Lars Gullik Bjønnes wrote:
Jean-Marc Lasgouttes <[EMAIL PROTECTED]> writes:
"Lars" == Lars Gullik Bjønnes <[EMAIL PROTECTED]> writes:
| Lars> | May I kindly ask for the result of our discussion? | I
| Lars> answered all of your questions and now I cannot see how to
| Lars> proceed.
| Lars> For 1.
Martin Vermeer wrote:
> On Mon, Mar 21, 2005 at 08:52:30PM +0100, Andre Poenitz wrote:
>
>> So this indicates a bug in either doDispatch or getStatus in this area.
>> The most nested inset should be asked first and thus gets a chance to
>> give a definite answer first. If that does not happen auto
On Mon, Mar 21, 2005 at 08:52:30PM +0100, Andre Poenitz wrote:
> On Mon, Mar 21, 2005 at 05:56:42PM +0200, Martin Vermeer wrote:
>
> > From my testing it appears that for a true grid inset (which may be
> > nested inside an "untrue" one), control never even passes the
> > MathHullInset getStatus
On Mon, Mar 21, 2005 at 08:43:46PM +0100, Andre Poenitz wrote:
> On Mon, Mar 21, 2005 at 01:54:34PM +0200, Martin Vermeer wrote:
> > The problem is not that a MathGridInset would have only one row/col. The
> > problem is that MathGridInset's getStatus gets called even for insets
> > that in reality
Andre Poenitz wrote:
> Anyway. Angus, you may add my name to that list of yours (not the one of
> people owing you lots of beer, but that other one with poeple granting
> permission to licence their contributions to LyX under the Gnu General
> Public Licence, version 2 or later).
Great! Done so.
Andre Poenitz <[EMAIL PROTECTED]> writes:
| On Mon, Mar 21, 2005 at 05:17:53PM +0100, Lars Gullik Bjønnes wrote:
>> Andre Poenitz <[EMAIL PROTECTED]> writes:
>>
>> | It looks like I can't really resist here, however I am not happy at all
>> | about this "or later" thing. I am interested to keep m
On Mon, Mar 21, 2005 at 05:17:53PM +0100, Lars Gullik Bjønnes wrote:
> Andre Poenitz <[EMAIL PROTECTED]> writes:
>
> | It looks like I can't really resist here, however I am not happy at all
> | about this "or later" thing. I am interested to keep my code (well, at
> | least the part related to Ly
On Mon, Mar 21, 2005 at 09:16:24AM +0100, Georg Baum wrote:
> Martin Vermeer wrote:
>
> > This is a real bug, and probably easy to fix. In math_gridinset's
> > getStatus, there should be tests for nrows() <= 0 and ncols() <= 0 for
> > the cases add-(h/v)line-(above|below), just like there are for
On Mon, Mar 21, 2005 at 05:56:42PM +0200, Martin Vermeer wrote:
> I believe that asGridInset will not return the right answer: it will
> always return true if the inset is a MathGridInset in the C++ sense,
> irrespective whether it is in the role of, e.g., an eqnarray.
Indeed.
> From my testing
On Mon, Mar 21, 2005 at 01:54:34PM +0200, Martin Vermeer wrote:
> The problem is not that a MathGridInset would have only one row/col. The
> problem is that MathGridInset's getStatus gets called even for insets
> that in reality are _not_ array-type insets.
Every math inset is wrapped in a MathHul
On Mon, Mar 21, 2005 at 03:36:15PM +0100, Georg Baum wrote:
> There is one (btw for every math inset):
Not for every one, only for these that need extra tweaking in some form
or the other. Unfortunately there are quite a few of them. This is sort
of "poor man's RTTI".
> virtual MathGridInset
So the test file actually works. I am surprised that the runtime test
yields 'no'.
On another front, what happens when you edit the lyxrc.defaults file
to contain
\tex_allows_spaces true
?
Does typesetting of docs in directories-with-spaces-in-their-name
work?
Yes. (This is in lyx-136.)
In lyx-140,
John Spray wrote
> > Yes, I think so. This is also what the qt graphics dialog does.
>
> Hmm, here the qt graphics dialog initially displays 0%.
Hmm. you're right. This has changed since 1.3. 100% seems more sensible,
though (even if the result is the same).
Jürgen
On Mon, 2005-03-21 at 19:37 +0100, Juergen Spitzmueller wrote:
> John Spray wrote:
> > So would it be correct to say that if (scale == string() && width ==
> > LyXLength() && height == LyXLength()) then I should present the user
> > with UI representing 100% scaling?
>
> Yes, I think so. This is a
Juergen Spitzmueller <[EMAIL PROTECTED]> writes:
| Lars Gullik Bjønnes wrote:
>> Is it not possible to get rid of the '13'?
>
| Yes it is. See me updated patch (reattached to this mail).
Ok, thanks,
--
Lgb
John Spray wrote:
> So would it be correct to say that if (scale == string() && width ==
> LyXLength() && height == LyXLength()) then I should present the user
> with UI representing 100% scaling?
Yes, I think so. This is also what the qt graphics dialog does.
Jürgen
On Mon, 2005-03-21 at 19:26 +0100, Juergen Spitzmueller wrote:
> John Spray wrote:
> > Would it not be more sensible to have scale = "100"? or does this get
> > set elsewhere?
>
> The scale param is not written in this case. The backend outputs it in
> original size (i.e. 100%) then, so it's not
Lars Gullik Bjønnes wrote:
> Is it not possible to get rid of the '13'?
Yes it is. See me updated patch (reattached to this mail).
Jürgen
P.S.: I found out that 1/3 of the font maxAscent gives the best result for
different font sizes, but also to the default size.
Index: rowpainter.C
==
John Spray wrote:
> Would it not be more sensible to have scale = "100"? or does this get
> set elsewhere?
The scale param is not written in this case. The backend outputs it in
original size (i.e. 100%) then, so it's not necessary to set it explicitely
to 100%.
Regards,
Jürgen
Martin Vermeer wrote:
> The problem is that MathGridInset is "abused" by MathHullInset to draw
> grid-like display formulas of various kinds, which can do most of the
> things a "true" gridinset can (like swapping two rows or columns) but
> not everything. See the handling of rowChangeOK and colCh
Jean-Marc Lasgouttes <[EMAIL PROTECTED]> writes:
>> "Lars" == Lars Gullik Bjønnes <[EMAIL PROTECTED]> writes:
>
| Lars> | May I kindly ask for the result of our discussion? | I
| Lars> answered all of your questions and now I cannot see how to
| Lars> proceed.
>
| Lars> For 1.3. J-M is calling
Juergen Spitzmueller <[EMAIL PROTECTED]> writes:
| The striking-out line for erased text is misplaced in the first row of the
| main text, since this also includes the top margin, which is not honoured by
| the strikeout calculation. The attached patch fixes it.
>
| The offset is still there for
Andre Poenitz <[EMAIL PROTECTED]> writes:
| It looks like I can't really resist here, however I am not happy at all
| about this "or later" thing. I am interested to keep my code (well, at
| least the part related to LyX) free and open in the spirit of the
| current GPL.
Regardless of what happes
Jean-Marc Lasgouttes wrote:
> Angus, there are also some recent announcements in www-user/announce.
> You may want to link to them...
Great idea.
--
Angus
On Mon, 2005-03-21 at 16:36, Georg Baum wrote:
> Martin Vermeer wrote:
>
> > The problem is not that a MathGridInset would have only one row/col. The
> > problem is that MathGridInset's getStatus gets called even for insets
> > that in reality are _not_ array-type insets.
>
> Do you know why? It
Hi,
In InsetGraphicsParams::init, we have:
> scale = string(); // output scaling in percentage
> width = LyXLength();
> height = LyXLength();
Would it not be more sensible to have scale = "100"? or does this get
set elsewhere?
John
> "Angus" == Angus Leeming <[EMAIL PROTECTED]> writes:
Angus> José, do you have any patches to make old versions of LyX
Angus> compile against XForms 1.0?
Angus> If so, can you post them to me and I'll update
Angus> http://www.devel.lyx.org/~leeming/www-user/archaeology/lyx_timeline.php
Angu
> "Jose'" == Jose' Matos <[EMAIL PROTECTED]> writes:
Jose'> I will do it this week, and then I will send you the
Jose'> resulting bits.
It would be nice to have statistics on the growth of LyX among
versions (tar size, binary size, compile time, whatever...), too.
Especially since you have
> "Lars" == Lars Gullik Bjønnes <[EMAIL PROTECTED]> writes:
Lars> | May I kindly ask for the result of our discussion? | I
Lars> answered all of your questions and now I cannot see how to
Lars> proceed.
Lars> For 1.3. J-M is calling the shots.
I propose to commit the QuoteName changes to 1.3
On Monday 21 March 2005 13:28, Angus Leeming wrote:
> José,
>
> do you have any patches to make old versions of LyX compile against
> XForms 1.0?
Sort of... :-)
I have the original directories as well as the new, a recursive diff
should do it.
> If so, can you post them to me and I'll update
> "Angus" == Angus Leeming <[EMAIL PROTECTED]> writes:
Angus> Jean-Marc Lasgouttes wrote:
>> Angus, considering that we know how broken the preview code is in
>> this case, can I still apply my fix, which is unrelated to the
>> preview stuff?
Angus> Certainly. It fixes a clear problem.
Thank
> "Eitan" == Eitan <[EMAIL PROTECTED]> writes:
Eitan> --- Jean-Marc Lasgouttes <[EMAIL PROTECTED]> wrote:
>> Another possibility would be to just output the content of the
>> field without any label: + if (!annote.empty()) + result << "\n\n"
>> << annote;
>>
Eitan> Good enough for me...
I
Martin Vermeer wrote:
> The problem is not that a MathGridInset would have only one row/col. The
> problem is that MathGridInset's getStatus gets called even for insets
> that in reality are _not_ array-type insets.
Do you know why? It looks like my picture of this whole getStatus/dispatch
mechan
> "Martin" == Martin Vermeer <[EMAIL PROTECTED]> writes:
>> While you are at it: Shouldn't the other lfuns that exist in
>> do_dispatch() be handled in getStatus(), too?.
Martin> Only those that are connected to a menu entry or such, that
Martin> should be greyed out. I think.
No, all functi
José,
do you have any patches to make old versions of LyX compile against XForms
1.0?
If so, can you post them to me and I'll update
http://www.devel.lyx.org/~leeming/www-user/archaeology/lyx_timeline.php
--
Angus
Angus Leeming wrote:
> My take on José's LyX archaeology page
> http://www.devel.lyx.org/~leeming/www-user/archaeology/lyx_timeline.php
> isn't validating as XHTML 1.0 ATM because it contains multiple blocks:
>
>
> contents
>
>
> W3C's validator complains with lots of:
> Line 202, column 10:
My take on José's LyX archaeology page
http://www.devel.lyx.org/~leeming/www-user/archaeology/lyx_timeline.php
isn't validating as XHTML 1.0 ATM because it contains multiple blocks:
contents
W3C's validator complains with lots of:
Line 202, column 10: ID "a" first defined here
Line 232, col
On Mon, 2005-03-21 at 10:16, Georg Baum wrote:
> Martin Vermeer wrote:
>
> > This is a real bug, and probably easy to fix. In math_gridinset's
> > getStatus, there should be tests for nrows() <= 0 and ncols() <= 0 for
> > the cases add-(h/v)line-(above|below), just like there are for the
> > other
> "Angus" == Angus Leeming <[EMAIL PROTECTED]> writes:
Angus> The point being that the core would explicitly tell individual
Angus> (named) dialogs when it was illegal to Apply their contents.
Angus> Your suggesting (if I understand correctly) that each and every
Angus> open dialog should rec
Jean-Marc Lasgouttes wrote:
> Angus> Conceptually, doing so is easy. We just need to tell the Dialog
> Angus> that the buffer has become readOnly, no? That way, the OK and
> Angus> Apply buttons will be disabled.
>
> Angus> Here's an unchecked implementation of the idea. Frontend only.
> Angus> No
Jean-Marc Lasgouttes wrote:
> Angus, considering that we know how broken the preview code is in this
> case, can I still apply my fix, which is unrelated to the preview
> stuff?
Certainly. It fixes a clear problem.
--
Angus
> "Georg" == Georg Baum <[EMAIL PROTECTED]> writes:
Georg> This patch fixes the appendix problems Jean-Pierre reported. A
Georg> test file is also attached. OK to apply?
This looks good to me.
JMarc
Juergen Spitzmueller wrote:
> Andre Poenitz wrote:
>> So a good check is always math-in-math-in-note-in-table-in-table. ÂIf
>> something works in this scenario it is fairly safe to assume it works
>> everywhere.
>
> This works too.
Thanks Juergen, Martin and Andre. I'll commit this tonight (gott
> "Angus" == Angus Leeming <[EMAIL PROTECTED]> writes:
Angus> Jean-Marc Lasgouttes wrote:
>> This patch fixes the following bug:
>>
>> 1/ create an InsetInclude with type 'include'
>>
>> 2/ place the cursor after the inset and hit
>>
>> 3/ see how the button turned into a small white squar
> "Juergen" == Juergen Spitzmueller <[EMAIL PROTECTED]> writes:
Juergen> Yes, of course not (btw this one is useless on KDE systems,
Juergen> because KWin catches C-Tab).
Yes.
Juergen> Revised patches attached. OK now?
Yes.
JMarc
> "Angus" == Angus Leeming <[EMAIL PROTECTED]> writes:
Angus> Conceptually, doing so is easy. We just need to tell the Dialog
Angus> that the buffer has become readOnly, no? That way, the OK and
Angus> Apply buttons will be disabled.
Angus> Here's an unchecked implementation of the idea. Fron
John Spray wrote:
> Attached, update to reflect progress on gtk frontend.
Comitted. You should see the change in ½hour or so.
--
Angus
Alfredo Braunstein wrote:
> Shows as a disalignement for instance in selections of nested tables.
I've commited this.
Alfredo
Georg Baum wrote:
> Angus Leeming wrote:
>
>> Here's an unchecked implementation of the idea. Frontend only. No hooks
>> from the core. However, I'm sure that you'll be able to turn this into
>> working code ;-)
>
> I'll try to do that later.
>
> Thanks!
My pleasure. One refinement. It probabl
Johnathan Burchill wrote:
> Juergen Spitzmueller wrote:
> > Sun, 20 Mar 2005 08:26:53 -0800
> > The striking-out line for erased text is misplaced in the first row of
> > the main text, since this also includes the top margin, which is not
> > honoured by the strikeout calculation. The attached pat
Attached, update to reflect progress on gtk frontend.
John
Index: guii.php3
===
RCS file: /cvs/lyx/www-devel/guii.php3,v
retrieving revision 1.72
diff -u -p -r1.72 guii.php3
--- guii.php3 2004/11/15 15:35:27 1.72
+++ guii.php3 2005/0
Juergen Spitzmueller wrote:
> Sun, 20 Mar 2005 08:26:53 -0800
> The striking-out line for erased text is misplaced in the first row of the
> main text, since this also includes the top margin, which is not honoured by
> the strikeout calculation. The attached patch fixes it.
This works for the f
Angus Leeming wrote:
> Here's an unchecked implementation of the idea. Frontend only. No hooks
> from the core. However, I'm sure that you'll be able to turn this into
> working code ;-)
I'll try to do that later.
Thanks!
Georg
Martin Vermeer wrote:
> This is a real bug, and probably easy to fix. In math_gridinset's
> getStatus, there should be tests for nrows() <= 0 and ncols() <= 0 for
> the cases add-(h/v)line-(above|below), just like there are for the
> others.
I thought that every MathGridInset has at least one row
54 matches
Mail list logo
