[EMAIL PROTECTED] wrote:
> CVSROOT: /usr/local/lyx/cvsroot
> Module name: lyx-devel
> Repository: lyx-devel/src/frontends/gtk/
> Changes by: [EMAIL PROTECTED]04/04/08 00:26:12
>
> Modified files:
> lyx-devel/src/frontends/gtk/: ChangeLog GPainter.C
>
> Log message:
> Trivial co
On Thu, Apr 08, 2004 at 12:04:26AM +0100, Angus Leeming wrote:
> Question 1
> ==
> Is this safe? Ie, will it work if $FIG is empty?
> +if test "$FIG" = "xfig"; then
>
> The usual strategy is:
> +if test "x$FIG" = "xxfig"; then
>
> which is guaranteed to work with all versions of sh.
I n
Asger Kunuk Ottar Alstrup wrote:
> There is a ministry which maintains an enumeration of "preapproved"
> first names. You can find it on the internet at
>
> Girls names:
> http://www.hum.ku.dk/navnef/godkpige.htm
>
> Boys names:
> http://www.hum.ku.dk/navnef/godkdren.htm
So you can't have names
Georg Baum wrote:
> Am Montag, 5. April 2004 17:31 schrieb Angus Leeming:
>> Georg Baum wrote:
>> >> Ok. 'EditCommand'. I'm happy to kill that...
>> >
>> > This is easy. I have added an editor (similar to the viewer) to
>> > the formats mechanism in my 1.3 source.
>>
>> > I can adapt the patch t
On Wed, 7 Apr 2004, Jean-Marc Lasgouttes wrote:
> But all these new fancy names are not recognized by the danish
> administration, are they? Or do you mean that the Church _is_ the
> administrator of the country?
There is a ministry which maintains an enumeration of "preapproved"
first names. You
Andre Poenitz wrote:
>
> My last patch gives us an almost working 'real' \mbox inset for
> mathed I.e. creation and editing works, but nothing I/O related.
>
> The reason for this is that mathed and texted still use different
> I/O schemes (I don't mean the format here, but mainly the signatures
Andre Poenitz wrote:
> Fix the macro stuff for good.
>
> The macro stuff is (a) cool, (b) conceptionally broken since day
> one, (c) broken in current CVS.
My take on this: good! I do think that this is the right time to fix
the math macros correctly. As you say, they are currently broken
anyway.
Georg Baum wrote:
>> > Could you apply it, please?
>>
>> Done.
>
> Unfortunately I did not test the "Edit file" button, which does not
> work. The attached fixes that.
>
> Angus, could you again apply please? I ask reluctantly for simple
> things like this, but as you know, I can't do it myself
Andre Poenitz wrote:
> On Wed, Apr 07, 2004 at 05:37:46PM +0100, Angus Leeming wrote:
>> Remember, the Danes never had a bloody revolution in their country
>> to sweep away the old order. They didn't need to because they had
>> Martin Luther instead.
>
> Well, you've never had one yourselves as fa
Am Montag, 5. April 2004 17:31 schrieb Angus Leeming:
> Georg Baum wrote:
> >> Ok. 'EditCommand'. I'm happy to kill that...
> >
> > This is easy. I have added an editor (similar to the viewer) to the
> > formats mechanism in my 1.3 source.
>
> > I can adapt the patch to 1.4 if you like. I always
Am Dienstag, 6. April 2004 23:07 schrieb Angus Leeming:
> Georg Baum wrote:
> > Could you apply it, please?
>
> Done.
Unfortunately I did not test the "Edit file" button, which does not work.
The attached fixes that.
Angus, could you again apply please? I ask reluctantly for simple things
like
I've just commited this.
It's very long, but mainly because it changes some signatures (and adds two
new files). *
I've added some bits here and there to dispatch functions to finally have an
update-less cursor movement (!). I hope I didn't break anything. Those who
are familiar with that code pl
John Levon wrote:
> On Wed, Apr 07, 2004 at 10:16:48PM +0200, Andre Poenitz wrote:
>
>> That'd be indeed a shame. But as all the change tracking code sits
>> around in a corner somewhere it 'only' takes someone who knows how it
>> used to work to cobble it together...
>
> Last I looked, fixing i
On Wed, Apr 07, 2004 at 10:16:48PM +0200, Andre Poenitz wrote:
> That'd be indeed a shame. But as all the change tracking code sits
> around in a corner somewhere it 'only' takes someone who knows how it
> used to work to cobble it together...
Last I looked, fixing it would require some non-trivi
On Wed, Apr 07, 2004 at 07:51:52PM +0200, Alfredo Braunstein wrote:
> Andre Poenitz wrote:
>
> > On Wed, Apr 07, 2004 at 02:18:38PM +0100, Angus Leeming wrote:
> >> Also, if you enter using the keyboard arrow keys, then the cursor is
> >> displayed elsewhere,
> >
> > I don't see this 'elsewhere'
On Wed, Apr 07, 2004 at 06:07:24PM +0100, John Levon wrote:
> Change tracking is unfortunately badly broken for quite some time, it
> would be a shame if this wasn't brought back into a working state by
> release time
That'd be indeed a shame. But as all the change tracking code sits
around in a c
On Wed, Apr 07, 2004 at 05:56:47PM +0200, Jean-Marc Lasgouttes wrote:
>
> I get this with gcc 3.2.2:
>
> g++ -DHAVE_CONFIG_H -I. -I../../lyx-devel/src -I. -I../../lyx-devel/boost
> -I/afs/inria.fr/rocq/home/preval/lasgoutt/.common//include -I/usr/X11R6/include
> -fno-exceptions -W -Wall -c
On Wed, Apr 07, 2004 at 05:39:01PM +0100, John Levon wrote:
> On Wed, Apr 07, 2004 at 04:57:55PM +0100, Angus Leeming wrote:
>
> > >> Works for me for the "tabular create" dialog in both lyx 13x and
> > >> 14x.
> > >
> > > Well, not here. This is fvwm2 but I doubt this is the reason.
>
> Sounds
On Wed, Apr 07, 2004 at 05:56:47PM +0200, Jean-Marc Lasgouttes wrote:
> Andre', how come you do not have the same problem?
Should be fixed now I hope.
Andre'
My last patch gives us an almost working 'real' \mbox inset for mathed
I.e. creation and editing works, but nothing I/O related.
The reason for this is that mathed and texted still use different I/O
schemes (I don't mean the format here, but mainly the signatures of
I/O functions).
In mathed we
On Wed, Apr 07, 2004 at 06:00:08PM +0100, Angus Leeming wrote:
> Incidentally, How is life with you? Did you find your "Challenging
> position in systems software development"? Can we tempt you back to
> writing code for LyX?
Or just check that everything is in order with change tracking, for
st
On Wed, Apr 07, 2004 at 05:37:46PM +0100, Angus Leeming wrote:
> Remember, the Danes never had a bloody revolution in their country to
> sweep away the old order. They didn't need to because they had Martin
> Luther instead.
Well, you've never had one yourselves as far as I can tell.
This Cromwe
Andre Poenitz wrote:
> On Wed, Apr 07, 2004 at 02:18:38PM +0100, Angus Leeming wrote:
>> Also, if you enter using the keyboard arrow keys, then the cursor is
>> displayed elsewhere,
>
> I don't see this 'elsewhere' except for macros which still are broken.
>
> Alfredo: Would you think we could a
John Levon wrote:
>> Fair enough. It's been a frustrating time, but things *are* getting
>> better.
>
> That's good to hear. I look forward to trying to work out how lyx
> works all over again :)
>
> Change tracking is unfortunately badly broken for quite some time,
> it would be a shame if this
On Wed, Apr 07, 2004 at 06:09:50PM +0100, Angus Leeming wrote:
> Interesting. Based here or in the US?
Both really :) I'm still in Manchester but practically everybody I work
with lives in California.
> Fair enough. It's been a frustrating time, but things *are* getting
> better.
That's good t
John Levon wrote:
>> Incidentally, How is life with you? Did you find your "Challenging
>> position in systems software development"?
>
> Sure did. I'm busy hacking Solaris for Sun.
Interesting. Based here or in the US?
>> Can we tempt you back to writing code for LyX?
>
> I'm waiting for lyx t
On Wed, Apr 07, 2004 at 06:00:08PM +0100, Angus Leeming wrote:
> Granted. But that doesn't have an "Ok" button, just a "Close".
True.
> Incidentally, How is life with you? Did you find your "Challenging
> position in systems software development"?
Sure did. I'm busy hacking Solaris for Sun.
>
John Levon wrote:
>> >> Works for me for the "tabular create" dialog in both lyx 13x and
>> >> 14x.
>> > Well, not here. This is fvwm2 but I doubt this is the reason.
> Sounds to me like Andre is talking about the Tabular edit dialog not
> Create, which disallows Return->OK for good reason
Grant
On Wed, Apr 07, 2004 at 04:19:40PM +0200, Jean-Marc Lasgouttes wrote:
> > "Andre" == Andre Poenitz <[EMAIL PROTECTED]> writes:
>
> Andre> Could we please move a large portion of src/Changelog and
> Andre> src/insets/Changelog into, say, */Changelog.OLD.
>
> Andre> Updating these over a modem
On Wed, Apr 07, 2004 at 04:12:20PM +0100, Angus Leeming wrote:
> Angus Leeming wrote:
> > Also, if you enter using the keyboard arrow keys, then the cursor is
> > displayed elsewhere, even though the inset is 'active'. Typing new
> > stuff into the inset works fine.
>
> Ok, Andre's latest patch pa
On Wed, Apr 07, 2004 at 04:57:55PM +0100, Angus Leeming wrote:
> >> Works for me for the "tabular create" dialog in both lyx 13x and
> >> 14x.
> >
> > Well, not here. This is fvwm2 but I doubt this is the reason.
Sounds to me like Andre is talking about the Tabular edit dialog not
Create, which
Jean-Marc Lasgouttes wrote:
> Asger> The good people at the church office looked a bit strange at
> me Asger> when I explained this great name transformation to them,
> but Asger> they looked it up in their books, and had to admit that
> Asger> everything was fine :-)
>
> But all these new fancy n
> "Asger" == Asger Kunuk Ottar Alstrup <[EMAIL PROTECTED]> writes:
Asger> I don't know about Sweden, since I am a Dane after all, but I
Asger> am probably weird anyway.
So it is not the same ? [ducks and runs]
Asger> The good people at the church office looked a bit strange at me
Asger> whe
../../lyx/src/text2.C:114:2: warning: #warning broken?
../../lyx/src/text2.C:1338:2: warning: #warning This will not work anymore when we
have multiple views of the same buffer
../../lyx/src/text2.C: In member function `bool
LyXText::deleteEmptyParagraphMechanism(LCursor&, const LCursor&)':
..
Andre Poenitz wrote:
>> Also, if you enter using the keyboard arrow keys, then the cursor
>> is displayed elsewhere,
> I don't see this 'elsewhere' except for macros which still are
> broken.
No, you've fixed this. What is broken is entering a math inset with a
mouse press.
--
Angus
I get this with gcc 3.2.2:
g++ -DHAVE_CONFIG_H -I. -I../../lyx-devel/src -I. -I../../lyx-devel/boost
-I/afs/inria.fr/rocq/home/preval/lasgoutt/.common//include -I/usr/X11R6/include
-fno-exceptions -W -Wall -c -o text2.o `test -f '../../lyx-devel/src/text2.C' || echo
'../../lyx-devel/src/'`
Andre Poenitz wrote:
>> > Is there a possibility to invoke the OK button in dialogs like
>> > Insert->Tabular with the key?
>>
>> Yes.
>>
>> > The picture <-' on the button seems to indicate that somebody was
>> > thinking of this possibility at some point of time, but it does
>> > not work for
On Wed, Apr 07, 2004 at 02:18:38PM +0100, Angus Leeming wrote:
> Also, if you enter using the keyboard arrow keys, then the cursor is
> displayed elsewhere,
I don't see this 'elsewhere' except for macros which still are broken.
Alfredo: Would you think we could afford a traversal of the whole
do
On Wed, Apr 07, 2004 at 02:35:30PM +0100, Angus Leeming wrote:
> Andre Poenitz wrote:
> > Is there a possibility to invoke the OK button in dialogs like
> > Insert->Tabular with the key?
>
> Yes.
>
> > The picture <-' on the button seems to indicate that somebody was
> > thinking of this possibi
Angus Leeming wrote:
> Also, if you enter using the keyboard arrow keys, then the cursor is
> displayed elsewhere, even though the inset is 'active'. Typing new
> stuff into the inset works fine.
Ok, Andre's latest patch partially cured this. The cursor is now
displayed at the expected position w
Jean-Marc Lasgouttes wrote:
>> "Georg" == Georg Baum <[EMAIL PROTECTED]> writes:
> Georg> The only remaining bit of bug 605 is now the fix for
> Georg> LaTeXFeatures::buffer_, which awaits a decision by Jean-Marc.
>
> Sorry. That's OK, of course. Shall I apply it?
Yes, please.
Georg
> "Andre" == Andre Poenitz <[EMAIL PROTECTED]> writes:
Andre> Could we please move a large portion of src/Changelog and
Andre> src/insets/Changelog into, say, */Changelog.OLD.
Andre> Updating these over a modem line is a pain...
I think the GNU use is to move ChangeLog to ChangeLog.1 (or
Cha
> "Andre" == Andre Poenitz <[EMAIL PROTECTED]> writes:
Andre> Btw, is there a method to use the autotools and yet get a
Andre> 'flat' Makefile (i.e. just a toplevel Makefile with possible
Andre> inclusion of other Makefiles but no recursive calls to make)?
Andre> I've seen to projects with 'm
> "Georg" == Georg Baum <[EMAIL PROTECTED]> writes:
Georg> Angus Leeming wrote:
>> Georg Baum wrote:
>>> Could you apply it, please?
>> Done.
Georg> Thanks.
Georg> The only remaining bit of bug 605 is now the fix for
Georg> LaTeXFeatures::buffer_, which awaits a decision by Jean-Marc.
Sorr
Andre Poenitz wrote:
> On Wed, Apr 07, 2004 at 10:52:38AM +0200, Alfredo Braunstein wrote:
>> Comments?
>
> With exception of the unusual indentation in
>
> + /// draw a line from point to point
> + Painter & line(int, int, int, int, LColor_color,
> + line_style
On Tue, 6 Apr 2004, Jean-Marc Lasgouttes wrote:
> Asger> In Denmark, you can just add any number of first names you
> Asger> would like.
>
> That's fascinating. Do a lot of people play this game in Sweden, or
> are you just weird?
I don't know about Sweden, since I am a Dane after all, but I am p
On Wed, Apr 07, 2004 at 01:38:45PM +0100, Angus Leeming wrote:
> > No, notifyCursorLeaves should be the place. However, I am pretty
> > sure this is almost never called when necessary. So every place
> > modifying the 'real' cursor (there aren't too many of them, the
> > MOUSE lfuns and Cursor/Buff
On Wed, Apr 07, 2004 at 02:11:50PM +0200, Georg Baum wrote:
> Angus Leeming wrote:
>
> > The display of subscripts in this file created by lyx 13x is foobarred
> > when viewed with lyx 14x. Andre, I suspect that this is your
>
> Not only the view is wrong, the same stuff is written to the .lyx fi
On Wed, Apr 07, 2004 at 01:41:53PM +0100, Angus Leeming wrote:
> Andre Poenitz wrote:
> > Could we please move a large portion of src/Changelog and
> > src/insets/Changelog into, say, */Changelog.OLD.
>
> Why not
> src/ChangeLog.pre14x
> src/insets/ChangeLog.pre14x
>
> Then just r
Andre Poenitz wrote:
> Is there a possibility to invoke the OK button in dialogs like
> Insert->Tabular with the key?
Yes.
> The picture <-' on the button seems to indicate that somebody was
> thinking of this possibility at some point of time, but it does not
> work for me neither in 1.4.0cvs n
Is there a possibility to invoke the OK button in dialogs like
Insert->Tabular with the key?
The picture <-' on the button seems to indicate that somebody was
thinking of this possibility at some point of time, but it does not
work for me neither in 1.4.0cvs nor 1.3.5cvs.
Andre'
Also, if you enter using the keyboard arrow keys, then the cursor is
displayed elsewhere, even though the inset is 'active'. Typing new
stuff into the inset works fine.
This is current cvs, including Andre's 1 line change to
math_scriptinset.C.
--
Angus
Andre Poenitz wrote:
> Could we please move a large portion of src/Changelog and
> src/insets/Changelog into, say, */Changelog.OLD.
Why not
src/ChangeLog.pre14x
src/insets/ChangeLog.pre14x
Then just rename the ChangeLog after we release. Seems scalable to
me...
> Updating these
Andre Poenitz wrote:
> On Tue, Apr 06, 2004 at 10:49:43PM +0100, Angus Leeming wrote:
>> Andre,
>>
>> the trigger used to generate a preview of a math inset is the
>> cursor leaving the inset. In the old world I used this:
>>
>> void InsetFormulaBase::insetUnlock(BufferView * bv)
>> {
>>
Angus Leeming wrote:
> The display of subscripts in this file created by lyx 13x is foobarred
> when viewed with lyx 14x. Andre, I suspect that this is your
Not only the view is wrong, the same stuff is written to the .lyx file if
you save it.
> re-definition of which cell means what. Have you i
On Wed, Apr 07, 2004 at 10:52:38AM +0200, Alfredo Braunstein wrote:
> Comments?
With exception of the unusual indentation in
+ /// draw a line from point to point
+ Painter & line(int, int, int, int, LColor_color,
+ line_style = line_solid, line_width = line_thin)
Could we please move a large portion of src/Changelog and
src/insets/Changelog into, say, */Changelog.OLD.
Updating these over a modem line is a pain...
Andre'
On Wed, Apr 07, 2004 at 10:29:36AM +0100, Angus Leeming wrote:
> The display of subscripts in this file created by lyx 13x is foobarred
> when viewed with lyx 14x. Andre, I suspect that this is your
> re-definition of which cell means what. Have you introduced a format
> change?
No. It's still LaT
On Wed, Apr 07, 2004 at 09:32:42AM +0100, Angus Leeming wrote:
> Using --enable-maintainer-mode I find that (as happened yesterday?)
> typing 'make' causes autogen.sh and configure to be re-run. Fine. The
> strangeness, however, is that the order of directory traversal is
> something like:
>
> int
Angus Leeming wrote:
> Georg Baum wrote:
>> Could you apply it, please?
>
> Done.
Thanks.
The only remaining bit of bug 605 is now the fix for LaTeXFeatures::buffer_,
which awaits a decision by Jean-Marc.
Georg
This fixes a crash when changing font for a whole table cell and brings
a bit of selection by mouse back. Not perfect, though...
Andre'
--
Index: BufferView_pimpl.C
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/BufferView_pimpl
On Tue, Apr 06, 2004 at 10:49:43PM +0100, Angus Leeming wrote:
> Andre,
>
> the trigger used to generate a preview of a math inset is the cursor
> leaving the inset. In the old world I used this:
>
> void InsetFormulaBase::insetUnlock(BufferView * bv)
> {
> if (mathcursor) {
>
The display of subscripts in this file created by lyx 13x is foobarred
when viewed with lyx 14x. Andre, I suspect that this is your
re-definition of which cell means what. Have you introduced a format
change?
--
Angus
trial.lyx
Description: application/lyx
Angus Leeming wrote:
> Alfredo Braunstein wrote:
>> This introduces a dumb NullPainter
>
>> Comments?
>
> Looks good to me. Trivial points: nullpainter.C is missing a header
> blurb and a #include . Why do you #include "LColor.h"
> before "nullpainter.h"?
Will take car of that.
> nullpainter.h
Alfredo Braunstein wrote:
> This introduces a dumb NullPainter
> Comments?
Looks good to me. Trivial points: nullpainter.C is missing a header
blurb and a #include . Why do you #include "LColor.h"
before "nullpainter.h"?
nullpainter.h declares itself incorrectly in the blurb as '\file
Painter.h'
Andre Poenitz wrote:
> Yes. I saw this mail too late. It seems to be simpler and faster.
Third shot.
This introduces a dumb NullPainter : public Painter which does nothing. The
NullPainter has infinite paperHeight, this means we can leave alone the
optimization of not drawing out-of-screen, whic
Using --enable-maintainer-mode I find that (as happened yesterday?)
typing 'make' causes autogen.sh and configure to be re-run. Fine. The
strangeness, however, is that the order of directory traversal is
something like:
intl
boost
src
Entering src leads to config.h being rebuilt, the upshot of wh
67 matches
Mail list logo
