From Fedora Project Wiki

previous meeting | next meeting

Agenda

Time:2009-11-25 0500 UTC

Place: #fedora-i18n at Freenode

Summary

  • reviewed ibus buglists

Actions

  • move long-term ibus future RFE bugs to upstream issues

Log

juhp shall we do the IM meeting? Nov 25 15:06
juhp ping tagoh3, fujiwarat, phuang, paragan Nov 25 15:07
paragan pong juhp Nov 25 15:07
tagoh3 hi Nov 25 15:07
phuang pong Nov 25 15:07
juhp shall we do a catchup on IM issues? Nov 25 15:07
juhp I don't have any particular agenda... and forgot to update the meeting page Nov 25 15:08
fujiwarat hi Nov 25 15:08
juhp tagoh3: btw what does your imsettings update do? didn't have a chance to look yet Nov 25 15:09
juhp not sure but think dychen might have finished makerdialog? Nov 25 15:09
tagoh3 juhp: including a fix of crash on XFCE Nov 25 15:09
juhp aha Nov 25 15:09
dychen juhp: Yes, I am checking the package. Nov 25 15:10
juhp cool Nov 25 15:10
juhp fujiwarat: see your new anthy icon Nov 25 15:11
fujiwarat phuang: probably I'll need a new branch on google for fedora12 for icon enhancement. I'm not sure if we need to bump to 1.3. Nov 25 15:11
juhp I was trying to remember what we talked about at the meeting Nov 25 15:11
juhp fujiwarat: google code? Nov 25 15:11
juhp fujiwarat: I guess the new icon is not ideal as underlay for status say though Nov 25 15:12
phuang fujiwarat, If you want, I think you could bump ibus-anthy version separately. Nov 25 15:12
juhp fujiwarat: you want to branch 1.2 you mean? Nov 25 15:13
fujiwarat juhp: I mean trunk on ibus community github.com. Oh not google. Nov 25 15:13
juhp yeah Nov 25 15:13
juhp ibus branch for ui work? Nov 25 15:13
juhp fujiwarat: I think you can do it in your tree? Nov 25 15:14
fujiwarat phuang: OK, I'll create 1.3 for ibus enhancement and glade to gtkbuilder. Nov 25 15:14
phuang fujiwarat, ibus or ibus-anthy? Nov 25 15:14
fujiwarat I mean currently almost ibus module version is 1.2.x. Nov 25 15:14
fujiwarat phuang: i mean ibus-anthy. Nov 25 15:15
juhp fujiwarat: not sure what you want - think you can worry about version later?? Nov 25 15:15
phuang fujiwarat, ok Nov 25 15:15
juhp fujiwarat: seems too much to bump major version just for icon - depends on your version scheme I guess? Nov 25 15:16
juhp anyway... Nov 25 15:16
juhp hmm Nov 25 15:17
juhp tagoh3: did you see the anthy icon? Nov 25 15:17
tagoh3 juhp: yep Nov 25 15:17
juhp fujiwarat: another comment - y is off the key but guess it is ok maybe Nov 25 15:18
juhp bottom of y Nov 25 15:18
juhp (y of anthy to clarify) Nov 25 15:18
tagoh3 looks good - as you said it may be not sufficient for the status icon right. but I thought we are trying to replace the crown icon yes? so it should works enough I think. Nov 25 15:18
juhp okay Nov 25 15:18
juhp let's try it in rawhide then Nov 25 15:19
fujiwarat juhp: I imply "Anthy" string is not drawn with the key image. Nov 25 15:19
juhp (I requested imsettings-0.107.4-3.fc12 to stable) Nov 25 15:19
juhp fujiwarat: ok I see Nov 25 15:20
juhp fujiwarat: it might be the best anthy icon so far Nov 25 15:20
juhp I might still prefer ibus-off icon + あ (or status) later Nov 25 15:21
fujiwarat Today I also sent hidegit the icon. Nov 25 15:21
juhp like pidgin or some other notification icons Nov 25 15:21
juhp aha Nov 25 15:21
juhp any feedback? Nov 25 15:21
juhp you made it yourself? :) Nov 25 15:22
fujiwarat No just 30 minutes ago :). Nov 25 15:22
juhp ok Nov 25 15:22
juhp cool Nov 25 15:22
fujiwarat juhp: If other icons could be changed to ibus-off + something, I'll back to that design. If not, it might be a little strange for me when ibus-anthy only uses ibus-off. Nov 25 15:25
juhp fujiwarat: of course they should all be changed :) Nov 25 15:25
juhp which why I asked about ibus tree :) Nov 25 15:25
-->pravins (n=psatpute@nat/redhat/session) has joined #fedora-i18n Nov 25 15:26
fujiwarat juhp: OK, I understood. Nov 25 15:26
juhp fujiwarat: I think you could just add the icon in fedora build for now even Nov 25 15:26
juhp I18N/InputMethods/Bugs Nov 25 15:27
fujiwarat juhp: OK, I'll keep 1.2. Nov 25 15:28
juhp read bug 530976 Nov 25 15:30
juhp ^I Nov 25 15:30
juhp bug 530976 Nov 25 15:30
juhp .bug 530976 Nov 25 15:30
zodbot juhp: Bug 530976 [Indic] Default keymap need to set for various languages - https://bugzilla.redhat.com/show_bug.cgi?id=530976 Nov 25 15:30
phuang this bug has been fixed Nov 25 15:31
juhp so I see Nov 25 15:32
juhp was curious Nov 25 15:32
juhp launguaes <- pretty good misspelling ;) Nov 25 15:33
juhp I am looking the IME list first... Nov 25 15:33
paragan .bug 538270 Nov 25 15:33
zodbot paragan: Bug 538270 Typo in ibus-m17n.spec - https://bugzilla.redhat.com/show_bug.cgi?id=538270 Nov 25 15:33
juhp yes Nov 25 15:34
paragan can be fixed in next release Nov 25 15:34
paragan or phuang can even commit it in spec only for now Nov 25 15:34
juhp was about to... Nov 25 15:34
juhp done Nov 25 15:40
juhp let's fix the trivial bugs early... Nov 25 15:40
juhp fujiwarat: still a few ibus-anthy error bugs Nov 25 15:43
fujiwarat juhp: really? Nov 25 15:43
juhp in bugzilla yes Nov 25 15:43
juhp fujiwarat: would you like bug 504735 ? Nov 25 15:45
juhp looking at ibus bugs now Nov 25 15:45
juhp .bug 504735 Nov 25 15:45
zodbot juhp: Bug 504735 reflect IME toolbar UI in systray applet menu - https://bugzilla.redhat.com/show_bug.cgi?id=504735 Nov 25 15:45
juhp hmm menu? Nov 25 15:45
juhp ah that is different Nov 25 15:46
juhp thought it was status in panel Nov 25 15:46
juhp phuang: planning to work on any ibus bugs? Nov 25 15:47
phuang juhp, yeah. Nov 25 15:47
-->krishnababu (n=kkrothap@nat/redhat/session) has joined #fedora-i18n Nov 25 15:47
phuang I am fixing some abrt crash problem Nov 25 15:47
juhp aha Nov 25 15:48
juhp .bug 529920: "I just hit Ctrl-space accidentally (that happens uncomfortably often to me...)" Nov 25 15:49
zodbot juhp: Error: '529920:' is not a valid integer. Nov 25 15:49
juhp .bug 529920 : "I just hit Ctrl-space accidentally (that happens uncomfortably often to me...)" Nov 25 15:49
zodbot juhp: (bug <number>) -- Return the name and URL of a Red Hat Bugzilla ticket. Nov 25 15:49
juhp .bug 529920 Nov 25 15:49
zodbot juhp: Bug 529920 language panel pops up on the wrong monitor - https://bugzilla.redhat.com/show_bug.cgi?id=529920 Nov 25 15:49
juhp clever bot Nov 25 15:49
phuang Maybe he used extension mode for second monitor. Nov 25 15:53
juhp extension? Nov 25 15:53
juhp phuang: I say it on my laptop too during meeting Nov 25 15:53
juhp saw Nov 25 15:53
juhp and also on this box... Nov 25 15:54
juhp phuang: you can reproduce? Nov 25 15:54
juhp can't Nov 25 15:54
phuang ibus get the desktop size from window manager, and put the panel on the right bottom Nov 25 15:55
juhp I noticed something I hadn't realised while testing ru kbd setting from gdm Nov 25 15:56
phuang probably window manager consider the two monitor as one big desktop Nov 25 15:56
juhp phuang: heh Nov 25 15:56
juhp hmm Nov 25 15:56
juhp phuang: can't it ask for bottom of screen? Nov 25 15:56
phuang So maybe ibus should not use desktop size Nov 25 15:57
juhp right Nov 25 15:57
juhp it shouldn't I guess Nov 25 15:57
juhp for multiscreen Nov 25 15:57
phuang Need some research on it Nov 25 15:57
juhp ok Nov 25 15:57
juhp matthias might be able to give some clues on it Nov 25 15:57
phuang ok Nov 25 15:58
juhp there are various RFEs that we have had lying around too long... we should open issues upstream and close the fedora bugs Nov 25 15:58
juhp I guess Nov 25 15:59
phuang yeah Nov 25 15:59
juhp let's try to do that before next time Nov 25 15:59
paragan hmm Nov 25 15:59
juhp ? Nov 25 15:59
juhp talking about FutureFeatures Nov 25 16:00
juhp kind of things Nov 25 16:00
paragan yes Nov 25 16:00
>paragan<sounds like disagreement :) Nov 25 16:00
>paragan<sorry "hmm" sounds like disagreement :) Nov 25 16:01
paragan we need to work on RFEs Nov 25 16:01
juhp .bug 532405 Nov 25 16:02
zodbot juhp: Bug 532405 ibus has no cursors-following-up in QT-based applications - https://bugzilla.redhat.com/show_bug.cgi?id=532405 Nov 25 16:02
juhp paragan: yes Nov 25 16:02
juhp tagoh3: want to talk about any imsettings stuff? Nov 25 16:03
tagoh3 not particularly Nov 25 16:03
paragan juhp, I too have observed above bug Nov 25 16:03
juhp what is it? Nov 25 16:03
paragan I have given my observations here https://bugzilla.redhat.com/show_bug.cgi?id=532405#c3 Nov 25 16:04
juhp .bug 504735 Nov 25 16:04
zodbot juhp: Bug 504735 reflect IME toolbar UI in systray applet menu - https://bugzilla.redhat.com/show_bug.cgi?id=504735 Nov 25 16:04
juhp eg bugs like this? Nov 25 16:04
paragan looks like need to fix in ibus-qt I guess Nov 25 16:04
*juhp looks at screenshot Nov 25 16:04
juhp was referring to FutureFeatures Nov 25 16:05
paragan ok Nov 25 16:05
juhp completely don't understand "no cursors-following-up" Nov 25 16:06
paragan have you looked at screenshot Nov 25 16:06
juhp 光标跟随? Nov 25 16:06
tagoh3 me neither - it looks good though? Nov 25 16:06
juhp wtf? excuse my english! Nov 25 16:06
paragan I guess reporter want to say move cursor only when character is committed from candidate window Nov 25 16:07
juhp phuang: ? Nov 25 16:07
paragan oops! phuang closed that bug as NOTABUG Nov 25 16:07
juhp oh Nov 25 16:07
juhp paragan: I think your comment is different - bug is about movement of candidate window? Nov 25 16:08
tagoh3 paragan: hmm, from the screenshot you took, didn't this issue also happen on kwrite no? Nov 25 16:09
juhp phuang: caspar was using xim? Nov 25 16:09
paragan What behaviour I can see using ml-swanlekha.mim keymap in gedit is not same in kwrite Nov 25 16:09
phuang juhp, yeah Nov 25 16:09
*juhp is a bit lost Nov 25 16:09
juhp phuang: could you please change the summary Nov 25 16:09
phuang ok Nov 25 16:10
juhp it is complete nonsense, sorry Nov 25 16:10
phuang juhp, qt4 does set cursor position via xim Nov 25 16:10
juhp cool Nov 25 16:10
juhp paragan, phuang: so what is indic issue in kwrite? Nov 25 16:11
juhp paragan: I think better to file a separate bug Nov 25 16:11
paragan juhp, correct bug is movement of candidate window but another side effect of same bug is that screenshot Nov 25 16:11
juhp phuang: czhang is in QA? Nov 25 16:11
paragan juhp, ok sorry for confusing comment in bug Nov 25 16:12
juhp dunno Nov 25 16:12
juhp I think the bug is so confused so better to start new one but maybe it is just me :) Nov 25 16:12
phuang :) Nov 25 16:13
tagoh3 that would be good to attach another screenshot, which is correct on gedit say. Nov 25 16:13
juhp ok so original report is for with ibus-qt Nov 25 16:13
juhp erm Without Nov 25 16:13
juhp ibus-qt fixed it except opera... => NOTABUG Nov 25 16:14
juhp and you guys are seeing something similar (but different?) in kwrite Nov 25 16:14
juhp what is my sumamry anyway Nov 25 16:14
juhp that Nov 25 16:14
juhp tagoh3: yet Nov 25 16:14
juhp yep Nov 25 16:14
juhp or you could hijack the bug now that it is closed ;) Nov 25 16:15
juhp I suggest a new - clone and edit if you wish Nov 25 16:15
phuang paragan, in kwrite, it is correct Nov 25 16:15
juhp paragan: Nov 25 16:15
phuang paragan, Some qt4 applications do not set correct position of cursor in preedit Nov 25 16:16
juhp aha Nov 25 16:16
phuang paragan, So I have to put the candidates window in the beginning of the preedit Nov 25 16:16
paragan but in gedit character selected is immediately committed and can be seen Nov 25 16:16
paragan phuang, ooh! I see Nov 25 16:17
juhp paragan: so it should not be using preedit?? Nov 25 16:17
juhp anyway we should probably finish soon - I think we are over time Nov 25 16:17
juhp anything else? Nov 25 16:18
tagoh3 phuang: that's why gtk+ gave up to use the over-the-spot but moving to on-the-spot when gtk2 was out. Nov 25 16:19
juhp can qt use on-the-spot? Nov 25 16:20
juhp can't Nov 25 16:20
juhp Releases/13/Schedule Nov 25 16:20
tagoh3 juhp: they can. dunno what's the default in qt Nov 25 16:20
juhp by default I mean yes Nov 25 16:20
phuang tagoh, you mean qt4 Nov 25 16:20
juhp 2010-01-12 Feature Submission Deadline Nov 25 16:21
juhp 2010-01-26Feature Freeze--Planning & Development Ends Nov 25 16:21
juhp 2010-02-02 Alpha Freeze Nov 25 16:21
juhp tagoh3: so that is decided by qt not the immodule? Nov 25 16:21
juhp or the toolkit rather Nov 25 16:22
tagoh3 juhp: should be in the xim immodule. Nov 25 16:22
juhp phuang: so maybe we should rfe qt4 to use on-the-spot but default? Nov 25 16:22
tagoh3 thought it's configurable in qtconfig Nov 25 16:22
tagoh3 though Nov 25 16:22
juhp yep Nov 25 16:22
juhp tagoh3: and what about other immodules? Nov 25 16:23
phuang Actually, some qt4 applications do not set the correct cursor position Nov 25 16:23
juhp phuang: is it reported? Nov 25 16:23
phuang yeah Nov 25 16:23
juhp ok Nov 25 16:23
phuang I think I reported in kde mail list Nov 25 16:23
phuang Some applications set the coordinate (x,y) of the cursor Nov 25 16:24
phuang some applications set the index in the preedit Nov 25 16:24
tagoh3 juhp: dunno. no detailed styles in immodules afaik Nov 25 16:24
tagoh3 aren't we talking about xim in qt? Nov 25 16:25
juhp ibus-qt Nov 25 16:25
juhp I thought Nov 25 16:26
<--noriko has quit (Read error: 104 (Connection reset by peer)) Nov 25 16:26
juhp phuang: so what does ibus-qt use? Nov 25 16:26
tagoh3 ibus-qt has the fix but not for Opera because it links qt4 library statically, thus NOTABUG. no? Nov 25 16:26
juhp okay Nov 25 16:27
juhp gotcha Nov 25 16:27
-->noriko_ (n=noriko@d58-106-140-120.sbr4.nsw.optusnet.com.au) has joined #fedora-i18n Nov 25 16:27
juhp me finds *-the-spot confusing Nov 25 16:27
tagoh3 unless I'm missing the context. Nov 25 16:27
-->jassy (n=jassy__@nat/redhat/session) has joined #fedora-i18n Nov 25 16:27
juhp so it is just the misplacement of kwrite then? Nov 25 16:27
juhp phuang: so it is kde bug right? Nov 25 16:28
phuang I think we are talking about different things Nov 25 16:28
juhp that is what I was wondering... Nov 25 16:28
tagoh3 paragan: is that screenshot with ibus-qt or without ibus-qt? Nov 25 16:28
juhp okay let's continue talking but close the log here if that's ok Nov 25 16:31
paragan tagoh3, with ibus-qt if I remember correctly Nov 25 16:31
tagoh3 aha Nov 25 16:32
paragan brb Nov 25 16:32
phuang Qt::ImCursorPosition2The logical position of the cursor within the text surrounding the input area (see ImSurroundingText). If any text is selected, the position returned will be at the logical end of the selection, even if the real cursor is located at the logical start. Nov 25 16:32
phuang ibus-qt use it to query the cursor position Nov 25 16:32
phuang Some return cursor index in preedit Nov 25 16:33
phuang some return coordinate in screen Nov 25 16:33
juhp hmm Nov 25 16:33
juhp surroundingtext? Nov 25 16:33
juhp nm Nov 25 16:34
-->noriko_1 (n=noriko@58.106.140.120) has joined #fedora-i18n Nov 25 16:35
<--noriko_1 has quit (Read error: 131 (Connection reset by peer)) Nov 25 16:36
-->noriko_1 (n=noriko@d58-106-140-120.sbr4.nsw.optusnet.com.au) has joined #fedora-i18n Nov 25 16:36
phuang http://bugs.kde.org/show_bug.cgi?id=164428 Nov 25 16:36
tagoh3 phuang: roughly looking at uim, they use the different way to determine the cursor position. it may be worth have a look? Nov 25 16:37
juhp aha Nov 25 16:38
dychen tagoh3: ibus-qt works with qt4-static-linked opera but not shared linked. Nov 25 16:38
juhp hm Nov 25 16:38
phuang dychen, ? Nov 25 16:39
tagoh3 dychen: aha. I was just repeating a comment at bz :) Nov 25 16:39
dychen phuang: Didn't we test it a few weeks ago? Nov 25 16:40
phuang phuang, I think ibus-qt should work with shared version Nov 25 16:40
phuang s/phuang/dychen Nov 25 16:40
dychen But anyway the default (the one that end-users will like be downloaded is share-qt3-build. Nov 25 16:41
juhp which we don't support with ibus immodule... Nov 25 16:42
juhp but getting OT Nov 25 16:42
juhp ok thanks guys! Nov 25 16:42
juhp #meeting closed Nov 25 16:42

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!