[Pkg-ime-devel] Bug#645729: Bug#645729: Bug#645729: checking for ibus support in gtk2 and gtk3 separately

[LI Daobing]

I think the correct patch is as following, and I don't know whethre
this patch works in multiarch env.


$ git diff
diff --git a/debian/xinput/ibus b/debian/xinput/ibus
index e2fe0ef..49c78ba 100644
--- a/debian/xinput/ibus
+++ b/debian/xinput/ibus
@@ -1,7 +1,9 @@
 XIM=ibus
 XIM_PROGRAM=/usr/bin/ibus-daemon
 XIM_ARGS="--xim"
-if [ -e /usr/lib/gtk-*.0/*/immodules/im-ibus.so ]; then
+if [ -e /usr/lib/gtk-3.0/*/immodules/im-ibus.so ] \
+   && [ -e /usr/lib/gtk-2.0/*/immodules/im-ibus.so ]
+then
     GTK_IM_MODULE=ibus
 else
     GTK_IM_MODULE=xim


On Tue, Oct 25, 2011 at 01:23, Steve Langasek
<steve.langasek at canonical.com> wrote:
> Hi there,
>
> On Sun, Oct 23, 2011 at 10:24:44PM +0900, Osamu Aoki wrote:
>> Thanks for fix of multi-arch for im-switch.  I am about to fix in
>> im-config package side.  (Since im-switch is not used for ibus now, I do
>> not know use of this as is but if we make changes as below, it is useful
>> for wheezy.)
>
> Ah; in Ubuntu, im-switch is used by default instead of im-config.  I don't
> know why this is, I would have to dig into the package history to see.
>
>> > If only one of ibus-gtk or ibus-gtk3 is installed, I don't know what the
>> > desired behavior would be here, but I think we should fall back to xim in
>> > that case because it's safer.
>
>> This is very deep question.
>
>> But practically thinking, control file needs update to fix this problem:
>
>> Now we have:
>> Depends: ${shlibs:Depends}, ${misc:Depends}, ${python:Depends},
>>  python-ibus (= ${source:Version}), python-xdg, librsvg2-common, python-notify, im-config
>> Recommends: ibus-gtk | ibus-qt4 | ibus-clutter
>> Conflicts: ibus-anthy (<< 1.2),
>>  ibus-table (<< 1.2),
>>  ibus-pinyin (<< 1.2.99),
>>  ibus-m17n (<< 1.2.99),
>>  ibus-hangul (<< 1.2.99),
>>  ibus-chewing (<< 1.2.99)
>
>> I think correct thing to do is:
>> Depends: ${shlibs:Depends}, ${misc:Depends}, ${python:Depends},
>>  python-ibus (= ${source:Version}), python-xdg, librsvg2-common, python-notify
>> Recommends: ibus-gtk3, ibus-gtk, ibus-qt4, ibus-clutter, im-config
>> Conflicts: ibus-anthy (<< 1.2),
>>  ibus-table (<< 1.2),
>>  ibus-pinyin (<< 1.2.99),
>>  ibus-m17n (<< 1.2.99),
>>  ibus-hangul (<< 1.2.99),
>>  ibus-chewing (<< 1.2.99)
>
>> So they all get installed as defualt, if admin choses not to do so, he
>> should deal with it.
>
> That makes sense to me.
>
> On Sun, Oct 23, 2011 at 11:41:05PM +0900, Osamu Aoki wrote:
>> On Sun, Oct 23, 2011 at 09:52:56PM +0800, Aron Xu wrote:
>> > We tried to workaround it in Fcitx by recommending GTK2 IM Module in
>> > the GTK3 package, and do the inverse in the GTK2 package. In this
>> > case, users are very probably getting both IM Modules installed at the
>> > same time, so they can avoid weird misbehavior of installing only one
>> > of the two.
>
>> Is having dependency loop OK for recommend?  I think this is bad idea.
>> See "the circular dependency loop" in recent policy:
>
> Yes, circular recommends are allowed.  It's even allowed for p1 to depend on
> p2 while p2 recommends p1.
>
> The restriction on circular dependencies is because the package manager
> doesn't know which order to configure the packages.  But a Recommends does
> not affect package configuration.
>
> Cheers,
> --
> Steve Langasek                   Give me a lever long enough and a Free OS
> Debian Developer                   to set it on, and I can move the world.
> Ubuntu Developer                                    http://www.debian.org/
> slangasek at ubuntu.com                                     vorlon at debian.org
>
> _______________________________________________
> Pkg-ime-devel mailing list
> Pkg-ime-devel at lists.alioth.debian.org
> http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/pkg-ime-devel
>



-- 
Best Regards
LI Daobing