Am 16.12.2010 um 09:47 schrieb Abdelrazak Younes:
On 12/16/2010 07:25 AM, Stephan Witt wrote:
Am 16.12.2010 um 00:44 schrieb Pavel Sanda:
Stephan Witt wrote:
Index: src/frontends/Clipboard.h
===
--- src/frontends
Am 19.12.2010 um 20:50 schrieb Peter Baumgartner:
2.) I couldn't bring to work the new spell-checking on the fly feature ...
I solved this problem: In previous versions I used to use Aspell; so I had
now to change the spellchecker engine from Aspell to Native.
Ok, but LyX2.0 supports Aspell
I'm trying to fix 7081 currently.
http://www.lyx.org/trac/ticket/7081
The attached patch does it to some extend.
The idea is to skip drawing the misspelled marker
when at cursor position. It's a little bit more work
as it looks at the first glance because some misspelled
words are drawn in
Am 20.12.2010 um 14:19 schrieb Vincent van Ravesteijn:
On Mon, Dec 20, 2010 at 2:06 PM, Pavel Sanda sa...@lyx.org wrote:
Stephan Witt wrote:
Question: what's the best way to solve that?
Should the cursor shift operation redraw the current row unconditionally?
if we have to pay next redraw
Am 21.12.2010 um 00:48 schrieb Pavel Sanda:
hi,
time to think about beta3.
there is already a couple of things which need to be released
and tested so we should start freezing at certain time.
are there some particular issues you want to push in trunk
before next beta?
As you know, I
... I get here.
* CompileC src/mathed/MathData.cpp normal i386 c++ com.apple.compilers.gcc.4_2
src/mathed/MathData.cpp: In member function 'void
lyx::MathData::detachMacroParameters(lyx::DocIterator*, size_t)':
src/mathed/MathData.cpp:508: warning: negative integer implicitly converted to
Am 20.12.2010 um 15:10 schrieb Pavel Sanda:
Stephan Witt wrote:
Question: what's the best way to solve that?
Should the cursor shift operation redraw the current row unconditionally?
if we have to pay next redraw on cursor movement for the sake of this bug
i vote for wontfix. its damn hard
When in pedantic mode the compiler issues a warning for hexToInt() in
src/support/lstrings.cpp.
It doesn't like passing an integer to sscanf when format is %x - this is for
conversion of a pointer.
There are two to fix that:
1) the C-way - use of strtol()
2) the C++-way - using a
Am 21.12.2010 um 18:00 schrieb Bernhard Roider:
Just a guess: Does unsignd int work with scanf?
Yes, of course.
% man sscanf
...
The following conversions are available:
...
x, X Matches an optionally signed hexadecimal integer; the next pointer
must be a pointer to unsigned int.
Am 22.12.2010 um 01:53 schrieb Pavel Sanda:
Stephan Witt wrote:
What's your opinion regarding the SystemCall patch?
I have it at hand. The idea was to be able to collect the
standard and error output of external startscript().
thought i already commented on this...
the patch itself
Am 22.12.2010 um 01:59 schrieb Pavel Sanda:
Stephan Witt wrote:
When in pedantic mode the compiler issues a warning for hexToInt() in
src/support/lstrings.cpp.
It doesn't like passing an integer to sscanf when format is %x - this is
for conversion of a pointer.
unless it fixes some
Am 22.12.2010 um 21:44 schrieb Bernhard Roider:
But if you want to avoid the warning you can use an unsigned for scanf and do
a type cast to int
Yes, one can do that.
The problem I have with it isn't the warning.
The compiler warning was a hint to think about that piece of code.
The only
Am 27.12.2010 um 00:33 schrieb sa...@lyx.org:
Author: sanda
Date: Mon Dec 27 00:33:23 2010
New Revision: 36999
URL: http://www.lyx.org/trac/changeset/36999
Log:
Resave spellcheck ui in older designer
Modified:
lyx-devel/trunk/src/frontends/qt4/ui/SpellcheckerUi.ui
Modified:
Am 27.12.2010 um 12:05 schrieb Pavel Sanda:
Stephan Witt wrote:
What's the difference between the versions?
I can see only whitespace and in the order of xml attributes...
hopefully no real difference. the point is to prepare it for clear diff.
if you dig often in commit history it is hard
Am 27.12.2010 um 17:00 schrieb John McCabe-Dansted:
I get the assertion below when I do the following
1) Ensure continuous spellchecking is on
2) Type e.g. Uw to create a word with red underlining
3) Press Ctrl-M to enter mathmode.
FYI, I am using ASpell.
Hi John,
thank you for the
See http://www.lyx.org/trac/ticket/7099
I encountered the problem yesterday too after recompile and restart without
updating the resources.
So I tried to understand the cause and I don't understand the code in
GuiView::restoreLayout().
As I read the code the call of initToolbars() depends on
Am 03.01.2011 um 09:26 schrieb Vincent van Ravesteijn:
I encountered the problem yesterday too after recompile and restart without
updating the resources.
So I tried to understand the cause and I don't understand the code in
GuiView::restoreLayout().
As I read the code the call of
Am 03.01.2011 um 11:28 schrieb Vincent van Ravesteijn:
But somehow there is a connection between visibility flags and the
toolbar type, isn't it? The allowauto_ flag depends on visibility.
This looks a little bit weird.
Yes, this is just an example of bad programming practice.
In the
Am 03.01.2011 um 12:42 schrieb v...@lyx.org:
Author: vfr
Date: Mon Jan 3 12:42:01 2011
New Revision: 37068
URL: http://www.lyx.org/trac/changeset/37068
Log:
Remove GuiToolbar::allowauto_ and define this property in the
Toolbars.h::Visibility enum.
Now, there is the bug that
Am 03.01.2011 um 13:13 schrieb Vincent van Ravesteijn:
Hmm.. gmail sent the mail to early:
void GuiToolbar::restoreSession()
{
QSettings settings;
setVisibility(settings.value(sessionKey() + /visibility).toInt());
}
void GuiToolbar::restoreSession()
{
Am 03.01.2011 um 13:56 schrieb Abdelrazak Younes:
On 01/03/2011 09:26 AM, Vincent van Ravesteijn wrote:
Furthermore, I'm not sure yet what causes the bug, because it seems
that if the user sets the toolbar visibility from within LyX, this
setting is lost again on the new start-up of LyX. Why
Am 03.01.2011 um 15:25 schrieb Vincent van Ravesteijn:
Ha! That's it.
The combination of changed ui file and already stored settings is the
culprit. Thanks!
In GuiApplication::readUIFile() the settings for views gets removed if the
ui file is newer than the saved timestamp in the
Am 03.01.2011 um 17:10 schrieb Vincent van Ravesteijn:
In GuiApplication::readUIFile() the settings for views gets removed if the
ui file is newer than the saved timestamp in the settings.
Later the non existing settings will be used to initialize the visibility.
The default value for a non
Am 03.01.2011 um 20:11 schrieb LyX Ticket Tracker:
#7151: Math/Table autoshow broken in OS X (2.0b2)
-+--
Reporter: metrics |Owner: lasgouttes
Type: defect | Status: closed
Priority: normal
Am 03.01.2011 um 18:40 schrieb LyX Ticket Tracker:
#1720: LyX/Mac: TOC Menus entries are not correctly disabled when a dialog has
focus
Comment:
Stephan,
Can you share your thoughts about this bug ?
Sorry, I didn't had this one at the list.
I'll have a look and answer ASAP.
Stephan
Am 03.01.2011 um 18:40 schrieb LyX Ticket Tracker:
#1720: LyX/Mac: TOC Menus entries are not correctly disabled when a dialog has
focus
--+-
Reporter: lasgouttes|Owner: lasgouttes
Type: defect|
Am 04.01.2011 um 21:21 schrieb Vincent van Ravesteijn:
I have to admit that I don't understand the problem.
I don't understand what version trunk refers in the ticket
comments - 1.6?
No .. trunk is 2.0 and branch is 1.6.x.
I got confused by the question if the changeset 30557 (including
Am 05.01.2011 um 17:47 schrieb Stephan Witt:
Am 04.01.2011 um 21:21 schrieb Vincent van Ravesteijn:
The biggest problem I see currently is that on Linux a change of focus
from one document to another updates the document settings contents
while on Mac this does not happen! (on LyX 2.0
Am 05.01.2011 um 01:00 schrieb BH:
On Tue, Jan 4, 2011 at 3:21 PM, Vincent van Ravesteijn v...@lyx.org wrote:
I have to admit that I don't understand the problem.
I don't understand what version trunk refers in the ticket
comments - 1.6?
No .. trunk is 2.0 and branch is 1.6.x.
And I
While looking at #7209 I stumbled over Cursor::normalize().
Is it really a good idea to use the cursor inset as math whithout
any check to write it to debug output?
Stephan
Possible patch:
Index: src/Cursor.cpp
===
---
Am 06.01.2011 um 10:08 schrieb Vincent van Ravesteijn:
On Thu, Jan 6, 2011 at 8:20 AM, Stephan Witt st.w...@gmx.net wrote:
While looking at #7209 I stumbled over Cursor::normalize().
Is it really a good idea to use the cursor inset as math whithout
any check to write it to debug output
Am 06.01.2011 um 10:49 schrieb Stephan Witt:
Am 06.01.2011 um 10:08 schrieb Vincent van Ravesteijn:
Second, we should not have a broken cursor.
You mean one should call Cursor::fixIfBroken() if a broken cursor is possible?
It gets broken after a replace operation of last word in paragraph
Am 06.01.2011 um 11:47 schrieb Vincent van Ravesteijn:
normalize() is called only from the LFUN_CUT case of
InsetMathNest::doDispatch().
It corrects only the pos of the top cursor slice (if needed), AFAICS.
Fact is, we should apparently not call normalize() from a non-math
inset, so there
Am 06.01.2011 um 16:13 schrieb BH:
This is a follow-up to problems we've had with on-screen display of
fonts, where the spacing between words and cursor position was not
calculated correctly due to ligatures, swooshes, or other font
features. My guess is that the fix to that problem was too
Am 06.01.2011 um 19:27 schrieb Kornel:
Am Donnerstag, 6. Januar 2011 schrieb kuem...@lyx.org:
Log:
cmake: use tabs
That's, what I don't like much. Too many people define the tab value
different.
For leading tabs any value results in perfectly aligned rows.
So, I don't understand your
Am 06.01.2011 um 19:21 schrieb BH:
On Thu, Jan 6, 2011 at 12:03 PM, Stephan Witt st.w...@gmx.net wrote:
Am 06.01.2011 um 16:13 schrieb BH:
Yes -- current svn with Qt-4.7.1 (Carbon, 32 bit). I'm on OS X 10.6.5,
using the 10.5 SDK.
OS X 10.6.5, Qt-4.7.1 (Carbon, 32 bit), 10.5 SDK.
It's all
Am 06.01.2011 um 22:26 schrieb Peter Kümmel:
On 06.01.2011 22:18, Peter Kümmel wrote:
On 06.01.2011 22:02, BH wrote:
On Thu, Jan 6, 2011 at 3:55 PM, Stephan Witt st.w...@gmx.net wrote:
Am 06.01.2011 um 19:21 schrieb BH:
On Thu, Jan 6, 2011 at 12:03 PM, Stephan Witt st.w...@gmx.net wrote
Am 07.01.2011 um 08:08 schrieb Kornel:
Am Freitag, 7. Januar 2011 schrieb Kornel:
Am Donnerstag, 6. Januar 2011 schrieb sw...@lyx.org:
Author: switt
Date: Thu Jan 6 21:58:19 2011
New Revision: 37139
URL: http://www.lyx.org/trac/changeset/37139
Log:
cmake: Server.cpp is
Am 07.01.2011 um 08:00 schrieb Kornel:
Am Donnerstag, 6. Januar 2011 schrieb Peter Kuemmel:
But ATMit is mixed in the cmake files, and in lyx's code also tabs are
used.
Peter
Mixing is a consequence having tab stop != indent.
Kornel
I've no problem switching
Am 07.01.2011 um 09:17 schrieb Abdelrazak Younes:
On 01/07/2011 09:11 AM, Stephan Witt wrote:
Am 07.01.2011 um 08:08 schrieb Kornel:
Am Freitag, 7. Januar 2011 schrieb Kornel:
Am Donnerstag, 6. Januar 2011 schrieb sw...@lyx.org:
Author: switt
Date: Thu Jan 6 21:58:19 2011
New Revision
Am 07.01.2011 um 09:51 schrieb Kornel:
Am Freitag, 7. Januar 2011 schrieb Stephan Witt:
Am 07.01.2011 um 08:00 schrieb Kornel:
Am Donnerstag, 6. Januar 2011 schrieb Peter Kuemmel:
But ATMit is mixed in the cmake files, and in lyx's code also tabs
are used.
Peter
Am 06.01.2011 um 22:55 schrieb Stephan Witt:
Am 06.01.2011 um 22:26 schrieb Peter Kümmel:
On 06.01.2011 22:18, Peter Kümmel wrote:
On 06.01.2011 22:02, BH wrote:
On Thu, Jan 6, 2011 at 3:55 PM, Stephan Witt st.w...@gmx.net wrote:
Am 06.01.2011 um 19:21 schrieb BH:
On Thu, Jan 6, 2011
Am 08.01.2011 um 18:56 schrieb Peter Kuemmel:
I've also managed it to build LyX on OSX.
Fine.
I've installed the 4.7.1 dmg for Cocoa from the Trolls, 10.6.6, and GCC 4.2.
(There was never a other Qt on the machine)
What I see is something like this when using
- Times New Roman:
Am 10.01.2011 um 01:06 schrieb Pavel Sanda:
Abdel,
CXXUndo.o
Undo.cpp: In member function 'void lyx::Undo::clear()':
Undo.cpp:250: warning: statement has no effect
pavel
I'd say it looks like a real problem.
Stephan
Index: src/Undo.cpp
Am 09.01.2011 um 10:19 schrieb Peter Baumgartner:
If I moved the outline window in a position where LyX tries to adapt for a
docking position on the *right* side of the document a crash occurred all
the time.
I know of an old ticket (http://www.lyx.org/trac/ticket/7029) that was
closed
Am 10.01.2011 um 10:19 schrieb Vincent van Ravesteijn:
Note: the old ticket mentions a problem with docking outliner at the *left*
side.
I'm happy now but I thought I should report this problem as other people
might experience the same problem.
(I'm using Mac OS 1.6.5 and
Am 10.01.2011 um 20:37 schrieb Pavel Sanda:
Stephan Witt wrote:
To me that looked strange at first. But I've found another factor to
control the misbehavior of LyX + Qt 4.7.1: change the screen zoom factor.
maybe report it on qt bugzilla?
I'm not sure... I read some parts of the qt sources
Am 12.01.2011 um 09:23 schrieb Guenter Milde:
On 2011-01-11, Murat Yildizoglu wrote:
2011/1/11 Stephan Witt st.w...@gmx.net:
ok - LyX can find the thesaurus files. I presume the document doc/Intro.lyx
is not writable. Then the thesaurus (unfortunately) does not work properly.
Just like you
Am 12.01.2011 um 09:54 schrieb Vincent van Ravesteijn:
But one could rename and save the help-file thus making it writable
Wolfgang
Or type buffer-toggle-read-only in the command buffer.
Or compile with DEVEL_VERSION.
The real problem is: why shouldn't I be able to look up
something
Am 12.01.2011 um 13:14 schrieb Vincent van Ravesteijn:
The real problem is: why shouldn't I be able to look up
something with thesaurus when viewing an read-only lyx file.
What is _exactly_ the problem then ?
(I can't test the Thesaurus right now)
If the file is writable no problem.
Am 12.01.2011 um 15:57 schrieb Stephan Witt:
Am 12.01.2011 um 13:14 schrieb Vincent van Ravesteijn:
The real problem is: why shouldn't I be able to look up
something with thesaurus when viewing an read-only lyx file.
What is _exactly_ the problem then ?
(I can't test the Thesaurus
Am 12.01.2011 um 20:21 schrieb Pavel Sanda:
Stephan Witt wrote:
I tried to verify this... it seems the flow of control is too tricky.
The thesaurus dialog cannot apply (Dialog::canApply() return false).
Therefore when buffer is readonly the Dialog::checkStatus() method
doesn't call GuiDialog
Am 12.01.2011 um 23:15 schrieb Pavel Sanda:
Stephan Witt wrote:
Am 12.01.2011 um 20:21 schrieb Pavel Sanda:
Stephan Witt wrote:
I tried to verify this... it seems the flow of control is too tricky.
The thesaurus dialog cannot apply (Dialog::canApply() return false).
Therefore when buffer
Am 13.01.2011 um 08:57 schrieb Vincent van Ravesteijn:
Op 13-1-2011 8:48, Pavel Sanda schreef:
Stephan Witt wrote:
The metrics cache computes the width of every single character -
inclusive cacheing of it (to get better performance). But the
drawing is done with complete text strings
Am 13.01.2011 um 09:05 schrieb Vincent van Ravesteijn:
i know nothing about windows, sorry.
How can you live like that ???
It's a really good live.
(If it's about Microsoft Windows :-))
Stephan
Am 13.01.2011 um 09:07 schrieb Abdelrazak Younes:
On 01/13/2011 08:48 AM, Pavel Sanda wrote:
Stephan Witt wrote:
The metrics cache computes the width of every single character -
inclusive cacheing of it (to get better performance). But the
drawing is done with complete text strings
Am 13.01.2011 um 09:28 schrieb Abdelrazak Younes:
On 01/13/2011 09:23 AM, Stephan Witt wrote:
Am 13.01.2011 um 09:07 schrieb Abdelrazak Younes:
On 01/13/2011 08:48 AM, Pavel Sanda wrote:
Stephan Witt wrote:
The metrics cache computes the width of every single character -
inclusive
Am 13.01.2011 um 09:33 schrieb Pavel Sanda:
Abdelrazak Younes wrote:
To be on the save side - if nothing else is changed - you have to draw
letter by letter all the times anyway.
The option can set to true by default.
i also wonder what the 'few more cpu cycles' mean. we are already
Am 13.01.2011 um 09:40 schrieb Pavel Sanda:
Stephan Witt wrote:
What I do not understand:
why depends the call of GuiDialog::updateView() on the result of canApply()?
these two commits may reveal something
r22049 (younes, 'Factorize general code out of GuiDialog...')
Ah... the original
Am 13.01.2011 um 12:49 schrieb Pavel Sanda:
Stephan Witt wrote:
Ah... the original version simply was:
DialogView::checkStatus() { updateView(); }
So alternatively we may override Dialog::checkStatus() for GuiThesaurus.
Possible patch is attached.
this one looks better, since
Am 13.01.2011 um 09:07 schrieb Abdelrazak Younes:
On 01/13/2011 08:48 AM, Pavel Sanda wrote:
Stephan Witt wrote:
The metrics cache computes the width of every single character -
inclusive cacheing of it (to get better performance). But the
drawing is done with complete text strings
Am 15.01.2011 um 00:20 schrieb Pavel Sanda:
Stephan Witt wrote:
Use case: When reading the Manual in a foreign language (maybe because
there is no translation to ones mothertongue) a thesaurus can help to get
the meaning of less common words.
this is joke? :) why not use translator?
I
Am 15.01.2011 um 12:18 schrieb Vincent van Ravesteijn:
Sorry. I'll dive into it when I'm able to.
Thanks, meanwhile I checked my patch in.
Stephan
Op 15 jan 2011 00:15 schreef Stephan Witt st.w...@gmx.net:
Am 13.01.2011 um 12:49 schrieb Pavel Sanda:
Stephan Witt wrote:
Ah
Am 14.01.2011 um 18:41 schrieb LyX Ticket Tracker:
#7149: Crash when issuing commands through pipes with minimized view
--+-
Reporter: vfr | Owner: lasgouttes
Type: defect| Status: new
Am 16.01.2011 um 14:25 schrieb Peter Kümmel:
On 15.01.2011 20:31, Joost Verburg wrote:
On 1/14/2011 11:23 PM, Pavel Sanda wrote:
unless somebody is going to make detail review, this is too late for 2.0.
it can bring regressions since the code touches common routine...
Some time ago I
Am 16.01.2011 um 16:58 schrieb Joost Verburg:
On 1/16/2011 4:12 PM, Pavel Sanda wrote:
so it depends whether the guys are satisfied with this.
pavel
Stephan's patch looks fine to me.
Indeed it fixes the single replace issue. But...
In addition a call to expandPath should be added for
Am 16.01.2011 um 17:41 schrieb Joost Verburg:
On 1/16/2011 5:11 PM, Stephan Witt wrote:
In addition a call to expandPath should be added for path_prefix.
... I wonder why that is of any use.
If I change my path_prefix in the preferences I don't expect it gets
replaced immediately
Am 16.01.2011 um 21:25 schrieb Joost Verburg:
On 1/16/2011 8:41 PM, Stephan Witt wrote:
LyX uses QProcess to create child processes.
I cannot tell for sure if the PATH environment with %LyXDir% included works.
I could not find it when reading the Qt docs here:
http://doc.qt.nokia.com/latest
Am 16.01.2011 um 21:25 schrieb Joost Verburg:
On 1/16/2011 8:41 PM, Stephan Witt wrote:
LyX uses QProcess to create child processes.
I cannot tell for sure if the PATH environment with %LyXDir% included works.
I could not find it when reading the Qt docs here:
http://doc.qt.nokia.com/latest
Am 17.01.2011 um 03:01 schrieb Enrico Forestieri:
On Mon, Jan 17, 2011 at 08:45:45AM +0800, hzluo wrote:
In addition to this exact path_prefix problem, it causes another
problem on Windows:
If the external uitility is a .bat/.cmd wrapper, it will fail. For
example, ps2pdf13 is a wrapper in
Am 17.01.2011 um 08:38 schrieb Stephan Witt:
I claim that 1) and 2) are fixed with my patch and should help the installer
a lot. The environment variables of the other preference path items are
replaced
already on read of the lyxrc files. To make use of it the installer has to
make
sure
Am 17.01.2011 um 08:38 schrieb Stephan Witt:
To sum it up, I see four problems now:
1) bad environment variable expansion in replaceEnvironmentPath()
2) missing environment variable expansion for path_prefix
3) too early environment variable expansion on saving preferences
4) windows
Am 17.01.2011 um 10:24 schrieb Joost Verburg:
On 1/17/2011 8:38 AM, Stephan Witt wrote:
You are the devils advocate? :-)
Seriously, it would help for external .bat/.cmd utilities on windows.
But in general it's a good move to refrain from using system().
I don't see a problem with batch
Am 17.01.2011 um 10:22 schrieb Joost Verburg:
On 1/17/2011 8:48 AM, Stephan Witt wrote:
Sorry, this was not verbose enough.
Here I propose to use LyXDir in lyxrc.dist of windows. E.g.
\path_prefix = $LyXDir/bin;$LyXDir/ImageMagick;$LyXDir/python
Note the $LyXDir notation.
The same holds
Am 17.01.2011 um 11:16 schrieb Joost Verburg:
On 1/17/2011 11:07 AM, Stephan Witt wrote:
I don't get it. LyX has no windows command prompt :-)
E.g. to typeset something LyX executes latex...
If it's an executable it works obviously. But does it work too,
if I make a latex.bat and place
Am 17.01.2011 um 12:30 schrieb Jean-Marc Lasgouttes:
Le 17/01/2011 11:29, Joost Verburg a écrit :
On 1/17/2011 11:16 AM, Jean-Marc Lasgouttes wrote:
The reason why I never did it after me promises is that I stopped at
the question what is LyXDir exactly? I failed to find an answer
that is
Am 17.01.2011 um 12:41 schrieb Jean-Marc Lasgouttes:
Le 17/01/2011 12:39, Stephan Witt a écrit :
mac: /Application/LyX20.app
The binary lives in /Application/LyX20.app/Contents/MacOSX
Maybe LyXDir should be /Application/LyX20.app then...
Oops, I meant to add the Contents there. Since all
Am 17.01.2011 um 15:44 schrieb Richard Heck:
On 01/17/2011 08:23 AM, Vincent van Ravesteijn wrote:
Author: rgheck
Date: Sun Jan 16 21:28:43 2011
New Revision: 37233
URL: http://www.lyx.org/trac/changeset/37233
-int LyXRC::read(FileName const filename)
+bool LyXRC::read(FileName const
Am 17.01.2011 um 15:52 schrieb Jean-Marc Lasgouttes:
Le 17/01/2011 15:44, Richard Heck a écrit :
Prefs2prefs need python to run. Therefore Windows needs to add the
path to python to the environment path before. However, the path is
stored in the preferences, so we first need to read the
Am 17.01.2011 um 13:57 schrieb Jean-Marc Lasgouttes:
Le 17/01/2011 12:51, Stephan Witt a écrit :
Do you plan to make the required changes? That would be fine with me.
My output in terms of LyX code is almost zero these days. I can always
promise, but it is safer to find someone else to do
Am 16.01.2011 um 16:12 schrieb Pavel Sanda:
Stephan Witt wrote:
Am 16.01.2011 um 14:25 schrieb Peter Kümmel:
The patch complete removes the regex logic in replaceEnvironmentPath because
it seems to be buggy, could someone have a look at it, I think it would be
better to fix the regex code
Am 17.01.2011 um 13:57 schrieb Jean-Marc Lasgouttes:
Le 17/01/2011 12:51, Stephan Witt a écrit :
Do you plan to make the required changes? That would be fine with me.
My output in terms of LyX code is almost zero these days.
I can always promise, but it is safer to find someone else to do
Am 17.01.2011 um 20:14 schrieb Pavel Sanda:
Bernhard Roider wrote:
just an idea: the method QTextCharFormat::setFontKerning can enable/disable
kerning. Could that help?
i dont know if QTextCharFormat is related to our painting,
It isn't, but perhaps it's good idea anyway. Thanks, Bernhard.
Am 17.01.2011 um 21:10 schrieb Stephan Witt:
Am 17.01.2011 um 20:14 schrieb Pavel Sanda:
Bernhard Roider wrote:
just an idea: the method QTextCharFormat::setFontKerning can enable/disable
kerning. Could that help?
i dont know if QTextCharFormat is related to our painting,
It isn't
Am 17.01.2011 um 16:07 schrieb Stephan Witt:
Am 16.01.2011 um 16:12 schrieb Pavel Sanda:
Stephan Witt wrote:
Am 16.01.2011 um 14:25 schrieb Peter Kümmel:
The patch complete removes the regex logic in replaceEnvironmentPath
because
it seems to be buggy, could someone have a look
Am 17.01.2011 um 22:59 schrieb Pavel Sanda:
Stephan Witt wrote:
At the first glance it looks not really helpful. :(
1) Constructor of GuiFontInfo in GuiFontLoader.cpp start with
GuiFontInfo::GuiFontInfo(FontInfo const f)
: metrics(QFont())
{
font.setKerning(false);
2) I
Am 17.01.2011 um 23:49 schrieb Vincent van Ravesteijn:
Does the user have to set this himself, or is it set by the installer,
or? As I said in a previous message, JMarc and Stephan seemed to suggest
that it ought to be set in lyxrc.dist or lyxrc.defaults, which we do not try
to convert.
the ${VARNAME} variant only to expand
variables and
stop supporting the alternative short form $VARNAME. But that's another dirty
hack.
Op 17 jan 2011 16:07 schreef Stephan Witt st.w...@gmx.net:
Am 16.01.2011 um 16:12 schrieb Pavel Sanda:
Stephan Witt wrote:
Am 16.01.2011 um 14:25
Am 18.01.2011 um 07:31 schrieb Stephan Witt:
Am 18.01.2011 um 07:03 schrieb Vincent van Ravesteijn:
Because if the variable is not a variable, it might just be a filename
containing a $.
But the function replaceEnvironmentPath() don't know if the $ is in the
filename
Am 18.01.2011 um 11:17 schrieb Christian Wilhelmsen:
I juse LyX and Dropbox in collaboration with my team mates, and the problem
is that when you have opened a file, another user edits it and sync it through
dropbox, LyX wont open the new version, it opens a buffered copy or
something.
Am 18.01.2011 um 13:49 schrieb Vincent van Ravesteijn:
I claim this is a better fix for ticket http://www.lyx.org/trac/ticket/4177
Other locations may affected to.
Stephan
If you think it's better, I believe you.
So you think I can apply the patch?
Is this still related to the
Am 18.01.2011 um 14:59 schrieb Vincent van Ravesteijn:
To make a portable LyX you need a proper working replaceEnvironmentPath().
(I.e. $LyXDir/component1/bin:$LyXDir/component2/bin in path_prefix must be
translated to /path/to/lyx/component1/bin:/path/to/lyx/component2/bin).
To fix this I
Am 18.01.2011 um 15:33 schrieb Richard Heck:
On 01/18/2011 09:30 AM, Peter Kümmel wrote:
On 18.01.2011 15:11, Vincent van Ravesteijn wrote:
Note that this is only an issue because your preferences file was out of
date, in the sense that it doesn't have a Format 1 tag at the beginning.
But,
Am 18.01.2011 um 15:23 schrieb Vincent van Ravesteijn:
Unfortunately it's not the only thing you did.
You introduced the regression that only the last occurrence of
a variable gets replaced. Your assumption was that boost returns
the first match - it's the last one instead!
Am 18.01.2011 um 16:42 schrieb rgh...@lyx.org:
Author: rgheck
Date: Tue Jan 18 16:42:27 2011
New Revision: 37253
URL: http://www.lyx.org/trac/changeset/37253
Log:
Update all of our bind and ui files to format 1.
Richard, do you have a proposal how I can test the transition
you are doing
Am 18.01.2011 um 18:03 schrieb Vincent van Ravesteijn:
The other things from the RELEASE_NOTES:
...
- \personal_dictionary, \spell_command, \use_input_encoding,
\use_personal_dictionary and \use_spell_lib (ispell support has been
removed)
But \use_personal_dictionary is a valid LyX 2.0
Am 18.01.2011 um 17:26 schrieb Richard Heck:
On 01/18/2011 11:14 AM, Stephan Witt wrote:
Am 18.01.2011 um 16:42 schrieb rgh...@lyx.org:
Author: rgheck
Date: Tue Jan 18 16:42:27 2011
New Revision: 37253
URL: http://www.lyx.org/trac/changeset/37253
Log:
Update all of our bind and ui
Am 19.01.2011 um 11:46 schrieb LyX Ticket Tracker:
#5211: lyx doesn't control if the opened document is changed by other process
-+--
Reporter: zak.mc.kraken@… |Owner: lasgouttes
Type: defect
Am 22.01.2011 um 15:33 schrieb Enrico Forestieri:
On Sat, Jan 22, 2011 at 01:53:40AM +0100, Pavel Sanda wrote:
this patch should help at least for symlinks.
Why not restricting the workaround to the case where the file actually
is stored on an ext4 file system?
It's copied for symlink
Am 22.01.2011 um 16:10 schrieb Peter Kümmel:
On 22.01.2011 14:36, Abdelrazak Younes wrote:
On 22/01/2011 12:04, Peter Kümmel wrote:
Before we set LyX 2.0 in stone by releasing it
I've cleaned up a bit the LYX_ option naming,
to make it more consistent:
- only positive names (replace NO_
701 - 800 of 4472 matches
Mail list logo
