| *** paulproteus_ has quit IRC | 02:51 | |
| *** paulproteus has joined #schooltool | 02:58 | |
| *** khildin has joined #schooltool | 08:33 | |
| *** yvl has joined #schooltool | 10:33 | |
| *** khildin has quit IRC | 14:16 | |
| *** ignas has joined #schooltool | 14:35 | |
| th1a_ | hi yvl... | 16:31 |
|---|---|---|
| yvl | hey th1a_ | 16:31 |
| *** th1a_ is now known as th1a | 16:31 | |
| th1a | Now you're the only one here... | 16:32 |
| th1a | So, how was your break? | 16:32 |
| yvl | a bit empty today | 16:32 |
| yvl | pretty good, thanks :) | 16:32 |
| yvl | is replaceafill missing, or?.. | 16:32 |
| yvl | (menesis is wandering around the office somewhere) | 16:33 |
| * th1a checks the dates for PyCon.... | 16:33 | |
| th1a | That's not until the 13th. | 16:33 |
| th1a | I'd say missing. | 16:33 |
| yvl | maybe the internets blew up again | 16:34 |
| th1a | Yes. | 16:34 |
| yvl | I looked briefly at https://answers.launchpad.net/schooltool/+question/219868 | 16:35 |
| yvl | it seems that they use cn to "make" the username, and it is most likely unicode | 16:35 |
| th1a | Ah. | 16:35 |
| yvl | so - unicode usernames | 16:35 |
| * yvl was not sure how to respond, so left the question open | 16:36 | |
| th1a | And that's not trivial for us to support, right? | 16:36 |
| yvl | nope | 16:36 |
| yvl | though it may be quite easy | 16:36 |
| yvl | or not | 16:36 |
| yvl | all we know is that most places handle it, but there are some that crash horribly | 16:37 |
| th1a | OK. If they restrict themselves to ASCII usernames? or what? It will work? | 16:37 |
| yvl | well | 16:37 |
| yvl | they would need ASCII usernames as LDAP attribute somewhere | 16:37 |
| yvl | uid tends to be ascii, but not always | 16:37 |
| th1a | OK... does this mean LDAP doesn't reall work, or that LDAP works if you restrict your usernames somehow? | 16:38 |
| *** menesis has joined #schooltool | 16:38 | |
| *** menesis has quit IRC | 16:38 | |
| *** menesis1 has joined #schooltool | 16:38 | |
| yvl | works if you restrict usernames to ASCII somehow | 16:38 |
| yvl | it is loosely configurable, so... | 16:38 |
| *** menesis has joined #schooltool | 16:38 | |
| th1a | The easiest way to restrict them is just to create them that way? | 16:39 |
| yvl | hmm | 16:39 |
| yvl | well | 16:39 |
| yvl | depends | 16:39 |
| yvl | too many implementations | 16:39 |
| yvl | they can add their own attributes in some | 16:39 |
| menesis1 | hi | 16:39 |
| yvl | or maybe set them in implementations that have them | 16:39 |
| th1a | hi menesis. | 16:40 |
| *** menesis1 has left #schooltool | 16:40 | |
| yvl | cn would be your "first name last name" | 16:40 |
| yvl | so it's not likely they'll restrict that | 16:40 |
| th1a | OK, they can map a different LDAP attribute to our username? | 16:41 |
| yvl | yes | 16:41 |
| yvl | I can query windows AD user on schooltool, but I need to add a record on uid attribute to match the "default_login_attr" attribute. Can I use "default_login_attr cn" to match cn attribute cause AD use cn(cn=common name) to locate the user, thanks. | 16:41 |
| yvl | probably it would be best to finally fix unicode username support and be done with that | 16:42 |
| th1a | Yes. | 16:42 |
| menesis | the bug is https://bugs.launchpad.net/zope.app.publication/+bug/646019 | 16:42 |
| menesis | and I have assigned it to me.. | 16:43 |
| menesis | but never got back to it | 16:43 |
| yvl | linked | 16:43 |
| yvl | thanks | 16:43 |
| th1a | Why can't I edit that bug? | 16:43 |
| menesis | it is a bug in zope | 16:44 |
| th1a | OK. | 16:44 |
| th1a | Hrm. | 16:44 |
| th1a | OK. | 16:44 |
| th1a | So... setting that aside. | 16:44 |
| th1a | Where do we stand with Celery? | 16:45 |
| yvl | th1a, maybe we can work around that | 16:45 |
| yvl | principle id is somewhat under our control | 16:45 |
| yvl | as for celery | 16:46 |
| menesis | I looked at the AD/ldap question and found an attribute sAMAddressName, but not sure if it is the username | 16:46 |
| yvl | usualy it's simply uid, menesis | 16:47 |
| yvl | in posix style accounts or however that's called | 16:47 |
| yvl | I'm done with celery set up, | 16:47 |
| yvl | mostly done with report task submission | 16:48 |
| yvl | some 1/3 done with report notifications | 16:48 |
| yvl | and... that's it unless I forgot something | 16:48 |
| th1a | OK, thanks yvl. | 16:49 |
| yvl | should I take a quick look at unicode usernames? | 16:50 |
| yvl | or postpone | 16:50 |
| th1a | I think I may have mentioned once or twice that getting Celery in this release is the top priority. ;-) | 16:50 |
| th1a | So... | 16:50 |
| th1a | Let's discuss that with menesis first. | 16:51 |
| yvl | ok | 16:52 |
| menesis | I haven't seen yvl's latest celery branch | 16:54 |
| th1a | menesis: My sense of this process is that it is better to rush something before feature freeze and then fix the bugs than miss the freeze. | 16:54 |
| th1a | OK, so that's step one. | 16:54 |
| menesis | yes | 16:54 |
| menesis | only uploading bugfixes after feature freeze is easier | 16:55 |
| menesis | I have uploaded the latest releases to raring already | 16:55 |
| menesis | but this does not include celery and related infrastructure changes | 16:56 |
| th1a | Right. | 16:56 |
| th1a | What we need here is a big push to the deadline. | 16:57 |
| menesis | yes | 16:58 |
| th1a | :-) | 16:58 |
| yvl | so... I'l put up a branch with infrastructure, but no actual celery tasks in use | 16:59 |
| th1a | So do you guys need to sit down together? | 16:59 |
| yvl | will that help? | 16:59 |
| yvl | worst case scenario, we can put reports themselves or the imports in the bugfix phase | 17:00 |
| yvl | that would make hard for menesis to test of course | 17:01 |
| th1a | Giving him a chance to get started now would help, I'm sure. | 17:01 |
| th1a | Better than waiting. | 17:01 |
| yvl | allright then, we can start from there | 17:03 |
| yvl | if there are no objections ;) | 17:03 |
| menesis | the important part for me is the configuration files and scripts. supervisor, celery, etc. | 17:03 |
| menesis | does not matter if it is not used for anything | 17:03 |
| yvl | great | 17:03 |
| th1a | OK. That's the top priority, and actually getting the reports working. | 17:05 |
| yvl | ok | 17:06 |
| th1a | Guess that's it. | 17:07 |
| th1a | I'm actually checking in additions to the book and generally over my writer's block. | 17:08 |
| menesis | and reports don't work in Ubuntu 13.10 . have to debug with versions reportlab = 2.6 z3c.rml = 2.0.0 and fix/workaround it | 17:08 |
| th1a | OK. | 17:11 |
| th1a | Anything else? | 17:13 |
| menesis | I have updated buildout/distribute versions | 17:14 |
| menesis | if you get a VersionConflict: We already have ... | 17:14 |
| menesis | error | 17:14 |
| menesis | try make update | 17:15 |
| menesis | or rm bin/buildout | 17:15 |
| menesis | then the new version of buildout will be bootstrapped. | 17:16 |
| yvl | ້ທທ | 17:16 |
| yvl | whoops | 17:16 |
| yvl | from the first glance, quick fix for unicode usernames works | 17:17 |
| th1a | What does the quick fix entail? | 17:17 |
| yvl | encoding principal.id with punycode | 17:18 |
| yvl | that's it. | 17:19 |
| th1a | Do we have to change views to decode it? | 17:19 |
| yvl | actually - no | 17:20 |
| yvl | at least no, if they work | 17:20 |
| yvl | need to check everything everywhere to be sure | 17:20 |
| menesis | :) | 17:21 |
| yvl | but I don't think anything else stores principal.id directly, so there should be no backwards compatibility problems | 17:21 |
| yvl | well, maybe when looking at the ZODB transaction history, it will display, say: | 17:21 |
| yvl | "sb.manager-" in new transactions and "sb.manager" in old ones | 17:22 |
| th1a | OK. I"ve got you penciled in for bugfixes the rest of the month yvl, so you'll have time to work on that. | 17:23 |
| yvl | menesis - http://pastebin.com/NCZjRYKD if you want to play around | 17:23 |
| menesis | I hate punycode for that- | 17:24 |
| * yvl too | 17:24 | |
| th1a | :-D | 17:24 |
| th1a | It is a one line change? | 17:25 |
| yvl | yes | 17:25 |
| yvl | who knew? | 17:25 |
| th1a | That's pretty funny. | 17:25 |
| th1a | Do you hate it for aesthetic reasons? | 17:25 |
| yvl | that little "-" saves lives! | 17:26 |
| yvl | but anyway - maybe somebody will want to play around | 17:27 |
| yvl | maybe Douglas | 17:27 |
| th1a | OK. | 17:27 |
| menesis | maybe encode('utf-8') would be better | 17:27 |
| yvl | just change some default users' usernames manually | 17:27 |
| yvl | I wouldn't do that after today's chat with mgedmin :) | 17:27 |
| yvl | rule of thumb - don't store UTF-8 in ZODB | 17:28 |
| yvl | or pain. | 17:28 |
| yvl | (python 3) | 17:28 |
| th1a | OK, let's wrap this up gentlemen. | 17:30 |
| th1a | Let's have a good push here through Thursday. | 17:30 |
| th1a | Work together! | 17:31 |
| th1a | Thanks guys! | 17:31 |
| * th1a drops the bag of gravel. | 17:31 | |
| yvl | thanks guys | 17:32 |
| yvl | http://localhost:7080/persons/%E0%BA%AB%E0%BA%B0%E0%BA%B5%E0%BA%81%E0%BA%B3%E0%BA%B7%E0%BA%B0 | 17:39 |
| th1a | replaceafill has a cold. I missed his email for some reason. | 18:03 |
| *** paulproteus has quit IRC | 18:04 | |
| *** paulproteus has joined #schooltool | 18:13 | |
| *** menesis has left #schooltool | 19:34 | |
| *** menesis2 has joined #schooltool | 19:35 | |
| *** menesis2 has left #schooltool | 19:35 | |
| *** menesis has joined #schooltool | 19:36 | |
| *** menesis has quit IRC | 19:44 | |
| *** ignas has quit IRC | 20:47 | |
Generated by irclog2html.py 2.15.1 by Marius Gedminas - find it at mg.pov.lt!