On Fri, Jun 28, 2002 at 11:45:45AM +0100, Angus Leeming wrote:
> > Which I believe is to late. It should be done lazily (i.e the first
> > screenfull of text should come up really fast) but not too lazy.
>
> Vapourware, vapourware...
?
> As you have found with the maths preview, a document wit
On Fri, Jun 28, 2002 at 01:58:57PM +0300, Dekel Tsur wrote:
> Try the following:
> insert ${a\choose b}$ to a document, and then
> do save, revert, save, revert etc.
> The number of braces increases at each iteration.
> We can solve this particular problem by changing MathBinomInset::write to
>
On Friday 28 June 2002 11:35 am, Andre Poenitz wrote:
> > There is no load up penalty with the graphics inset as I understand it
> > because we don't startLoading() until a draw request is received.
>
> Which I believe is to late. It should be done lazily (i.e the first
> screenfull of text should
On Fri, Jun 28, 2002 at 01:33:49PM +0300, Dekel Tsur wrote:
> The old graphics inset stored the width & height of an image (on the LyX
> display) in the LyX file.
> This prevented changes in display when loading the images.
> Perhaps this should be implemented in the new graphics inset.
Sounds se
On Fri, Jun 28, 2002 at 11:40:40AM +, [EMAIL PROTECTED] wrote:
> Modified files:
> lyx-devel/src/mathed/: ChangeLog math_charinset.h math_inset.h
> math_parser.C math_sizeinset.h
> math_symbolinset.h
>
> Log message:
> r
On Fri, Jun 28, 2002 at 10:59:38AM +0100, Angus Leeming wrote:
> > Bad idea. Konni was already complaining about that features for graphics.
> > In old LyX you had just to wait some time and were sure that everything was
> > loaded. Now you can't even scroll anymore through a document without the
On Fri, Jun 28, 2002 at 11:38:58AM +0200, Andre Poenitz wrote:
> > In InsetGraphics, I do not startLoading() until the inset recieves a draw
> > request. Perhaps you could use a similar approach?
>
> Bad idea. Konni was already complaining about that features for graphics.
> In old LyX you had j
On Fri, Jun 28, 2002 at 12:01:02PM +0200, David Kastrup wrote:
> You are of course free to ignore the experiences we made while
> developing preview-latex (I described this a few times already), and
> free to ignore the LaTeX styles, PostScript code and other
> infrastructure we developed to make
On Friday 28 June 2002 10:38 am, Andre Poenitz wrote:
> On Fri, Jun 28, 2002 at 10:08:14AM +0100, Angus Leeming wrote:
> > I don't think that's correct. Strictly speaking, loading stalls until you
> > have started off all the conversion processes.
>
> Why is that?
>
> Can't we have something like
On Friday 28 June 2002 11:01 am, David Kastrup wrote:
> Angus Leeming <[EMAIL PROTECTED]> writes:
> > Alternatively, perhaps we should think of setting off a single
> > forked process that will result in multiple image files when loading
> > the document. This is definitely possible in the case of
On Fri, Jun 28, 2002 at 11:49:15AM +0200, Helge Hafting wrote:
> The patch is here: http://www.aitel.hist.no/~helgehaf/sw/
Does not look too bad, certainly a step in the right direction.
What exactly is the problem why you can't use this \subimport* directly?
Is this non-standard?
Andre'
--
Angus Leeming <[EMAIL PROTECTED]> writes:
> Alternatively, perhaps we should think of setting off a single
> forked process that will result in multiple image files when loading
> the document. This is definitely possible in the case of the mathed
> preview I believe and is essentially what David
The patch is here: http://www.aitel.hist.no/~helgehaf/sw/
On Fri, Jun 28, 2002 at 10:08:14AM +0100, Angus Leeming wrote:
> I don't think that's correct. Strictly speaking, loading stalls until you
> have started off all the conversion processes.
Why is that?
Can't we have something like a "qun queue" and put things just there and
have _everything_ don
On Friday 28 June 2002 10:01 am, Andre Poenitz wrote:
> Angus,
>
> there is still something fishy with the loading.
>
> When I have a doc with a only few formulas, it can loads quickly and
> I can see that the preview renderings are done asynchrounously. Fine.
>
> However, when I start with a larg
On Friday 28 June 2002 06:50, Andre Poenitz wrote:
> On Fri, Jun 28, 2002 at 12:49:43AM +0200, Lars Gullik Bjønnes wrote:
> > | It changes all the time. Every build it seems to be different. It's
> > | pretty useless as a piece of information in a diff.
> >
> > _what_ changes in it?
>
> See atta
Angus,
there is still something fishy with the loading.
When I have a doc with a only few formulas, it can loads quickly and
I can see that the preview renderings are done asynchrounously. Fine.
However, when I start with a larger doc, loading stalls until all preview
stuff is done.
I had the
On Fri, Jun 28, 2002 at 09:28:10AM +0100, Angus Leeming wrote:
> > The '.H' looks a bit strange. I would have expected a '.h'
>
> No, that's correct.
I see.
Andre'
--
Those who desire to give up Freedom in order to gain Security,
will not have, nor do they deserve, either one. (T. Jefferson)
> No, that's correct.
>
> fdesign -convert generates .c and .h files.
> We run these through sed scripts to generate .C and .H files.
> Finally we compare the .H file to the original .h file and if it's
> different we overright said .h file (saves unnecessary recompilations).
>
> I've just realise
On Friday 28 June 2002 6:49 am, Andre Poenitz wrote:
> On Thu, Jun 27, 2002 at 01:38:35PM -0700, Kayvan A. Sylvan wrote:
> > #==
> > # Modify the .h file for use by LyX
> > HIN=${DIRNAME}/${BASENAME}.h
> > + HIN=./form_aboutlyx.h
> > HPATCH=${DIRNAME}/${BASENAME}.h.
On Thursday 27 June 2002 9:38 pm, Kayvan A. Sylvan wrote:
> Here is a log of "sh -xv fdfix.sh form_aboutlyx.fd"
Very neat and clever. I didn't know about "-xv"
So the question is, why aren't you creating form_aboutlyx.H?
Do you need to modify the script, so:
# Note that the REAL magic is to b
21 matches
Mail list logo
