2008-09-27T00:05:51 * dreimark looks again at the windows py.tests
2008-09-27T00:15:01 <dreimark> ALT 092 is \
2008-09-27T00:23:02 <dreimark> why is windows telling C:\Python25\lib\zipfile.py:259: DeprecationWarning: struct integer overflow mask
2008-09-27T00:23:06 <dreimark> and not linux
2008-09-27T00:24:52 <CIA-25> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 4333:b52a579db72d 1.9/MoinMoin/security/antispam.py: fix crash in antispam code
2008-09-27T00:25:32 <ThomasWaldmann> dreimark: i regularly see zipfile warning when creating the language packages
2008-09-27T00:28:15 <dreimark> xorAxAx: py.test on a windows box seems to test attachments in underlay
2008-09-27T00:31:01 <ThomasWaldmann> dreimark: the sourcecode test?
2008-09-27T00:31:27 <dreimark> yes
2008-09-27T00:37:59 <ThomasWaldmann> that's same as on linux
2008-09-27T00:38:40 <dreimark> hmm I don't get this file tested on my installation
2008-09-27T00:39:02 <ThomasWaldmann> touch it
2008-09-27T00:43:47 <dreimark> no change
2008-09-27T00:57:28 <ThomasWaldmann> http://test19.moinmo.in
2008-09-27T01:09:13 <ThomasWaldmann> zenhase: file upload -> crash
2008-09-27T01:25:20 <dreimark> looks like windows XP does not know image/svg+xml
2008-09-27T01:25:38 <ThomasWaldmann> zenhase: AttributeError: 'AllContext' object has no attribute 'getBaseURL'
2008-09-27T01:42:00 <dreimark> gn
2008-09-27T10:56:31 <dreimark> moin
2008-09-27T11:45:33 <ThomasWaldmann> moin
2008-09-27T15:34:39 <CIA-25> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 4139:2bd88d32a0a1 1.8/ (4 files in 4 dirs): merged moin/1.7
2008-09-27T15:57:22 <ThomasWaldmann> byeongweon_linux: http://dev.fckeditor.net/ticket/2333 did you see that?
2008-09-27T16:04:17 <ThomasWaldmann> byeongweon_linux: fckeditor roadmap tells 10 days until 2.6.4, so we maybe can use that when it is ready.
2008-09-27T16:10:56 <byeongweon_gtx> ThomasWaldmann: ok. I will ready for that.
2008-09-27T16:12:18 <byeongweon_gtx> I was very busy this month about company and school problem. I'm sorry for didn't concentrate moin works.
2008-09-27T16:24:37 <ThomasWaldmann> no problem :)
2008-09-27T16:25:52 <ThomasWaldmann> would be nice if we could support opera and safari gui editing in 1.8
2008-09-27T16:37:33 <ThomasWaldmann> // login focus
2008-09-27T16:37:33 <ThomasWaldmann> if (document.forms['loginform']) {
2008-09-27T16:37:33 <ThomasWaldmann> document.forms['loginform'].elements['name'].focus();
2008-09-27T16:37:33 <ThomasWaldmann> }
2008-09-27T16:38:04 <ThomasWaldmann> i think we should have a more generic function to put the focus on some specific form element
2008-09-27T16:54:12 <CIA-25> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 4140:c1a7c986afa5 1.8/wiki/htdocs/common/js/common.js: some cosmetic changes to the javascript code (whitespace only)
2008-09-27T18:11:49 <dreimark> re
2008-09-27T18:12:07 <dreimark> btw that IE preview problem is in 1.7/1.8/1.9
2008-09-27T18:12:26 <dreimark> and I believe currently it has do something with sessions
2008-09-27T18:56:42 <grzywacz> Back to 15" :S
2008-09-27T20:09:10 * ThomasWaldmann still at 10"
2008-09-27T20:32:25 <dreimark> 24" rocks
2008-09-27T20:50:02 <ThomasWaldmann> dreimark: can you test a patch for the "IE hangs too long" bug?
2008-09-27T20:54:18 <dreimark> yes
2008-09-27T20:54:25 <dreimark> ThomasWaldmann:
2008-09-27T20:55:59 <ThomasWaldmann> dreimark: ok, then please start measuring some comparison values for a reference page with the current code
2008-09-27T20:56:07 <ThomasWaldmann> (put them onto that bug page)
2008-09-27T20:56:21 <ThomasWaldmann> IE, Opera, FF if possible
2008-09-27T20:56:30 <ThomasWaldmann> and moin 1.8
2008-09-27T21:00:58 <dreimark> ThomasWaldmann: http://forums.microsoft.com/MSDN/ShowPost.aspx?PostID=774350&SiteID=1
2008-09-27T21:01:14 <dreimark> in the middle search for <anonymousIdentification
2008-09-27T21:01:43 <dreimark> someone is explaining that ie7 needs domain property for session cookies
2008-09-27T21:02:28 <dreimark> can I try this if I set a cookie_domain in the config
2008-09-27T21:16:08 <ThomasWaldmann> dreimark: what of our bugs is this about?
2008-09-27T21:17:49 <dreimark> I think this is related to the preview one
2008-09-27T21:18:15 <ThomasWaldmann> i dont see the connection
2008-09-27T21:18:56 <ThomasWaldmann> we don't have a problem with being logged off, we have a problem with IE not displaying the POST response, sometimes
2008-09-27T21:20:31 <dreimark> if you step through the code you see the Preview correct and a step over the session code the page changes it content to the bug page
2008-09-27T21:21:36 <dreimark> so the preview works and then it is killed somehow
2008-09-27T21:21:42 <ThomasWaldmann> didn't you tell it happens when </html> is sent?
2008-09-27T21:22:21 <dreimark> I thought this before but haven't examined note enough today in the morning
2008-09-27T21:22:38 <dreimark> s/note/this not/
2008-09-27T21:23:08 <ThomasWaldmann> so when exactly does the page vanish?
2008-09-27T21:26:26 <ThomasWaldmann> note that </html> is the very last thing moin emits. i don't think anything happens in moin that should influence the client's display of the respone html after that.
2008-09-27T21:26:39 <dreimark> me thought this too
2008-09-27T21:27:03 <ThomasWaldmann> (the server maybe just closes the connection)
2008-09-27T21:27:30 <ThomasWaldmann> ok, can we try the hand thing now?
2008-09-27T21:27:31 <dreimark> I don't have the exact place yet but i can tell that I have a breakpoint in session.finish at the if
2008-09-27T21:27:36 <ThomasWaldmann> ehrm, HANG :)
2008-09-27T21:27:45 <dreimark> doing this already
2008-09-27T21:28:15 <dreimark> just want to tell if I continue from there the page is preview vanished
2008-09-27T21:28:28 <dreimark> I have no idea how we can influence ie
2008-09-27T21:28:47 <dreimark> s/is/in
2008-09-27T21:29:03 <dreimark> I need a stop watch
2008-09-27T21:30:48 <ThomasWaldmann> static.moinmo.in/temp/js.diff < some modification to the js, utilizing some getElementsByClass stuff I found on the web
2008-09-27T21:34:40 <ThomasWaldmann> dreimark: time read
2008-09-27T21:37:29 <dreimark> that is a real speed up for ie6
2008-09-27T21:37:41 <dreimark> and it fixes a bug in the eyes of ie6
2008-09-27T21:39:04 <ThomasWaldmann> can you put the results into a table on the bug page?
2008-09-27T21:49:02 <dreimark> done but I was wrong it looks like ie loads it fast but the next action takes long to be executed
2008-09-27T21:49:45 <dreimark> I thought I measure from loading until edit is executed
2008-09-27T21:52:13 <dreimark> ThomasWaldmann:
2008-09-27T21:52:54 <dreimark> on the local repo ie6 tells about an object failure in the html code which it doesn't show on MM
2008-09-27T21:57:55 <ThomasWaldmann> what are the timings in that table?
2008-09-27T21:58:42 <ThomasWaldmann> and how is it related to edit? the bug report is just about rendering the page.
2008-09-27T21:59:53 <dreimark> the bug report is about that the page is hanging
2008-09-27T22:00:47 <dreimark> how could I determine then if it does hang or not?
2008-09-27T22:01:12 <dreimark> you can point the cursor over the Edir link too
2008-09-27T22:01:23 <dreimark> and it does not change for 35 secs
2008-09-27T22:04:03 <dreimark> you don't get a new message shown in the status bar over that time while the browser already tells "Fertig"
2008-09-27T22:06:48 <ThomasWaldmann> hanging just means it takes very long. rendering time is until it does not hang any more.
2008-09-27T22:09:43 <ThomasWaldmann> (except if it hangs forever :)
2008-09-27T22:13:45 <dreimark> the problem is how to find out if it has rendered all and is ready for the next user request.
2008-09-27T22:14:05 <dreimark> I can't move the IE window in that time it looks frozen
2008-09-27T22:15:24 <ThomasWaldmann> maybe you could add some alert("finished!"); at the end of load() in common.js
2008-09-27T22:16:43 <dreimark> that patch you showed me is that added to MM ? I have done the tests there. This looks now I should do it on the standalone server
2008-09-27T22:17:02 <dreimark> +that
2008-09-27T22:20:44 <ThomasWaldmann> you need to do all tests locally, on same machine, with and without patch
2008-09-27T22:21:34 <ThomasWaldmann> (and the browser also need to run on same machine for the tests, otherwise you can't compare afterwards)
2008-09-27T22:22:02 <dreimark> alert does not work
2008-09-27T22:23:22 <dreimark> have done it before on localhost too but i can't test ie7 there
2008-09-27T22:23:46 <dreimark> need a bit of time
2008-09-27T22:41:55 <dreimark> :)
2008-09-27T22:47:00 <dreimark> ThomasWaldmann: http://moinmo.in/MoinMoinBugs/LargePagesCauseIEToHang/ReferenceTime
2008-09-27T22:51:05 <dreimark> the alert is shown shortly before the page
2008-09-27T23:07:33 <CIA-25> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 4141:5c88b27f31e5 1.8/wiki/htdocs/common/js/common.js: speed up javascript comments processing on IE by getElementsByClassName()
2008-09-27T23:07:33 <CIA-25> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 4142:5c29df15cd90 1.8/wiki/htdocs/common/js/common.js: some cosmetic changes to the javascript code (whitespace only)
MoinMoin: MoinMoinChat/Logs/moin-dev/2008-09-27 (last edited 2008-09-26 22:15:02 by IrcLogImporter)