| ../irclogs/#mantishelp.2009-12-21.log | ||
| --- scribe started --- | 00:00 | |
| nexact_ | hey all, is there any existing theme for mantis ? | 00:17 |
|---|---|---|
| dhx_m | hey | 00:21 |
| dhx_m | well you can play with CSS to change colours ,etc | 00:21 |
| dhx_m | someone made a theme a little while ago (search for bugs related to that topic and you should find it) | 00:21 |
| dhx_m | but it is still quite similar to the default Mantis look | 00:21 |
| dhx_m | and isn't very polished IMO | 00:22 |
| nexact_ | mkay, thanks ;} | 00:22 |
| nexact_ | is it possible to modify and/or remove some criteria when creating a new report ? | 00:22 |
| dhx_m | nexact_: yes, in 1.2.x | 00:23 |
| nexact_ | ex: in my context, its not necessary to specify the Platform so i'd like to remove it | 00:23 |
| nexact_ | okay, great, is it something complex ? | 00:23 |
| dhx_m | nexact_: check config_defaults_inc.php for options to disable those sorts of fields from being used | 00:23 |
| dhx_m | nexact_: nah, it's simple on/off sort of stuff... documented in config_defaults_inc.php | 00:24 |
| dhx_m | but only for 1.2.x, not 1.1.x | 00:24 |
| nexact_ | awesome | 00:25 |
| nexact_ | another question, i think the answer is yes but, is it possible to automatically send an email with ticket info to the person i've assigned when i created my ticket ? | 00:27 |
| dhx_m | yeah there is a notification setting in config_defaults_inc.php | 00:38 |
| dhx_m | hmmm actually maybe not | 00:38 |
| dhx_m | you'd have to send it to all users of a particular access level or higher afaik | 00:38 |
| nexact_ | mkay | 00:45 |
| nexact_ | mm, i've modified $g_project_status_enum_string to have different status but it's not reflecting any changes in http://localhost/manage_proj_create_page.php | 00:48 |
| dhx_m | you tried adding a new status? | 00:53 |
| dhx_m | there is more to it than changing just that setting | 00:53 |
| dhx_m | check the user docs | 00:53 |
| nexact_ | dhx_m, i think there's a bug... | 02:09 |
| nexact_ | k nvm | 02:10 |
| dhx_m | ok | 02:11 |
| kirillka | Hi all | 07:49 |
| kirillka | paul__: around? | 07:49 |
| kirillka | giallu_: mo | 07:54 |
| kirillka | giallu_: did you receive my mail in mantisbt-dev maillist? | 07:55 |
| giallu_ | kirillka, about LDP? | 08:01 |
| giallu_ | LDAP | 08:01 |
| kirillka | giallu: yes. Ok. I don't receive. | 08:02 |
| giallu | when you have doubts, check archives :) | 08:03 |
| giallu | so, another great commit... | 08:03 |
| dhx_m | why \x00 and \00.... | 08:32 |
| kirillka | dhx_m: lo | 08:32 |
| dhx_m | kirillka: hey | 08:32 |
| kirillka | dhx_m: I don't kow :) | 08:32 |
| kirillka | know | 08:32 |
| dhx_m | ohh | 08:33 |
| dhx_m | LDAP must treat \12 as 0x12 | 08:33 |
| dhx_m | so we're converting from say 0x00 to \00 for LDAP, as it doesn't accept the real null character | 08:33 |
| dhx_m | mind you, I don't use LDAP so don't know much about it/don't care :p | 08:34 |
| CIA-27 | Mantisbt: s.mazeland * r1b21ad18f278 / (37 files in 2 dirs): Localisation update from translatewiki.net (2009-12-21) | 08:54 |
| paul__ | dhx_m: it should be $p_string not $t_string | 08:55 |
| paul__ | anyway | 08:55 |
| dhx_m | oh yeah | 08:56 |
| paul__ | I spotted that ~4 minutes after the commit | 08:56 |
| paul__ | on saturday | 08:56 |
| paul__ | or friday or whenver | 08:57 |
| paul__ | you know we were talking about columns the other day? | 08:58 |
| dhx_m | yep | 08:58 |
| paul__ | I also spotted our current column implementation for editing is somewhat broken | 08:58 |
| paul__ | we call config_Get with project/user params the wrong way around | 08:59 |
| paul__ | In LDAP filters these 5 characters should be escaped with the backslash escape character, following by the two digit ASCII hex representation of the character. The following table documents this: | 09:01 |
| paul__ | * \2A | 09:01 |
| paul__ | ( \28 | 09:02 |
| paul__ | first hit on google for 'ldap escape' | 09:02 |
| paul__ | ) \29 | 09:02 |
| paul__ | \ \5C | 09:02 |
| paul__ | NUL \00 | 09:02 |
| CIA-27 | Mantisbt: hickseydr * r4491908ade49 /core/ldap_api.php: Fix typo in ldap_escape_string that broke LDAP auth | 09:03 |
| * paul__ hopes git is good with merges ;p | 09:03 | |
| dhx_m | git :) | 09:06 |
| CIA-27 | Mantisbt: paul master-1.2.x * ref44819f5a10 /core/ldap_api.php: Update ldap_escape_string function to escape / and \x00 | 09:06 |
| dhx_m | nice commit :p | 09:06 |
| paul__ | still not that sure what to do about ldap api ;/ | 09:10 |
| dhx_m | neither ;) | 09:13 |
| CIA-27 | Mantisbt: hickseydr master-1.2.x * r39ea07ffe49b /library/ (317 files in 41 dirs): Revert "Update libraries to match those of 1.3.x branch" | 09:13 |
| CIA-27 | Mantisbt: hickseydr master-1.2.x * r0cf2a36fad03 /library/ (244 files in 32 dirs): Resync libraries to versions used in the 1.3.x branch | 09:13 |
| paul__ | there's 2 reasons I can't use it at work in it's current state (ignoring my own patches/additions that should be merged back in) | 09:14 |
| paul__ | 1) it stores md5 hash of password | 09:14 |
| paul__ | 2) it auto creates users | 09:14 |
| paul__ | and there's no config option for either | 09:15 |
| paul__ | and I don't particularly want to add a config option as the config option would be deprecated as soon as we make auth plugins | 09:15 |
| dhx_m | well there has been some discussion on the ML/wiki about a new group/auth system | 09:19 |
| paul__ | groups are kinda seperate | 09:22 |
| paul__ | to auth plugins | 09:22 |
| paul__ | iirc, even you said this | 09:22 |
| dhx_m | well the problem was that if you want groups and LDAP at once | 09:22 |
| dhx_m | but yeah, that's more access control I would have thought | 09:23 |
| CIA-27 | Mantisbt: jreese master-1.2.x * r7fdb964b2b3b /core/filter_api.php: Fix edge cases with search terms | 15:23 |
| CIA-27 | Mantisbt: jreese * rbd9fa3cff991 /core/filter_api.php: Fix edge cases with search terms | 15:23 |
| paul__ | nice | 15:34 |
| paul__ | nuclear_eclipse: dhx sync'd libs this morning | 15:35 |
| paul__ | I need to learn how to use git more ;/ | 15:35 |
| nuclear_eclipse | I noticed | 15:35 |
| paul__ | it was giving me strange errors when i tried to cherry pick them | 15:35 |
| nuclear_eclipse | probably was giving conflicts | 15:35 |
| nuclear_eclipse | cherry-pick in git is just a simple "try to apply commit as patch on top of existing code" | 15:36 |
| paul__ | right | 15:36 |
| paul__ | which should have worked | 15:36 |
| paul__ | as I tried to pick the post-revert patches of the libs | 15:36 |
| nuclear_eclipse | ah | 15:37 |
| nuclear_eclipse | who knows | 15:37 |
| paul__ | so in theory should have been the original | 15:37 |
| paul__ | anyway | 15:37 |
| paul__ | pls read through the ldap discussion if you haven't | 15:37 |
| nuclear_eclipse | I did | 15:38 |
| paul__ | initial thoughts? would you agree we've got the current (new) behaviour slightly wrong? | 15:39 |
| nuclear_eclipse | I have no clue -- ldap is really beyond most of my understanding, and I tend to not have the time/energy to figure it all out | 15:39 |
| paul__ | the way I see it is... | 15:40 |
| paul__ | most popular ldap is probably MS Active Directory | 15:40 |
| paul__ | you'd have a domain with 2-3+ domain controllers | 15:40 |
| paul__ | in theory, only time ldap auth should fail is if 3 dc's are down | 15:41 |
| nuclear_eclipse | right, we should def support AD in some format, be it as a separate selection from ldap, or as a config option to the current ldap support | 15:41 |
| paul__ | if your domain controllers are all down | 15:41 |
| nuclear_eclipse | you're pretty hosed anyways :P | 15:41 |
| paul__ | you'd probably not have dns/ad etc | 15:41 |
| paul__ | so we dont really need an md5 pass fallback | 15:41 |
| paul__ | therefore storing md5 fallback ==> | 15:41 |
| nuclear_eclipse | that I could pretty much agree with | 15:41 |
| paul__ | a) potential security risk as the md5 hash might be less secure then how AD stores data | 15:42 |
| paul__ | b) possible redundancy/desync issues | 15:42 |
| nuclear_eclipse | depends on whether AD uses NTLANMAN :P | 15:42 |
| paul__ | I'm also then thinking: | 15:42 |
| paul__ | if we consider plugins' | 15:42 |
| paul__ | you might want a situation where you have | 15:42 |
| paul__ | MD5 for some users, LDAP for others | 15:42 |
| paul__ | i.e. 'try ldap auth, then try md5' | 15:43 |
| paul__ | and potentially want to allow different passwords or whatever | 15:43 |
| nuclear_eclipse | yeah | 15:43 |
| nuclear_eclipse | that's starts to get into really complicated setups | 15:43 |
| paul__ | I think it's probably 'ok' to sync realname | 15:43 |
| paul__ | (although, it's possible that realname in AD could be longer then what mantis allows | 15:44 |
| paul__ | so therefore you might not want to sync but allow people to override the realname | 15:44 |
| paul__ | and in addition, we now auto create users by default | 15:45 |
| paul__ | in my case, where I work in a school - staff have logons, students dont | 15:45 |
| paul__ | so I we need some params for the auto creation | 15:46 |
| paul__ | (incidently, auto creation fails if a user in ldap doesn't have an email address set, and stores the md5(pass) again) | 15:46 |
| paul__ | I'm not really that happy about this as it really needs fixing for 1.2 now (imo) | 15:47 |
| paul__ | and the ideal time to fix it is as part of thinking about auth plugin's | 15:47 |
| paul__ | gonna drive home before it snows more | 15:48 |
| paul__ | bbi15m | 15:48 |
| nuclear_eclipse | paul__: can you take a few moments and review the patch on bug 11293? | 15:54 |
| paul__ | 11293 | 16:22 |
| paul__ | well i'm home safely :P | 16:22 |
| paul__ | I dont get why we need both parts | 16:25 |
| paul__ | -if( !isset( $g_cache_config_project ) ) { | 16:25 |
| paul__ | +if( !isset( $g_cache_config_project ) || ( isset( $g_project_override ) && !in_array( $g_project_override, $g_cache_config_project ) ) ) | 16:25 |
| paul__ | i guess i'll need to read wtf this override shit does | 16:26 |
| paul__ | which from the fact you've asked me to look probably means it's something retarded: ( | 16:26 |
| nuclear_eclipse | from what I gather, the point of this is that when viewing a bug that's in a sub-project of the currently selected project, if the sub-project has config options set differently from the parent, the override is supposed to use the sub-project's settings when viewing the bug, but it currently uses the parent's settings because the override is ignored | 16:27 |
| nuclear_eclipse | but since I don't know enough about how config_api is supposed to work, I wanted you to look at it to make sure it's going down the correct path | 16:29 |
| paul__ | we use project override in 39 places | 16:29 |
| paul__ | including the process bugnote callback functions | 16:29 |
| * paul__ sighs | 16:30 | |
| paul__ | so basically | 16:30 |
| paul__ | we need to review all 39 cases | 16:30 |
| paul__ | for config_get calls before the override is set | 16:31 |
| nuclear_eclipse | potentially, yes | 16:31 |
| paul__ | bug_assign_reporter.php: | 16:32 |
| paul__ | access_ensure_bug_level( config_get( 'update_bug_threshold' ), $f_bug_id ); | 16:32 |
| paul__ | $t_bug = bug_get( $f_bug_id, true ); | 16:32 |
| paul__ | if( $t_bug->project_id != helper_get_current_project() ) { | 16:32 |
| paul__ | # in case the current project is not the same project of the bug we are viewing... | 16:32 |
| paul__ | # ... override the current project. This to avoid problems with categories and handlers lists etc. | 16:32 |
| paul__ | $g_project_override = $t_bug->project_id; | 16:32 |
| paul__ | oh dear | 16:32 |
| * paul__ thinks :) | 16:32 | |
| paul__ | I guess that's wrong then | 16:32 |
| paul__ | for the config_get change | 16:36 |
| mantisbot | Bug 11293 - cmfitch1 - open - new | 17:03 |
| mantisbot | View report page is ignoring project specific config settings - http://www.mantisbt.org/bugs/view.php?id=11293 | 17:03 |
| CIA-27 | Mantisbt: s.mazeland * rdc5bb9d053cd /lang/strings_english.txt: Fix #6816: disambiguate "Save login" | 17:32 |
| CIA-27 | Mantisbt: s.mazeland * r0ddfc76a22f1 /admin/ (check.php index.php): Fix #11294: Fix PHP notices. Patch contributed by Krasnov Kirill | 17:47 |
| Professeur | Hi, just a question, do I need a server to install Mantis or is there an other solution | 22:17 |
Generated by irclog2html.py