Andre Poenitz wrote:
> At least a fat 'HACK' would be in order. Maybe even a debug message.
>
> For a proper solution we'd need something more general handling
> uncached metrics data.
But isn't the check needed anyway? At least it can not harm, even if we fix
the cache problems (which we should)
Angus Leeming wrote:
Can you see why it doesn't work:
\begin{flushleft}\includegraphics[%
width=0.9\columnwidth,
height=0.5\columnwidth]{/string"C:/Documents and Settings/Angus/Local
Settings/Temp/lyx_tmpdir5900a04256/lyx_tmpbuf0/J__MinSYS_home_Angus_ekkehart2_im_ages_picture/string".eps}
Andre Poenitz wrote:
I was hoping to solicit your comments, and ask a couple of questions.
1) Do you think this is feasable?
The equivalent of LyX's formula editor can be coded up in three weeks of
work once you know what you are doing... To know what to do starting
from scratch will probably
Jean-Marc,
I just noticed that the left margin for some label strings is too small
(at least when using LyX 1.3.6 on Windows). For instance, open a
"dinbrief"; label "Unterschrift" is clipped. I will try to fix this
during the week..
Regards, Michael
Angus Leeming wrote:
Georg Baum wrote:
Angus Leeming wrote:
Than please commit it and I'll fire up the Windows box.
Done. I really hope that it works now :-) If it does, I'll prepare a
similar patch for 1.4.
I'll try and find some time this evening to compile it and the latest
Qt/Win Free
On Sun, Jul 03, 2005 at 11:03:23AM +0200, Georg Baum wrote:
> Am Samstag, 2. Juli 2005 20:43 schrieb Andre Poenitz:
>
> > For grid-like structures it is sometimes convienient to extract row
> > and col from idx. So there are some helper functions. And sometimes
> > it is useful to give the user th
On Sat, Jul 02, 2005 at 11:23:44PM +0300, Martin Vermeer wrote:
> While most of what you write has merit, I respectfully want to play the
> devil's advocate with the attached patch, which just bites the bullet
> and re-writes all the idx()-as-lvalue statements. It eliminates idx_ and
> makes row_,
On Fri, Jul 01, 2005 at 05:20:20PM +0200, Georg Baum wrote:
> Jean-Marc Lasgouttes wrote:
>
> > I am not sure I follow the logic by reading the patch. Could you tell
> > me what the different situations may be?
>
> 1. file version is 221 -> don't call lyx2lyx
> 2. file version is not 221 -> det
On Wed, Jun 29, 2005 at 10:46:18AM +0300, Martin Vermeer wrote:
> BTW Andre, I think this bug, and the ugly looks of this brute-force
> patch, tell us something about the choice of cursor position
> representation in cursor_slice...
Could as well be the lack of beer at certain times.
> I see tha
On Fri, Jul 01, 2005 at 03:15:47PM +0100, Angus Leeming wrote:
> Isn't
> foo\end{env}
> semantically the same as
> foo%
> \end{env}
>
> ? If so, why not do that?
It is, even with additonal tabs and spaces at the beginning
of the second line.
Andre'
On Fri, Jul 01, 2005 at 01:17:52PM -0400, David Wilson wrote:
> I am working on (for Google's Summer of Code) a plugin for Gaim to
> allow conversations featuring math notation. For this plugin, I want
> to use a handy graphical interface and immediate WYSIWYG
> response...much like LyX. I stand li
On Wed, Jun 29, 2005 at 03:51:32PM +0200, Georg Baum wrote:
> Martin Vermeer wrote:
>
> > Here is the patch, armed with comments for posterity.
>
> It works here, too, but I guess that there are similar problems in
> insettabular.
> I noticed that undo does not work for the copy-column operation:
On Mon, Jun 27, 2005 at 07:33:09AM -0300, João Luis Meloni Assirati wrote:
> Hello,
>
> These small patches add the option to boldsymbol font in math panel for both
> lyx 1.3.6 and 1.4. They are for qt only, is this acceptable?
m->insertItem(qt_("Roman\\mathrm"), 1);
m->i
On Fri, Jul 01, 2005 at 02:45:03PM +0200, Juergen Spitzmueller wrote:
> http://bugzilla.lyx.org/show_bug.cgi?id=1793
>
> Just checks if the cells are actually in the cache.
Index: math_nestinset.C
===
RCS file: /usr/local/lyx/cvsroo
On Fri, Jul 01, 2005 at 05:01:09PM +0200, Georg Baum wrote:
> Jean-Marc Lasgouttes wrote:
>
> > I think that the case of mathed is much simpler, though.
>
> Yes, see the attached untested patch. Uwe, put that in your private tree and
> test it if you really want that to be included in some future
David Wilson wrote:
> Hello all,
>
> Forgive me if this isn't entirely within the purview of this list.
>
> I am working on (for Google's Summer of Code) a plugin for Gaim to
> allow conversations featuring math notation.
Hi, David. Sounds fun, but why don't I see your project here:
http://
Georg Baum wrote:
>> If we want to use stdin, stdout or stderr properly, then we need to
>> give LyX some additional capabilities to write to and read from
>> pipes. Something else entirely.
> Or use the -e switch of lyx2lyx to write errors to a temporary file
> and read that. The drawback of that
Angus Leeming wrote:
> If we want to use stdin, stdout or stderr properly, then we need to
> give LyX some additional capabilities to write to and read from
> pipes. Something else entirely.
Or use the -e switch of lyx2lyx to write errors to a temporary file and read
that. The drawback of that is
Juergen Spitzmueller wrote:
>>It looks like strftime will "just work" on Windows too.
>>http://msdn.microsoft.com/library/en-us/vclib/html/_crt_strftime.2c_.wcsftime.asp
> Cool (you could verify with m-x date-insert).
I'm not going to try and compile LyX 1.4.x on Windows using g++ ATM.
Last time
First of all, since I am the host, I get to choose I have chosen the
lousy name. So this year it will be lyx::paris.
Then, the latest grid of who's coming and when is
13 14 15 16 17 18 19Laptop
JMarcX X X X X X X X (and a PC and more if needed)
Lars X
John Levon wrote:
> > + struct tm * chng_tm = localtime(&c.changetime);
> > + char date[50];
> > + strftime(date, sizeof(date),
> > + lyxrc.date_insert_format.c_str(), chng_tm);
> > + return string(date);
>
> I can't help but feel that this should be in a support functio
John Levon wrote:
> On Mon, Jul 04, 2005 at 05:28:07PM +0200, Juergen Spitzmueller
> wrote:
>
>> +struct tm * chng_tm = localtime(&c.changetime);
>> +char date[50];
>> +strftime(date, sizeof(date),
>> +lyxrc.date_insert_format.c_str(), chng_tm);
>> +return string(date)
> "Angus" == Angus Leeming <[EMAIL PROTECTED]> writes:
Angus> The error code is what is returned by lyx2lyx when it exits:
Angus> int main() { return my_error_code; } We already have access to
Angus> this information within LyX:
Angus> but don't make much use of it. Usually, all we're interes
Jose' Matos wrote:
> On Monday 04 July 2005 12:50, Jean-Marc Lasgouttes wrote:
>>
>> OK, I think you can apply your patch for now, but I continue to
>> think that lyx2lyx should return more descriptive error codes.
>
> It is quite easy for lyx2lyx to return descriptive errors, the
> structure
On Mon, Jul 04, 2005 at 05:28:07PM +0200, Juergen Spitzmueller wrote:
> + struct tm * chng_tm = localtime(&c.changetime);
> + char date[50];
> + strftime(date, sizeof(date),
> + lyxrc.date_insert_format.c_str(), chng_tm);
> + return string(date);
I can't help but feel
Angus Leeming wrote:
> It looks like strftime will "just work" on Windows too.
> http://msdn.microsoft.com/library/en-us/vclib/html/_crt_strftime.2c_.wcsfti
>me.asp
Cool (you could verify with m-x date-insert).
The attached implementation works here, using the date settings from the
preferences.
On Monday 04 July 2005 12:50, Jean-Marc Lasgouttes wrote:
>
> OK, I think you can apply your patch for now, but I continue to think
> that lyx2lyx should return more descriptive error codes.
It is quite easy for lyx2lyx to return descriptive errors, the structure
for most part of it is already
> "João" == João Luis Meloni Assirati <[EMAIL PROTECTED]> writes:
João> Hello, These small patches add the option to boldsymbol font in
João> math panel for both lyx 1.3.6 and 1.4. They are for qt only, is
João> this acceptable?
This seems reasonable enough, since xforms does not have this me
> "João" == João Luis Meloni Assirati <[EMAIL PROTECTED]> writes:
João> Hi again, A quick comparison of 1.3.x and 1.4 sources suggested
João> the attached patch to the problem. It seems to work well.
I applied your patch to 1.3.6cvs.
JMarc
Helge Hafting wrote:
> I can insert a label (insert label) into inlined math, and then
> I may reference that label using the cross reference dialog.
>
> Unfortunately, such a label disappears and screws up output.
> Save the document, reload it, and the label is gone from lyx too!
>
> Now, perh
Helge Hafting wrote:
> Document settings have something called "branches", which
> wasn't self-explanatory to me. The word wasn't mentioned in
> the userguide either.
These are environments (insets) that can be shown/hidden. E.g., define a
branch "results" in document branches, then you can inse
I thought ctime returns translated strings if LC_TIME is set
correctly? Obviously not the case.
>> strftime is the right thing, probably.
> Yes, I'm just playing with it. Then at least the lyxrc.date_format
> gets some more sense again.
It looks like strftime will "just work" on Window
Helge Hafting wrote:
> Document settings have something called "branches", which
> wasn't self-explanatory to me. The word wasn't mentioned in
> the userguide either.
"Branches" are something new to LyX 1.4 and their creator, Martin
Vermeer should really add some documentation to the User Guide..
John Levon: wrote
> > > I thought ctime returns translated strings if LC_TIME is set
> > > correctly? Obviously not the case.
>
> strftime is the right thing, probably.
Yes, I'm just playing with it. Then at least the lyxrc.date_format gets some
more sense again.
Jürgen
On Mon, Jul 04, 2005 at 01:37:21PM +0100, Angus Leeming wrote:
> > I thought ctime returns translated strings if LC_TIME is set
> > correctly? Obviously not the case.
strftime is the right thing, probably.
regards
john
Juergen Spitzmueller wrote:
> Angus Leeming wrote:
>> > (Only the actual date is still untranslated)
>> I don't think that there's anything we can do about the translation
>> of strings returned by system functions (here ctime).
> I thought ctime returns translated strings if LC_TIME is set
> cor
Document settings have something called "branches", which
wasn't self-explanatory to me. The word wasn't mentioned in
the userguide either.
I believe I have seen the topic on this list. Unfortunately,
searching for it turns up tons of messages about the
development branches (1.4 and 1.3), drown
> "Helge" == Helge Hafting <[EMAIL PROTECTED]> writes:
Helge> This time with the newly available vspace translations,
Helge> as well as a few more xforms translations.
Thanks. Applied.
JMarc
Angus Leeming wrote:
> > (Only the actual date is still untranslated)
>
> I don't think that there's anything we can do about the translation of
> strings returned by system functions (here ctime).
I thought ctime returns translated strings if LC_TIME is set correctly?
Obviously not the case.
Jü
This time with the newly available vspace translations,
as well as a few more xforms translations.
Helge Hafting
Index: no.po
===
RCS file: /cvs/lyx/lyx-devel/po/no.po,v
retrieving revision 1.97
diff -u -p -u -r1.97 no.po
--- no.po
Juergen Spitzmueller wrote:
> Angus Leeming wrote:
>> This isn't the right way to internationalise this, is it? The right
>> way is something like:
>> text += boost::format(_("Change by %s\n\n"), author);
>
> OK, I'll commit the attached, if you feel comfortable with it.
It looks good to me.
> (
Georg Baum wrote:
> Angus Leeming wrote:
>> Than please commit it and I'll fire up the Windows box.
> Done. I really hope that it works now :-) If it does, I'll prepare a
> similar patch for 1.4.
I'll try and find some time this evening to compile it and the latest
Qt/Win Free which should fix ke
> "Juergen" == Juergen Spitzmueller <[EMAIL PROTECTED]> writes:
Juergen> Jean-Marc Lasgouttes wrote: Yes. See my second patch.
>> I saw that after sending my message. I prefer this version for
>> 1.3.6cvs.
Juergen> Which? The second?
Juergen> Does that mean the (second) patch can be committ
Jean-Marc Lasgouttes wrote:
> Juergen> Yes. See my second patch.
>
> I saw that after sending my message. I prefer this version for
> 1.3.6cvs.
Which? The second?
Does that mean the (second) patch can be committed?
Jürgen
> "Georg" == Georg Baum <[EMAIL PROTECTED]> writes:
>> Isn't it possible to run lyx2lyx and check whether there has been
>> an error? I think having lyx2lyx return an error code would be much
>> more robust than parsing its output.
Georg> That would require to change lyx2lyx to return differ
> "Juergen" == Juergen Spitzmueller <[EMAIL PROTECTED]> writes:
Juergen> Jean-Marc Lasgouttes wrote:
>> What about floats().getType(type).name() where floats is a member
>> of the textclass?
Juergen> Yes. See my second patch.
I saw that after sending my message. I prefer this version for
1
Angus Leeming wrote:
> This isn't the right way to internationalise this, is it? The right way is
> something like:
> text += boost::format(_("Change by %s\n\n"), author);
OK, I'll commit the attached, if you feel comfortable with it.
(Only the actual date is still untranslated)
Jürgen
I
Angus Leeming wrote:
> Than please commit it and I'll fire up the Windows box.
Done. I really hope that it works now :-) If it does, I'll prepare a similar
patch for 1.4.
Georg
Jean-Marc Lasgouttes wrote:
> What about
> floats().getType(type).name()
> where floats is a member of the textclass?
Yes. See my second patch.
> Juergen> Jean-Marc, this could also be a candidate for 1.3.6.
>
> I would really like to see a patch where only the type of float is
> used internall
Georg Baum wrote:
>> Don't we therefore need:
>> \includegraphics{\string"my dir/my picture\string".png}
> Yes. And the patch should exactly do that! If it does not, please shout!
Than please commit it and I'll fire up the Windows box.
--
Angus
Forwarding this so that there's a record that cntl-@ wouldn't work
with Qt4.
Angus
--- Begin Message ---
> --- Ursprüngliche Nachricht ---
> Von: Angus Leeming <[EMAIL PROTECTED]>
> An: "Christian Ehrlicher" <[EMAIL PROTECTED]>
> Betreff: Re: KeyEvent handling
> Datum: Mon, 4 Jul 2005 11:37:18 +0
> "Juergen" == Juergen Spitzmueller <[EMAIL PROTECTED]> writes:
Juergen> http://bugzilla.lyx.org/show_bug.cgi?id=1870 The problem was
Juergen> that we compared two different string types (GuiName and
Juergen> type), one of which is translatable, moreover.
Juergen> The attached patch fixes it.
Angus Leeming wrote:
> Hm. I read this as outputting
> \includegraphics{\string"my dir/my picture\string"}
No. The magic happens in latex_path: strip extension, add quotes, add
extension.
> I thought that missing out the extension would break latex?
Yes.
> Don't we
> therefore need:
>
Paul A. Rubin wrote:
> Don't know which of these are known so far:
>
> In an existing Beamer presentation (works fine with 1.3.5) containing
> .pdf and .png graphics, calling View->PDF (pdflatex) gets me a half
> dozen error messages, all "Unknown graphics extension: .XXX" where XXX
> is either pd
>>>Angus> Qt/Win Free doesn't handle Alt-foo key presses. Maybe we could
>>>Angus> get them to backport something from the just-released Qt 4.0?
>>>That is a big problem indeed. Didn't it use to work?
>>Not with Qt/Win Free. It works with the Qt non-commercial that Ruurd
>>used, but I'm not going
Georg Baum wrote:
> Conclusion: We can't make spaces work with pdfetex from tetex, but we can
> support them on miktex. Therefore I propose the following patch for 1.3.6
> (I'll send a similar one for 1.4 if this is OK).
Hm. I read this as outputting
\includegraphics{\string"my dir/my pic
> "Helge" == Helge Hafting <[EMAIL PROTECTED]> writes:
Helge> I see the patch is in - but still no strings in no.po. I guess
Helge> there is something I should do to have the strings pulled into
Helge> the po file, but what?
Try something like:
rm lyx.pot
make lyx.pot
make no.po-update
If y
Juergen Spitzmueller wrote:
> Helge Hafting wrote:
>> menu "merge changes"
>> The window starts out blank. It'd be better if it started with the
>> first change. Now I have to press the button "Next Change" as
>> the first thing I do - the dialog ought to do that
>> automatically.
>> Not a serio
> "Helge" == Helge Hafting <[EMAIL PROTECTED]> writes:
Helge> A few more updates to no.po. Some error messages, and a few
Helge> corrections/refinements on my earlier work.
Applied.
JMarc
A few more updates to no.po. Some error messages, and a few
corrections/refinements
on my earlier work.
Helge Hafting
Index: no.po
===
RCS file: /cvs/lyx/lyx-devel/po/no.po,v
retrieving revision 1.95
diff -u -p -u -r1.95 no.po
---
Jean-Marc Lasgouttes wrote:
Helge> - The string "LyX keyboard maps (*.kmap)" is also missing, and
Helge> it surely isn't a qt string. It isn't broken into shorter
Helge> pieces either, the word "maps" does not exist in no.po at all.
It is a bug, fixed by the attached patch that will be committe
Juergen Spitzmueller wrote:
> The attached patch fixes it. It would be cleaner if there was a method to
> get the GuiName from the type, but I have found no such method.
Attached is a more general, IMO better solution. For 1.3.6, maybe the first
patch would do.
Jürgen
Index: toc.C
==
I can insert a label (insert label) into inlined math, and then
I may reference that label using the cross reference dialog.
Unfortunately, such a label disappears and screws up output.
Save the document, reload it, and the label is gone from lyx too!
Now, perhapsh it doesn't make sense to have
David Wilson wrote:
Hello all,
Forgive me if this isn't entirely within the purview of this list.
I was hoping to solicit your comments, and ask a couple of questions.
1) Do you think this is feasable?
It is possible, but not easy. It will take a long time if you want to
integrate this in an
64 matches
Mail list logo
