Re: Re[2]: Python script conversion

[Bo Peng]

> What will happen if I try to send an attachment to the mailing list?
> Is it allowed?
>

Please send the detailed instructions on how to do this to the list.
After we get rid of perl and mingw and have python scripts compiled
like this, we will be much closer to an everything-included lyx
distribution ( convert.exe and aspell.exe should be easy to handle,
and packing miktex is not all that necessary. ), although many people
do not like this idea.

Cheers,
 Bo

Re: Grumpy me

[Juergen Spitzmueller]

Jean-Marc Lasgouttes wrote:
> There is also a bug that I never filed :) Insert>File is dead broken.

In which respect?

Jürgen

Re[2]: Python script conversion

[Alex]

Dear Mike,


>> > The downside is that you wind up with one copy of Python installed for
>> > every script. If you've only got one script, this is ok. If you've got
>> > more than one, you probably want to consider another approach.
>> It will be the same, using several scripts, too.
>> 
>> I have tried, I can do the following variant:
>> - place all the python scripts into the script dir
>> - create .exe files for the scripts
>> - create an archive file containing python
>> 
>> In this case, you will have only one python packaged , but several small
>> scrits converted to .exe.

> Unless you're doing something strange, the process of creating the
> .exe files will put the python dll and copies of all the library files
> into the .exe. Are you doing something to prevent that? And if so -
> I'd be interested in knowing what.
Yes, I do a little different. Have you read the py2exe documentation?
Also have you looked at
http://wiki.lyx.org/uploads/LyX/Windows/lyx2lyx-1.4.0cvs-win32-bin.zip this
link. This conversion using my approach. Placing the necessary thins into an 
arcive
file, thus making the .exe file small enough to load the rest.

It contains similar things. I could build small executables with python
libraries in an archive file.
I don't know anything about python. I could read documentation, I could try to
understand how python could work, but currently that's all.

>> A few converted scritps look like this:
>> PYTHON23 DLL   979 005  05.02.08  16.23 python23.dll
>> W9XPOPEN EXE16 384  05.02.08  16.24 w9xpopen.exe
>> LIBRARY  ZIP   311 060  05.10.12  19.30 library.zip
>> LYX2LYX  EXE40 960  05.10.12  19.30 lyx2lyx.exe
>> GENERA~1 EXE28 672  05.10.12  19.30 general_command_wrapper.exe
>> PIC2AS~1 EXE28 672  05.10.12  19.30 pic2ascii.exe
>> 
>> So, they can use the same interpreter.

> This amounts to simply bundling the Python interpreter with LyX. I
> think that's a great option - in fact, I think someone should bundle
> up a complete system, with all the dependencies included - but I'm
> pretty sure that shouldn't be the only binary offering available.

For early lyxes this was the case. Ruurd Reitsma made the same, he was packaged
python into the windows installer (so made me troubles after installing another
one) :). Therfore, I prefer this .exe variant.

I have played a lot with this conversion tool. You can customise it to fit
your needs. It is very flexibile.

What will happen if I try to send an attachment to the mailing list?
Is it allowed?

-- 
Best regards,
 Alexmailto:[EMAIL PROTECTED]

Re[2]: Python script conversion

[Alex]

Dear Bo,

>> > I found that I could convert all the python script to .exe.
>> > Thus I can remove Pyhthon dependecy.
>>
>> Are you sure?

> Also, there are some dynamically generated sh scripts that might be
> translated to dynamically generated python script later so an
> interpreter is needed.

Is that means that we could decrease the necessary surrounding programs to
approximately one. That should supply as a script platform for all the 
configuration
and other (eg. conversion) purposes?

I would prefer a prepackaged python.
If we supply python with lyx, the user will feel comfortable, because (s)he wont
need to install other software (worring about correct path env. var.) to get 
the software
completly working.

From the opposite side, in case of python 2.4.x, also there is some trouble with
license. When I do package the scripts, it will copy the "MSVCR71.dll" next to 
the
archive. This file has not so good licensing conditions, for details see px2exe
mailing list.

Also the 2.4.x python.dll is quite larger, making slower to load
at runtime. Therefore I had python 2.3.5 for packaging (to reduce donwload
time, and to avoid licensing conditions).

-- 
Best regards,
 Alexmailto:[EMAIL PROTECTED]

Re: ICU - uneasy feeling

[Angus Leeming]

Lars Gullik Bjønnes wrote:
> I have been trying to look at the ICU api, but I find the
> documentation utterly confusing and hard to get a clear understanding
> on how it works. (Probably caused be me not finding a "Hello World"
> code snippet)
> 
> Also, I must say, some of this is based on really old (before 2000)
> ideas on how to write portable C++. And at least to me some of this
> feel antiquated. Makes me feel a bit uneasy.
> 
> Would be nice if some of you could have a look at this lib as well,
> and see what you think of it. I know it is _The_ Unicode lib to use,
> but still...

Perhaps it would help us all if you tried to map out exactly where you
propose to use ICU? If I understand correctly, we'd have:

LyX file format in UTF-8 --ICU--> data in USC-4 encoded std::wstrings

data in memory --ICU--> LyX file

data in memory --ICU--> Multibyte char strings used by XForms

data in memory --> QString, no conversion needed?

If that's correct, then effectively LyX doesn't see the unicode at all. It
continues to operate on a char_type-based string, where char_type is now
wchar_t, or uint or whatever is portable.

Again, if that's correct, then I don't think that the ICU interface matters
at all. Wrap it up in your own interface and leave ICU as an
implementation detail.

-- 
Angus

ICU - uneasy feeling

[Lars Gullik Bjønnes]

I have been trying to look at the ICU api, but I find the
documentation utterly confusing and hard to get a clear understanding
on how it works. (Probably caused be me not finding a "Hello World"
code snippet)

Also, I must say, some of this is based on really old (before 2000)
ideas on how to write portable C++. And at least to me some of this
feel antiquated. Makes me feel a bit uneasy.

Would be nice if some of you could have a look at this lib as well,
and see what you think of it. I know it is _The_ Unicode lib to use,
but still...

-- 
Lgb

Re: 1.4.0pre2 on cygwin

[Kayvan A. Sylvan]

On Wed, Oct 12, 2005 at 09:50:53PM +, Luis Rivera wrote:
> > Are you trying to compile for xforms or for qt?
> 
> I did it with xforms.  
> 
> I'm trying with qt...
> 
> Now I'll try --with-included-gettext (thanks to Lars!)
> 
> > I do both all the time with LyX CVS. Maybe I can help you.
> 
> Do you? On cygwin?!

Yes, on cygwin. I have no trouble at all.

There are a couple of vagaries:

1) Any compiler more recent than gcc-3.3.3 seems to screw me up. Even
   if I recompile xforms and qt-3.

2) The qt-3 build needs you to use a linker script in the final ld phase.

I have attached my build script. I use this with the latest CVS lyx
almost every day.

My directories are set up like this:

build_stage/
   lyx/ <--- This is the CVS lyx directory
   qtbuild/ <--- Where the QT version is built
   build/ <--- for the xforms version
   build_lyx <--- The build script attached to this email.

I do this:

  cd build_stage/qtbuild
  ../build-lyx all-qt

And QT lyx is installed in /usr/local/bin

  cd ../build
  ../build_lyx all

to install xforms lyx.

You will probably have to change some paths in the script (it was a quick
and dirty thing --- not really meant to be a generic build script).

I hope you find it useful.

---Kayvan
-- 
Kayvan A. Sylvan  | Proud husband of   | Father to my kids:
Sylvan Associates, Inc.   | Laura Isabella Sylvan  | Katherine Yelena (8/8/89)
http://sylvan.com/~kayvan | "crown of her husband" | Robin Gregory (2/28/92)
#!/bin/sh

LYXSRC=$LYXSRC # from env
if [ "x"$LYXSRC = "x" ]
then
  LYXSRC=../lyx
  if [ -z "$MKLYXDBG" ]
  then
echo "No LYXSRC environment, assuming $LYXSRC"
MKLYXDBG=1
export MKLYXDBG
  fi
fi

LYXBUILD=`pwd`

QT_LD_SCRIPT=/home/kayvan/src/qt-3/mkspecs/cygwin-g++/i386pe.x-no-rdata

usage() {
  echo "Usage: $0 [help|cvs|config|debug|config-qt|debug-qt|build|install]";
  echo "You can also use \"all\" or \"all-qt\" for combined actions."
  exit 0;
}

while [ $# -ge 1 ]
do
  arg=$1; shift;
  case $arg in
  all)
$0 cvs config build install
;;
  all-qt)
$0 cvs config-qt build install
;;
  help)
usage
;;
  cvs)
cd $LYXSRC; cvs update -dP 2>&1 | grep -v 'cvs server'
;;
  config)
cd $LYXSRC; ./autogen.sh; cd $LYXBUILD
$LYXSRC/configure --disable-debug --enable-optimization=-O2 \
--with-frontend=xforms \
--with-extra-lib=/usr/local/lib \
--with-extra-inc=/usr/local/include
;;
  debug)
cd $LYXSRC; ./autogen.sh; cd $LYXBUILD
$LYXSRC/configure --enable-debug --enable-optimization=-g \
--with-frontend=xforms \
--with-extra-lib=/usr/local/lib \
--with-extra-inc=/usr/local/include
;;
  config-qt)
cd $LYXSRC; ./autogen.sh; cd $LYXBUILD
LDFLAGS="-L/usr/local/lib  -Wl,--export-all-symbols \
 -Wl,--enable-runtime-pseudo-reloc \
 -Wl,--script,$QT_LD_SCRIPT" \
$LYXSRC/configure --disable-debug --enable-optimization=-O2 \
--with-extra-lib=/usr/local/lib \
--with-extra-inc=/usr/local/include \
--with-frontend=qt --with-qt-dir=$QTDIR \
--with-qt-includes=$QTDIR/include --with-qt-libraries=$QTDIR/lib 
;;
  debug-qt)
cd $LYXSRC; ./autogen.sh; cd $LYXBUILD
LDFLAGS="-L/usr/local/lib  -Wl,--export-all-symbols \
 -Wl,--enable-runtime-pseudo-reloc \
 -Wl,--script,$QT_LD_SCRIPT" \
$LYXSRC/configure --enable-debug --enable-optimization=-g \
--with-extra-lib=/usr/local/lib \
--with-extra-inc=/usr/local/include \
--with-frontend=qt --with-qt-dir=$QTDIR \
--with-qt-includes=$QTDIR/include --with-qt-libraries=$QTDIR/lib 
;;
  build)
make
;;
  install) 
make install
;;
  *)
usage
;;
  esac
done

Re: 1.4.0pre2 on cygwin

[Luis Rivera]

Kayvan A. Sylvan <[EMAIL PROTECTED]> writes:

> 
> On Wed, Oct 12, 2005 at 07:31:22PM +, Luis Rivera wrote:
> > Jean-Marc Lasgouttes  ...> writes:
> > 
> > > 
> > > > "Luis" == Luis Rivera  ...> writes:
> > > 
> > > Luis> And then it failed almost at the end of compilation. I attach
> > > Luis> config.log and error.log reporting the errors...
> > > 
> > > Didn't you forget them?
> > > 
> > 
> > Yep.  I realized as soon as I clicked on `send'... 
> 
> I don't have the logs in front of me.
> 

They're on the mailing archives...

> Are you trying to compile for xforms or for qt?
> 

I did it with xforms.  

I'm trying with qt...

Now I'll try --with-included-gettext (thanks to Lars!)

> I do both all the time with LyX CVS. Maybe I can help you.

Do you? On cygwin?!

Any help is welcome!

Luis.

Re: [patch] fix bug 2088

[Jose' Matos]

Lars Gullik Bjønnes wrote:

> When did we have 220?

  1.2 as the header says. :-)

  To know those relations either visit the archaeology page or read
LyX.py in lyx2lyx

> Lgb

-- 
José Matos

Re: Python script conversion

[Bo Peng]

> > I found that I could convert all the python script to .exe.
> > Thus I can remove Pyhthon dependecy.
>
> Are you sure?

Also, there are some dynamically generated sh scripts that might be
translated to dynamically generated python script later so an
interpreter is needed.

Bo

Your Account with PayPal

[Paypal]

Re: New bug 2094

[Juergen Spitzmueller]

Angus Leeming wrote:
> > Juergen had a fix there.
>
> Great!

But it needs very good testing. I just have enabled a function I found. I 
don't know if it has any side effects (and why André left it commented out).

Jürgen

Re: LyX 1.4 on Windows (MinGW)

[Michael Gerz]

Martin Vermeer wrote:


Neither am I... but I did it anyway, just to get the stuff working. But
your idea is better.

All you have to do, at start-up, is

1) Check that a locale is defined, and if not, define en_US;
2) remove any encoding suffixes like fi_FI.utf-8 --> fi_FI.

Doing it in one place at start-up (and only using, say, LC_ALL or LANG,
and making the others conform) would allow a lot of special code to be
taken out. In theory.

I wonder if this is 1.4 stuff. Perhaps not. So I say, do the minimal
patch based on Michael's idea.
 



I had a very short look at the code this evening. AFAICS we need all the 
stuff because we want to control the translation based on the language 
of the current document. In other words, you can have a Finnish UI with 
German layout styles (labels). The code in messages.C seems to do the 
right thing.


Nevertheless, my patch changes it in two ways:

- It checks a couple of environment variables to determine lang_. This 
is needed for MinGW on the Windows platform.
- It moves a couple of checks out of the getter method and into the 
constructor. This is more efficient but not necessary in principle.


But, of course, it would be nice if someone with more gettext experience 
could have a look at the code as well.


Regards, Michael

Re: 1.4.0pre2 on cygwin

[Jean-Marc Lasgouttes]

> "Luis" == Luis Rivera <[EMAIL PROTECTED]> writes:

Luis> Jean-Marc Lasgouttes <[EMAIL PROTECTED]> writes:
>>  > "Luis" == Luis Rivera <[EMAIL PROTECTED]> writes:
>> 
Luis> And then it failed almost at the end of compilation. I attach
Luis> config.log and error.log reporting the errors...
>>  Didn't you forget them?
>> 

Luis> Yep. I realized as soon as I clicked on `send'...

Luis> They're on the lists now, but as binaries weird...

Luis> Did you get them? Otherwise, I'll try again using .txt

I got them, but then had to leave. From the brief look I had, it seems
you should add
#include 
at the beginning of src/message.C

JMarc

Re: 1.4.0pre2 on cygwin

[Lars Gullik Bjønnes]

Luis Rivera <[EMAIL PROTECTED]> writes:

| I assume it is my fault, since I don't know exactly how to install nls
| support.  However, I think ./configure should report it, to keep people
| away from assuming nls will be working.
| 
| Anyway, I went ahead and tried 
| 
| ./configure --with-frontend=xforms --with-extra-prefix=/usr/local
| --without-latex-config --disable-pch --disable-nls
| 
| ...and bingo!  It compiled, installed, ran, imported/exported, and even
| bindist!  The problem: internationalization was removed.  I don't mind
| using the English GUI, but somebody would...

You can try with --with-included-gettext as well (instead of
--disable-nls)

-- 
Lgb

Re: 1.4.0pre2 on cygwin

[Kayvan A. Sylvan]

On Wed, Oct 12, 2005 at 07:31:22PM +, Luis Rivera wrote:
> Jean-Marc Lasgouttes <[EMAIL PROTECTED]> writes:
> 
> > 
> > > "Luis" == Luis Rivera <[EMAIL PROTECTED]> writes:
> > 
> > Luis> And then it failed almost at the end of compilation. I attach
> > Luis> config.log and error.log reporting the errors...
> > 
> > Didn't you forget them?
> > 
> 
> Yep.  I realized as soon as I clicked on `send'... 

I don't have the logs in front of me.

Are you trying to compile for xforms or for qt?

I do both all the time with LyX CVS. Maybe I can help you.

---Kayvan

Re: [patch] fix bug 2088

[Lars Gullik Bjønnes]

Georg Baum <[EMAIL PROTECTED]> writes:

| Lars Gullik Bjønnes wrote:
| 
| > So where did that file come from?
| 
| See http://bugzilla.lyx.org/show_bug.cgi?id=2088. It was appereantly
| generated by an old LyX version. It is in format 220.

When did we have 220?

-- 
Lgb

Re: 1.4.0pre2 on cygwin

[Luis Rivera]

Jean-Marc Lasgouttes <[EMAIL PROTECTED]> writes:

> 
> > "Luis" == Luis Rivera <[EMAIL PROTECTED]> writes:
> 
> Luis> And then it failed almost at the end of compilation. I attach
> Luis> config.log and error.log reporting the errors...
> 
> Didn't you forget them?
> 

Yep.  I realized as soon as I clicked on `send'... 

They're on the lists now, but as binaries weird...

Did you get them?  Otherwise, I'll try again using .txt

Luis.

Re: [patch] fix bug 2059 (cut and paste in mathed)

[Georg Baum]

Am Mittwoch, 12. Oktober 2005 18:22 schrieb Jean-Marc Lasgouttes:
> > "Georg" == Georg Baum <[EMAIL PROTECTED]> writes:
> 
> Georg> Daniel Watkins wrote:
> >> This one works for me, in conjunction with the undo patch.
> 
> Georg> Can the two patches go in?
> 
> Yes. Just remove the "MathGridInset: PASTE:" lyxerr message, while you
> are at it.

Both done. I initially wanted to remove the error message, but then forgot 
it.


Georg

Re: Python script conversion

[Mike Meyer]

In <[EMAIL PROTECTED]>, [EMAIL PROTECTED] typed:
> > The downside is that you wind up with one copy of Python installed for
> > every script. If you've only got one script, this is ok. If you've got
> > more than one, you probably want to consider another approach.
> It will be the same, using several scripts, too.
> 
> I have tried, I can do the following variant:
> - place all the python scripts into the script dir
> - create .exe files for the scripts
> - create an archive file containing python
> 
> In this case, you will have only one python packaged , but several small
> scrits converted to .exe.

Unless you're doing something strange, the process of creating the
.exe files will put the python dll and copies of all the library files
into the .exe. Are you doing something to prevent that? And if so -
I'd be interested in knowing what.

> A few converted scritps look like this:
> PYTHON23 DLL   979 005  05.02.08  16.23 python23.dll
> W9XPOPEN EXE16 384  05.02.08  16.24 w9xpopen.exe
> LIBRARY  ZIP   311 060  05.10.12  19.30 library.zip
> LYX2LYX  EXE40 960  05.10.12  19.30 lyx2lyx.exe
> GENERA~1 EXE28 672  05.10.12  19.30 general_command_wrapper.exe
> PIC2AS~1 EXE28 672  05.10.12  19.30 pic2ascii.exe
> 
> So, they can use the same interpreter.

This amounts to simply bundling the Python interpreter with LyX. I
think that's a great option - in fact, I think someone should bundle
up a complete system, with all the dependencies included - but I'm
pretty sure that shouldn't be the only binary offering available.

  http://www.mired.org/consulting.html
Independent Network/Unix/Perforce consultant, email for more information.

Re[2]: Python script conversion

[Alex]

Dear Mike,

> Are you sure? The only way I know of to convert a python script to
> .exe is with py2exe, which does it by creating an executable archive,
> and putting all the scripts that make up the program and the python
> interpreter in the archive. It doesn't remove the Python dependency so
> much as hide it.
Sure, you are right. If we convert the scripts to .exe, the user will not
need to install a newer (and possible incopatible) python, but a good tested
version. Therfore, we can say that the depency is removed.

> The downside is that you wind up with one copy of Python installed for
> every script. If you've only got one script, this is ok. If you've got
> more than one, you probably want to consider another approach.
It will be the same, using several scripts, too.

I have tried, I can do the following variant:
- place all the python scripts into the script dir
- create .exe files for the scripts
- create an archive file containing python

In this case, you will have only one python packaged , but several small
scrits converted to .exe.

A few converted scritps look like this:
PYTHON23 DLL   979 005  05.02.08  16.23 python23.dll
W9XPOPEN EXE16 384  05.02.08  16.24 w9xpopen.exe
LIBRARY  ZIP   311 060  05.10.12  19.30 library.zip
LYX2LYX  EXE40 960  05.10.12  19.30 lyx2lyx.exe
GENERA~1 EXE28 672  05.10.12  19.30 general_command_wrapper.exe
PIC2AS~1 EXE28 672  05.10.12  19.30 pic2ascii.exe

So, they can use the same interpreter.

This is my offer.

-- 
Best regards,
 Alex  mailto:[EMAIL PROTECTED]

Re: Python script conversion

[Alex]

Dear Mike,

> Are you sure? The only way I know of to convert a python script to
> .exe is with py2exe, which does it by creating an executable archive,
> and putting all the scripts that make up the program and the python
> interpreter in the archive. It doesn't remove the Python dependency so
> much as hide it.
Sure, you are right. If we convert the scripts to .exe, the user will not
need to install a newer (and possible incopatible) python, but a good tested
version. Therfore, we can say that the depency is removed.

> The downside is that you wind up with one copy of Python installed for
> every script. If you've only got one script, this is ok. If you've got
> more than one, you probably want to consider another approach.
It will be the same, using several scripts, too.

I have tried, I can do the following variant:
- place all the python scripts into the script dir
- create .exe files for the scripts
- create an archive file containing python

In this case, you will have only one python packaged , but several small
scrits converted to .exe.

A few converted scritps look like this:
PYTHON23 DLL   979 005  05.02.08  16.23 python23.dll
W9XPOPEN EXE16 384  05.02.08  16.24 w9xpopen.exe
LIBRARY  ZIP   311 060  05.10.12  19.30 library.zip
LYX2LYX  EXE40 960  05.10.12  19.30 lyx2lyx.exe
GENERA~1 EXE28 672  05.10.12  19.30 general_command_wrapper.exe
PIC2AS~1 EXE28 672  05.10.12  19.30 pic2ascii.exe

So, they can use the same interpreter.

This is my offer.

-- 
Best regards,
 Alex  mailto:[EMAIL PROTECTED]

Re: LyX-140pre2: Minor Mac issues

[Bennett Helm]

On Oct 12, 2005, at 6:05 AM, Jean-Marc Lasgouttes wrote:


"Angus" == Angus Leeming <[EMAIL PROTECTED]> writes:



Angus> Lars Gullik Bjønnes wrote:


| Second (and very minor), I noticed that on a clean installation,
the | LyX user's directory is ~/Library/Application\ Support/lyx;
it should | be ~/Library/Application\ Support/LyX -- both for
consistency with | 1.3.6 and to conform to Mac conventions for
naming application | support folders.

This I wont touch. (i.e. I leave it for somebody else)



Angus> Ok, looks like some things fell between the cracks. The
Angus> attached pair of patches sync the two LyX versions. I'll commit
Angus> both as they're obviously correct.

Well, the 1.4.x version is not obviously correct. I have decided in
head to set explicitly PACKAGE to LyX when needed. Unfortunately, my
code was untested and wrong. The patch below (which reverts yours)
should fix all the problems.

Bennett, Angus, I would appreciate if you could test it. I think the
code to handle package in configure is much cleaner now.


Works for me.

Bennett

LyX/Mac 1.4.0pre2 crash on open new file

[Bennett Helm]

I reported this bug before, and Andreas developed a patch to fix it.  
However, the patch has apparently fallen through the cracks, and the  
bug is still present in 1.4.0pre2.


The problem is this: LyX-140 on Mac will crash *when using key  
bindings* for File > Open, File > New, and File > New from Template.  
It will not crash when using the mouse to select these from the  
menus, nor will it crash dragging and dropping files into the LyX  
window.


The successful patch can be found here: .


Bennett

Re: lyx: Error detected by libpthread: Destroying locked mutex.

[Jeremy C. Reed]

On Wed, 21 Sep 2005, Angus Leeming wrote:


Anyway, trying to unlock a mutex that is already unlocked will result in
undefined behaviour when using POSIX threads. See, for example Section 3.3.2
"Locking and unlocking a mutex" of "Programming with POSIX threads" by David
R Butenhof.

So, I'm not going to apply your fix; sorry.


I was also told in http://thread.gmane.org/gmane.editors.lyx.devel/46629 
that "The problem is that QApplication::locked() is lying to you. That's a 
Qt problem, so the solution must be: fix Qt on NetBSD."


So I filed a bug report with trolltech. They gave me some sample code to 
run and I did several tests. This was their reply:


 Right. You are sure that it's the destruction of the QApplication mutex
 that's causing this? Anyway, I'm not sure I see the reason why they're
 trying to unlock the QApplication mutex, you're supposed to lock it
 from other threads than the thread where QApplication was created, and
 you're not allowed to unlock a mutex from a thread that did not
 previously lock it. I also noticed the following paragraph from the
 QMutex::locked() documentation that could be related:

Warning: Due to different implementations of recursive mutexes on the
supported platforms, calling this function from the same thread that
previously locked the mutex will give undefined results.

Does this help at all?

According to the test I was given by trolltech to run on NetBSD, it 
behaved correctly.


I will be glad to test other ideas.

Thanks,

 Jeremy C. Reed

Re: Back to basics.

[Angus Leeming]

Jean-Marc Lasgouttes wrote:
> Jose'>   It is a shame that Angus work has been lost. :-(
> Yes...

That'll learn me not to commit to a repository :(

The work got lost when Lars kicked me off baywatch and I didn't think
before `rm -r ~/public_html/www-user` ...

-- 
Angus

Re: 1.4.0pre2 on cygwin

[Jean-Marc Lasgouttes]

> "Luis" == Luis Rivera <[EMAIL PROTECTED]> writes:

Luis> And then it failed almost at the end of compilation. I attach
Luis> config.log and error.log reporting the errors...

Didn't you forget them?

JMarc

Re: [patch] fix bug 2059 (cut and paste in mathed)

[Jean-Marc Lasgouttes]

> "Georg" == Georg Baum <[EMAIL PROTECTED]> writes:

Georg> Daniel Watkins wrote:
>> This one works for me, in conjunction with the undo patch.

Georg> Can the two patches go in?

Yes. Just remove the "MathGridInset: PASTE:" lyxerr message, while you
are at it.

JMarc

Re: [patch] math undo fixes in 1.4

[Jean-Marc Lasgouttes]

> "Georg" == Georg Baum <[EMAIL PROTECTED]> writes:

Georg> Open a new document. Open a new math formula with C-M, and type
Georg> \sqrt. The result is of course an empty sqrt inset. Now
Georg> undo and get a nice crash. If you insert the sqrt via the math
Georg> panel you get no crash, but have to press undo 4 times (instead
Georg> of 2 times) in order to get an empty document again.

Georg> The reason for the crash is an invalid cursor, caused by too
Georg> many recorded undo steps. The attached patch fixes both
Georg> problems. OK to apply?

Yes.

JMarc

Re: Back to basics.

[Jean-Marc Lasgouttes]

> "Jose'" == Jose' Matos <[EMAIL PROTECTED]> writes:

Jose'> On Wednesday 12 October 2005 11:25, Jean-Marc Lasgouttes wrote:
>>  What about posting this on the news section of www.lyx.org?

Jose'>   I don't have write cvs access to the www-users module. :-)

Lars?

Jose'>   It is quite interesting to add, as well as converting the
Jose'> archaeology to the site. :-)

I guess it would not be too difficult to adapt, if the thing is put in
its own archaeology/ directory.

Jose'>   It is a shame that Angus work has been lost. :-(

Yes...

JMarc

Re: Bug 2090: Failure to handle old 1.3 documents with single-quoting style

[Juergen Spitzmueller]

Jose' Matos wrote:
> > Additionally, we could force \quotes_times to 2 between 229 and 230.
>
>   What is the difference if you don't do it?

Files up to the forthcoming version might have the described bug (single 
quotes enabled, which does not make sense and basically breaks quotation mark 
handling).

Jürgen

1.4.0pre2 on cygwin

[Luis Rivera]

Hello,

I want to report that LyX 1.4.0pre2 compiles and runs succesfully on
cygwin+xwin+MiKTeX...

However, there seems to be a problem with pch and nls.

A first shot at it, running 

./configure --with-frontend=xforms --with-extra-prefix=/usr/local
--without-latex-config

reported something missing with precompiled headers.  So I removed it
and tried 

./configure --with-frontend=xforms --with-extra-prefix=/usr/local
--without-latex-config --disable-pch

And then it failed almost at the end of compilation. I attach
config.log and error.log reporting the errors...

I assume it is my fault, since I don't know exactly how to install nls
support.  However, I think ./configure should report it, to keep people
away from assuming nls will be working.

Anyway, I went ahead and tried 

./configure --with-frontend=xforms --with-extra-prefix=/usr/local
--without-latex-config --disable-pch --disable-nls

...and bingo!  It compiled, installed, ran, imported/exported, and even
bindist!  The problem: internationalization was removed.  I don't mind
using the English GUI, but somebody would...

End of report.

Cheers,

Luis.


===

Please avoid sending me Word, Excel, or PowerPoint attachments.
Word, Excel, and PowerPoint files are unreliable, unmaintainable, and unsafe.
Try some open format instead.
See http://en.wikipedia.org/wiki/open_format

J.L.Rivera



__ 
Yahoo! Music Unlimited 
Access over 1 million songs. Try it free.
http://music.yahoo.com/unlimited/

Re: Back to basics.

[Jose' Matos]

On Wednesday 12 October 2005 11:25, Jean-Marc Lasgouttes wrote:
>
> What about posting this on the news section of www.lyx.org?

  I don't have write cvs access to the www-users module. :-)

  It is quite interesting to add, as well as converting the archaeology to the 
site. :-)

  It is a shame that Angus work has been lost. :-(

> JMarc

-- 
José Abílio

[PATCHes] adapt i18n stats to new devel site.

[Jean-Marc Lasgouttes]

The following pair of patches adapts postats.sh to the new site.

Committing now, along with some small changes to the site itself.

JMarc

Index: po/ChangeLog
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/po/ChangeLog,v
retrieving revision 1.156.2.109
diff -u -p -r1.156.2.109 ChangeLog
--- po/ChangeLog	3 Oct 2005 12:50:58 -	1.156.2.109
+++ po/ChangeLog	12 Oct 2005 15:43:55 -
@@ -1,3 +1,8 @@
+2005-10-12  Jean-Marc Lasgouttes  <[EMAIL PROTECTED]>
+
+	* Makefile.in.in:
+	* postats.sh: adapt to new developers site.
+
 2005-10-03  Michael Gerz  <[EMAIL PROTECTED]>
 
 	* de.po: tiny improvement by Hartmut Haase
Index: po/Makefile.in.in
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/po/Makefile.in.in,v
retrieving revision 1.28.2.9
diff -u -p -r1.28.2.9 Makefile.in.in
--- po/Makefile.in.in	27 Apr 2005 15:23:56 -	1.28.2.9
+++ po/Makefile.in.in	12 Oct 2005 15:43:55 -
@@ -308,7 +308,7 @@ $(srcdir)/default_ui_l10n.pot: $(top_src
 } \
 ' ../lib/ui/default.ui > default_ui_l10n.pot
 
-i18n.php3: $(POFILES)
+i18n.php: $(POFILES)
 	(cd $(srcdir) ; ./postats.sh $(POFILES)) >$@
 
 # Tell versions [3.59,3.63) of GNU make not to export all variables.
Index: po/postats.sh
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/po/postats.sh,v
retrieving revision 1.6.2.14
diff -u -p -r1.6.2.14 postats.sh
--- po/postats.sh	3 Oct 2005 12:51:41 -	1.6.2.14
+++ po/postats.sh	12 Oct 2005 15:43:55 -
@@ -13,12 +13,12 @@
 # and generates a PHP web page.
 #
 # Invocation:
-#postats.sh po_files > "pathToWebPages"/i18n.php3
+#postats.sh po_files > "pathToWebPages"/i18n.php
 
 # modifiy this when you change version
 # Note that an empty lyx_branch variable (ie cvs HEAD)
 # will "do the right thing".
-lyx_version=1.3.6
+lyx_version=1.3.7cvs
 lyx_branch=BRANCH_1_3_X
 
 
@@ -163,10 +163,10 @@ cat <
@@ -307,7 +307,7 @@ while (list(\$foo,\$info) = each(\$podat
 
 
 
 EOF
 }
Index: po/ChangeLog
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/po/ChangeLog,v
retrieving revision 1.233
diff -u -p -r1.233 ChangeLog
--- po/ChangeLog	8 Oct 2005 23:59:17 -	1.233
+++ po/ChangeLog	12 Oct 2005 15:43:44 -
@@ -1,3 +1,8 @@
+2005-10-12  Jean-Marc Lasgouttes  <[EMAIL PROTECTED]>
+
+	* Makefile.in.in:
+	* postats.sh: adapt to new developers site.
+
 2005-10-09  Lars Gullik Bjonnes  <[EMAIL PROTECTED]>
 
 	* rename no.po to nb.po, remerge
Index: po/Makefile.in.in
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/po/Makefile.in.in,v
retrieving revision 1.53
diff -u -p -r1.53 Makefile.in.in
--- po/Makefile.in.in	15 Sep 2005 15:15:52 -	1.53
+++ po/Makefile.in.in	12 Oct 2005 15:43:44 -
@@ -516,7 +516,7 @@ ui_l10n.pot: $(top_srcdir)/lib/ui/*.ui
 		}' \
 	${top_srcdir}/lib/ui/*.ui > $@
 
-i18n.php3: $(POFILES)
+i18n.php: $(POFILES)
 	(cd $(srcdir) ; ./postats.sh $(POFILES)) >$@
 
 force:
Index: po/postats.sh
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/po/postats.sh,v
retrieving revision 1.10
diff -u -p -r1.10 postats.sh
--- po/postats.sh	6 Sep 2005 17:20:44 -	1.10
+++ po/postats.sh	12 Oct 2005 15:43:44 -
@@ -13,7 +13,7 @@
 # and generates a PHP web page.
 #
 # Invocation:
-#postats.sh po_files > "pathToWebPages"/i18n.php3
+#postats.sh po_files > "pathToWebPages"/i18n.php
 
 # modifiy this when you change version
 # Note that an empty lyx_branch variable (ie cvs HEAD)
@@ -163,10 +163,10 @@ cat <
@@ -307,7 +307,7 @@ while (list(\$foo,\$info) = each(\$podat
 
 
 
 EOF
 }

Re: New bug 2094

[Angus Leeming]

Martin Vermeer wrote:
>> Detailed description and recipe to recreate here:
>> http://bugzilla.lyx.org/show_bug.cgi?id=2094

> Isn't 2083 the same bug?

Looks like. I've marked 2094 as a duplicate.

> Juergen had a fix there.

Great!

-- 
Angus

Re: New bug 2094

[Martin Vermeer]

On Wed, 2005-10-12 at 16:12 +0100, Angus Leeming wrote:
> Navigation with the arrow keys through a math inset when a float is open on
> the same screen leads to a crash.
> 
> Assertion triggered in virtual void InsetCollapsable::cursorPos(const
> CursorSlice&, bool, int&, int&) const by failing check "status() !=
> Collapsed"
> in file ../../../src/insets/insetcollapsable.C:207
> 
> Detailed description and recipe to recreate here:
> http://bugzilla.lyx.org/show_bug.cgi?id=2094

Isn't 2083 the same bug?

Juergen had a fix there.

- Martin



signature.asc
Description: This is a digitally signed message part

New bug 2094

[Angus Leeming]

Navigation with the arrow keys through a math inset when a float is open on
the same screen leads to a crash.

Assertion triggered in virtual void InsetCollapsable::cursorPos(const
CursorSlice&, bool, int&, int&) const by failing check "status() !=
Collapsed"
in file ../../../src/insets/insetcollapsable.C:207

Detailed description and recipe to recreate here:
http://bugzilla.lyx.org/show_bug.cgi?id=2094

-- 
Angus

Re: Math fonts look fuzzy in lyx_1.4.0cvs

[Angus Leeming]

Anand Rangarajan wrote:
> For a while now (about a year), I've noticed that math fonts look
> fuzzy in lyx_1.4.0cvs. In contrast, they look fine for lyx_1.3.x on the
> same machine. I've attached a screenshot.

Yes, I get this too on both a FC3 i686 machine and on an Alpha running
Debian unstable.

Bad old bitmap fonts.

-- 
Angus

Re: Python script conversion

[Mike Meyer]

In <[EMAIL PROTECTED]>, [EMAIL PROTECTED] typed:
> Dears,
> 
> I found that I could convert all the python script to .exe.
> Thus I can remove Pyhthon dependecy.

Are you sure? The only way I know of to convert a python script to
.exe is with py2exe, which does it by creating an executable archive,
and putting all the scripts that make up the program and the python
interpreter in the archive. It doesn't remove the Python dependency so
much as hide it.

The downside is that you wind up with one copy of Python installed for
every script. If you've only got one script, this is ok. If you've got
more than one, you probably want to consider another approach.

  http://www.mired.org/consulting.html
Independent Network/Unix/Perforce consultant, email for more information.

Re: RFC: GThesaurus.C et al.

[Angus Leeming]

Bernhard Reiter wrote:

> first things first...
> 
>> One thing before I move on to the code. Can I have your formal agreement
>> that you agree to licence your contributions under the terms of the GNU
>> General Public Licence (GPL) version 2 or later? It's enough to reply to
>> this mail. I'll add a reference to your "yes" to
>> http://www.lyx.org/about/blanket-permission.php
>> 
> 
> Sure:
> I grant permission to license any and all contributions I've made to
> LyX under the Gnu GPL version 2 or later.

Thank you.

-- 
Angus

Re: Bug 2090: Failure to handle old 1.3 documents with single-quoting style

[Juergen Spitzmueller]

Jose' Matos wrote:
> > Should I prepare a patch with another file format change?
>
>   Yes.

Patch attached. Please review.
Am I right that norhing has to be done on lyx2lyx reversion?

Jürgen
Index: development/FORMAT
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/development/FORMAT,v
retrieving revision 1.42
diff -p -u -r1.42 FORMAT
--- development/FORMAT	28 Sep 2005 09:40:46 -	1.42
+++ development/FORMAT	12 Oct 2005 12:33:44 -
@@ -1,6 +1,11 @@
 LyX file-format changes
 ---
 
+2005-10-12  Jürgen Spitzmüller  <[EMAIL PROTECTED]>
+
+	* format incremented to 245. The \quotes_times parameter
+	has been removed.
+
 2005-09-24  Jürgen Spitzmüller  <[EMAIL PROTECTED]>
 
 	* format incremented to 244. Rename '\InsetSpace \,'
Index: lib/lyx2lyx/LyX.py
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/lib/lyx2lyx/LyX.py,v
retrieving revision 1.26
diff -p -u -r1.26 LyX.py
--- lib/lyx2lyx/LyX.py	28 Sep 2005 14:29:23 -	1.26
+++ lib/lyx2lyx/LyX.py	12 Oct 2005 12:33:49 -
@@ -46,7 +46,7 @@ format_relation = [("0_10",  [210], ["0.
("1_1_6fix3", [218], ["1.1.6fix3","1.1.6fix4","1.1"]),
("1_2", [220], ["1.2.0","1.2.1","1.2.3","1.2.4","1.2"]),
("1_3", [221], ["1.3.0","1.3.1","1.3.2","1.3.3","1.3.4","1.3.5","1.3.6","1.3"]),
-   ("1_4", range(222,245), ["1.4.0cvs","1.4"])]
+   ("1_4", range(222,246), ["1.4.0cvs","1.4"])]
 
 
 def formats_list():
@@ -481,7 +481,6 @@ class NewFile(LyX_Base):
 "\\paragraph_separation indent",
 "\\defskip medskip",
 "\\quotes_language english",
-"\\quotes_times 2",
 "\\papercolumns 1",
 "\\papersides 1",
 "\\paperpagestyle default",
Index: lib/lyx2lyx/lyx_1_4.py
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/lib/lyx2lyx/lyx_1_4.py,v
retrieving revision 1.53
diff -p -u -r1.53 lyx_1_4.py
--- lib/lyx2lyx/lyx_1_4.py	28 Sep 2005 09:40:48 -	1.53
+++ lib/lyx2lyx/lyx_1_4.py	12 Oct 2005 12:33:54 -
@@ -2258,6 +2258,13 @@ def remove_paperpackage(file):
 file.header[i] = "\\papersize default"
 
 
+def remove_quotestimes(file):
+i = find_token(file.header, '\\quotes_times', 0)
+if i == -1:
+return
+del file.header[i]
+
+
 ##
 # Convertion hub
 #
@@ -2286,9 +2293,11 @@ convert = [[222, [insert_tracking_change
[241, [convert_ert_paragraphs]],
[242, [convert_french]],
[243, [remove_paperpackage]],
-	   [244, [rename_spaces]]]
+	   [244, [rename_spaces]],
+	   [245, [remove_quotestimes]]]
 
-revert =  [[243, [revert_space_names]],
+revert =  [[244, []],
+	   [243, [revert_space_names]],
 	   [242, []],
[241, []],
[240, [revert_ert_paragraphs]],
Index: src/buffer.C
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/buffer.C,v
retrieving revision 1.628
diff -p -u -r1.628 buffer.C
--- src/buffer.C	28 Sep 2005 09:40:49 -	1.628
+++ src/buffer.C	12 Oct 2005 12:43:34 -
@@ -144,7 +144,7 @@ extern BufferList bufferlist;
 
 namespace {
 
-int const LYX_FORMAT = 244;
+int const LYX_FORMAT = 245;
 
 } // namespace anon
 
Index: src/bufferparams.C
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/bufferparams.C,v
retrieving revision 1.100
diff -p -u -r1.100 bufferparams.C
--- src/bufferparams.C	17 Jul 2005 11:17:12 -	1.100
+++ src/bufferparams.C	12 Oct 2005 12:43:36 -
@@ -105,25 +105,6 @@ QuotesLangTranslator const & quoteslangt
 }
 
 
-// Quote times
-typedef Translator QuotesTimesTranslator;
-
-
-QuotesTimesTranslator const init_quotestimestranslator()
-{
-	QuotesTimesTranslator translator(1, InsetQuotes::SingleQ);
-	translator.addPair(2, InsetQuotes::DoubleQ);
-	return translator;
-}
-
-
-QuotesTimesTranslator const & quotestimestranslator()
-{
-	static QuotesTimesTranslator translator = init_quotestimestranslator();
-	return translator;
-}
-
-
 // Paper size
 typedef Translator PaperSizeTranslator;
 
@@ -307,7 +288,6 @@ BufferParams::BufferParams()
 {
 	paragraph_separation = PARSEP_INDENT;
 	quotes_language = InsetQuotes::EnglishQ;
-	quotes_times = InsetQuotes::DoubleQ;
 	fontsize = "default";
 
 	/*  PaperLayout */
@@ -460,10 +440,6 @@ string const BufferParams::readToken(LyX
 		string quotes_lang;
 		lex >> quotes_lang;
 		quotes_language = quoteslangtranslator().find(quotes_lang);
-	} else if (token == "\\quotes_times") {
-		int qtimes;
-		lex >> qtimes;
-		quotes_times = quotestimestranslator().find(qtimes);
 	} else if (token == "\\papersize") {
 		string ppsize;
 		lex >> ppsize;
@@ -663,9 +639,7 @@ void BufferParams::writeFile(ostream & o
 	   << string_paragraph_separation[paragrap

Re: RFC: GThesaurus.C et al.

[Bernhard Reiter]

first things first... 

> One thing before I move on to the code. Can I have your formal agreement
> that you agree to licence your contributions under the terms of the GNU
> General Public Licence (GPL) version 2 or later? It's enough to reply to
> this mail. I'll add a reference to your "yes" to
> http://www.lyx.org/about/blanket-permission.php 
> 

Sure:
I grant permission to license any and all contributions I've made to 
LyX under the Gnu GPL version 2 or later.

> > working on this, i'm afraid i stumbled over a bug in the thesaurus
> > controller (but didn't do enough research to be able to cure it):
> > 1. highlight some word in the main window
> > 2. open the thesaurus dialog, choose a synonym and click replace
> > 3. choose another synonym, click replace again - it won't work
> > 
> > i checked this with the qt frontend. same thing, so it's apparently in
> > the controller's replace function... (didn't look any further into that,
> > admittedly...)
> 
> If you do look in the controller (I did for the first time just now!)
> you'll see that oldstr_, the string you're replacing, is set when the
> dialog is opened, in initialiseParams.
> 
> Call replace(newstr). oldstr_ is replaced throughout your document.
> Call replace(newstr2). There's no oldstr_ left to replace.
> 
> I'd suppose that oldstr_ should be replaced by newstr at the end of the
> first call to replace(newstr), but I'll leave it to you to think about
> this further.
> 

as far as i can see, only the current occurence of oldstr_ is
replaced... 
assigning oldstr_ = newstr afterwards solves the problem if there is
only one occurence of oldstr_, but is not sufficient if there are more
than one: in (anon)::replace (lyxfind.C) -- which is indirectly called
by LFUN_WORD_REPLACE -- ::find is called, so the next occurence of
oldstr_ is selected automatically. there seems to have been a parameter
'bool once' in 1.3.6's LyxFind, but it looks like it has vanished
since... 
as far as i can see, changing back the thesaurus's behavior would
require such a parameter in the replace LFUN -- and i'm not sure if i'm
in the position to solely decide about interface changes of that kind...
other options i can think of are leaving the thesaurus's behavior this
spellchecker-like way (but i doubt this is desirable) or calling the
find LFUN with fw=false (which i'd consider a waste of time and
resources). i don't know if there are any ready-to-use LFUNs (instead of
LFUN_WORD_REPLACE, that is), though.

> > is gtk support planned for 1.4.0?
> 
> No. The frontend is clearly a work-in-progress.
> 
> > and, one more question: is there a way to enable compilation for all
> > three frontends? i used the qt ui/sources as a template, but seeing it
> > in live action would have been quite useful in parallel.
> 
> Here I use:
>configure --with-version-suffix \
>  --disable-stdlib-debug \
>  --with-frontend='qt xforms gtk' \
>  --with-qt-includes='/usr/include/qt3'
> 

thanks, that helped a lot. 

> On to the code...
> 
> I've never looked in GTK's Dialogs.C before. First thing I see when I do so
> is that the code initially sets the ButtonController to be the XForms one
> and then overwrites this in the creation of many of the individual
> dialogs. I think that I'd move
> 
> dialog->bc().view(new xformsBC(dialog->bc()));
> 
> to those dialogs that actually need it...
> 

done that for now. 

> By the way, your GThesaurus.patch has a CVS conflict in the ChangeLog.
> That's those <<<  == ...  bits. CVS is unsure
> what's going on here so adds the placeholders as pointers for you to
> examine.
> 

oops, seems i messed around with cvs update and diff a little too
much... i'll take better care in the future, i promise :)

> GThesaurus.h: 
> /** This class provides a GTK+ implementation of the FormSearch Dialog.
> I don't think it does...
> 

oops again. couldn't that damn clipboard or gedit be more intelligent
(than me)? 

> You might add some more doxygen comments (start /// or /**) to describe
> what individual functions do. Most are obvious I guess, but I'd need to
> look at the implementation to guess what the three non-virtual functions
> do.
> 

ok, i added some, hopefully sufficient.

> GThesaurus.C:
> no real comments.
> A stylistic one: use tab-indentation please.
> What happens if you remove the applylock_ and readonly lines entirely? (I'm
> unfamiliar with them although I do know the button controller code.)
> 
> I don't believe that you need 'bc().refreshReadOnly();'. In face, I'd
> remove applylock, readonly and bc().valid(false) entirely from doBuild.
> Their settings should be set, if at all, in update().
> 
> Have a look in controllers/Dialogs.C to get a handle on the overall logic
> behind the Controller-View split. Without this overview, you're fighting
> with one arm behind your back and one eye closed.
> 
> Well done!
> 

thanks for the warm welcome. i hope i'll gradually require less
atten

Re: LyX-140pre2: Minor Mac issues

[Jean-Marc Lasgouttes]

> "Angus" == Angus Leeming <[EMAIL PROTECTED]> writes:

Angus> Ahh, Jean-Marc, you sly old dog!

I am still not sure whether I should take offense at "sly". But at
least you got me to look it up in the dictionary...

JMarc

Re: LyX-140pre2: Minor Mac issues

[Angus Leeming]

Jean-Marc Lasgouttes wrote:
> Angus> Ok, looks like some things fell between the cracks. The
> Angus> attached pair of patches sync the two LyX versions. I'll commit
> Angus> both as they're obviously correct.
> 
> Well, the 1.4.x version is not obviously correct. I have decided in
> head to set explicitly PACKAGE to LyX when needed. Unfortunately, my
> code was untested and wrong. The patch below (which reverts yours)
> should fix all the problems.
> 
> Bennett, Angus, I would appreciate if you could test it. I think the
> code to handle package in configure is much cleaner now.

Ahh, Jean-Marc, you sly old dog!

I'll give it a go.

-- 
Angus

Re: [patch] bug 2066: C-i for opening/closing insets only works when cursor is at the beginning of the inset

[Juergen Spitzmueller]

Jean-Marc Lasgouttes wrote:
> Could you explain what makes tabular special?

If I only knew.

> If special code is needed, what about putting it in
> InsetTabular::dispatch, or in some other InsetTabular method?

That does not work, but don't ask me why.

Jürgen

Re: Back to basics.

[Jean-Marc Lasgouttes]

> "Jose'" == Jose' Matos <[EMAIL PROTECTED]> writes:

Jose'> Lets us say no more, let us abandon 1.4, no need to
Jose'> have such release. Soon celebrating the 10 years of the first
Jose'> still available source code in the lyx ftp site I propose a new
Jose'> path.

What about posting this on the news section of www.lyx.org?

JMarc

Re: [patch] bug 2066: C-i for opening/closing insets only works when cursor is at the beginning of the inset

[Jean-Marc Lasgouttes]

> "Juergen" == Juergen Spitzmueller <[EMAIL PROTECTED]> writes:

Juergen> http://bugzilla.lyx.org/show_bug.cgi?id=2066 The attached
Juergen> patch does not only restore the behaviour of C-i like it was
Juergen> in 1.3, but also fixes two crashes with C-i and
Juergen> all-insets-toggle when tabular insets were involved.

Could you explain what makes tabular special?

If special code is needed, what about putting it in
InsetTabular::dispatch, or in some other InsetTabular method?

JMarc

Re: Minor Mac patches for 1.4.0 and 1.3.x

[Jean-Marc Lasgouttes]

> "Bennett" == Bennett Helm <[EMAIL PROTECTED]> writes:

Bennett> 2. mac-lyxeditor.diff updates the shell script, which is used
Bennett> to enable LyX/Mac to interact with .lyxpipe, to use the new
Bennett> location of the LyX user's directory. This should apply to
Bennett> both 1.4.0 and 1.3.x (the patch works for both).

I applied it to 1.3.x.

JMarc

Re: LyX 1.4 on Windows (MinGW)

[Martin Vermeer]

On Wed, 2005-10-12 at 11:54 +0200, Georg Baum wrote:
> Martin Vermeer wrote:
> 
> > On Wed, 2005-10-12 at 09:07 +0200, Georg Baum wrote:
> >> Also true. But I have to admit that I don't understand the current
> >> machinery. For example, why not simply strip the suffix from the
> >> environment variables once on program start, set the new environment
> >> variables and leave the frontend code as it was before Paris?
> > 
> > Sounds OK. Could you do that?
> 
> I am not sure whether I understand this locale stuff well enough, so I think
> that I am not the right person for this task. I was simply thinking out
> load.

Neither am I... but I did it anyway, just to get the stuff working. But
your idea is better.

All you have to do, at start-up, is

1) Check that a locale is defined, and if not, define en_US;
2) remove any encoding suffixes like fi_FI.utf-8 --> fi_FI.

Doing it in one place at start-up (and only using, say, LC_ALL or LANG,
and making the others conform) would allow a lot of special code to be
taken out. In theory.

I wonder if this is 1.4 stuff. Perhaps not. So I say, do the minimal
patch based on Michael's idea.

- Martin



signature.asc
Description: This is a digitally signed message part

Re: Bug 2090: Failure to handle old 1.3 documents with single-quoting style

[Jose' Matos]

On Wednesday 12 October 2005 10:58, Juergen Spitzmueller wrote:
> Should I prepare a patch with another file format change?

  Yes.

> Additionally, we could force \quotes_times to 2 between 229 and 230.

  What is the difference if you don't do it?

> Jürgen

-- 
José Abílio

Re: LyX-140pre2: Minor Mac issues

[Jean-Marc Lasgouttes]

> "Angus" == Angus Leeming <[EMAIL PROTECTED]> writes:

Angus> Lars Gullik Bjønnes wrote:
>> | Second (and very minor), I noticed that on a clean installation,
>> the | LyX user's directory is ~/Library/Application\ Support/lyx;
>> it should | be ~/Library/Application\ Support/LyX -- both for
>> consistency with | 1.3.6 and to conform to Mac conventions for
>> naming application | support folders.
>> 
>> This I wont touch. (i.e. I leave it for somebody else)

Angus> Ok, looks like some things fell between the cracks. The
Angus> attached pair of patches sync the two LyX versions. I'll commit
Angus> both as they're obviously correct.

Well, the 1.4.x version is not obviously correct. I have decided in
head to set explicitly PACKAGE to LyX when needed. Unfortunately, my
code was untested and wrong. The patch below (which reverts yours)
should fix all the problems. 

Bennett, Angus, I would appreciate if you could test it. I think the
code to handle package in configure is much cleaner now.

JMarc

Index: ChangeLog
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/ChangeLog,v
retrieving revision 1.1033
diff -u -p -r1.1033 ChangeLog
--- ChangeLog	30 Sep 2005 09:44:36 -	1.1033
+++ ChangeLog	12 Oct 2005 10:02:13 -
@@ -1,3 +1,8 @@
+2005-10-12  Jean-Marc Lasgouttes  <[EMAIL PROTECTED]>
+
+	* configure.ac: call LYX_USE_PACKAGING early. Sanitize PACKAGE
+	setting. 
+
 2005-09-30  Angus Leeming  <[EMAIL PROTECTED]>
 
 	* configure.ac: check for NewAPIs.h. If found, define the
Index: configure.ac
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/configure.ac,v
retrieving revision 1.64
diff -u -p -r1.64 configure.ac
--- configure.ac	10 Oct 2005 13:33:03 -	1.64
+++ configure.ac	12 Oct 2005 10:02:13 -
@@ -7,9 +7,12 @@ AM_CONFIG_HEADER([src/config.h])
 
 AC_CONFIG_AUX_DIR(config)
 
-PACKAGE=lyx${program_suffix}
+# first the version
 VERSION="1.4.0cvs"
 LYX_CHECK_VERSION
+# Check how the files should be packaged
+LYX_USE_PACKAGING
+LYX_VERSION_SUFFIX
 
 dnl default maintainer mode to true for development versions
 if test "${enable_maintainer_mode+set}" != set; then
@@ -18,9 +21,7 @@ fi
 AM_MAINTAINER_MODE
 AC_CANONICAL_TARGET
 
-LYX_VERSION_SUFFIX
-
-AM_INIT_AUTOMAKE($lyxname, $VERSION)
+AM_INIT_AUTOMAKE($PACKAGE, $VERSION)
 
 ### Set the execute permissions of the various scripts correctly
 for file in config/install-sh config/mkinstalldirs lib/configure ; do
@@ -227,8 +228,6 @@ dnl qt build will fail without moc or ui
   esac
 done
 
-### Check how the files should be packaged
-LYX_USE_PACKAGING
 # fix the value of the prefixes.
 test "x$prefix" = xNONE && prefix=$default_prefix
 test "x$exec_prefix" = xNONE && exec_prefix='${prefix}'
Index: config/ChangeLog
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/config/ChangeLog,v
retrieving revision 1.175
diff -u -p -r1.175 ChangeLog
--- config/ChangeLog	29 Sep 2005 13:25:36 -	1.175
+++ config/ChangeLog	12 Oct 2005 10:02:13 -
@@ -1,3 +1,9 @@
+2005-10-12  Jean-Marc Lasgouttes  <[EMAIL PROTECTED]>
+
+	* lyxinclude.m4 (LYX_USE_PACKAGING): set PACKAGE depending on
+	packaging setting.
+	(LYX_VERSION_SUFFIX): honor PACKAGE value.
+
 2005-09-29  Angus Leeming  <[EMAIL PROTECTED]>
 
 	* lyxinclude.m4: add a new MSYS_AC_CANONICAL_PATH macro that
Index: config/lyxinclude.m4
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/config/lyxinclude.m4,v
retrieving revision 1.116
diff -u -p -r1.116 lyxinclude.m4
--- config/lyxinclude.m4	29 Sep 2005 13:25:36 -	1.116
+++ config/lyxinclude.m4	12 Oct 2005 10:02:13 -
@@ -28,7 +28,7 @@ dnl Define the option to set a LyX versi
 dnl
 dnl
 AC_DEFUN([LYX_VERSION_SUFFIX],[
-AC_MSG_CHECKING([for install target ... ])
+AC_MSG_CHECKING([for install target])
 RPM_VERSION_SUFFIX='""'
 AC_ARG_WITH(version-suffix,
   [  --with-version-suffix[=]  install lyx files as lyx],
@@ -37,12 +37,11 @@ AC_ARG_WITH(version-suffix,
  withval="-$VERSION"
  ac_configure_args=`echo $ac_configure_args | sed "s,--with-version-suffix,--with-version-suffix=$withval,"`
fi
-   lyxname="lyx$withval"
+   PACKAGE="$PACKAGE$withval"
program_suffix=$withval
-   RPM_VERSION_SUFFIX="--with-version-suffix=$withval"],
-  [lyxname=lyx])
+   RPM_VERSION_SUFFIX="--with-version-suffix=$withval"])
 AC_SUBST(RPM_VERSION_SUFFIX)
-AC_MSG_RESULT([$lyxname])
+AC_MSG_RESULT([$PACKAGE])
 ])
 
 
@@ -587,20 +586,21 @@ AC_ARG_WITH(packaging,
 AC_MSG_RESULT($lyx_use_packaging)
 case $lyx_use_packaging in
macosx) AC_DEFINE(USE_MACOSX_PACKAGING, 1, [Define to 1 if LyX should use a MacOS X application bundle file layout])
-	   test $PACKAGE = "lyx" && PACKAGE=LyX
+	   PACKAGE=LyX
 	   default_prefix="/Applications/LyX.app"
 	   bindir='${prefix}/Contents/MacOS'
 	   libdir='${prefix}/Contents/Resources'
 	   datadi

Re: LyX 1.4 on Windows (MinGW)

[Georg Baum]

Martin Vermeer wrote:

> On Wed, 2005-10-12 at 09:07 +0200, Georg Baum wrote:
>> Also true. But I have to admit that I don't understand the current
>> machinery. For example, why not simply strip the suffix from the
>> environment variables once on program start, set the new environment
>> variables and leave the frontend code as it was before Paris?
> 
> Sounds OK. Could you do that?

I am not sure whether I understand this locale stuff well enough, so I think
that I am not the right person for this task. I was simply thinking out
load.


Georg

Re: Bug 2090: Failure to handle old 1.3 documents with single-quoting style

[Juergen Spitzmueller]

Helge Hafting wrote:
> Create a document with lyx 1.3  Set the quoting style to single quotes
> instead of the usual doubles.
>
> Open this document in lyx-1.4  Notice how it is impossible to turn
> single-quoting off again, as lyx-1.4 don't have this in the GUI
> anymore.  The problem persists as the file is saved by lyx-1.4.

Apparently I forgot to change the file format when introducing the new quote 
insets :-(
File format was 229 back then and has been increased a month later. 
\quotes_times has to be removed anyway.

Should I prepare a patch with another file format change? Additionally, we 
could force \quotes_times to 2 between 229 and 230. 

Jürgen

Re: [patch] fix bug 2088

[Georg Baum]

Lars Gullik Bjønnes wrote:

> So where did that file come from?

See http://bugzilla.lyx.org/show_bug.cgi?id=2088. It was appereantly
generated by an old LyX version. It is in format 220.


Georg

Re: LyX 1.4 on Windows (MinGW)

[Martin Vermeer]

On Wed, 2005-10-12 at 09:07 +0200, Georg Baum wrote:
> Martin Vermeer wrote:
> 
> > On Wed, 2005-10-12 at 00:42 +0200, Lars Gullik Bjønnes wrote:
> >> [EMAIL PROTECTED] (Lars Gullik Bjønnes)
> >> writes:
> >> 
> >> | To be quite honest I belive we should revert messages.C to the simple
> >> | state it was in before this summers meeting.
> >> 
> >> But this is me being grumpy again. Just ignore it.
> > 
> > Actually I am a bit afraid that have a kernel of truth there.
> 
> This seems indeed to be the case.
> 
> > OTOH we cannot really go back there, as all newer Linux installations
> > have a utf-8 locale nowadays. The cutting off of the encoding suffix is
> > necessary as long as LyX is non-unicode.
> 
> Also true. But I have to admit that I don't understand the current
> machinery. For example, why not simply strip the suffix from the
> environment variables once on program start, set the new environment
> variables and leave the frontend code as it was before Paris?

Sounds OK. Could you do that?

- Martin



signature.asc
Description: This is a digitally signed message part

Re: [patch] fix bug 2088

[Lars Gullik Bjønnes]

Georg Baum <[EMAIL PROTECTED]> writes:

| Lars Gullik Bjønnes wrote:
| 
| > Georg Baum <[EMAIL PROTECTED]>
| > writes:
| > 
| > | I did not know that an empty \layout is valid, but both 1.3 and 1.4 can
| > | read that, so I made lyx2lyx understand it, too.
| > | 
| > | OK to commit?
| > 
| > Where did you find those empty \layout ?
| 
| In this case in the example file in the bug report. I never encountered it
| myself, but since LyX accepts it I thought it was valid.

So where did that file come from?

| 
| > It is a bug that they exist in the first place.
| 
| You mean we should do nothing and declare such .lyx files as broken?

Yes. Untill we know where this file actually comes from.

What generated it?
What converted it?

-- 
Lgb

Re: [patch] fix bug 2088

[Georg Baum]

Lars Gullik Bjønnes wrote:

> Georg Baum <[EMAIL PROTECTED]>
> writes:
> 
> | I did not know that an empty \layout is valid, but both 1.3 and 1.4 can
> | read that, so I made lyx2lyx understand it, too.
> | 
> | OK to commit?
> 
> Where did you find those empty \layout ?

In this case in the example file in the bug report. I never encountered it
myself, but since LyX accepts it I thought it was valid.

> It is a bug that they exist in the first place.

You mean we should do nothing and declare such .lyx files as broken?


Georg

Re: [patch] fix bug 2088

[Georg Baum]

Jose' Matos wrote:

> We do it in several places and this would fix all of them. This should be 
> done for 1.4.1 at least.
> 
>   We have also other parts of the code where this test is done.

Really? I glanced over lyx_1_4.py and could not find anything.

>   FWIW I consider this to be a bug from lyx. It a subject where two wrongs
> make one right. :-)
> 
>   The first bug is for lyx to output and empty layout, the second is to
>   read
> it.

True. We should probably add a new format where this is not allowed anymore,
and change LyX to not accept it anymore.


Georg

Re: [patch] fix bug 2088

[Jose' Matos]

On Wednesday 12 October 2005 09:37, Lars Gullik Bjønnes wrote:
>
> Where did you find those empty \layout ?

  I have found them in previous versions but I thought this was not possible 
anymore.

> It is a bug that they exist int he first place.

  Agreed.

-- 
José Abílio

Re: Bug 2091: Wrong typography when mixing single & double quotes (1.4cvs)

[Lars Gullik Bjønnes]

Helge Hafting <[EMAIL PROTECTED]> writes:

| Discovered by Jim Osborn
| <[EMAIL PROTECTED]> who seems to understand
| better than me what kind of latex ought to be generated.
| 
| Some quoting styles have problems with this sentence:
| "The word is 'foo'", he said.

Look at the "Ligature Break" in Insert->Special Formatting

Not the best of solutions perhaps, but we have a solution for the
problem

-- 
Lgb

Re: [patch] fix bug 2088

[Lars Gullik Bjønnes]

Georg Baum <[EMAIL PROTECTED]> writes:

| I did not know that an empty \layout is valid, but both 1.3 and 1.4 can read
| that, so I made lyx2lyx understand it, too.
| 
| OK to commit?

Where did you find those empty \layout ?

It is a bug that they exist int he first place.

-- 
Lgb

Re: [patch] fix bug 2088

[Jose' Matos]

On Wednesday 12 October 2005 09:01, Georg Baum wrote:
> I did not know that an empty \layout is valid, but both 1.3 and 1.4 can
> read that, so I made lyx2lyx understand it, too.

  It think that this function should be generalized and sent to parser tools. 
We do it in several places and this would fix all of them. This should be 
done for 1.4.1 at least.

  We have also other parts of the code where this test is done.

> OK to commit?

  For 1.4.0 certainly, it fixes the bug and it is the minimal change.

> Georg

  FWIW I consider this to be a bug from lyx. It a subject where two wrongs 
make one right. :-)

  The first bug is for lyx to output and empty layout, the second is to read 
it.

-- 
José Abílio

Re: Two usability problems in 1.4.0pre2

[Jean-Marc Lasgouttes]

> "Helge" == Helge Hafting <[EMAIL PROTECTED]> writes:

Helge> I think sliders are a good idea - it can be done better than
Helge> the current dialog though. Vertical sliders, with the names of
Helge> the sectioning hierarchy to the left of the slider would be
Helge> easier to understand. The slider position would then correspond
Helge> directly with what gets included in the TOC. (Or what gets
Helge> numbered.)

We could have a vertical line radio buttons instead of a vertical
slider, maybe.

Helge> You may want to file a bug at http://bugzilla.lyx.org/
Helge> concerning this. I am not sure there is time to get this into
Helge> 1.4.0, but this way it won't be forgotten.

This is definitely not something we want to do for 1.4.0.

JMarc

[patch] fix bug 2088

[Georg Baum]

I did not know that an empty \layout is valid, but both 1.3 and 1.4 can read
that, so I made lyx2lyx understand it, too.

OK to commit?


GeorgIndex: lib/lyx2lyx/ChangeLog
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/lib/lyx2lyx/ChangeLog,v
retrieving revision 1.102
diff -u -p -r1.102 ChangeLog
--- lib/lyx2lyx/ChangeLog	28 Sep 2005 14:29:23 -	1.102
+++ lib/lyx2lyx/ChangeLog	12 Oct 2005 07:56:03 -
@@ -1,3 +1,7 @@
+2005-10-12  Georg Baum  <[EMAIL PROTECTED]>
+
+	* lyx_1_4.py (convert_breaks): handle empty layout (bug 2088)
+
 2005-09-28  José Matos  <[EMAIL PROTECTED]>
 
 	* LyX.py: fix relation_format reversion
Index: lib/lyx2lyx/lyx_1_4.py
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/lib/lyx2lyx/lyx_1_4.py,v
retrieving revision 1.53
diff -u -p -r1.53 lyx_1_4.py
--- lib/lyx2lyx/lyx_1_4.py	28 Sep 2005 09:40:48 -	1.53
+++ lib/lyx2lyx/lyx_1_4.py	12 Oct 2005 07:56:04 -
@@ -570,7 +570,11 @@ def convert_breaks(file):
 i = find_token(file.body, "\\begin_layout", i)
 if i == -1:
 return
-layout = split(file.body[i])[1]
+tokens = split(file.body[i])
+if len(tokens) > 1:
+layout = split(file.body[i])[1]
+else:
+layout = 'Standard'
 i = i + 1
 
 # Merge all paragraph parameters into a single line

Bug 2091: Wrong typography when mixing single & double quotes (1.4cvs)

[Helge Hafting]

Discovered by Jim Osborn <[EMAIL PROTECTED]> who seems to understand 
better than me what kind of latex ought to be generated.


Some quoting styles have problems with this sentence:
"The word is 'foo'", he said.

Everything is fine on screen, but the ' followed by " (with no space 
inbetween) goes wrong in the output.  The output looks more like "' 
instead. This may be due to the way latex handles ligatures (double 
qoute being a ligature of single quotes???) but lyx should "do the right 
thing" here as the user's intention is clear.


Note that the sentence have no problem when a different quoting style is 
used:

<>>, he said. That one comes out fine!

Also see messages at the user's list, titled:
space between single and right double quote marks

From the last one:

It does exactly as you say with the
  «The word was  » [I'll call F style]
quoting style, but not with the
 ``The word was `foo' ''[I'll call E style]
style.

Here's the lyx for my E style example:

  \begin_inset Quotes eld
  \end_inset
  
  The word was

  \begin_inset Quotes els
  \end_inset
  
  foo

  \begin_inset Quotes ers
  \end_inset
  
  \begin_inset Quotes erd

  \end_inset
  
   he said.


And here's the dvi:
...foo

The lyx for the F style has "fld" in place of "eld" etc., and it does
indeed produce this dvi with LyX 1.3.6:
...foo

So, apparently it's a bug, at least in LyX 1.3.6, with the E style quotes
but not the F style; I didn't test all the other possibilities.

If you have time, could you verify whether the bug exists in LyX 1.4
for the E style quotes?

End of included message

The 1.3 problem may (or may not) be slightly different.  There
is a problem in 1.4 too though.  It can be worked around by
manually inserting a thin space between the troublesome quotes.

Helge Hafting

Bug 2090: Failure to handle old 1.3 documents with single-quoting style

[Helge Hafting]

Create a document with lyx 1.3  Set the quoting style to single quotes 
instead of the usual doubles.


Open this document in lyx-1.4  Notice how it is impossible to turn 
single-quoting off again, as lyx-1.4 don't have this in the GUI 
anymore.  The problem persists as the file is saved by lyx-1.4.


The correct solution:
Lyx 1.4 have no GUI for singlequote-mode, because single quotes are now 
inserted as special characters.  So support for singlequote-mode in the 
.lyx should be removed too (no saving/loading of this option, and no 
internal state variable to keep this state either), and lyx2lyx should 
simply strip this option when converting old documents.


This would solve the problem. Old documents with single quotes should 
still work, and users will be forced to insert single quotes using the 
insert menu.


Alternative solution:  Reinstate the button that switches between 
single/double quote mode.  Everything else is in place already - it was 
never removed. (Ideally, in single-quote mode the insert menu should 
then offer the insertion of a double quote as an alternative to the 
default single.)  This is the best solution _if_ singlequotes are 
popular anywhere in the world.


Helge Hafting

Re: Back to basics.

[Jose' Matos]

On Tuesday 11 October 2005 01:36, Luis Rivera wrote:
>
> It DOES run on cygwin!
>
> Cute to see lil' LyX, like seeing baby pictures in an album!
>
> I wonder how did it grow so big! ;-)
>
> Anyway, good experiment for after lunch...

  :-)

  Notice that there is a bug there. For some reason the vertical coordinates 
are inverted. That is why you see the menus in the  bottom and the minibuffer 
in the top.

  This comes for the fact that it was designed to run with a different version 
of xforms. Lars told me that he remembers of a switch in the coordinate 
system for xforms at some point between 0.8 and 0.10. He is right since the 
inversion does not show in 0.10.

  I will release soon the patch to run 0.8. :-)
  It is a funny patch:

  - memory allocated with *alloc released with delete
  - memory allocated with new[]  released with delete
  - memory allocated with newreleased with delete[]
  - memory allocated with newreleased with free

  Impressive. ;-)

> Cheers,
>
> Luis.

-- 
José Abílio

Re: [patch] fix bug 2059 (cut and paste in mathed)

[Georg Baum]

Daniel Watkins wrote:

> This one works for me, in conjunction with the undo patch.

Fine. Did anybody else test it?
Can the two patches go in?


Georg

Re: LyX 1.4 on Windows (MinGW)

[Georg Baum]

Martin Vermeer wrote:

> On Wed, 2005-10-12 at 00:42 +0200, Lars Gullik Bjønnes wrote:
>> [EMAIL PROTECTED] (Lars Gullik Bjønnes)
>> writes:
>> 
>> | To be quite honest I belive we should revert messages.C to the simple
>> | state it was in before this summers meeting.
>> 
>> But this is me being grumpy again. Just ignore it.
> 
> Actually I am a bit afraid that have a kernel of truth there.

This seems indeed to be the case.

> OTOH we cannot really go back there, as all newer Linux installations
> have a utf-8 locale nowadays. The cutting off of the encoding suffix is
> necessary as long as LyX is non-unicode.

Also true. But I have to admit that I don't understand the current
machinery. For example, why not simply strip the suffix from the
environment variables once on program start, set the new environment
variables and leave the frontend code as it was before Paris?
 

Georg