On Thu, Jun 05, 2003 at 08:02:57PM +0100, John Levon wrote:
> I'm considering turning on the extra toolbar by default. Comments ?
What extra toolbar?
[But I guess that answers the question...]
Btw: It would be nice to re-evaluate the default.ui changes.
I fail to see why e.g Edit->ParagraphSet
> "Angus" == Angus Leeming <[EMAIL PROTECTED]> writes:
Angus> Would anyone find this useful? Shall I put it in lib/scripts?
Can it be directly used by LyX? If not, I am not sure that lib/scripts
is the best place for it...
Nice idea, BTW. Note that recent gnuplot cvs versions do have support
Jose' Matos wrote:
> On Thursday 05 June 2003 13:17, Angus Leeming wrote:
>> Attached is a shell script to run a gnuplot command file through
>> gnuplot and create an xfig file. This file is then manipulated to
>> ensure that all text is marked as "special". Sample usage:
>>
>> Would anyone find t
On Thursday 05 June 2003 13:17, Angus Leeming wrote:
> Attached is a shell script to run a gnuplot command file through
> gnuplot and create an xfig file. This file is then manipulated to
> ensure that all text is marked as "special". Sample usage:
>
> Would anyone find this useful? Shall I put it
Attached is a shell script to run a gnuplot command file through
gnuplot and create an xfig file. This file is then manipulated to
ensure that all text is marked as "special". Sample usage:
$ gnuplot2fig q8.gnuplot q8.fig
$ cat q8.gnuplot
a=324
b=0.0534
c=10
f(x)=a * exp(-b * x) + c
set xrange
And incidently, I just found one of the 'off by some' [;-}] error when
having a look at setLayout()'s undo...
Note that there is quite some potential to remove code in text2.C along the
lines of:
- ParagraphList::iterator endpit = boost::next(send_cur.par());
- ParagraphLi
On Thu, Jun 05, 2003 at 02:05:38PM +0200, Jean-Marc Lasgouttes wrote:
> > "Andre" == Andre Poenitz <[EMAIL PROTECTED]> writes:
>
> Andre> The current implementation stores only full outer level
> Andre> paragraphs, so this is more expensive than the previous
> Andre> solution. This matters for
> "Andre" == Andre Poenitz <[EMAIL PROTECTED]> writes:
Andre> The current implementation stores only full outer level
Andre> paragraphs, so this is more expensive than the previous
Andre> solution. This matters for large tables etc.
Do you have a plan for restoring this capability? I think it
Here's a rough and unready start on insetcode. Try it with "code-insert"
Issues :
o extra newline between pars. I suspect we need some handling in
TeXOnePar but that sounds hacky
o we can trivially allow insets such as insetref OK I think
o footnote is more difficult due to the "%\n" string.
On Thu, Jun 05, 2003 at 12:38:07PM +0200, Lars Gullik Bjønnes wrote:
> | I had a look at it last night. It is a bit outdated, but the ideas fit
> | nicely in the two stage drawing system. So a plan could be as follows:
>
> Before we do any of this, we should try to get Undo into "perfect"
> stat
Andre Poenitz <[EMAIL PROTECTED]> writes:
| On Wed, Jun 04, 2003 at 07:37:06PM +0200, Lars Gullik Bjønnes wrote:
| > | I don't understand the rebreaking point. Do you mean breaking a par into
| > | two or some re-calculation of which part of a par belongs to which row?
| > | If the latter, I think
This patch (including the previous free spacing one plus a fix) splits
most of the ERT code into a parent class for the benefit of an alltt
based insetcode (http://bugzilla.lyx.org/show_bug.cgi?id=802).
InsetCode should hopefully be rather simple after this.
Comments ?
regards
john
Index: buf
On Wed, Jun 04, 2003 at 12:32:02PM +, Angus Leeming spake thusly:
> Andre Poenitz wrote:
>
> > On Wed, Jun 04, 2003 at 01:20:27PM +0200, Lars Gullik Bjønnes wrote:
> >> Hmm the bag says: "Blended & Packed by R. Twinning & Company Limited,
> >> London, England."
> >>
> >> So I guess it you w
Andre Poenitz wrote:
> On Thu, Jun 05, 2003 at 10:26:59AM +0200, Jean-Marc Lasgouttes
> wrote:
>> > "Andre" == Andre Poenitz <[EMAIL PROTECTED]> writes:
>>
>> Andre> Hm... now I have to use ./configure --with-included-string
>> Andre> --without-debug --with-version-suffix=1.4.0cvs
>> Andre> -
Jean-Marc Lasgouttes wrote:
>> "Angus" == Angus Leeming <[EMAIL PROTECTED]> writes:
>
> Angus> This is the last piece of nastiness associated with the
> Angus> external inset:
>
> [...]
>
> Angus> I propose to replace it with a new LFUN:
> [...]
> Angus> (which also means I can get rid of t
Red Hat 6.2
gcc gcc-2.95.2-7cl
autoconf-2.53-8
automake-1.6.3-4cl
Clean build:
./autogen.sh
configured by ./configure, generated by GNU Autoconf 2.53, with options
\"'--with-extra-lib=/home/mv/xforms-1.0/lib:/home/mv/xforms-1.0/image'
'--with-extra-inc=/home/mv/xforms-1.0/exports/include'
'--wit
On Thu, Jun 05, 2003 at 10:26:59AM +0200, Jean-Marc Lasgouttes wrote:
> > "Andre" == Andre Poenitz <[EMAIL PROTECTED]> writes:
>
> Andre> Hm... now I have to use ./configure --with-included-string
> Andre> --without-debug --with-version-suffix=1.4.0cvs
> Andre> --enable-maintainer-mode instead
On Thu, Jun 05, 2003 at 10:25:39AM +0200, Juergen Vigna wrote:
> >But the contained InsetTexts do, so I am not spanking 'slow InsetTabular as
> >created by Juergen' but 'InsetTabular as container of lots InsetTexts'
> >[Although I'd think that there is some unnecessary fat in InsetTabular,
> >too..
> "Andre" == Andre Poenitz <[EMAIL PROTECTED]> writes:
Andre> Hm... now I have to use ./configure --with-included-string
Andre> --without-debug --with-version-suffix=1.4.0cvs
Andre> --enable-maintainer-mode instead of a simple ./configure.
But you just have to do it once, remember...
JMarc
Andre Poenitz wrote:
But the contained InsetTexts do, so I am not spanking 'slow InsetTabular as
created by Juergen' but 'InsetTabular as container of lots InsetTexts'
[Although I'd think that there is some unnecessary fat in InsetTabular,
too...]
You may have a point here, althought I don't know
On Thu, Jun 05, 2003 at 10:06:58AM +0200, Jean-Marc Lasgouttes wrote:
> Andre> The new(?) fourth item in textclass.lst leads to
> Andre> incomaptibility between 1.3 and 1.4 both ways, i.e. one always
> Andre> has to run Reconfigure when switching version which is
> Andre> especially annoying as LyX
On Thu, Jun 05, 2003 at 09:03:16AM +0200, Juergen Vigna wrote:
> Andre Poenitz wrote:
> >This 'when needed' is basically the time between first and second drawing
> >phase. We need the rebreak to figure out the height of the InsetText,
> >so it is needed in the metrics() phase. And once we've done
> "Andre" == Andre Poenitz <[EMAIL PROTECTED]> writes:
Andre> The new(?) fourth item in textclass.lst leads to
Andre> incomaptibility between 1.3 and 1.4 both ways, i.e. one always
Andre> has to run Reconfigure when switching version which is
Andre> especially annoying as LyX 'needs to be rest
> "Angus" == Angus Leeming <[EMAIL PROTECTED]> writes:
Angus> This is the last piece of nastiness associated with the
Angus> external inset:
[...]
Angus> I propose to replace it with a new LFUN:
[...]
Angus> (which also means I can get rid of that klunky
Angus> InsetExternal::cache method.)
On Wed, Jun 04, 2003 at 07:37:06PM +0200, Lars Gullik Bjønnes wrote:
> | I don't understand the rebreaking point. Do you mean breaking a par into
> | two or some re-calculation of which part of a par belongs to which row?
> | If the latter, I think the solution is consequent two-phase drawing with
Andre Poenitz wrote:
This 'when needed' is basically the time between first and second drawing
phase. We need the rebreak to figure out the height of the InsetText,
so it is needed in the metrics() phase. And once we've done it there there
is no reason not to save the results up to the next draw()
I'd like to get insettext/insettabular on par with the rest of insets/*
which I believe is in comparatively good shape nowadays.
The attached patch removes InsetText::InnerCache and tells all users of it
that there is not anything valid in the cache (i.e. assume find(...) ==
end() and remove all
The new(?) fourth item in textclass.lst leads to incomaptibility between
1.3 and 1.4 both ways, i.e. one always has to run Reconfigure when
switching version which is especially annoying as LyX 'needs to be
restarted' afterwards.
This should be fixed at least for the upgrade 1.3 -> 1.4.
[Who inv
On Wed, Jun 04, 2003 at 09:28:43PM +, Angus Leeming wrote:
> Of the files in lib/scripts, these are now redundant:
> [..]
> Ok to remove the redundant ones?
Could you imagine any reason why it wouldn't be ok?
If not...
Ander'
--
Those who desire to give up Freedom in order to
This is the last piece of nastiness associated with the external inset:
void ControlExternal::editExternal()
{
lyx::Assert(params_.get());
dialog().view().apply();
InsetExternal inset;
inset.setParams(*params_, kernel().buffer()->filePath());
inset.cache(ke
Of the files in lib/scripts, these are now redundant:
fig2pstex.py pic2ascii.py pic2png_eps.py general_command_wrapper.py
pic2png_eps.sh fig2png.py
Leaving these as stuff used by LyX:
convertDefault.sh fig2pstex.sh fen2ascii.py fig2pdftex.sh TeXFiles.sh
lyxpreview2bitmap.sh listerrors
Ok to rem
Now I have got the external inset working as I'd like, I guess that it is
time to actually kill the RasterImage template. The attached patch to
lyx2lyx replaces any RasterImage external inset with a graphics inset.
Ok?
--
Angus? lib/reLyX-safe
Index: lib/ChangeLog
=
Jose' Matos wrote:
> Yes, it can, but that is not the present behaviour.
> For the present scheme to work you need to create a new file
> lyxconvert_223.py and put there the convertion stuff.
See attached. Works perfectly.
> My question remains, why to change now? Was that intentional?
>
Andre Poenitz wrote:
> "More than one lyxview" would need some generic solution as e.g. mathed
> can't cope with that anyway (and I doubt graphics will - or anything
> caching a bv_ privately for that matter). That's solidly a few years in
> the future, if at all...
The graphics loading stuff does
> Ruurd> To get back at my original question: Can I put it on the lyx
> Ruurd> ftp site when I'm done?
>
> Well, I still do not know what is involved... Of course, the storage
> on the server is not a problem. What do you have in your distribution?
> Didn't I read something about bits of perl or py
On Wed, Jun 04, 2003 at 07:37:06PM +0200, Lars Gullik Bjønnes wrote:
> One other idea when I read through the lazy_lyxtext doc, is to never
> cache the lyx rows at all, but always generate them when needed. That
> might however be too slow... I don't know.
This 'when needed' is basically the time
On Wednesday 04 June 2003 17:47, Angus Leeming wrote:
> Jose' Matos wrote:
>
> I don't claim to understand the magic that lyx2lyx uses here.
I am asking what is the criterion for the version jump. I elaborate this
bellow.
> Which bit shouldn't be bumped? lyx2lyx or buffer.C?
>
> Why doesn't it
Andre Poenitz <[EMAIL PROTECTED]> writes:
| On Wed, Jun 04, 2003 at 05:23:06PM +0200, Lars Gullik Bjønnes wrote:
| > | Wouldn't it be much simpler if each Paragraph contained only its own part
| > | of the RowList (sort of 'drawing cache') and is completely ignorant of the
| > | others? Paragraphs
On Wed, Jun 04, 2003 at 06:12:00PM +0200, Andre Poenitz wrote:
> caching a bv_ privately for that matter). That's solidly a few years in
> the future, if at all...
Yep !
Note that per-paragraph rowlists is actually not very far away from
InsetParagraph
regards
john
On Wed, Jun 04, 2003 at 05:23:06PM +0200, Lars Gullik Bjønnes wrote:
> | Wouldn't it be much simpler if each Paragraph contained only its own part
> | of the RowList (sort of 'drawing cache') and is completely ignorant of the
> | others? Paragraphs should contain an integer number of full rows, sho
Jose' Matos wrote:
> On Wednesday 04 June 2003 17:25, Angus Leeming wrote:
>> Adds a lyx2lyx conversion routine and;
>> bumps the LyX format to 224.
>
>Do we need this?
>Even if we do it we should do it in a different way, or else lyx2lyx
>will
> not be able to read 223 format documen
On Wednesday 04 June 2003 17:25, Angus Leeming wrote:
> Adds a lyx2lyx conversion routine and;
> bumps the LyX format to 224.
Do we need this?
Even if we do it we should do it in a different way, or else lyx2lyx will
not be able to read 223 format documents.
> Any problems with me applying
This removes the unusable InsetExternal::Params::parameters variable;
Rewrites the InsetExternal read and write methods to in/output the new
format elegantly;
Adds a lyx2lyx conversion routine and;
bumps the LyX format to 224.
Aside to John: my attempts to make an elegant External dialog in Qt ha
Andre Poenitz <[EMAIL PROTECTED]> writes:
| On Wed, Jun 04, 2003 at 05:08:24PM +0200, Andre' Poenitz wrote:
| > This fixes part of the cursor positioning after undo.
| > It even seems to work sometimes in insets.
|
| And here's the patch...
go for it.
--
Lgb
Andre Poenitz <[EMAIL PROTECTED]> writes:
| This fixes part of the cursor positioning after undo.
| It even seems to work sometimes in insets.
|
| It still crashes at a pretty high rate, most often in redoParagraphs().
|
| ---
|
| I really wonder what this function is good for anyway.
|
| Can
On Wed, Jun 04, 2003 at 05:08:24PM +0200, Andre' Poenitz wrote:
> This fixes part of the cursor positioning after undo.
> It even seems to work sometimes in insets.
And here's the patch...
Index: bufferview_funcs.C
===
RCS file: /usr
This fixes part of the cursor positioning after undo.
It even seems to work sometimes in insets.
It still crashes at a pretty high rate, most often in redoParagraphs().
---
I really wonder what this function is good for anyway.
Can anybody explain me in simple words the relationship between Pa
> "John" == John Levon <[EMAIL PROTECTED]> writes:
John> Load extended.lyx and find the first Itemize list, then increase
John> the depth of one of the items. Dead slow. Has to be fixed I
John> think
I did some profiling with gprof. The culprit seems to be:
[43]12.90.000.30
48 matches
Mail list logo
