ReleaseStatus Minutes 2008-01-28 IRC log

(14:59:56) ja_: hi
(15:00:02) Kai_A: huhu
(15:00:23) mdamboldt: Moin
(15:01:03) _Nesshof_: moin
(15:01:19) _Nesshof_: Kai_A: will you replace kai in this meeting ?
(15:01:43) Kai_A: _Nesshof_: Yes, I will.
(15:02:16) MechtiIde: hello
(15:02:24) Kai_A: A short introduction to me: I'm the software engineering manager (Sun) / ProjectLead (OOo) for the graphics area
(15:02:57) paveljanik: Hi
(15:03:11) _Nesshof_: ok, lets have a look at the 2.4 release status
(15:03:17) _Nesshof_: http://www.openoffice.org/issues/showdependencytree.cgi?id=84957
(15:03:19) Kai_A: In the Sun role, I also manage the Base (database) and extras team here in Hamburg
(15:04:14) _Nesshof_: issue 85457 doesn't seem to have an real owner yet
(15:04:21) IZBot: Spreadsheet DEFECT NEW P2 Crash opening .xls http://qa.openoffice.org/issues/show_bug.cgi?id=85457
(15:04:39) _Nesshof_: same for issue 85639
(15:04:44) IZBot: Database access DEFECT NEW P2 With ODBC-connection no data input possible http://qa.openoffice.org/issues/show_bug.cgi?id=85639
(15:04:55) ***_rene_ doesn't have Issue 85462 , but /me had .utf8 locales, not .UTF-8
(15:05:02) IZBot: framework DEFECT STARTED P2 Office refuses to start with UTF-8 locales http://qa.openoffice.org/issues/show_bug.cgi?id=85462
(15:05:02) _Nesshof_: and issue 85371
(15:05:07) IZBot: Database access DEFECT NEW P1 Base: Can't copy any table, query, form in database browser http://qa.openoffice.org/issues/show_bug.cgi?id=85371
(15:05:34) MechtiIde: _Nesshof_, I'm working on it to give an example DB
(15:05:37) _rene_: yeah, and this one is even worse
(15:05:53) _rene_: because any distri will have problems building a sane OOo
(15:06:05) _rene_: most people don't ship ancient flex anymore
(15:06:16) _rene_: see Issue 85508
(15:06:25) IZBot: Database access DEFECT NEW P3 Connecting to databases fails if flex 2.5.33 was used http://qa.openoffice.org/issues/show_bug.cgi?id=85508
(15:06:26) ***_rene_ happily has flex-old available, but...
(15:06:58) ***_rene_ thinks that should be a blocker...
(15:08:00) Kai_A: #85731: Frank still answers that this task ist not reproducable by him
(15:08:02) IZBot: no issue with number 85731
(15:08:20) Kai_A: #85371, of course
(15:08:26) IZBot: Database access DEFECT NEW P1 Base: Can't copy any table, query, form in database browser http://qa.openoffice.org/issues/show_bug.cgi?id=85371
(15:11:03) _Nesshof_: Kai_A: you argue that this is not a blocker because non-reproducable ?
(15:12:06) Kai_A: _Nesshof_: no, I don't, but that we need a valid test scenario as Mechtilde is already creating
(15:12:54) _rene_: he just updated the issue
(15:12:58) _rene_: "
(15:13:01) _rene_: bah
(15:13:11) _rene_: "Okay, I can reproduce the problem that the query cannot be opened in / switched
(15:13:14) _rene_: to SQL view. I change the summary accordingly, and target the issue to 2.4."
(15:13:25) MechtiIde: no it is very difficult to create it
(15:13:40) MechtiIde: the hsql example is wrong sorry for that noise
(15:15:03) _Nesshof_: Kai_A: do we have an estimation for the flex fix ?
(15:16:15) Kai_A: _Nesshof_: No, not at all...
(15:16:55) _Nesshof_: Kai_A: if that fix is not too complicated I would like to take it for 2.4
(15:17:40) _Nesshof_: but if there is too much effort and/or riskI would target 3.0 because there is a workaround: use old flex for release
(15:17:47) _Nesshof_: _rene_: :-(
(15:17:56) MechtiIde: Kai_A, if someone from the base-team has a little bit time to discuss the problem with me in German maybe on #dba.openoffice.org-de
(15:17:59) _rene_: _Nesshof_: the problem is that many linus distris probla yo not have the ancient flex anymore
(15:18:08) _rene_: _Nesshof_: and therefore can't use it for building
(15:18:16) _rene_: _Nesshof_: so also with Mac OS
(15:19:04) _rene_: _Nesshof_: *anyone* who builds on a current system gets serious problems... Note that flex 2.5.3x is years old already, so it's not even "new" flex
(15:19:06) _Nesshof_: if they are able to build OOo, they also should be able to build and use an old flex version for this
(15:19:38) _rene_: so they need to build an own flex in their package building scripts before the build OOo? come on...
(15:19:51) _rene_: that's not really practicable...
(15:20:12) Kai_A: Frank and Ocke will take a closer look, but Frank mentions that it most likely seems to be a flex bug
(15:20:24) _Nesshof_: Kai_A: and regression
(15:20:42) _rene_: Kai_A: yes, but it did work in 2.3.1 *with* the "new" flex
(15:20:53) ***_Nesshof_ agrees that this is not very practical, but ...
(15:20:56) xiuzhi [n=Administ@125.96.8.152] hat den Raum betreten.
(15:21:22) Kai_A: Frank agreed to get in contact with Mechtilde by PM
(15:21:51) _rene_: Kai_A: so even if it was a flex bug, it did work in 2.3.1 and some 2.4 change broke it
(15:22:33) Kai_A: So, they should fix it.
(15:22:54) _rene_: yes, the OOo code should be fixed. my point :)
(15:23:05) Fridrich: +1
(15:23:08) _rene_: since a change there caused it :)
(15:23:29) _Nesshof_: lets wait for the effort esstimation
(15:23:30) Kai_A: agreed
(15:24:14) Fridrich: _Nesshof_: the problem is that changing drastically build requirements is not the best practice. Unless one does not care :(
(15:24:24) _Nesshof_: I will continue to retarget the remaining issues to a later target
(15:24:30) _Nesshof_: Fridrich: agreed
(15:24:45) _rene_: ok, next thing: jsc20 remove dthe DevGuide from the tree and (supposedly) replaced the links to the wiki
(15:25:09) _rene_: still; there are files which reference e.g DevelopersGuide/DevelopersGuide.xhtml from the local tree, which of course are not existing anymore
(15:25:20) _rene_: e.g. affected: DevelopersGuide_into.html, the starting point.
(15:25:25) _rene_: imho should be fixed for 2.4
(15:25:41) _Nesshof_: _rene_: is there an issue for this ?
(15:25:47) _rene_: yes, http://www.openoffice.org/issues/show_bug.cgi?id=82046 (my comment to that topic ignored)
(15:25:54) IZBot: issue 82046: sdk TASK REOPENED P3 remove Developer's guide from SDK and replce links into the wiki
(15:28:48) _Nesshof_: _rene_: does the issue contain all references which need to be modified
(15:28:53) _rene_: no
(15:29:11) _rene_: I just saw _intro.html and stopped.
(15:29:19) _rene_: (the examples/ one was wrong)
(15:29:46) _Nesshof_: _rene_: we should know about how many necessary changes we are speaking
(15:29:52) _rene_: _Nesshof_: I have no time to look over the whole tree, but for jsc or someone who knows the internals it should be quick and he should know :)
(15:30:01) _rene_: _Nesshof_: but as he didn't even comment...
(15:30:41) _rene_: _Nesshof_: well, it should be able by some perl running anyway I think (don't know the exact new URL format, though) :)
(15:31:24) _rene_: what I *do* know is that it looks *extremely* bad if the front page of the SDK contains of a TOC which just consists of non-working links :)
(15:31:46) _Nesshof_: yes
(15:32:31) _Nesshof_: the intro file seems to be the only one
(15:32:57) _Nesshof_: I will try to take care of this one
(15:33:49) _rene_: ok, thanks
(15:34:10) _rene_: then we have the ICU security fix which should be merged (erack brought it up on releases@):
(15:34:14) _rene_: http://www.openoffice.org/issues/show_bug.cgi?id=85632
(15:34:19) IZBot: issue 85632: utilities PATCH STARTED P3 Apply ICU regex patch
(15:35:08) _Nesshof_: _rene_: since this one is already assign to er I think this will go its way
(15:35:29) _rene_: _Nesshof_: yes, but it shouldn't be overseen
(15:35:38) _rene_: (I am just going sure...)
(15:35:43) _Nesshof_: :-)
(15:36:14) _rene_: and last but not least for me (at least for today):
(15:36:15) _rene_: http://www.openoffice.org/issues/show_bug.cgi?id=83670
(15:36:20) IZBot: issue 83670: Word processor DEFECT CLOSED WORKSFORME P2 [CWS langstatusbar] Letter wizard not working and office can not be closed normally thereafter
(15:36:37) ja_: closed worksforme ?
(15:36:44) _rene_: I can reproduce this for my builds at will, and it's *not* a timing issue (it can't be given the error message)
(15:37:51) _rene_: and even cd admits he was able to reproduce it at one point...
(15:38:13) _rene_: I don't think we should ignore this (which might happen if we leave it at its current state...)
(15:38:16) _Nesshof_: _rene_: then you should reopen the issue
(15:38:41) _rene_: wrote a comment a few mins ago, waiting for a reaction from him...
(15:38:49) _rene_: but feel free to reopen it, too :)
(15:38:59) _Nesshof_: overall, I think we are on track for doing a release candidate after last cws integration after Feb 7th
(15:39:36) _rene_: ja_, _Nesshof_: ok, I'll reopen it :)
(15:40:07) ja_: _rene_: thanks
(15:40:34) _rene_: done
(15:41:02) rtimm: FYI, builds this week: OOH680 m6 will be done by Oliver (obo), SRC680 m245 by me (rt)
(15:41:56) _Nesshof_: ok
(15:42:21) _rene_: ok, I forgot something
(15:42:25) _Nesshof_: anything else for today ?
(15:42:44) _rene_: hyphenexternal *STILL* need windows people (or noone told me he did windows work, but I don't see commits either)
(15:43:14) _rene_: and it fixed the hyph-en-us-has-no-license-thus-its-all-rights-reserved-so-no-distributable issue
(15:43:32) _rene_: what is with the people who were promised me to help me there?
(15:44:10) _Nesshof_: sorry, I didn't found a vlunteer until now
(15:48:01) ja_: .
(15:48:09) _rene_: alternative: someone please give me the tools I need to try (although I fear I won't succed, no windows build knowledge) it, yes, I know you can get it from M$, but I do not want to register there :)
(15:49:03) _Nesshof_: _rene_: what about a windows build training during your next vist in Hamburg in three weeks ?
(15:49:51) _Nesshof_: ok, by
(15:49:57) _rene_: if we've time, why not?
(15:50:24) _rene_: (my flight goes at 20:00 on Tue, though)
(15:52:02) Kai_A hat den Raum verlassen (quit: "ChatZilla 0.9.79 [Firefox 2.0.0.11/2007120410]").
(15:52:12) ja_: bye
(15:52:23) UweL: bye
(15:52:27) UweL hat den Raum verlassen (quit: "ChatZilla 0.9.77 [Firefox 1.5.0.12/2007050813]").