
[KK]KirillMerge made by recursive.08:19
[KK]Kirillcore/string_api.php |   36 ++++++++++++++++++++++++++++++++++--08:19
[KK]Kirill1 files changed, 34 insertions(+), 2 deletions(-)08:19
[KK]KirillI delete all branch and create new08:19
dhx_mthat looks better :)08:20
[KK]Kirillthis I can test?08:20
mantisbotNew bug: Bug 10360 - karthisudar - open - new08:21
mantisbotNew bug: test - http://www.mantisbt.org/bugs/view.php?id=1036008:21
mantisbotNew bug: Bug 10361 - karthisudar - open - new08:21
mantisbotNew bug: test - http://www.mantisbt.org/bugs/view.php?id=1036108:21
mantisbotNew bug: Bug 10362 - karthisudar - open - new08:21
mantisbotNew bug: test - http://www.mantisbt.org/bugs/view.php?id=1036208:21
mantisbotNew bug: Bug 10363 - karthisudar - open - new08:21
mantisbotNew bug: test - http://www.mantisbt.org/bugs/view.php?id=1036308:21
mantisbotNew bug: Bug 10364 - karthisudar - open - new08:21
mantisbotNew bug: test - http://www.mantisbt.org/bugs/view.php?id=1036408:21
[KK]Kirillbun him08:21
[KK]Kirillban him08:21
dhx_myou can do: "git archive --format=tar mastertest | gzip -9 /some/path/mastertest.tar.gz" to export it08:21
dhx_moh well, it'll give some devs here a fun job to do later :)08:22
[KK]Kirill:) thanks08:23
dhx_mthen you can extract from the archive to test it08:25
dhx_mor you can send the archive to a non-git user to test out08:25
mantisbotNew bug: Bug 10365 - karthisudar - open - new08:26
mantisbotNew bug: test - http://www.mantisbt.org/bugs/view.php?id=1036508:26
mantisbotNew bug: Bug 10366 - karthisudar - open - new08:26
mantisbotNew bug: test - http://www.mantisbt.org/bugs/view.php?id=1036608:26
dhx_mwhat could you possibly need to test in 15 bug reports?08:26
[KK]KirillI can.08:28
[KK]KirillI just upload string_api.php08:32
dhx_mah, straight copy from your working directory?08:32
dhx_mto your test site?08:32
[KK]Kirillnot to work site08:32
dhx_mor did you push the whole thing to a repo?08:33
dhx_m(was wondering if there is anything to look at on mantisforge) :)08:33
[KK]KirillI has'n projects on mantisforge08:33
[KK]KirillI have support site on my work08:34
dhx_mah ok :)08:34
AndyHNuclear Eclipse, I found the option to start my bug ID's from say more than 100+, using myphpadmin fine, thanks for the advice,08:37
AndyHbut :(  ... I've got another quesion! :P08:37
AndyHHas anyone used Mantis with Microsoft SQL server? I'm having issueing with the admin username and password, as mine arn't being acepted08:38
dhx_mah common... having bug id's < 100 is just a sign that your developers are experts and never make mistakes :p08:38
AndyHi've tried formatting my username into DOMAIN\USERNAME or just USERNAME but neither work. Any ideas?08:38
dhx_m... can't help there sorry08:39
AndyHLol, no, that wasn't the intention. I was trying to start my Bug ID's FROM 1000, as we've already used another bit of software to log <100008:39
AndyHwould be really nice if we only found <100 :)08:39
AndyHAny one else used or have any ideas on using Microsoft SQL server with autenticating usernames and passwords ??? --- ???08:40
* giallu stays away from MS stuff08:44
dhx_mMantis uses AdoDB as a backend for all database access08:44
dhx_mso I assume there must be some AdoDB documentation on the topic...?08:45
dhx_mlike giallu said, most Mantis developers probably stay a long way away from MSSQL :)08:45
gialludhx_m, IIRC paulr made some fixes for MSSQL08:46
dhx_m:o so someone does use it08:46
giallubut he tends to work just on trunk08:46
dhx_mlol yeah08:46
dhx_mare you much of a git expert?08:46
gialluso if 1.1.7 is broken on MSSQL you will probably need to test 1.208:47
dhx_mI was trying to merge my 10274 branch into a copy of the current master08:47
dhx_mand had a merge conflict... which is easy to fix08:47
dhx_mand git rebase is probably not a good thing if anyone has been using my branch (it is publicly available)08:47
dhx_mhowever, I assume in most projects, you want to be able to say "please pull from ...." without the other person having to resolve conflicts08:49
dhx_mso now I have the dilemma of "do I rebase or not" :)08:49
dhx_mactually rebasing should be ok... it just means downstream people have to also rebase08:51
gialludhx_m, we already got bitten git git rebase...08:52
giallubitten by08:53
dhx_myeah it is bad if others are depending on your branch08:53
dhx_m(in my case I'm pretty sure no one has extended my branch)08:54
gialluanyway, being that _your_ own branch, feel free to decide...09:18
[KK]Kirilldhx_m: what you want  for me testing?09:19
dhx_m10330 :)09:27
dhx_mI think paul gave it a look over09:27
dhx_mand noted that there needs to be some upgrade path for configuration changes?09:27
dhx_mie. renaming some constants09:27
dhx_mnot really sure if that is possible though09:27
dhx_mgiallu: it is more a case of not wanting to make a branch that is impossible/too hard to merge into the mantisbt master09:29
dhx_mI tried merging master into my branch to resolve the conflict... see http://git.mantisforge.org/w/mantisbt/dhx.git?a=shortlog;h=refs/heads/1027409:30
dhx_mbut now it is hard to actually review what changed?09:30
mantisbotNew bug: Bug 10367 - cas - open - new09:31
mantisbotNew bug: Moving attachments while cloning issue - http://www.mantisbt.org/bugs/view.php?id=1036709:31
mantisbotNew bug: Bug 10368 - jonathh - open - new09:31
mantisbotNew bug: Overall Roadmap? - http://www.mantisbt.org/bugs/view.php?id=1036809:31
gialludhx_m, we probably need to find a smart way to warn the user the old constant names are deprecated09:34
dhx_mwhen the user tries to use a new version of Mantis against an old database09:35
dhx_mit takes them to an upgrade screen doesn't it?09:35
dhx_massuming of course that you get that far without PHP errors warning you that config_inc.php is referencing undefined constants09:35
dhx_mok it seems my merging went OK: http://git.mantisforge.org/w/mantisbt/dhx.git?a=shortlog;h=refs/heads/fyp09:48
dhx_mah.... git cherry-pick would have probably been a better choice10:04
[KK]KirillI need help11:45
[KK]Kirillwho around?11:45
[KK]Kirillpaul, you here?11:48
[KK]Kirillgiallu: dhx_m: you here?11:50
dhx_myep11:59
[KK]Kirilldhx_m: I have trouble with verify link12:04
[KK]KirillI switch off self-plugins12:05
dhx_mI'm confused with what you're describing?12:05
[KK]KirillI switch-on MantisCore Formating12:05
[KK]Kirillbut I recieve this mail12:05
[KK]Kirillhttp://localhost/bug/verify.php?id=29&amp;confirm_hash=a1496b3b227859c4da7a1c8165df6e1412:06
[KK]KirillUsername:&#32;Test12:06
[KK]KirillRemote IP address:&#32;10.0.128.16912:06
[KK]KirillDO NOT REPLY TO THIS MESSAGE12:06
[KK]Kirillhttp://localhost/bug/verify.php?id=29&confirm_hash=a1496b3b227859c4da7a1c8165df6e1412:07
[KK]KirillUsername: Test12:07
[KK]KirillRemote IP address: 10.0.128.16912:07
[KK]KirillDO NOT REPLY TO THIS MESSAGE12:07
[KK]KirillI want recieve last version12:07
[KK]Kirillbut html character not transform12:07
dhx_mI know some fixes were made to that stuff recently... might be best to ask paul?12:08
[KK]KirillI see this task, but if I understand they have error with \n -line separator12:09
nuclear_eclipsemorning all12:40
dhx_mhi12:41
[KK]Kirillhi John12:45
[KK]KirillHelp me, John12:45
[KK]Kirill:(12:47
[KK]Kirilldaryn: hi12:58
darynhello12:59
[KK]KirillJohn, bug 1036913:03
mantisbotBug 10369 - Kirill - open - new13:03
mantisbotVerification link not work - http://www.mantisbt.org/bugs/view.php?id=1036913:03
[KK]Kirillnuclear_eclipse: Thanks13:07
mantisbotNew bug: Bug 10369 - Kirill - open - assigned13:07
mantisbotNew bug: Verification link not work - http://www.mantisbt.org/bugs/view.php?id=1036913:07
mantisbotNew bug: Bug 10370 - jonathh - open - new13:47
mantisbotNew bug: Max field length not respected. - http://www.mantisbt.org/bugs/view.php?id=1037013:47
nowotnyhello... I noticed mantis supports some html tags in bugnotes but not all... is there a list what it does and does not support somewhere...?14:30
giallunowotny, yes, in config_defaults_inc.php14:32
nowotnyI mean... from the user's point of view...14:32
gialluyou can select what to allow and what not14:32
gialluthe admin choose what's allowed14:33
nowotnybut there are some defaults, right...?14:33
gialluright, in config_defaults_inc.php14:33
paul_lo17:36
nuclear_eclipsehowdy17:36
paul_howdy17:37
paul_i'm surprised you say hi to me :)17:39
nuclear_eclipsewell, I am rather bored :P17:39
paul_oh?17:39
nuclear_eclipsebored != not busy...17:40
paul_wanna discuss subprojects? :)17:40
nuclear_eclipseperhaps17:41
paul_we seem to do big unncessary loops17:41
nuclear_eclipseprobably17:41
paul_I started looking at caching17:42
nuclear_eclipseI remember being put off by the general level of complexity/redundancy when I was working near it, but didn't feel like touching it more than I already had to17:42
paul_but decided all i'm doing there is caching big loops17:42
paul_I want to improve projects for atif ;P17:42
paul_although17:42
paul_he seems to have gone quiet and disappeared17:43
* daryn_away is away: Gone away for now17:46
mantisbotNew bug: Bug 10371 - kec161 - open - new18:24
mantisbotNew bug: Added ability for reminders to come from the user who sent them rather than the from_email - http://www.mantisbt.org/bugs/view.php?id=1037118:24
EduardRGreetings - checked docs and forums and can't tell whether it's ok to use Postgres (marked experimental). Does it work fine or stick with MySQL for now? Thanks.18:40
paul_mysql is less hassle18:41
CIA-11Mantisbt: jreese * rbf6e5ef1cb08 / (config_defaults_inc.php core/print_api.php): Support attached text files with no file extension.18:42
paul_jreese: we shouldn't expand files with no extension by default18:42
paul_nuclear_eclipse:  even18:43
nuclear_eclipsewhy not?  I've never seen a binary file with no extension, but *lot's* of people use text files with no extension, and it would greatly simplify previewing of patch files on the tracker...18:44
* nuclear_eclipse points at bug 1037118:44
mantisbotBug 10371 - kec161 - open - new18:44
mantisbotAdded ability for reminders to come from the user who sent them rather than the from_email - http://www.mantisbt.org/bugs/view.php?id=1037118:44
* paul_ nods18:45
nuclear_eclipsetwo patches attached, neither with an extension, and I've seen that happen many many times18:45
paul_but at the same time18:45
nuclear_eclipseand I hate having to save the files to my system just to glance at them...18:45
paul_if I do git bundle > bundle18:45
paul_and upload that18:45
paul_you've got a binary file that's gonna get output to html18:46
nuclear_eclipseytf would you upload a bundle rather than a set of formatted patches?18:46
paul_tis an example18:46
nuclear_eclipseand how is that any different than a user renaming an image file to .txt and uploading that?18:46
paul_true18:46
paul_but with .txt you know what you are expecting by default :)18:46
nuclear_eclipseeither way, you get a whole bunch of gibberish, and at least you can preview it and realize that it's not a text file...18:47
paul_true18:47
nuclear_eclipseit's not a security risk to preview a binary file, just ugly18:47
paul_we seem to be starting to get problem reports about plugins into tracker ;/18:48
nuclear_eclipseI know...18:48
nuclear_eclipseand they aren't even plugins we use...18:48
nuclear_eclipseI could understand getting reports about something like the twitter or source integration plugins, but for something that's obviously barely even supported by the author, I'm not sure why people would ever post bugs against them on our tracker...18:49
mantisbotNew bug: Bug 10372 - kec161 - open - new18:49
mantisbotNew bug: Don't allow reminders to be sent if the user doesn't have an email address specificed - http://www.mantisbt.org/bugs/view.php?id=1037218:49
nuclear_eclipseuh, do we even allow users to not specify an email address?18:50
paul_:)18:50
paul_hmm18:50
paul_lets ses18:50
nuclear_eclipseand yth would you want to throw an error because the user's email isn't set?18:51
nuclear_eclipseit would be far better to just silently ignore the problem and continue sending emails...18:51
paul_anyway18:51
paul_ignore those18:51
nuclear_eclipseyep18:51
paul_we need to think about projects18:51
nuclear_eclipsethink about it quickly, I need to leave for class in a little bit18:52
paul_we have:18:53
paul_project_hierarchy_cache -> which seems to select from project+hierachy table, and store a cache of project_hierarchy + project_inheritance id's18:53
paul_we have project_hierarchy_get_all_subprojects which seems to get nested subprojects all the way down18:54
paul_and project_hierarchy_get_subprojects which gets first level of subprojects?18:54
nuclear_eclipsethat sounds about right18:55
paul_then we have18:55
paul_user_get_accessible_projects, which stores a cache of accessible projects and either:18:55
paul_a) if someone has private_project_threshold globally, calls project_hierarchy_get_subprojects18:56
paul_(shouldn't that be _get_all?)18:56
* paul_ shrugs18:56
paul_or18:56
paul_b) does it's own select of hierachy+project+project_user_list table18:56
nuclear_eclipseI think the subproject aggregating is a total mess at the moment, and should be taken out back and beaten, and then replaced with something simpler and clearer18:57
paul_but looking for PUBLIC || PRIVATE+user is on project_user_list table18:57
nuclear_eclipsehowever, seeing that it actually works right now, I'd rather focus attention elsewhere18:58
nuclear_eclipsequite frankly, I feel that Atif is most likely insane for devising a 30,000 project hierarchy....18:58
paul_yea18:59
paul_at the same time18:59
paul_what we do atm is a complete mess18:59
nuclear_eclipseagreed, but it still works, which is more than we can say for some other pieces of the app... :P18:59
paul_function user_get_all_accessible_projects( $p_user_id, $p_project_id )18:59
paul_unction user_get_all_accessible_subprojects( $p_user_id, $p_project_id )18:59
paul_function user_get_accessible_subprojects( $p_user_id, $p_project_id, $p_show_disabled = false )19:00
paul_function user_get_accessible_projects( $p_user_id, $p_show_disabled = false )19:00
paul_it's just confusing :)19:00
paul_we could fix 'atif's issue quite easily:19:00
paul_by loading hierarchy table on it's own19:01
thraxispThe whole implementation of project based rights has some big holes anyways.19:05
thraxisppaul_: The functions you listed are just variants of each other.19:06
nuclear_eclipsedhx_m: there?19:06
paul_thraxisp: with various levels of caching19:06
paul_(different caching)19:06
paul_we cache the hierachy19:07
paul_but then dont use it ;/19:07
* paul_ thinks he might try to reorganise it ;/19:10
* nuclear_eclipse is off to class19:19
nuclear_eclipsecheers all19:19
thraxispDoesn't MySQL do caching anyways? Most of our caching seems to be "read the table, and avoid queries".19:20
paul_well, we dont just support mysql but:19:21
paul_we try not to read the same data >119:21
paul_it's likely a db query is going to be slower then doing some maths in php19:21
w00terHow do i put the options for status_enum_string in the array field of Configuration Report (which on file looks like; $s_status_enum_string = '10:Gelogd,20:Geprioriteerd,30:Impact Analyse,40:Bouw,50:SRT Test,60:GAT,80:Op Productie,90:Gesloten'; )19:55
dhx_mhi nuclear_eclipse22:49
dhx_mthe import/export plugin throws up this error:23:26
dhx_mAPPLICATION WARNING #300: String "manage_import_issues_link" not found.23:26
dhx_mplugin.php?page=XmlImportExport%2Fimport23:26

Generated by irclog2html.py