Team Chat Logs

September 16, 2010

2010 8
Mo Tu We Th Fr Sa Su
    1 2 3 4 5
6 7 8 9 10 11 12
13 14 15 16 17 18 19
20 21 22 23 24 25 26
27 28 29 30      

[01:32:39.775742]<osimons>Hodge*: one down. goodie :-)
[01:36:22.387706]<Hodgework>osimons: Yay. :)
[01:36:36.514171]<Hodgework>I think #577 is quite easy to close too.
[01:36:51.520189]*Hodgework pings davidfraser.
[01:38:40.804612]*davidfraser will look shortly
[01:41:26.107050]<Hodgework>davidfraser: There isn't much to look at. I think it's just a case of deciding whether the int(...) is important.
[01:42:00.134453]<davidfraser>Hodgework: Right, that means actually running it though - I seem to remember adding it for a reason
[01:42:12.331638]<Hodgework>I can see arguments both ways for which x-axis spacing is better.
[01:42:27.049404]<Hodgework>davidfraser: Ah. Testing. :D
[01:46:59.907071]<davidfraser>Testing complete :)
[01:47:20.503322]<davidfraser>Hodgework/osimons: Can one of you backport that, or tell me the svn incantation to backport in the recommended way?
[01:47:27.561322]<davidfraser>r894, that is, when it shows up
[01:47:48.425456]<CIA-10>r894 by dfraser in bitten/report/coverage.py: Don't convert to `int` here - this won't work for mercurial etc, and isn't necessary. See #577 (also remove the handling of the exception that can't ...
[01:48:15.556264]<osimons>cd /path/to/0.6.x
[01:48:33.255996]<osimons>svn mergeinfo --show-revs eligible ^/trunk # just to show revisions
[01:48:47.124151]<osimons>svn merge -c 894 ^/trunk
[01:48:54.062184]<davidfraser>thanks osimons
[01:49:04.677771]<osimons>svn ci -m "0.6dev: Merged [894] from trunk."
[01:49:19.967700]<osimons>piece of cake - once you got the snippet stored :-)
[01:49:37.279590]*davidfraser saves it as a pseudo-script in his 0.6.x directory
[01:53:16.262500]<CIA-10>r895 by dfraser in branches/ (0.6.x/bitten/report/coverage.py 0.6.x): 0.6dev: Merged[894] from trunk.
[01:54:23.347581]<Hodgework>osimons: Ooh. I wasn't aware of the ^ syntax.
[01:55:19.662843]<osimons>new in 1.5.x iirc
[02:15:06.396531]<Hodgework>I get used to using my tools and then I forget to look for new features when updates come along. :)
[02:36:27.709985]<CIA-10>r896 by osimons in doc/ (reports.txt commands.txt index.txt): Moving the wiki:ReportFormats page into the repository.
[02:39:36.089579]<CIA-10>r897 by osimons in branches/ (4 files in 2 dirs): 0.6dev: Merged [896] from trunk.
[02:43:27.043585]<CIA-10>r898 by osimons in doc/reports.txt: Aaargh. Mime-type for wiki rendering...
[02:44:47.729281]<CIA-10>r899 by osimons in branches/ (0.6.x 0.6.x/doc/reports.txt): 0.6dev: Merged [898] from trunk.
[03:16:59.881536]<osimons>Hodge*: the keepalive code was walter, was it not?
[03:18:37.447890]<osimons>Hodgework: ^^
[03:18:37.525807]<Hodgework>osimons: Yes.
[03:18:40.891017]<osimons>ah
[03:18:50.796814]<osimons>you do notice the Hodge* - goodie :-)
[03:19:57.455179]<Hodgework>It's a bit odd that I got flagged by Hodge* -- I have 'hodge' set as one of the things to ping me on. :)
[03:20:33.136944]<Hodgework>I've seen some non-critical problems with the keepalive code (mostly not handling exceptions for HTTP requests somewhere) but haven't had a chance to dig down into it.
[03:20:45.807944]<Hodgework>Lunch.
[03:22:58.538600]<osimons>Hodgework: like this that i just commented perhaps? http://bitten.edgewall.org/ticket/607#comment:2
[03:25:22.707159]<osimons>scfe: want to pick this up, refresh the patch and verify it? if so i'll have no problems committing it - it seems reasonable.
[03:26:27.911206]<scfe>osimons: sorry, I'm a bit out of the loop. Which patch?
[03:26:30.554126]<osimons>scfe: sorry - forgot link :-) http://bitten.edgewall.org/ticket/610
[03:27:35.521428]<osimons>mostly done, just needs to be verified and updated in case it no longer applies cleanly
[03:27:47.452743]<osimons>- it will make your trac look prettier, eh? :-)
[03:27:50.075144]<scfe>osimons: I'm happy to do that but due to time constraints I can't promise that it'll be done before the weekend
[03:28:01.517099]<scfe>osimons: you know how to get people to work ;-)
[03:28:25.030259]<osimons>no probs. thanks!
[03:28:43.093166]<osimons>i'll make a note on the ticket too - perhaps the author is still keen.
[03:28:53.076805]<osimons>see you are on cc, so you'll be in the loop
[03:31:03.870306]<osimons>oh. Hodge*, that is your ticket i see. presume you agree...
[03:50:41.216257]<CIA-10>r900 by osimons in bitten/report/coverage.py: Fix context menu 'coverage' linking. Closes #578. Thanks to simohe@... for report and patch.
[03:53:11.188241]<CIA-10>r901 by osimons in branches/ (0.6.x/bitten/report/coverage.py 0.6.x): 0.6dev: Merged [900] from trunk.
[04:02:02.030301]<Hodgework>osimons: I do.
[04:02:33.498672]<Hodgework>trbs == scfe?
[04:03:22.315851]<osimons>no. scfe = felix on cc for that ticket (and that provided patch for mercurial that i applied the other day)
[04:04:32.499524]<Hodgework>Ah.
[04:04:36.232665]<Hodgework>Hello scfe!
[04:05:07.355525]<scfe>Hodgework: hi
[04:05:27.227113]<Hodgework>osimons: On an unrelated note I realised while doing the Bitten port that it was silly of me to implement my on xi:include for Bitten when I could just make the build recipes Genshi templates.
[04:05:39.770918]<Hodgework>s/Bitten/Genshi/
[04:05:39.779885]<evil_twin>hodgework meant: osimons: On an unrelated note I realised while doing the Genshi port that it was silly of me to implement my on xi:include for Bitten when I could just make the build recipes Genshi templates.
[04:05:55.414367]<Hodgework>wb evil_twin/
[04:06:12.169915]<osimons>that isn't included now, is it? i can't remember anymore...
[04:06:51.687501]<osimons>Hodgework: what was that ticket id again?
[04:08:03.111302]<osimons>remembered - http://bitten.edgewall.org/ticket/301 scheduled for 0.7. got time to improve that then, Hodge* :-)
[04:09:13.675979]<Hodgework>Yep. Not included yet.
[04:09:56.747800]<Hodgework>It exposes some interesting things about build recipes -- namely that the XML stored in the database and the XML sent to the slave are not the same. :)
[04:21:58.976083]<scfe>Hodgework: so can we have conditional build steps then?
[04:24:13.308017]<Hodgework>scfe: Yep.
[04:26:04.809156]<scfe>Hodgework: cool, I'm interested :-)
[05:10:27.494377]<osimons>oh bummer. text template syntax has changed with 0.11 -> 0.12. i've forgotten all about that, and now we need conditional template usage based on trac version... aaargh!
[05:24:49.947303]<Hodgework>osimons: Where do we use text templates?
[05:25:29.179665]<osimons>build notification
[05:26:07.620657]<osimons>just the for/if loops and such syntax. the common ${} substitutions is the same, but the expression syntax has changed
[05:26:44.665717]*osimons digs in for compat features...
[05:28:32.422961]<Hodgework>We could just import the template class we need directly from Genshi?
[05:28:50.358294]<Hodgework>It kept the old text template class around for compatibility.
[05:32:51.115430]<osimons>Hodgework: i know, but we don't render it so we need to change in the Trac classes
[05:32:58.776443]<osimons>that is perhaps the best way around this
[05:33:31.080075]<davidfraser>Changing the text template syntax is not that hard...
[05:51:02.055592]<osimons>got it now, thanks
[05:51:57.503918]<osimons>the problem is that the loading is all hidden inside Trac NotifyEmail and Chrome classes - we don't load any templates, so we need to override..
[05:56:21.248049]<osimons>davidfraser, Hodgework: this makes sense? http://scratchpad.cmlenz.net/cdf1c0d0295e0f13d2689264250002ab/
[05:56:37.649848]<osimons>- even wrote a test before converting to make sure it passed... :-)
[05:57:02.016880]<osimons>this uses new syntax even for 0.11
[05:57:31.239674]<davidfraser>osimons: Yes, nice
[05:58:36.877679]<osimons>i'll just commit it then
[06:01:48.907824]<Hodgework>osimons: Looks good to me too.
[06:02:04.634899]<CIA-10>r902 by osimons in bitten/ (3 files in 3 dirs): Changed notification to always use Genshi `NewTextTemplate` which lets us keep just one template for both Trac 0.11 and 0.12+.
[06:02:40.822729]<Hodgework>osimons: Does the "not isinstance" mean the template is loaded twice in some cases?
[06:03:19.352618]<CIA-10>r903 by osimons in branches/ (4 files in 4 dirs): 0.6dev: Merged [902] from trunk.
[06:04:20.261588]<osimons>Hodgework: shouldn't think so - we just replace the self.template loading a new, i believe? do it through the cache in Chrome and TemplateLoader from Genshi, and my guess would be that filepath is the key for that cache?
[06:04:33.220926]<osimons>- all assumptions and hearsay though :-)
[06:05:19.356219]<osimons>now, i'm sure i keep some other plugins that notify as well... hmm...
[06:05:25.861612]*osimons digs into project archive...
[06:09:25.737299]*osimons is relieved, and happy he settled for the 'outsourced' pattern of "for listener in self.listeners: ...." in other plugins...
[06:20:41.055891]<osimons>davidfraser, Hodgework: i've looked through all the tickets, and the 0.6 queue now says 11 open tickets that we should handle somehow before a release.
[06:21:20.513413]<osimons>i've committed what i have at the moment, and also updted some of the change docs at http://bitten.edgewall.org/milestone/0.6
[06:22:36.094964]<osimons>not a bad days work all in all. now family - back later.
[06:23:22.106311]<davidfraser>osimons: Great progress
[06:23:32.081069]<davidfraser>does that allow us to close "Is this project dead?"
[07:03:01.651563]<davidfraser>trac/osimons: See the interesting trac crash I discovered in #525
[07:03:15.839551]<davidfraser>sorry, s/trac/Hodgework/ - clearly you're becoming a piece of software
[07:36:27.281489]*Hodgework goes to look.