Enrico Forestieri wrote:
The attached patch against 1.4 adds support for clean_dvi to the new
configure.py script, so images in paths with spaces that are included in
preamble or ERT work again.
This is also relevant for cygwin, so please check also for
sys.platform == 'cygwin'
Updated
Bo Peng a écrit :
Please test the attached patch. I changed the names of included
boost_* and intl to included_boost_* and included_intl, just to be
safe. Other changes are trivial.
All my problems are solved Bo. While you are maybe right about -L
ordering I think this renaming is a safe bet
Abdelrazak Younes a écrit :
There is no caching either in qt3. I guess we've found a nice
optimisation to do :-)
I'll send a patch soon!
Font caching was not done in qt3 and qt4, this patch fixes this. Please
test on linux/qt[34] and Mac/qt[34]
I am going to commit the qt4 part of this
Am Sonntag, 21. Mai 2006 09:56 schrieb Joost Verburg:
Enrico Forestieri wrote:
The attached patch against 1.4 adds support for clean_dvi to the new
configure.py script, so images in paths with spaces that are included
in
preamble or ERT work again.
This is also relevant for cygwin,
Some remarks:
- The name of the draft dvi format needs to be dvi2. The number at the end
marks it as child format of dvi, and it inherits the dvi viewer setting.
- I adjusted the test a bit so that we don't need to list the latex
programs twice
Then clean_dvi script could also be used on
Am Sonntag, 21. Mai 2006 11:36 schrieb Joost Verburg:
Thanks. Do you also have a patch for 1.4? I have no 1.5 tree to test.
Here it is. Only one hunk in configure.py is different.
Gorg
Index: lib/configure.py
===
---
Unless QFileDialog does some magic in addToolButton() we have a memory
leak here. Can some Qt expert please tell whether this patch is OK?
Georg
Index: src/frontends/qt3/FileDialog_private.h
===
---
Juergen Spitzmueller wrote:
Updated patch attached. OK now?
Hmpf, I should have had a closer look.
Correct patch attached.
Jürgen
Index: src/insets/insettabular.C
===
--- src/insets/insettabular.C (Revision 13887)
+++
Georg Baum a écrit :
Unless QFileDialog does some magic in addToolButton() we have a memory
leak here. Can some Qt expert please tell whether this patch is OK?
I am not sure of that but I think I have read somewhere that it is safe
to use pointers on Qt objects this way. If that is not the
Abdelrazak Younes wrote:
Georg Baum a écrit :
Unless QFileDialog does some magic in addToolButton() we have a memory
leak here. Can some Qt expert please tell whether this patch is OK?
I am not sure of that but I think I have read somewhere that it is safe
to use pointers on Qt objects this
hi,
atm pplatex is not recognized by configure.py (at least under windows)
perhaps someone knowledgable can have a peek? would be nice if this just
works after putting pplatex.exe in ./bin and dvipost.sty in ./tex
thanks, ed.
Georg Baum wrote:
- The name of the draft dvi format needs to be dvi2. The number at the end
marks it as child format of dvi, and it inherits the dvi viewer setting.
DraftDVI is not intended to be viewed, so there should be no associated
viewer.
The others things in your patch work fine.
Hi,
I'm trying to convert a document with tables to another document class.
LyX 1.4.2svn however crashes.
To reproduce:
* Start a new document
* Insert a table
* Insert some text (a single character is enough) to one of the cells
* Go to Document Settings and select another document class
*
Joost Verburg wrote:
I'm trying to convert a document with tables to another document class.
LyX 1.4.2svn however crashes.
To reproduce:
* Start a new document
* Insert a table
* Insert some text (a single character is enough) to one of the cells
* Go to Document Settings and select
On Sat, May 20, 2006 at 10:03:38PM -0500, Bo Peng wrote:
2) nls=yes gettext=included is correctly taken into account but -lintl
is passed. So the mingw intl dll is used as opposed as libs/libintl.a,
Does the order of -L matter? I notice that -Ldebug/libs is the first
in the link
On Sun, May 21, 2006 at 10:35:30AM +0200, Abdelrazak Younes wrote:
Everything compiled with -03 -w in 35 minutes. That's twice as fast as
autotools (with -j3) :-)
Do you really mean -03 or is it -O3? ;-)
Then scons install restarted the compilation from the beginning. The
full scons call
1) a very minor glitch when compiling qt4: Both
-ID:\program\Qt\4.1\include\QtCore4 and
-ID:\program\Qt\4.1\include\QtCore are passed (same for QtGui). The
first one is not useful (and would never exists), please remove. The
trailing '4' is only possible for libraries.
This is a qt4.py bug and
Yes, I noticed that I have always to use fast_start=no, too.
I spent days on this option! :-)
After scons is stabilized (no frequent env change), this feature is
supposed to be useful and is non-intrusive. The first time when scons
is used, a file called env.cache is created and cache all test
Juergen Spitzmueller wrote:
It might be another consequence of the problematic BufferView cache situation
in tabulars. Georg, do you have an idea?
Something seems to be seriously wrong with tables. I'm working on quite
some tables today and I get more crashes. Maybe it's all related to the
On Thu, May 18, 2006 at 10:48:33PM -0500, Bo Peng wrote:
That was the problem. Downloaded 0.96.92's src.rpm, did an rpmbuild
--rebuild
and installed it and it works like a charm.
In the script, I can only check for scons 0.96, not 1 or 92.
Now I have to figure out how to do the rpm stuff
Also sprach Joost Verburg:
Something seems to be seriously wrong with tables. I'm working on quite
some tables today and I get more crashes. Maybe it's all related to the
same issue.
Please report all of them.
Jürgen
Juergen Spitzmueller wrote:
Something seems to be seriously wrong with tables. I'm working on quite
some tables today and I get more crashes. Maybe it's all related to the
same issue.
Please report all of them.
When I can reproduce one I'll report it, but it's not always easy to
find the
On Sat, 20 May 2006, Jean-Marc Lasgouttes wrote:
Joost == Joost Verburg [EMAIL PROTECTED] writes:
Joost Hi, The current LyX website looks outdated, both in terms of
Joost design and content. Especially now LyX is available for Windows
Joost users as well, a modern and easy-to-use website is
Am Sonntag, 21. Mai 2006 15:06 schrieb Joost Verburg:
DraftDVI is not intended to be viewed, so there should be no associated
viewer.
I think it should be viewable, because clean_dvi.py sometimes fails (IIRC
with some hyperref constructs).
Georg
We have a new crash bug of LyX:
http://bugzilla.lyx.org/show_bug.cgi?id=2609
It seems that it occurs only in the Windows version. Could somebody
check this with a Lunix version.
thanks and regards
Uwe
Georg Baum wrote:
I think it should be viewable, because clean_dvi.py sometimes fails (IIRC
with some hyperref constructs).
I've never had any problems with hyperref and clean_dvi. Can you try to
reproduce it?
In my opinion a DraftDVI menu item is very confusing.
Joost
Am Sonntag, 21. Mai 2006 16:14 schrieb Juergen Spitzmueller:
It might be another consequence of the problematic BufferView cache
situation
in tabulars. Georg, do you have an idea?
I got one after a bit of debugging. The problem is not the BufferView
cache, but the use of std::swap() instead
Am Sonntag, 21. Mai 2006 12:42 schrieb Edwin Leuven:
When you create a QObject with another object as parent, it's added to
the parent's children() list, and is deleted when the parent is.
I knew there must have been something special, so the current code is
alright. I don't do qt too often,
The only thing that is necessary is to add a dest_dir= option
for sons install so that the installation can be done in another
directory tree.
It is called prefix, the same name used by autotools.
Bo
On Sun, May 21, 2006 at 01:22:12PM -0500, Bo Peng wrote:
The only thing that is necessary is to add a dest_dir= option
for sons install so that the installation can be done in another
directory tree.
It is called prefix, the same name used by autotools.
Not quite. scons prefix=/tmp/foo will
That is how the current configure-based RPM spec file works. After configuring
with prefix=/usr, we do:
make DESTDIR=${RPM_BUILD_ROOT} install
Does this make sense?
This is beyond my understanding of this option. I always do configure
--prefix and make install without DESTDIR. Using scons, I
On Sunday 21 May 2006 21:53, Bo Peng wrote:
I would like to see other's opinion regarding a separate dest_dir option.
That is the standard procedure for rpm (and also deb) building. prefix is
the place where the package will be installed and run, destdir is the place
where the package is
Please test the attached patch. It introduces dest_dir.
Bo
Index: development/scons/SConscript
===
--- development/scons/SConscript (revision 13895)
+++ development/scons/SConscript (working copy)
@@ -1235,26 +1235,17 @@
if
On Sun, May 21, 2006 at 04:59:28PM -0500, Bo Peng wrote:
Please test the attached patch. It introduces dest_dir.
I have not tested your patch but DESTDIR is simpler than you
think, perhaps. You should do everything as specified by PREFIX
and then, at install time, copy to $DESTDIR/$PREFIX if
I have not tested your patch but DESTDIR is simpler than you
think, perhaps. You should do everything as specified by PREFIX
and then, at install time, copy to $DESTDIR/$PREFIX if DESTDIR
is defined, otherwise everything goes as before, ie, copy to $PREFIX.
DESTDIR is useful because you can
Enrico Forestieri wrote:
The attached patch against 1.4 adds support for clean_dvi to the new
configure.py script, so images in paths with spaces that are included in
preamble or ERT work again.
This is also relevant for cygwin, so please check also for
sys.platform == 'cygwin'
Updated
Bo Peng a écrit :
> Please test the attached patch. I changed the names of included
> boost_* and intl to included_boost_* and included_intl, just to be
> safe. Other changes are trivial.
All my problems are solved Bo. While you are maybe right about -L
ordering I think this renaming is a safe
Abdelrazak Younes a écrit :
There is no caching either in qt3. I guess we've found a nice
optimisation to do :-)
I'll send a patch soon!
Font caching was not done in qt3 and qt4, this patch fixes this. Please
test on linux/qt[34] and Mac/qt[34]
I am going to commit the qt4 part of this
Am Sonntag, 21. Mai 2006 09:56 schrieb Joost Verburg:
> Enrico Forestieri wrote:
> >> The attached patch against 1.4 adds support for clean_dvi to the new
> >> configure.py script, so images in paths with spaces that are included
in
> >> preamble or ERT work again.
> >
> > This is also
Some remarks:
- The name of the draft dvi format needs to be dvi2. The number at the end
marks it as child format of dvi, and it inherits the dvi viewer setting.
- I adjusted the test a bit so that we don't need to list the latex
programs twice
Then clean_dvi script could also be used on
Am Sonntag, 21. Mai 2006 11:36 schrieb Joost Verburg:
> Thanks. Do you also have a patch for 1.4? I have no 1.5 tree to test.
Here it is. Only one hunk in configure.py is different.
Gorg
Index: lib/configure.py
===
---
Unless QFileDialog does some magic in addToolButton() we have a memory
leak here. Can some Qt expert please tell whether this patch is OK?
Georg
Index: src/frontends/qt3/FileDialog_private.h
===
---
Juergen Spitzmueller wrote:
> Updated patch attached. OK now?
Hmpf, I should have had a closer look.
Correct patch attached.
Jürgen
Index: src/insets/insettabular.C
===
--- src/insets/insettabular.C (Revision 13887)
+++
Georg Baum a écrit :
Unless QFileDialog does some magic in addToolButton() we have a memory
leak here. Can some Qt expert please tell whether this patch is OK?
I am not sure of that but I think I have read somewhere that it is safe
to use pointers on Qt objects this way. If that is not the
Abdelrazak Younes wrote:
Georg Baum a écrit :
Unless QFileDialog does some magic in addToolButton() we have a memory
leak here. Can some Qt expert please tell whether this patch is OK?
I am not sure of that but I think I have read somewhere that it is safe
to use pointers on Qt objects this
hi,
atm pplatex is not recognized by configure.py (at least under windows)
perhaps someone knowledgable can have a peek? would be nice if this just
works after putting pplatex.exe in ./bin and dvipost.sty in ./tex
thanks, ed.
Georg Baum wrote:
- The name of the draft dvi format needs to be dvi2. The number at the end
marks it as child format of dvi, and it inherits the dvi viewer setting.
DraftDVI is not intended to be viewed, so there should be no associated
viewer.
The others things in your patch work fine.
Hi,
I'm trying to convert a document with tables to another document class.
LyX 1.4.2svn however crashes.
To reproduce:
* Start a new document
* Insert a table
* Insert some text (a single character is enough) to one of the cells
* Go to Document > Settings and select another document class
Joost Verburg wrote:
> I'm trying to convert a document with tables to another document class.
> LyX 1.4.2svn however crashes.
>
> To reproduce:
>
> * Start a new document
> * Insert a table
> * Insert some text (a single character is enough) to one of the cells
> * Go to Document > Settings and
On Sat, May 20, 2006 at 10:03:38PM -0500, Bo Peng wrote:
> > > > > 2) nls=yes gettext=included is correctly taken into account but -lintl
> > > > > is passed. So the mingw intl dll is used as opposed as libs/libintl.a,
> > > >
>
> Does the order of -L matter? I notice that -Ldebug/libs is the
On Sun, May 21, 2006 at 10:35:30AM +0200, Abdelrazak Younes wrote:
> Everything compiled with "-03 -w" in 35 minutes. That's twice as fast as
> autotools (with -j3) :-)
Do you really mean "-03" or is it "-O3"? ;-)
> Then "scons install" restarted the compilation from the beginning. The
> full
1) a very minor glitch when compiling qt4: Both
-ID:\program\Qt\4.1\include\QtCore4 and
-ID:\program\Qt\4.1\include\QtCore are passed (same for QtGui). The
first one is not useful (and would never exists), please remove. The
trailing '4' is only possible for libraries.
This is a qt4.py bug and
Yes, I noticed that I have always to use "fast_start=no", too.
I spent days on this option! :-)
After scons is stabilized (no frequent env change), this feature is
supposed to be useful and is non-intrusive. The first time when scons
is used, a file called env.cache is created and cache all
Juergen Spitzmueller wrote:
It might be another consequence of the problematic BufferView cache situation
in tabulars. Georg, do you have an idea?
Something seems to be seriously wrong with tables. I'm working on quite
some tables today and I get more crashes. Maybe it's all related to the
On Thu, May 18, 2006 at 10:48:33PM -0500, Bo Peng wrote:
> >That was the problem. Downloaded 0.96.92's src.rpm, did an rpmbuild
> >--rebuild
> >and installed it and it works like a charm.
>
> In the script, I can only check for scons 0.96, not 1 or 92.
>
> >Now I have to figure out how to do
Also sprach Joost Verburg:
> Something seems to be seriously wrong with tables. I'm working on quite
> some tables today and I get more crashes. Maybe it's all related to the
> same issue.
Please report all of them.
Jürgen
Juergen Spitzmueller wrote:
Something seems to be seriously wrong with tables. I'm working on quite
some tables today and I get more crashes. Maybe it's all related to the
same issue.
Please report all of them.
When I can reproduce one I'll report it, but it's not always easy to
find the
On Sat, 20 May 2006, Jean-Marc Lasgouttes wrote:
"Joost" == Joost Verburg <[EMAIL PROTECTED]> writes:
Joost> Hi, The current LyX website looks outdated, both in terms of
Joost> design and content. Especially now LyX is available for Windows
Joost> users as well, a modern and easy-to-use
Am Sonntag, 21. Mai 2006 15:06 schrieb Joost Verburg:
> DraftDVI is not intended to be viewed, so there should be no associated
> viewer.
I think it should be viewable, because clean_dvi.py sometimes fails (IIRC
with some hyperref constructs).
Georg
We have a new crash bug of LyX:
http://bugzilla.lyx.org/show_bug.cgi?id=2609
It seems that it occurs only in the Windows version. Could somebody
check this with a Lunix version.
thanks and regards
Uwe
Georg Baum wrote:
I think it should be viewable, because clean_dvi.py sometimes fails (IIRC
with some hyperref constructs).
I've never had any problems with hyperref and clean_dvi. Can you try to
reproduce it?
In my opinion a DraftDVI menu item is very confusing.
Joost
Am Sonntag, 21. Mai 2006 16:14 schrieb Juergen Spitzmueller:
> It might be another consequence of the problematic BufferView cache
situation
> in tabulars. Georg, do you have an idea?
I got one after a bit of debugging. The problem is not the BufferView
cache, but the use of std::swap()
Am Sonntag, 21. Mai 2006 12:42 schrieb Edwin Leuven:
> "When you create a QObject with another object as parent, it's added to
> the parent's children() list, and is deleted when the parent is."
I knew there must have been something special, so the current code is
alright. I don't do qt too
The only thing that is necessary is to add a dest_dir= option
for "sons install" so that the installation can be done in another
directory tree.
It is called prefix, the same name used by autotools.
Bo
On Sun, May 21, 2006 at 01:22:12PM -0500, Bo Peng wrote:
> >The only thing that is necessary is to add a dest_dir= option
> >for "sons install" so that the installation can be done in another
> >directory tree.
>
> It is called prefix, the same name used by autotools.
Not quite. "scons
That is how the current configure-based RPM spec file works. After configuring
with prefix=/usr, we do:
make DESTDIR=${RPM_BUILD_ROOT} install
Does this make sense?
This is beyond my understanding of this option. I always do configure
--prefix and make install without DESTDIR. Using scons, I
On Sunday 21 May 2006 21:53, Bo Peng wrote:
>
> I would like to see other's opinion regarding a separate dest_dir option.
That is the standard procedure for rpm (and also deb) building. prefix is
the place where the package will be installed and run, destdir is the place
where the package is
Please test the attached patch. It introduces dest_dir.
Bo
Index: development/scons/SConscript
===
--- development/scons/SConscript (revision 13895)
+++ development/scons/SConscript (working copy)
@@ -1235,26 +1235,17 @@
if
On Sun, May 21, 2006 at 04:59:28PM -0500, Bo Peng wrote:
> Please test the attached patch. It introduces dest_dir.
I have not tested your patch but DESTDIR is simpler than you
think, perhaps. You should do everything as specified by PREFIX
and then, at install time, copy to $DESTDIR/$PREFIX if
I have not tested your patch but DESTDIR is simpler than you
think, perhaps. You should do everything as specified by PREFIX
and then, at install time, copy to $DESTDIR/$PREFIX if DESTDIR
is defined, otherwise everything goes as before, ie, copy to $PREFIX.
DESTDIR is useful because you can
70 matches
Mail list logo
