Abdelrazak Younes wrote:
> Well, I've lost some battles in LyX development already. The problem is
> that most of the changes I envision would need full support of
> maintainers so I loose. My available time is not much so I can't afford
> to spend time on things that will not be accepted. I have t
Angus Leeming a écrit :
Abdelrazak Younes wrote:
Well, I've lost some battles in LyX development already. The problem is
that most of the changes I envision would need full support of
maintainers so I loose. My available time is not much so I can't afford
to spend time on things that will not be
On Sat, May 06, 2006 at 01:13:34PM +0300, Martin Vermeer wrote:
> It looks correct in Andre's commit. But this is still there and must be
> corrected:
>
> + { LFUN_DELETE_FORWARD_BACKWARD, "delete-backward",
> SingleParUpdate },
> + { LFUN_DELETE_FORWARD_BACKWARD_SKIP,
On Sat, May 06, 2006 at 12:07:54AM +0200, Edwin Leuven wrote:
> Angus Leeming wrote:
> >LOL! I think You are. You've confused this with "const &"
>
> and to add to the confusion (at least mine):
>
> why don't i see any invalid behavior or crashes with my original (wrong)
> patch?
Because 'undef
On Fri, May 05, 2006 at 10:28:14PM +0200, Abdelrazak Younes wrote:
> Andre Poenitz a écrit :
> >Does it really help do try to call all four
> >LFUN_OUTLINE_* functions "outline"?
>
> Because there are four outlining operations?
> There is only one outline function that behave differently depending
On Sat, May 06, 2006 at 01:10:08PM +0300, Martin Vermeer wrote:
> char const * using_this_returned_value_is_undefined_behaviour()
> {
> char const * cdata = std::string("data").c_str();
> return cdata;
> }
>
> I think I understand this. The problem is that cdata is made to point to
> the start
On Fri, May 05, 2006 at 10:25:36PM +0200, Abdelrazak Younes wrote:
> Andre Poenitz a écrit :
> >Pretty boring. I'll commit soonish.
>
> You are going to break outline with your patch. I am actually astonished
> that it compiles at all.
I rarely compile before committing.
Andre'
Hi Edwin,
I would propose that you split your patch into three if it's possible:
- the strictly cleanup stuff. I think you can safely commit that.
- the file dialog work would probably need a review from Andre. Could
you explain a bit what you have done?
- the coloritem stuff will probably inter
Andre Poenitz a écrit :
On Fri, May 05, 2006 at 10:25:36PM +0200, Abdelrazak Younes wrote:
Andre Poenitz a écrit :
Pretty boring. I'll commit soonish.
You are going to break outline with your patch. I am actually astonished
that it compiles at all.
I rarely compile befor
Edwin Leuven wrote:
> i would like to put it in. these things can always be changed later...
I'd be interested to know with what you are replacing the color items. I found
that a good ui solution, actually.
Jürgen
Edwin Leuven a écrit :
Abdelrazak Younes wrote:
Hi Edwin,
I would propose that you split your patch into three if it's possible:
- the strictly cleanup stuff. I think you can safely commit that.
it all is
- the file dialog work would probably need a review from Andre. Could
you explain a bi
Andre Poenitz wrote:
On Sat, May 06, 2006 at 01:13:34PM +0300, Martin Vermeer wrote:
It looks correct in Andre's commit. But this is still there and must be
corrected:
+ { LFUN_DELETE_FORWARD_BACKWARD, "delete-backward",
SingleParUpdate },
+ { LFUN_DELETE_FORWAR
Juergen Spitzmueller wrote:
Edwin Leuven wrote:
i would like to put it in. these things can always be changed later...
I'd be interested to know with what you are replacing the color items. I found
that a good ui solution, actually.
a vector of color names
i store the color names (rgb) whe
Edwin Leuven wrote:
> a vector of color names
>
> i store the color names (rgb) when initializing the dialog as
> prefcolors_ and make a copy newcolors_
>
> when changing i set the appropriate element in new colors
>
> after that applying is simple:
I mean in the browser widget.
Jürgen
Edwin Leuven wrote:
> Juergen Spitzmueller wrote:
> > I mean in the browser widget.
>
> screenshot attached...
I see, very good.
Please go ahead.
Jürgen
Edwin Leuven wrote:
i would like to put it in. these things can always be changed later...
it is in...
Andre,
if we cleanup the LFUNs, then let's do it right :-)
Here comes a patch that fixes some of the names. I also adjusted some
function names such that they comply to the LyX naming conventions.
Moreover, I changed line-delete-forward into line-delete because there
is no line-delete-backwar
Am Sonntag, 7. Mai 2006 11:24 schrieb Abdelrazak Younes:
> - the coloritem stuff will probably interest Georg as well. I have
> posted a class skeleton a while ago for Qt4 color handling (deriving
> from QColorItem), maybe we could use something like that.
I am not too interested in the colorit
Am Sonntag, 7. Mai 2006 12:04 schrieb Michael Gerz:
> Andre,
>
> if we cleanup the LFUNs, then let's do it right :-)
>
> Here comes a patch that fixes some of the names.
IMHO you can commit this part without further discussion, and André owes
you a beer.
> I also adjusted some
> function nam
Am Samstag, 6. Mai 2006 19:43 schrieb Enrico Forestieri:
> This patch works.
OK, it is in now. Edwin/Abdel: I added a FIXME for you in
src/frontends/qt4/QDelimiterDialog.C.
> I take that you mean the "backslash" problem.
Yes.
> Is there any glitch with my proposed fix for it?
It might hide an
Georg Baum wrote:
I also adjusted some
function names such that they comply to the LyX naming conventions.
Moreover, I changed line-delete-forward into line-delete because there
is no line-delete-backward.
Are these changes worth it to break compatibility? I tend to say yes, but
am not
Michael Gerz wrote:
Ok to commit everything except for "line-delete"?
Here comes a second patch. This time, I kept "line-delete-forward" but
used LFUN_LINE_DELETE internally. (Doesn't break backwards compatibility)
Georg, all: Ok?
Michael
Index: src/LyXAction.C
==
Abdelrazak Younes <[EMAIL PROTECTED]> writes:
| Following my config.h analysis, this patch gets rid of
|
| I am ready to do more of this but I would like to know first if this
| kind of patch will be accepted.
It won't. Don't remove
--
Lgb
the attached patch gets rid of the qt3 support library without changing
current functionality:
qiconpallete: replace old gridlayout with new one
browsebox: don't compile and comment out from BulletsModule (wasn't
working anyway)
...
put this in? or wait until replacement browsebox is progra
Hello LyXers,
I uploaded a new version of the german math manual "LyXMathebefehle" to
http://wiki.lyx.org/LyX/LyXMathebefehle
Version 3.0 is completely revisited to cover the new features of LyX
1.4. I also uploaded a bugfix version 2.10 that describes the math
features of LyX 1.3.7.
The do
Hi,
this is the second set of changes for the CT cleanup. The changes are
supposed to be purely syntactical (renaming of methods; removal of
superfluous methods)
FYI: There will be one or two more simple patches before I actually
start breaking CT...
Michael
Index: src/paragraph.h
While working on my math manual I discovered a lot of new bugs. Most of
them are cursor problems but some of them are very important. Here's a
list of them:
---
Loss of functionality, major bugs that should be fixed as soon as possible:
- Bug 2466, framebox and fbox can't be used when operator
Michael Gerz wrote:
Hi,
this is the second set of changes for the CT cleanup. The changes are
supposed to be purely syntactical (renaming of methods; removal of
superfluous methods)
FYI: There will be one or two more simple patches before I actually
start breaking CT...
Can I commit the
Abdelrazak Younes
Sat, 06 May 2006 14:17:35 -0700
Enrico Forestieri a écrit :
On Sat, May 06, 2006 at 09:33:50PM +0200, Abdelrazak Younes wrote:
Enrico Forestieri a écrit :
On Sat, May 06, 2006 at 07:41:03PM +0200, Abdelrazak Younes
wrote:
For those who
This email is a proposal for a new test procedure of LyX releases.
I started to use LyX when LyX 1.2 came out. While working with it I
started to write my math manual which was not finished when LyX 1.3 was
released. I reworked my document to cover the new features of lyX 1.3
and discovered lo
On Sun, May 07, 2006 at 05:00:23PM +0200, Uwe Stöhr wrote:
> Personally I use a test procedure for every new release of my
> LyXWinInstaller. Derived from this here's my proposal for new releases
> of LyX:
Some time ago I started a lyx-tests CVS repository for precisely this
regression-testing
Edwin Leuven a écrit :
the attached patch gets rid of the qt3 support library without changing
current functionality:
qiconpallete: replace old gridlayout with new one
browsebox: don't compile and comment out from BulletsModule (wasn't
working anyway)
...
put this in? or wait until replace
John Levon wrote:
- Test scrolling and editing (copy/paste, etc.) with a long and big
document like a PHD thesis.
I believe we knew about all the problems here, but released anyway.
Yes, that was definitely the problem. We locked ourself by a code freeze
period that started too early
Abdelrazak Younes wrote:
Please go ahead but put some comment as to what to do in the code.
No more Qt3Support, year!
Does it mean that the Qt4 frontend can be compiled or that it is fully
functional or what?
Michael
On Sun, May 07, 2006 at 06:04:13PM +0200, Michael Gerz wrote:
> Yes, that was definitely the problem. We locked ourself by a code freeze
> period that started too early (and lasted too long).
No, that wasn't the problem, it was too much changes allowed at once,
rather than small simple improveme
On Sunday 07 May 2006 15:10, Persio Barros wrote:
> Abdel,
Oi, :-)
> Sorry for the delay. Today is sunday and wright now I'm cooking a feijoada.
> And drinking some cachaça, of couse ;-).
After eating and drink like that do you still find strength to write a
message? Wow, I am impressed. ;-
Michael Gerz a écrit :
Abdelrazak Younes wrote:
Please go ahead but put some comment as to what to do in the code.
No more Qt3Support, year!
Does it mean that the Qt4 frontend can be compiled or that it is fully
functional or what?
Well qt4 frontend could be compiled from day 0 of svn pres
Persio Barros a écrit :
Abdelrazak Younes
Maybe, but Persio Barros has tested it with success on WinMe. Could
you test again Persio?
Abdel.
Abdel,
Sorry for the delay. Today is sunday and wright now I'm cooking a
feijoada. And drinking some cachaça, of couse ;-).
Lucky guy ;-)
Where do y
I got your last version of LyX-1.5. This one asks for the zlib1.dll,
unlike the previous one. Got the dll, and now it runs.
Yes, this is a new requirement, note that the dll is included in the two
packages and it is also available on my web site.
Abdel.
Lars Gullik Bjønnes a écrit :
Abdelrazak Younes <[EMAIL PROTECTED]> writes:
| Following my config.h analysis, this patch gets rid of
|
| I am ready to do more of this but I would like to know first if this
| kind of patch will be accepted.
It won't. Don't remove
That's old news Lars...
I
On Sat, May 06, 2006 at 12:51:59PM +0200, Joost Verburg wrote:
> Andre Poenitz wrote:
> >So this is arguably a breach of the GPL unless the installed
> >software could be considered as a non-integral part of the
> >installer.
> >
> >It's safer to make a Qt4 installer with the GPL'd Qt4/Win.
>
> I
On Sat, May 06, 2006 at 12:46:31PM +0200, Michael Gerz wrote:
> Andre Poenitz wrote:
>
> >>I don't know the details, but I think you need a commercial license if
> >>you develop with Qt3.
> >>
> >>
> >
> >In theory one could take the *nix GPL version and port it to Windows...
> >In practice th
Andre Poenitz wrote:
The problem is not Qt in this case but the GPL. Assuming the installed
software is considered an integral part of the installer, the installer
would have to be released with a GPL-compatible license. I am not sure
how this could be achieved with Qt (commercial).
Right, I un
Dear all,
The scons-based building system is now available from svn. Currently,
1. Under Linux (RHEL4), scons can build qt3, qt4 without problem.
Using commands like
$ scons( default to use qt3)
$ scons frontend=qt4 qt_dir=/path/to/qt4
You can also use options like
$ scons extra_inc_
BTW, please use SCons >= 0.96.92.
Bo
On Sun, May 07, 2006 at 06:00:42PM -0500, Bo Peng wrote:
> Windows troubles:
>
> 1. the ...rdata file used to build cygwin seems to be a libtool thing,
> I have tried to pass script=... option to g++ but failed. I guess one
> can run till the last step, get the link command from scons_lyx.log
> a
No, it is not. The "-Wl,--script,/path/to/i386pe.x-no-rdata" option
is indeed meant for libtool which then translates it to
"-Wl,--script -Wl,/path/to/i386pe.x-no-rdata" when calling g++.
Please, don't forget "-Wl,--enable-runtime-pseudo-reloc".
These options are passed.
Under cygwin, I just n
Abdelrazak Younes <[EMAIL PROTECTED]> writes:
| Abdel.
| Index: qttableview.C
| ===
| --- qttableview.C (revision 13796)
| +++ qttableview.C (working copy)
| @@ -1,36 +1,24 @@
| /**
48 matches
Mail list logo
