I edit a file that I was happily editing an update or two ago:
Layout 'Scrap' does not exist in textclass 'article'.
Trying to use default layout instead.
The document class is literate-article.
This happens on Linux, Solaris and Cygwin.
---Kayvan
--
Kayvan A. Sylvan
On Sat, Jun 21, 2003 at 01:26:14AM +0200, Alfredo Braunstein wrote:
> frontends/controllers/ControlError.C | 30 -
> frontends/controllers/ControlError.h | 40 -
At last !
Pre-approved of course
Thanks for all your work on this
If you're really proactive, you can close a
frontends/controllers/ControlError.C | 30 -
frontends/controllers/ControlError.h | 40 -
frontends/controllers/Makefile.am|2
frontends/qt2/Dialogs.C |6 -
frontends/qt2/Makefile.am|1
frontends/qt2/Makefile.dialogs |2
fr
Alfredo Braunstein <[EMAIL PROTECTED]> writes:
| Please comment, Alfredo
Looks good.
Please commit.
--
Lgb
John Levon wrote:
> Can't we remove unknown_tokens altogether then ?
> where it is used ?
Attached is the above +
o make use of the new signal to provide gui-less parse errors console
feedback (dumped to cerr)
o pimpled BufferView::addError
o removed Buffer::sgmlError, and use uniformly the s
John Levon wrote:
>> - if (unknown_tokens > 0) {
>
> Can't we remove unknown_tokens altogether then ?
> where it is used ?
Yes, I think it's not used anywhere else.
Alfredo
On Fri, Jun 20, 2003 at 05:10:52PM +0200, Alfredo Braunstein wrote:
> - I had eliminated the dialog showing the total number of unknown tokens. I
> don't think it's an useful information, but I can bring it back in some way
> if it is.
The errorlist dialog is a far superior solution anyway.
rega
Below is finally a patch to get rid of the last inseterror user.
Remarks:
- I had to invert the order of a) filling a paragraph and b) inserting it
into the parlist, to be able to have the definitive parid while parsing. I
think it may be a good thing anyway, as an empty par is faster to copy th
On Fri, Jun 20, 2003 at 03:52:22PM +0100, John Levon wrote:
> On Fri, Jun 20, 2003 at 04:51:56PM +0200, Andre Poenitz wrote:
>
> > I thought there is one?
>
> The branch tag is a different thing from a point tag. I didn't see what
> command you typed to branch off ...
As Lars suggested:
cvs ta
On Fri, Jun 20, 2003 at 04:51:56PM +0200, Andre Poenitz wrote:
> I thought there is one?
The branch tag is a different thing from a point tag. I didn't see what
command you typed to branch off ...
john
On Fri, Jun 20, 2003 at 03:40:53PM +0100, John Levon wrote:
> On Fri, Jun 20, 2003 at 04:36:50PM +0200, Andre Poenitz wrote:
>
> > So thius branching is essentially nothing else than a separate CVS
> > repository and the merging has to be done manually?
>
> Uh no. It's a bit more awkward than it
On Fri, Jun 20, 2003 at 04:36:50PM +0200, Andre Poenitz wrote:
> So thius branching is essentially nothing else than a separate CVS
> repository and the merging has to be done manually?
Uh no. It's a bit more awkward than it should be but it's definitely not
manual.
I'll do all the merges if you
On Fri, Jun 20, 2003 at 03:33:25PM +0100, John Levon wrote:
> On Fri, Jun 20, 2003 at 03:41:29PM +0200, Andre Poenitz wrote:
>
> > Is it ok to 'cvs commit' as usual after creating this branch or do I have
> > to do something special?
>
> The tricky bit is merging from the trunk onto the branch. W
Minor, but...
-Version 1.3.1 has been released, anyone interested in testing it should
+Version 1.3.2 has been released, anyone interested in testing it should
download it from the LyX FTP server.
--
Angus
On Fri, Jun 20, 2003 at 03:41:29PM +0200, Andre Poenitz wrote:
> Is it ok to 'cvs commit' as usual after creating this branch or do I have
> to do something special?
The tricky bit is merging from the trunk onto the branch. We should do
it regularly, and to do it right you need to move a tag alon
On Fri, Jun 20, 2003 at 01:13:25PM +0200, Andre Poenitz wrote:
> If there were a 'updateless' CVS branch, would anybody help working on it?
Yes. A branch means I have a playground to try things out for the bugs
I'm seeing.
I'm away for at least two weeks very soon though.
john
On Fri, Jun 20, 2003 at 10:54:05AM +, Angus Leeming wrote:
> What is the rule for multiple adjacent '/' in paths? I believe that they are
> treated as a single '/'. Am I right?
POSIXly, yes, except at the very beginning where "^//..." can indicate a
different namespace ;)
john
Andre Poenitz <[EMAIL PROTECTED]> writes:
| On Fri, Jun 20, 2003 at 01:56:02PM +0200, Lars Gullik Bjønnes wrote:
| > cvs tag BRANCH_POINT_NOUPDATE
| > cvs rtag -r BRANCH_POINT_NOUPDATE -b BRANCH_NOUPDATE lyx-devel
|
| cd ..
not needed
| > cvs co -r BRANCH_NOUPDATE
|
| cvs co -d lyx-devel-noup
Andre Poenitz <[EMAIL PROTECTED]> writes:
| Is it ok to 'cvs commit' as usual after creating this branch or do I have
| to do something special?
Just check that in the tree that you want to commit in, have a file
"Tag" in the CVS dir. This "Tag" contents should then be
"BRANCH_NOUPDATE"
--
> "Andre" == Andre Poenitz <[EMAIL PROTECTED]> writes:
Andre> Is it ok to 'cvs commit' as usual after creating this branch or
Andre> do I have to do something special?
You can work as usual.
JMarc
Angus Leeming wrote:
> The usual trick is to #include the qt stuff _after_ the rest.
>
> #include "buffer.h"
> #include "QDocument.h"
> etc
Thank, that indeed worked.
Alfredo
> "Angus" == Angus Leeming <[EMAIL PROTECTED]> writes:
Angus> (I tried using NormalisePath in tex_helpers.C's getTexFilesList
Angus> but end up with things like
Angus> /usr/share/texmf/tex/generic/xypic/movie.cls/ Note the trailing
Angus> slash.)
Yes, because it is supposed to be used on a pa
Andre Poenitz wrote:
> On Fri, Jun 20, 2003 at 03:39:50PM +0200, Alfredo Braunstein wrote:
>> How to proceed?
>
> Re-order includes?
That did it, thanks.
Alfredo
taking this to lyx.devel...
Alfredo Braunstein wrote:
> qt headers include a nice '#define signals xxx' (puaj) and boost seems to
> choke on that.
> if I #include "boost/signals/signal1.hpp" in buffer.h I get a clash in
> QDocument.C:
>
> How to proceed?
The usual trick is to #include the qt stu
On Fri, Jun 20, 2003 at 03:39:50PM +0200, Alfredo Braunstein wrote:
> How to proceed?
Re-order includes?
Andre'
--
Those who desire to give up Freedom in order to gain Security, will not have,
nor do they deserve, either one. (T. Jefferson or B. Franklin or both...)
qt headers include a nice '#define signals xxx' (puaj) and boost seems to
choke on that.
if I #include "boost/signals/signal1.hpp" in buffer.h I get a clash in
QDocument.C:
How to proceed?
In file included from /home/ab/lyx-cvs2/boost/boost/signals/slot.hpp:21,
from
/home/ab/lyx-
Jean-Marc Lasgouttes wrote:
>>> There is some code in TeXFiles.sh to remove // already, so I am
>>> surprised.
>
> Angus> This you mean? dirs=`kpsewhich --show-path=$kpsetype
> Angus> 2>/dev/null | tr "$SEP" " " | sed -e 's%///%/%' -e 's%//%/%g'
> Angus> -e 's%!!%%g'`
>
> Angus> s%///%/% needs a
Is it ok to 'cvs commit' as usual after creating this branch or do I have
to do something special?
Andre'
--
Those who desire to give up Freedom in order to gain Security, will not have,
nor do they deserve, either one. (T. Jefferson or B. Franklin or both...)
> "Angus" == Angus Leeming <[EMAIL PROTECTED]> writes:
>> There is some code in TeXFiles.sh to remove // already, so I am
>> surprised.
Angus> This you mean? dirs=`kpsewhich --show-path=$kpsetype
Angus> 2>/dev/null | tr "$SEP" " " | sed -e 's%///%/%' -e 's%//%/%g'
Angus> -e 's%!!%%g'`
Angus>
On Fri, 20 Jun 2003, Nirmal Govind wrote:
> I'm not sure why you resorted to the article class.. the CV class
> included with LyX can produce something very similar..
very short on time... I've been looking a bit at the CV class now.
>
> I created mine using the CV class.. it's at:
>
> www.pers
On Fri, Jun 20, 2003 at 01:56:02PM +0200, Lars Gullik Bjønnes wrote:
> cvs tag BRANCH_POINT_NOUPDATE
> cvs rtag -r BRANCH_POINT_NOUPDATE -b BRANCH_NOUPDATE lyx-devel
cd ..
> cvs co -r BRANCH_NOUPDATE
cvs co -d lyx-devel-noupdate -r BRANCH_NOUPDATE lyx-devel
>
> or something...
And when I have
Alfredo Braunstein <[EMAIL PROTECTED]> writes:
| Lars Gullik Bjønnes wrote:
|
| > Looks good.
|
| Updated patch below for if someone else wants to test it (would be nice).
| Tell me when you want me to commit.
I think you can do that right away.
--
Lgb
On Friday 20 June 2003 12:24 pm, Jean-Marc Lasgouttes wrote:
> > "Angus" == Angus Leeming <[EMAIL PROTECTED]> writes:
>
> Angus> You know how it is when a little change cascades all over the
> Angus> place? A small fix, but lots of cleaner code.
>
> It looks OK. Do we need the change to Normali
On Friday 20 June 2003 12:20 pm, Jean-Marc Lasgouttes wrote:
> What does
> kpsewhich --show-path=.tex
> tell you?
$ kpsewhich --show-path=.tex
.:!!/home/angus/texmf/tex/generic//:!!/usr/local/share/texmf/tex/generic//:!!/usr/share/texmf/tex/generic//:!!/home/angus/texmf/tex///:!!/usr/local/share
Lars Gullik Bjønnes wrote:
> Looks good.
Updated patch below for if someone else wants to test it (would be nice).
Tell me when you want me to commit.
Regards, Alfredo
Index: BufferView.C
===
RCS file: /usr/local/lyx/cvsroot/lyx-de
> "Angus" == Angus Leeming <[EMAIL PROTECTED]> writes:
Angus> You know how it is when a little change cascades all over the
Angus> place? A small fix, but lots of cleaner code.
It looks OK. Do we need the change to NormalizePath?
JMarc
> "Angus" == Angus Leeming <[EMAIL PROTECTED]> writes:
>> But why do our scripts generate such nonsense? Why is the same
>> path searched twice?
Angus> This is orthogonal to my suggestion to modify NormalizePaths.
Yes, but it is better to fix the problem at the root.
Angus> Nonetheless, it
Andre Poenitz wrote:
> + if (!data.empty())
> + data.clear();
> +
>
> Just data.clear() should suffice...
Thanks. Done.
--
Angus
On Fri, Jun 20, 2003 at 12:50:02PM +, Angus Leeming wrote:
> You know how it is when a little change cascades all over the place?
> A small fix, but lots of cleaner code.
+ if (!data.empty())
+ data.clear();
+
Just data.clear() should suffice...
Andre'
--
Those who de
Andre Poenitz wrote:
> On Fri, Jun 20, 2003 at 12:42:48PM +, Angus Leeming wrote:
>> Andre Poenitz wrote:
>> > [And tell me how to use it... cvs co -r noupdate lyx-devel?]
>>
>> Attached is Lars original post on this, suitably corrected ;-) I followed
>> the prescription faithfully when I bra
Andre Poenitz <[EMAIL PROTECTED]> writes:
| On Fri, Jun 20, 2003 at 01:28:22PM +0200, Lars Gullik Bjønnes wrote:
| > Andre Poenitz <[EMAIL PROTECTED]> writes:
| >
| > | If there were a 'updateless' CVS branch, would anybody help working on it?
| >
| > I cannot promise much help, but I will certa
You know how it is when a little change cascades all over the place?
A small fix, but lots of cleaner code.
Ok to apply?
Angus
$ diffstat cleanup.diff
lib/ChangeLog |4 +
lib/scripts/TeXFiles.sh|4 +
src/frontends/controllers/ChangeLog
On Fri, Jun 20, 2003 at 12:42:48PM +, Angus Leeming wrote:
> Andre Poenitz wrote:
> > [And tell me how to use it... cvs co -r noupdate lyx-devel?]
>
> Attached is Lars original post on this, suitably corrected ;-) I followed
> the prescription faithfully when I branched off the MCV and NATBIB
Andre Poenitz wrote:
> [And tell me how to use it... cvs co -r noupdate lyx-devel?]
Attached is Lars original post on this, suitably corrected ;-) I followed
the prescription faithfully when I branched off the MCV and NATBIB stuff
and it worked perfectly.
--
Angus
On Fri, Jun 20, 2003 at 01:34:59PM +0200, Alfredo Braunstein wrote:
> Andre Poenitz wrote:
>
> > If there were a 'updateless' CVS branch, would anybody help working on it?
>
> I can help you with testing, even if I have proven that I'm even not so
> useful for that... ;^)
It's not testing, it's
Andre Poenitz wrote:
> If there were a 'updateless' CVS branch, would anybody help working on it?
I can help you with testing, even if I have proven that I'm even not so
useful for that... ;^)
Regards, Alfredo
On Fri, Jun 20, 2003 at 01:28:22PM +0200, Lars Gullik Bjønnes wrote:
> Andre Poenitz <[EMAIL PROTECTED]> writes:
>
> | If there were a 'updateless' CVS branch, would anybody help working on it?
>
> I cannot promise much help, but I will certainly test it, and do stuff
> that will help this work.
Andre Poenitz <[EMAIL PROTECTED]> writes:
| If there were a 'updateless' CVS branch, would anybody help working on it?
I cannot promise much help, but I will certainly test it, and do stuff
that will help this work.
--
Lgb
Andre Poenitz wrote:
> If there were a 'updateless' CVS branch, would anybody help working on it?
> I have right now a completely disfunctional patch which I believe is more
> than halfway through so that I really do not want to lose it...
Not me. I am about to dissappear of the radar screens for
If there were a 'updateless' CVS branch, would anybody help working on it?
I have right now a completely disfunctional patch which I believe is more
than halfway through so that I really do not want to lose it...
Andre'
--
Those who desire to give up Freedom in order to gain Security, will not
On Friday 20 June 2003 10:25 am, Jean-Marc Lasgouttes wrote:
> > "Angus" == Angus Leeming <[EMAIL PROTECTED]> writes:
>
> Angus> Trying to clean up the Texinfo dialog, I find that although we
> Angus> try and eliminate duplicate entries in the list of TeX files,
> Angus> things like
>
> Angus>
> "Angus" == Angus Leeming <[EMAIL PROTECTED]> writes:
Angus> Trying to clean up the Texinfo dialog, I find that although we
Angus> try and eliminate duplicate entries in the list of TeX files,
Angus> things like
Angus> /usr/share/texmf/tex/generic/xypic/movie.cls
Angus> /usr/share/texmf/tex/
Angus Leeming <[EMAIL PROTECTED]> writes:
| Lars Gullik Bjønnes wrote:
| > | Should I modify filetools' NormalizePath to replace multiple adjacent
| > | '/' with a single one? It is as simple as:
| > |
| > | boost::RegEx regex("/{2,}");
| >
| > Why the comma?
|
| "Two or more".
ah...
Lars Gullik Bjønnes wrote:
> | Should I modify filetools' NormalizePath to replace multiple adjacent
> | '/' with a single one? It is as simple as:
> |
> | boost::RegEx regex("/{2,}");
>
> Why the comma?
"Two or more".
--
Angus
Angus Leeming <[EMAIL PROTECTED]> writes:
| Trying to clean up the Texinfo dialog, I find that although we try and
| eliminate duplicate entries in the list of TeX files, things like
|
| /usr/share/texmf/tex/generic/xypic/movie.cls
| /usr/share/texmf/tex//generic/xypic/movie.cls
|
| slip throug
On Fri, Jun 20, 2003 at 10:54:05AM +, Angus Leeming wrote:
> Trying to clean up the Texinfo dialog, I find that although we try and
> eliminate duplicate entries in the list of TeX files, things like
>
> /usr/share/texmf/tex/generic/xypic/movie.cls
> /usr/share/texmf/tex//generic/xypic/movie.
Trying to clean up the Texinfo dialog, I find that although we try and
eliminate duplicate entries in the list of TeX files, things like
/usr/share/texmf/tex/generic/xypic/movie.cls
/usr/share/texmf/tex//generic/xypic/movie.cls
slip through.
What is the rule for multiple adjacent '/' in paths?
Lars Gullik Bjønnes wrote:
> Alfredo Braunstein <[EMAIL PROTECTED]> writes:
>
> | It wasn't, but it doesn't make a difference. I got the same bt. Then
> | retrying under gdb, I got also this one:
>
> Actaully it makes a huge difference, no the line numbers and files
> match up.
Of course. I mea
Alfredo Braunstein <[EMAIL PROTECTED]> writes:
| It wasn't, but it doesn't make a difference. I got the same bt. Then
| retrying under gdb, I got also this one:
Actaully it makes a huge difference, no the line numbers and files
match up.
And no signal raised...
what is "it" here?
--
L
Angus Leeming wrote:
> Alfredo Braunstein wrote:
>
>> +bool readFile(Buffer * b, string const & s)
>> +{
>> + string ts(s);
>> + string e = OnlyPath(s);
>> + string a = e;
>> + // File information about normal file
>> + FileInfo fileInfo2(s);
>> +
>
> Being pedantic
Angus Leeming <[EMAIL PROTECTED]> writes:
| +/* \file buffer_funcs.h
| +/**
| + * Loads a LyX file \param filename into \param Buffer
| + * and \return success status.
| + */
| +bool loadLyXFile(Buffer *, string const & filename);
|
| Apparently these comments should be \c filename rather than
Alfredo Braunstein wrote:
> +bool readFile(Buffer * b, string const & s)
> +{
> + string ts(s);
> + string e = OnlyPath(s);
> + string a = e;
> + // File information about normal file
> + FileInfo fileInfo2(s);
> +
Being pedantic, why do you call it fileInfo2 and not
Lars Gullik Bjønnes wrote:
> Alfredo Braunstein <[EMAIL PROTECTED]> writes:
>
> | With current cvs, only by opening and closing some documents randomly,
> | at some point I get this abort.
> |
> | I doesn't happend very often (and doesn't depend on the particular
> | documents).
>
> Is this a n
63 matches
Mail list logo
