Fri 29 Nov 2013 05:12:41 PM UTC, comment #11:
Yes it is possible.
|
Fri 29 Nov 2013 05:03:54 PM UTC, comment #10:
OK, but please confirm that you can enter the rsup tag by typeing '(dead)^ space'. This seems reasonable to me, if ibus does not report the individual keypresses for '(dead)^'. So unless you contradict me, I indeed do not consider this to be a bug inside TeXmacs.
|
Fri 29 Nov 2013 04:35:37 PM UTC, comment #9:
Note: the bug is reproducible with an azerty keymap.
> If I understand things correctly, ibus preempts the deadkey ^.
> In other words, typing '(dead)^ space' should produce ^,
> whereas '(dead))^ a' should produce â.
Exactly.
> Now what does Qt send as charcters to TeXmacs when trying to produce ^ and â in the above way?
TeXmacs] debug-keyboard, Pressed ^ at 30147
TeXmacs] debug-keyboard, ^ -> ^ -> ^
TeXmacs] debug-keyboard, Pressed â at 32091
TeXmacs] debug-keyboard, â -> â -> â
And Qt does not send anything if '(dead)^' or '(dead)^(dead)^' is typed.
> Which symbols do you manage to insert, and which characters do not work?
A 'rsup tag in math mode is no more entered via the dead hat.
|
Fri 29 Nov 2013 04:11:39 PM UTC, comment #8:
OK, so please give me some more info.
If I understand things correctly, ibus preempts the deadkey ^.
In other words, typing '(dead)^ space' should produce ^,
whereas '(dead))^ a' should produce â.
Now what does Qt send as charcters to TeXmacs when trying to produce ^ and â in the above way? Which symbols do you manage to insert, and which characters do not work?
|
Fri 29 Nov 2013 03:10:39 PM UTC, comment #7:
I can reproduce it. But for me it's an ibus bug. Let's close it.
|
Fri 29 Nov 2013 02:53:51 PM UTC, comment #6:
I cannot reproduce this bug under Ubuntu (assuming a French keyboard). I guess that it has been fixed in the meantime. On the other hand, I am not sure whether ^ is directly understood as a dead key (it seems to be simply a hat, which is neverthless rewritten as a dead hat at an appropriate moment).
|
Tue 16 Oct 2012 12:03:11 PM UTC, comment #5:
Another precision : the bug isn't effectively due to the activation of Ibus, but only the (needed by Ibus) declaration of the environment variable : QT_IM_MODULE="ibus".
|
Tue 16 Oct 2012 11:56:54 AM UTC, comment #4:
In fact, without Ibus, when I type 'â', I get on the debugger :
Keyboard] Pressed hat at 27640
] $ hat -> $ hat -> $ hat
] hat -> hat -> accent:deadhat
Keyboard] Pressed a at 28800
] hat a -> hat a -> accent:deadhat a
] a -> a -> a
When I use Ibus, I don't have anymore the information of composition via dead key. I get directly the glyph in utf8 :
Keyboard] Pressed � at 1448593
] space � -> space � -> space
] � -> � ->
(the � displayed is due to cork encoding, but it is in fact a Cork 'â')
I suppose the problem is due to Ibus which preempt the effect of the dead key, and doesn't transmit the information to Qt. So, if it is the case, I think this bug is not at all due to TeXmacs.
|
Tue 16 Oct 2012 11:50:11 AM UTC, comment #3:
I confirm that a part of the problem was due to Ibus. With a newer version of Ibus, this bug is less problematic.
From now, it is possible, when Ibus is activated, to get "hated" letters, like â. It is also possible, to get exponents in math mode. But this doesn't work with the "dead" hat key; this work this the "direct" '^' key (obtained by composition).
|
Mon 10 Sep 2012 09:33:53 AM UTC, comment #2:
I think this can also due to Ibus. I experienced a different behavior before and after having installed it.
See http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=591459
|
Fri 30 Mar 2012 12:41:46 PM UTC, comment #1:
This is almost surely the same problem as (or at least it's very closely related to) https://savannah.gnu.org/bugs/?36037
|
Fri 30 Mar 2012 12:28:23 PM UTC, original submission:
Step to reproduce (under debian) :
- install ibus ibus-clutter ibus-qt4 ibus-table
- launch TeXmacs with QT4_IM_MODULE=ibus
- test to enter $x^a$
---> the 'hat' shortcut has no effect.
|