| ../irclogs/#mantishelp.2010-01-12.log | ||
| --- scribe started --- | 00:00 | |
| CIA-22 | Mantisbt: hickseydr * r94dc37c3a3c2 /view_all_set.php: Fix #11382: Typo in includes within view_all_set.php | 00:04 |
|---|---|---|
| CIA-22 | Mantisbt: Paul * ra47272749c91 / (174 files in 2 dirs): Improve indentation - left inconsistent by previous recent commits. | 00:22 |
| nuclear_eclipse | paul_____: here now | 02:06 |
| nuclear_eclipse | paul_____: what is the purpose of require_api and require_lib? | 02:10 |
| nuclear_eclipse | ie, what was "wrong" with just using require_once? | 02:11 |
| nuclear_eclipse | paul_____: also, if require_api/lib actually fixes something, should we then be backporting them to 1.2? | 02:14 |
| dwestfall | APPLICATION ERROR #200 driving me crazy | 03:12 |
| dwestfall | APPLICATION ERROR #200 driving me cray | 03:17 |
| dwestfall | [11-Jan-2010 23:33:45] APPLICATION ERROR #200 | 03:18 |
| dwestfall | [11-Jan-2010 23:33:45] APPLICATION ERROR #A required parameter to this page (reproducibility) was not found. | 03:18 |
| dwestfall | [11-Jan-2010 23:33:45] PHP Notice: Undefined index: bug_id in E:\wwwroot\wise\wise\issues\core\error_api.php on line 77 | 03:18 |
| dwestfall | [11-Jan-2010 23:33:45] APPLICATION ERROR # | 03:18 |
| dwestfall | [11-Jan-2010 23:33:45] APPLICATION ERROR USER WFT\SobczyRX | 03:18 |
| dwestfall | [11-Jan-2010 23:33:45] APPLICATION ERROR USER /wise/issues/bug_report.php | 03:18 |
| dwestfall | [11-Jan-2010 23:33:45] APPLICATION ERROR METHOD GET | 03:18 |
| dwestfall | [11-Jan-2010 23:33:45] APPLICATION ERROR QUERY STRING | 03:18 |
| Wraiyth | hey, anyone around that can help me with an account activation issue? | 03:23 |
| Wraiyth | guess not | 03:31 |
| Wraiyth | I have users that randomly don't get activation emails | 03:31 |
| Wraiyth | and I can't seem to find a way to manually activate an account. | 03:31 |
| dhx_m | dwestfall: which version of Mantis? | 04:47 |
| dhx_m | Wraiyth: have you got batch email sending on? | 04:48 |
| alemayo | is there any plugin-based way to add a new custom-filed type or would I just patch cf_def standard ? | 08:31 |
| kirillka | dhx_m: around? | 08:41 |
| dhx_m | kirillka: hi | 08:56 |
| dhx_m | alemayo: not afaik | 08:56 |
| kirillka | dhx_m: lo | 08:56 |
| kirillka | I have trouble with last updates | 08:56 |
| kirillka | I see blank page | 08:56 |
| dhx_m | alemayo: although you could add entries to the database (where custom field info is stored) directly from your plugin? | 08:57 |
| dhx_m | alemayo: there ^may^ by some API in custom_field_api.php that handles that for you, but I don't know much about it | 08:57 |
| dhx_m | kirillka: do you have PHP error reporting turned on? | 08:57 |
| kirillka | yes. no error | 08:58 |
| dhx_m | every page is affected? | 08:58 |
| dhx_m | which server, PHP version, etc? | 08:58 |
| kirillka | httpd, php 5.2.9 | 08:58 |
| alemayo | dhx_m: okay ... I just found that it is also quite handy to add new cf_defs ... just trying to figure out how to do that in detail | 08:58 |
| alemayo | I am trying to add a custom field that can hold a whole table of informations | 08:58 |
| dhx_m | alemayo: custom fields is something I'd like to rewrite :) | 08:59 |
| alemayo | we are using mantis here for helpdesk and also for tracking hardware orders and similar stuff (so we need to add 2 mices and 5 LCD screens to a issue) | 08:59 |
| dhx_m | alemayo: our current implementation is partly there, but isn't as straightforward to use as it could be (using OO classes, etc) | 08:59 |
| alemayo | a lot of stuff is not OO :) | 09:00 |
| dhx_m | it strikes me that you may not want to use custom fields then | 09:00 |
| alemayo | but also this cf-def approach is quite incomplete .. hm .. maybe because everything would be so much easier with OO :) | 09:00 |
| alemayo | dhx_m: it strikes me that you may not want to use custom fields then <--- what do you mean by that ? | 09:00 |
| alemayo | I will use CFs I think | 09:00 |
| dhx_m | well it sounds like you want a field like "relationships" where you can link a "bug" to types of hardware? | 09:01 |
| dhx_m | or would it just store non-orthogonal data (IDs, text supplied by the user, etc)? | 09:01 |
| dhx_m | kirillka: Apache 2? | 09:02 |
| kirillka | yes | 09:03 |
| dhx_m | kirillka: what is the first bad commit? | 09:03 |
| alemayo | dhx_m it is not really a relationship ... just a plain table to enter these items (there is no item database in the background to choose from certain types and so on) | 09:04 |
| alemayo | dhx_m: I think I will call my CF "Table" and define an own table to store the whole info (with colums like bugid, Type, Quantity, Unit Price) | 09:05 |
| dhx_m | alemayo: that sort of list data really then needs to be stored in the database within its own table | 09:05 |
| dhx_m | alemayo: so your plugin would handle that table by itself... and the integration with Mantis would be using it's API, and outputting HTML to the relevant pages where you want to show data | 09:05 |
| dhx_m | yep | 09:06 |
| alemayo | yes | 09:06 |
| dhx_m | however it wouldn't really have anything to do with custom fields | 09:06 |
| alemayo | dhx_m: I found the definition and stuff of the custom fields handy -- that's why we use it .. | 09:06 |
| dhx_m | ideally we'd have a built-in method for handling custom fields that are lists of data (orthogonal or not) | 09:06 |
| alemayo | I am happy to make it in a way that is easy to integrate and flexible ... | 09:07 |
| dhx_m | but at the moment, if you go the plugin route, it won't integrate with Mantis' search capabilities, columns on view_all_bug_page.php, etc | 09:07 |
| dhx_m | not that list data can really be put in a single row :) | 09:07 |
| alemayo | a long one maybe :-) | 09:08 |
| alemayo | is there any example to add own custom fields ? | 09:08 |
| alemayo | because there is already that flexible approach with a cf_mycompany in the cfdefs folder | 09:09 |
| alemayo | and then $g_custom_field_types[CUSTOM_FIELD_TYPE_TABLE] = 'mycompany'; | 09:09 |
| dhx_m | I'd advise you to look at what SourceIntegration (mantisforge.org) does | 09:09 |
| alemayo | but it looks like I also should modify $g_custom_field_type_enum_string in the next step :) | 09:09 |
| alemayo | and also the localization ... | 09:09 |
| dhx_m | also remember if you need to change something in Mantis' core... feel free to submit patches to the official MantisBT bug tracker | 09:11 |
| dhx_m | if it's a "good change" it can be committed to the repository to be supported for all future versions of Mantis :D | 09:11 |
| kirillka | dhx_m: this last commit witch merged without problems 317f809f8f83f4b3867bb93ea7a48af6242378c0 | 09:13 |
| kirillka | after I merge more time next | 09:14 |
| dhx_m | kirillka: so http://git.mantisbt.org/?p=mantisbt.git;a=commitdiff;h=ea3cd38976a5b2af0ff0615b91d2f5d51970d783 is bad for you? | 09:14 |
| kirillka | dhx_m: I think yes | 09:15 |
| alemayo | dhx_m: how can mantisforge help me with my custom fields? I am confused ... | 09:17 |
| alemayo | ah i should read the intro :) | 09:18 |
| dhx_m | alemayo: yep I just meant that there is a plugin on there which would be worthwhile you looking at first, to see how it does things | 09:18 |
| dhx_m | kirillka: that's interesting... the commit I mentioned wouldn't result in blank output | 09:18 |
| alemayo | I was looking alot for plugins but I only found that old stuff from that french website that did not work :) | 09:19 |
| dhx_m | alemayo: yeah that plugin stuff has nothing to do with Mantis 1.2.x :) | 09:19 |
| kirillka | dhx_m: How I can enabled error witch can help you? | 09:19 |
| dhx_m | kirillka: add to config_inc.php: | 09:19 |
| dhx_m | $g_show_detailed_errors = ON; | 09:19 |
| dhx_m | $g_display_errors = array(E_WARNING => 'halt', | 09:19 |
| dhx_m | E_NOTICE => 'halt', | 09:19 |
| dhx_m | E_USER_ERROR => 'halt', | 09:19 |
| dhx_m | E_USER_WARNING => 'halt', | 09:19 |
| dhx_m | E_USER_NOTICE => 'halt', | 09:19 |
| dhx_m | E_ALL => 'halt'); | 09:19 |
| kirillka | dhx_m: nothing - only blank page | 09:20 |
| dhx_m | kirillka: do you have any uncommitted patches/changes in your current MantisBT directory? | 09:21 |
| dhx_m | ah better question... can you checkout a fresh copy of the current 1.3.x branch (master) and see if it works out-of-the-box? | 09:22 |
| alemayo | wow ... PHPXref is cool | 09:26 |
| alemayo | why is there no huuuge link to mantisforge on mantisbt.org .... hmpf | 09:26 |
| dhx_m | oh sorry, I meant git.mantisforge.org :) | 09:27 |
| dhx_m | no wonder you had trouble finding it | 09:27 |
| dhx_m | alemayo: phpxref will be broken at present (we need to patch it to handle some changes I made to how includes are found) | 09:28 |
| alemayo | where is the actual code of a plugin, let's say: http://git.mantisforge.org/w/tgulacsi-unosoft.git ?? | 09:32 |
| * alemayo is out for lunch | 09:32 | |
| kirillka | dhx_m: fresh not work too :( | 09:32 |
| kirillka | dhx_m: this master branch with test data | 09:39 |
| dhx_m | alemayo: see http://git.mantisforge.org/w/source-integration.git?a=tree | 09:44 |
| dhx_m | kirillka: I can't reproduce... or see any reason why it'd only show a blank page.... can you view web server error logs? | 09:44 |
| kirillka | dhx_m: [12/Jan/2010:11:47:45 +0200] "GET /test/ HTTP/1.1" 200 26 | 09:48 |
| dhx_m | try view.php?id=1 | 09:48 |
| dhx_m | account_page.php, etc | 09:49 |
| dhx_m | a few different PHP files | 09:49 |
| kirillka | 10.0.129.235 - - [12/Jan/2010:11:49:16 +0200] "GET /test/view.php?id=1 HTTP/1.1" 200 26 | 09:49 |
| kirillka | 10.0.129.235 - - [12/Jan/2010:11:49:19 +0200] "GET /test/account_page.php HTTP/1.1" 200 26 | 09:49 |
| kirillka | but all blank page | 09:49 |
| dhx_m | have you checked php.ini to make sure error reporting is on/debug verbosity high/etc? | 09:50 |
| dhx_m | display_errors in particular | 09:52 |
| kirillka | display_errors = On | 09:53 |
| dhx_m | yep | 09:54 |
| kirillka | restart httpd | 09:54 |
| kirillka | but nothing in web and log | 09:54 |
| alemayo | dhx_m: http://git.mantisforge.org/w/tgulacsi-unosoft.git?a=tree --> 404 - Reading tree failed | 11:03 |
| dhx_m | alemayo: is that your repository? | 11:04 |
| alemayo | no | 11:05 |
| alemayo | :) | 11:05 |
| dhx_m | the reason you get that error is that they haven't uploaded a master branch | 11:05 |
| dhx_m | they've only pushed this: http://git.mantisforge.org/w/tgulacsi-unosoft.git?a=tree;h=refs/heads/tgulacsi-unosoft;hb=tgulacsi-unosoft | 11:05 |
| alemayo | dhx_m: looks like the whole mantis code ... | 11:15 |
| Wraiyth | dhx_m: wasn't here when you replied. Yes, email notifications are turned on, but it seems like the email confirmation failing is randomly happening. Could be an issue with our host, I'm not sure, but it seems intermittent. | 11:15 |
| dhx_m | back | 11:50 |
| dhx_m | alemayo: it is... it's a branch :) | 11:50 |
| dhx_m | alemayo: I was pointing you to the SourceIntegration plugin before which isn't the whole code... it's a plugin ;) | 11:50 |
| alemayo | yeah, got it :) | 11:51 |
| alemayo | I am just wondering whether I should make my customizations towards mantis 1.2.0rc2 or the latest git version | 11:57 |
| alemayo | but we plan to use that soon in our production environment so I guess the 1.2.0 is safer :) ? | 11:58 |
| dhx_m | against the 1.2.x branch for now | 12:00 |
| dhx_m | 1.3.x will be undergoing a lot of changes | 12:01 |
| alemayo | the OO-work :) ? | 12:02 |
| dhx_m | so there isn't really much point in developing plugins for it unless you plan to keep up to date with development | 12:02 |
| dhx_m | yeah | 12:02 |
| dhx_m | ... depending on who wants to help out with it :) | 12:02 |
| alemayo | wanting .. yes .. time .. no :) | 12:07 |
| alemayo | anyway .. .I will try to skill-up a PHP developer :) but first of all I have to do this on my own and my core job is normally networking and server infrastructure, not development :) | 12:07 |
| dhx_m | well as I said before, if you make changes to the core of Mantis, pass the patches on and we'll see if we can integrate them upstream | 12:19 |
| dhx_m | the advantage of open source is that you can push your changes upstream... at which point you don't need to care about maintaining those changes yourself :) | 12:19 |
| alemayo | I know :) | 12:21 |
| alemayo | Just was wondering if there is no DB update necessary from 1.2.0 RC 1 to RC 2 ? just copied my config and it directly works ... | 12:22 |
| alemayo | I think I will make a patch to extend custom fields sequence (negativ ID shows them on top of the report) | 12:25 |
| paul_____ | 02:15 < nuclear_eclipse> paul_____: what is the purpose of require_api and require_lib? | 12:26 |
| paul_____ | that's dhx's code | 12:26 |
| paul_____ | I dont really see why we need require_lib | 12:26 |
| paul_____ | seems redundant | 12:26 |
| dhx_m | nuclear_eclipse: avoids stat() calls from require_once() | 12:34 |
| dhx_m | nuclear_eclipse: in tests, it's probably twice as quick... on Linux, which caches stat() results | 12:34 |
| dhx_m | paul_____: require_lib/require_api look in the $g_library_path and $g_core_path directories respectively | 12:35 |
| dhx_m | paul_____: thus you don't have to worry about filename clashes, or require_once() having to search 10 different directories for a file when you already know where it'll be | 12:36 |
| paul_____ | I need to look at what I had before | 12:36 |
| paul_____ | as I vaguely recall looking at autoloading of those libs | 12:36 |
| paul_____ | (I recall a performance/memory improvement in adodb where it autoloads classes we dont use for example | 12:37 |
| * paul_____ thinsk | 12:37 | |
| dhx_m | autoloading is the ideal method to use | 12:37 |
| dhx_m | I'd love to kill off all require_* calls | 12:37 |
| dhx_m | we don't need to include kitchen sinks every time we want to call just a single function from access_api, for instance | 12:38 |
| paul_____ | anyway | 12:43 |
| paul_____ | require_lib I think we should *kill* | 12:43 |
| paul_____ | all our libs are classes right? | 12:43 |
| paul_____ | so can be engineered to autoload | 12:43 |
| dhx_m | I'm happy with that decision, if we fix autoload appropriately (ie. don't use require_* calls inside the code to load libraries) | 12:44 |
| paul_____ | biab | 12:46 |
| alemayo | how can I suppress those "Common subdirectories" messages when doing diff -u originalmantis/ mymantis/ | 13:30 |
| alemayo | oh, I should use -r .. :) | 13:33 |
| alemayo | dhx_m: if you have any suggestions on my first patch .. :) http://www.mantisbt.org/bugs/view.php?id=11385 | 13:38 |
| paul_____ | I probably wouldn't do it like that personally | 13:41 |
| paul_____ | the complete order of all fields should be customisable | 13:41 |
| alemayo | it is | 13:42 |
| alemayo | ah you mean also the build-in ones ... | 13:42 |
| paul_____ | I mean | 13:42 |
| paul_____ | yes | 13:42 |
| paul_____ | bbl - need to go back to work | 13:42 |
| alemayo | but that's harder to implement :) | 13:42 |
| alemayo | but you're right... | 13:42 |
| azneita | hi guys! i'm having problems with email | 15:21 |
| azneita | the email sectio of config_inc.php looks like this -> http://pastebin.com/d17bfd8bf | 15:23 |
| azneita | so far no email goes out | 15:23 |
| azneita | likewise, phpmailer is already on php's path | 15:24 |
| nuclear_eclipse | azneita: 1.1.x currently doesn not work properly with gmail | 15:24 |
| azneita | ouch | 15:24 |
| nuclear_eclipse | you'll need to download 1.2 (not yet finished) to use gmail as your smtp server | 15:25 |
| nuclear_eclipse | or wait a couple weeks for a final release | 15:25 |
| azneita | i tried using mail() but it's not working too | 15:25 |
| nuclear_eclipse | mail() would require setting up your php.ini appropriately | 15:26 |
| nuclear_eclipse | or having an MTA installed on your local machine | 15:26 |
| azneita | php.ini meaning getting phpmailer classes on php's path right? | 15:27 |
| nuclear_eclipse | no, I mean that mail() uses PHP's core configuration to determine how to send mail | 15:27 |
| nuclear_eclipse | so you'd have to setup your php.ini to list gmail's servers, etc | 15:27 |
| azneita | ah, i'm not that far ahead | 15:28 |
| azneita | my best option then is to use a local MTA | 15:28 |
| nuclear_eclipse | if you're running unix/linux, setting up a simple MTA is probably the best option | 15:29 |
| azneita | and the use phpmailer_method_sendmail | 15:29 |
| nuclear_eclipse | that, or leave it at the default mail(), and php will pick up the sendmail call | 15:30 |
| azneita | no other variables to setup right? | 15:30 |
| nuclear_eclipse | eg, I have postfix on my server, and never touched the phpmailer configs for mantis, and it just worked right out of the box | 15:30 |
| azneita | let me see my package list | 15:31 |
| azneita | nothing installed | 15:31 |
| nuclear_eclipse | there might be another MTA that would be easier to setup than postfix though | 15:32 |
| nuclear_eclipse | what distro are you on? | 15:32 |
| azneita | centos | 15:32 |
| nuclear_eclipse | that's what we use on mantisbt.org, but I'm not sure what MTA we have | 15:34 |
| azneita | out of the box, postfix should pick up anything mail() throws at it right? | 15:34 |
| nuclear_eclipse | I would expect it to | 15:35 |
| azneita | let me give it a try | 15:35 |
| nuclear_eclipse | you could always test it with a script | 15:35 |
| azneita | phpmailer's scripts? | 15:39 |
| azneita | everything working now | 15:45 |
| azneita | it seems like i just need the MTA | 15:45 |
| azneita | thanks nuclear_eclipse | 15:46 |
| paul_____ | dhx_m: you broke something | 19:52 |
Generated by irclog2html.py