Re: New Graphics dialog layout

2002-08-18 Thread Dekel Tsur

On Thu, Aug 15, 2002 at 03:46:54PM -0300, Garst R. Reese wrote:
   when I write a article/book then I do not know for some images
   what kind of output (b/w - gray - color) maybe the best.
   No problem with current lyx, I switch to different views
   and decide online to what I will convert the image later.
   or which one I have to create new.
  
  Why can't you use xdvi for this ? The in lyx display isn't high fidelity
  enough anyway.
 It takes more time. I have to go through the file, see which images I
 want to check, set parameters, display dvi, reset the parameters,
 display dvi. yuchh.

We can add an View/Edit button to the graphics dialog which can
run your favorite external tool on the image.



Re: [PrePatch] New Graphics dialog for Xforms; Help needed!

2002-08-18 Thread Dekel Tsur

On Sat, Aug 17, 2002 at 02:23:09AM +0900, Rob Lahaye wrote:
 The following is an excerpt from the LyX output file that has two graphics insets, 
the
 first simply with the graphics default values, the second with everthing non-default.
 
 --
 \layout Standard
 
 
 \begin_inset Graphics FormatVersion 1

Shouldn't the format be  1 ?
However, I don't see why we should store the format # in the inset.

   filename /home/lahaye/myGraph.xpm
   lyxscale 100

Since this is the default, it shouldn't be written to the file. 



Re: [PrePatch] New Graphics dialog for Xforms; Help needed!

2002-08-18 Thread John Levon

On Sun, Aug 18, 2002 at 04:52:47PM +0300, Dekel Tsur wrote:

 Shouldn't the format be  1 ?
 However, I don't see why we should store the format # in the inset.

I don't see why the output format needs changing. Instead we just
deprecate the bits we don't use any more.

regards
john

-- 
Someone turn off the good idea tap; we're drowning here!
- Rusty Russell



Re: Tooltips issues: Graphics, Preferences, Tooltips.[Ch] etc.

2002-08-18 Thread John Levon

On Sat, Aug 17, 2002 at 05:34:56PM +0200, Juergen Spitzmueller wrote:

 John Levon wrote:
  For tooltips it's not an option because the dialogs will differ, and
  they are encoded in the .ui file, and I don't want the Qt tooltips to be
  the George Bernard Shaw essays that xforms has. It would be possible for
 
 Thank God that it is not James Joyce ;-)

That is reserved for certain aspects of the core code. Except without
all the hidden jokes.

  what's this thingies like the quick description of how to enter index
  entries etc.
 
 Of course I mean what's this?, which is an equivalent to our long tooltips 
 (as I understand them). So the idea was to share the text strings for xforms' 
 tooltips, qt's what's this? and gtk's whatever, if that is technically 
 possible.

It's technically possible yes. It remains to be seen how many cases this
will be useful for. Since it requires no Qt changes as such, I don't
mind an experiment to see what could be shared.

regards
john
-- 
Someone turn off the good idea tap; we're drowning here!
- Rusty Russell



Re: [Bug 394] floats don't have a default placement

2002-08-18 Thread Dekel Tsur

On Thu, Aug 15, 2002 at 10:34:38PM -, [EMAIL PROTECTED] wrote:
 http://bugzilla.lyx.org/show_bug.cgi?id=394
 
 [EMAIL PROTECTED] changed:
 
What|Removed |Added
 
Target Milestone|1.2.1   |1.2.2
 
 
 
 --- Additional Comments From [EMAIL PROTECTED]  2002-08-16 10:34 ---
 Bumping 1.2.1 targetted bugs to 1.2.2

Why isn't it fixed in 1.2.1 ? 



Re: [Patch] Short title inset: Success!

2002-08-18 Thread Dekel Tsur

On Fri, Aug 16, 2002 at 05:09:41PM +0100, Angus Leeming wrote:

 + if (style-latexparam() == void) {
 + InsetShortTitle * it = shortTitleInset(*this);
 + if (it != 0)
 + it-latexOptionals(buf, os, false, false);
 + }

Wouldn't it be more general to call latexOptionals on every InsetShortTitle
inset in the paragraph ?
This will allow support for commands that can have more than one optional
argument.

Also, I think it would be better to call the inset InsetOptiobal.



Re: [PrePatch] New Graphics dialog for Xforms; Help needed!

2002-08-18 Thread Dekel Tsur

On Sun, Aug 18, 2002 at 04:03:55PM +0100, John Levon wrote:
 On Sun, Aug 18, 2002 at 04:52:47PM +0300, Dekel Tsur wrote:
 
  Shouldn't the format be  1 ?
  However, I don't see why we should store the format # in the inset.
 
 I don't see why the output format needs changing. Instead we just
 deprecate the bits we don't use any more.

Because the format was changed ?



Re: [PrePatch] New Graphics dialog for Xforms; Help needed!

2002-08-18 Thread John Levon

On Sun, Aug 18, 2002 at 06:41:03PM +0300, Dekel Tsur wrote:

 On Sun, Aug 18, 2002 at 04:03:55PM +0100, John Levon wrote:
  On Sun, Aug 18, 2002 at 04:52:47PM +0300, Dekel Tsur wrote:
  
   Shouldn't the format be  1 ?
   However, I don't see why we should store the format # in the inset.
  
  I don't see why the output format needs changing. Instead we just
  deprecate the bits we don't use any more.
 
 Because the format was changed ?

That's what I mean - why ? I admit I am too lazy to look at the patch
but I thought Rob was just removing stuff.

regards
john

-- 
Someone turn off the good idea tap; we're drowning here!
- Rusty Russell



Re: [PrePatch] New Graphics dialog for Xforms; Help needed!

2002-08-18 Thread Rob Lahaye

John Levon wrote:
 On Sun, Aug 18, 2002 at 06:41:03PM +0300, Dekel Tsur wrote:
 
 
On Sun, Aug 18, 2002 at 04:03:55PM +0100, John Levon wrote:

On Sun, Aug 18, 2002 at 04:52:47PM +0300, Dekel Tsur wrote:


Shouldn't the format be  1 ?
However, I don't see why we should store the format # in the inset.

I don't see why the output format needs changing. Instead we just
deprecate the bits we don't use any more.

Because the format was changed ?
 
 
 That's what I mean - why ? I admit I am too lazy to look at the patch
 but I thought Rob was just removing stuff.

Patch to the xforms graphics dialog is coming soon. I'm almost done.
John, I patched two Qt files, just to keep Qt compiling here, but the Qt graphics
dialog needs an update, that removes the redundant buttons and intput fields.
I would like to leave that up to you.

Indeed I'm removing stuff and reinterpreting the left-over stuff.
Backwards compatibility may be broken in some odd cases.

But that Format Id is not in 1.2.x, was it? If am right, there has not been yet
a release with the Format Id in the lyx file. So we still can modify the format in
1.3.0cvs, without changing the Format Id number.
After the release of 1.3.0, a format change should also imply push up of the Format Id.

Good point why it has to be in each graphics inset! Couldn't it be moved
to the header of the lyx file, to have this Format Id only once in the lyx file?

Regards,
Rob.




Re: [PrePatch] New Graphics dialog for Xforms; Help needed!

2002-08-18 Thread Dekel Tsur

On Sun, Aug 18, 2002 at 04:50:51PM +0100, John Levon wrote:
 
 That's what I mean - why ? I admit I am too lazy to look at the patch
 but I thought Rob was just removing stuff.

Even if he was only removing stuff, it should be  considered as a format
change.
But we already have the change size_kind-size_type.



Re: [PrePatch] New Graphics dialog for Xforms; Help needed!

2002-08-18 Thread John Levon

On Mon, Aug 19, 2002 at 01:18:30AM +0900, Rob Lahaye wrote:

 graphics
 dialog needs an update, that removes the redundant buttons and intput 
 fields.
 I would like to leave that up to you.

Sure.

 Indeed I'm removing stuff and reinterpreting the left-over stuff.
 Backwards compatibility may be broken in some odd cases.

Right.

 Good point why it has to be in each graphics inset! Couldn't it be moved
 to the header of the lyx file, to have this Format Id only once in the lyx 
 file?

We should bump the file minor version. Also I didn't check but do you
patch development/FORMAT with your changes (removed, added and changed)
?

I wonder if it would be useful to indicate whether lyx2lyx does/needs to
handle each entry in FORMAT

regards
john

-- 
Someone turn off the good idea tap; we're drowning here!
- Rusty Russell



Re: lyx-devel src/: BufferView_pimpl.C text.C ChangeLog

2002-08-18 Thread Dekel Tsur

On Sun, Aug 18, 2002 at 07:15:26PM +, [EMAIL PROTECTED] wrote:
 CVSROOT:  /usr/local/lyx/cvsroot
 Module name:  lyx-devel
 Repository:   lyx-devel/src/
 Changes by:   [EMAIL PROTECTED]  02/08/18 19:15:26
 
 Modified files:
   lyx-devel/src/: BufferView_pimpl.C text.C ChangeLog 
 
 Log message:
   RTL fixes

It should also be applied to 1.2.1cvs.
A pactch is attached.



patch.gz
Description: GNU Zip compressed data


Re: LFUNs, insets and the outside world

2002-08-18 Thread Lars Gullik Bjønnes

Angus Leeming [EMAIL PROTECTED] writes:

| On Friday 16 August 2002 9:09 am, Andre Poenitz wrote:
|  On Fri, Aug 16, 2002 at 09:50:53AM +0200, Lars Gullik Bjønnes wrote:
|   | Actually I think this implements _your_ idea of every interaction
|   | between frontend and core should go through an LFUN rather well.
|  
|   This is not how I read it... I read it more that all LFUNS go through
|   dialogs...
| 
|  Ok, than let's wait until Angus is back. I think he wants just to call
|  dispatch() from a dialog's apply() instead of hard-wiring the connection to
|  the inset by a Inset* member. [And I certainly like _this_ idea.]
| 
| That's exactly what I want to do. The frontends should call dispatch() and 
| then its up to the core to decide what to do with the data.
| 
| Lars, I can't see how you've reached your conclusion. It's exactly upside 
| down to what I mean!

the LFUN_OPEN, LFUN_CLOSE ting...

-- 
Lgb



Re: reducing compile times

2002-08-18 Thread Lars Gullik Bjønnes

Angus Leeming [EMAIL PROTECTED] writes:

| On Friday 16 August 2002 2:19 pm, Andre Poenitz wrote:
|  On Fri, Aug 16, 2002 at 01:57:32PM +0100, Angus Leeming wrote:
|   So we would lose lazy construction.
| 
|  What was this good for anyway? To reduce startup time?
| 
| Conceptual elegance? Reducing compile times? Lars knows.

Just back from a wonderfull weekend at the west coast of Norway; what
does Lars know?

-- 
Lgb



[PATCH] New Xforms Graphics Dialog

2002-08-18 Thread Rob Lahaye


Hi,

Attached is the patch that implements the new layout for the graphics dialog
with xforms. Two files in the Qt directory have been modified, only to keep
Qt compiling; the Qt Graphics dialog itself needs an update similar to the
xforms version of the dialog. John, can you give that a shot?

Few remarks:

1) Now a new graphics inset will always open the dialog in the file-tab,
irrespectively of what tab was active before.

2) Only non-default values for the inset keywords are saved to the lyx file.
See development/FORMAT for some more info.
Do we need to push up the format version?

3) What is this strange setting of defaultUnit in FormGraphics:
if (lyxrc.default_papersize  3) defaultUnit = in;
(? what is this magic 3 ?).

4) insetgraphics.C::InsetGraphics::readFigInset()
I commented out the keywords that do not exist anymore, but I
don't know if I have done the right thing here (and in general
in this file). Things seem to work here at my side, though.

5) Please test and check the backward compatibility for reasonable cases.

...and apply.

Cheers,
Rob.



NewGraphicsDialog.diff.gz
Description: application/gzip


fix for 570

2002-08-18 Thread Andre Poenitz


This applies to 1.2.1cvs, I'll have something similar in my 1.3.0 tree
which should get commited sometimes.

Jean-Marc, I think this is safe for 1.2.x.

Andre'


-- 
Those who desire to give up Freedom in order to gain Security,
will not have, nor do they deserve, either one. (T. Jefferson)


Index: mathed/math_hullinset.C
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/mathed/math_hullinset.C,v
retrieving revision 1.22.2.1
diff -u -p -r1.22.2.1 math_hullinset.C
--- mathed/math_hullinset.C 31 May 2002 10:38:48 -  1.22.2.1
+++ mathed/math_hullinset.C 19 Aug 2002 05:56:03 -
 -442,6 +442,8  void MathHullInset::appendRow()
 
 void MathHullInset::delRow(row_type row)
 {
+   if (nrows() = 1)
+   return;
MathGridInset::delRow(row);
nonum_.erase(nonum_.begin() + row);
label_.erase(label_.begin() + row);
 -480,6 +482,8  void MathHullInset::addCol(col_type col)
 
 void MathHullInset::delCol(col_type col)
 {
+   if (ncols() = 1)
+   return;
switch (getType()) {
case LM_OT_ALIGNAT:
case LM_OT_XALIGNAT:



Re: reducing compile times

2002-08-18 Thread Andre Poenitz

On Mon, Aug 19, 2002 at 12:04:47AM +0200, Lars Gullik Bjønnes wrote:
 Angus Leeming [EMAIL PROTECTED] writes:
 
 | On Friday 16 August 2002 2:19 pm, Andre Poenitz wrote:
 |  On Fri, Aug 16, 2002 at 01:57:32PM +0100, Angus Leeming wrote:
 |   So we would lose lazy construction.
 | 
 |  What was this good for anyway? To reduce startup time?
 | 
 | Conceptual elegance? Reducing compile times? Lars knows.
 
 Just back from a wonderfull weekend at the west coast of Norway; what
 does Lars know?

People say Lars should know what lazy dialog construction is good for.
My guess was better startup times, Angus thought of conceptual elegance.

Andre'

-- 
Those who desire to give up Freedom in order to gain Security,
will not have, nor do they deserve, either one. (T. Jefferson)



Re: New Graphics dialog layout

2002-08-18 Thread Dekel Tsur

On Thu, Aug 15, 2002 at 03:46:54PM -0300, Garst R. Reese wrote:
> > > when I write a article/book then I do not know for some images
> > > what kind of output (b/w - gray - color) maybe the best.
> > > No problem with current lyx, I switch to different views
> > > and decide "online" to what I will convert the image later.
> > > or which one I have to create new.
> > 
> > Why can't you use xdvi for this ? The in lyx display isn't high fidelity
> > enough anyway.
> It takes more time. I have to go through the file, see which images I
> want to check, set parameters, display dvi, reset the parameters,
> display dvi. yuchh.

We can add an View/Edit button to the graphics dialog which can
run your favorite external tool on the image.



Re: [PrePatch] New Graphics dialog for Xforms; Help needed!

2002-08-18 Thread Dekel Tsur

On Sat, Aug 17, 2002 at 02:23:09AM +0900, Rob Lahaye wrote:
> The following is an excerpt from the LyX output file that has two graphics insets, 
>the
> first simply with the graphics default values, the second with everthing non-default.
> 
> --
> \layout Standard
> 
> 
> \begin_inset Graphics FormatVersion 1

Shouldn't the format be > 1 ?
However, I don't see why we should store the format # in the inset.

>   filename /home/lahaye/myGraph.xpm
>   lyxscale 100

Since this is the default, it shouldn't be written to the file. 



Re: [PrePatch] New Graphics dialog for Xforms; Help needed!

2002-08-18 Thread John Levon

On Sun, Aug 18, 2002 at 04:52:47PM +0300, Dekel Tsur wrote:

> Shouldn't the format be > 1 ?
> However, I don't see why we should store the format # in the inset.

I don't see why the output format needs changing. Instead we just
deprecate the bits we don't use any more.

regards
john

-- 
"Someone turn off the good idea tap; we're drowning here!"
- Rusty Russell



Re: Tooltips issues: Graphics, Preferences, Tooltips.[Ch] etc.

2002-08-18 Thread John Levon

On Sat, Aug 17, 2002 at 05:34:56PM +0200, Juergen Spitzmueller wrote:

> John Levon wrote:
> > For tooltips it's not an option because the dialogs will differ, and
> > they are encoded in the .ui file, and I don't want the Qt tooltips to be
> > the George Bernard Shaw essays that xforms has. It would be possible for
> 
> Thank God that it is not James Joyce ;-)

That is reserved for certain aspects of the core code. Except without
all the hidden jokes.

> > what's this thingies like the quick description of how to enter index
> > entries etc.
> 
> Of course I mean "what's this?", which is an equivalent to our long "tooltips" 
> (as I understand them). So the idea was to share the text strings for xforms' 
> "tooltips", qt's "what's this?" and gtk's whatever, if that is technically 
> possible.

It's technically possible yes. It remains to be seen how many cases this
will be useful for. Since it requires no Qt changes as such, I don't
mind an experiment to see what could be shared.

regards
john
-- 
"Someone turn off the good idea tap; we're drowning here!"
- Rusty Russell



Re: [Bug 394] floats don't have a default placement

2002-08-18 Thread Dekel Tsur

On Thu, Aug 15, 2002 at 10:34:38PM -, [EMAIL PROTECTED] wrote:
> http://bugzilla.lyx.org/show_bug.cgi?id=394
> 
> [EMAIL PROTECTED] changed:
> 
>What|Removed |Added
> 
>Target Milestone|1.2.1   |1.2.2
> 
> 
> 
> --- Additional Comments From [EMAIL PROTECTED]  2002-08-16 10:34 ---
> Bumping 1.2.1 targetted bugs to 1.2.2

Why isn't it fixed in 1.2.1 ? 



Re: [Patch] Short title inset: Success!

2002-08-18 Thread Dekel Tsur

On Fri, Aug 16, 2002 at 05:09:41PM +0100, Angus Leeming wrote:

> + if (style->latexparam() == "void") {
> + InsetShortTitle * it = shortTitleInset(*this);
> + if (it != 0)
> + it->latexOptionals(buf, os, false, false);
> + }

Wouldn't it be more general to call latexOptionals on every InsetShortTitle
inset in the paragraph ?
This will allow support for commands that can have more than one optional
argument.

Also, I think it would be better to call the inset InsetOptiobal.



Re: [PrePatch] New Graphics dialog for Xforms; Help needed!

2002-08-18 Thread Dekel Tsur

On Sun, Aug 18, 2002 at 04:03:55PM +0100, John Levon wrote:
> On Sun, Aug 18, 2002 at 04:52:47PM +0300, Dekel Tsur wrote:
> 
> > Shouldn't the format be > 1 ?
> > However, I don't see why we should store the format # in the inset.
> 
> I don't see why the output format needs changing. Instead we just
> deprecate the bits we don't use any more.

Because the format was changed ?



Re: [PrePatch] New Graphics dialog for Xforms; Help needed!

2002-08-18 Thread John Levon

On Sun, Aug 18, 2002 at 06:41:03PM +0300, Dekel Tsur wrote:

> On Sun, Aug 18, 2002 at 04:03:55PM +0100, John Levon wrote:
> > On Sun, Aug 18, 2002 at 04:52:47PM +0300, Dekel Tsur wrote:
> > 
> > > Shouldn't the format be > 1 ?
> > > However, I don't see why we should store the format # in the inset.
> > 
> > I don't see why the output format needs changing. Instead we just
> > deprecate the bits we don't use any more.
> 
> Because the format was changed ?

That's what I mean - why ? I admit I am too lazy to look at the patch
but I thought Rob was just removing stuff.

regards
john

-- 
"Someone turn off the good idea tap; we're drowning here!"
- Rusty Russell



Re: [PrePatch] New Graphics dialog for Xforms; Help needed!

2002-08-18 Thread Rob Lahaye

John Levon wrote:
> On Sun, Aug 18, 2002 at 06:41:03PM +0300, Dekel Tsur wrote:
> 
> 
>>On Sun, Aug 18, 2002 at 04:03:55PM +0100, John Levon wrote:
>>
>>>On Sun, Aug 18, 2002 at 04:52:47PM +0300, Dekel Tsur wrote:
>>>
>>>
Shouldn't the format be > 1 ?
However, I don't see why we should store the format # in the inset.
>>>
>>>I don't see why the output format needs changing. Instead we just
>>>deprecate the bits we don't use any more.
>>
>>Because the format was changed ?
> 
> 
> That's what I mean - why ? I admit I am too lazy to look at the patch
> but I thought Rob was just removing stuff.

Patch to the xforms graphics dialog is coming soon. I'm almost done.
John, I patched two Qt files, just to keep Qt compiling here, but the Qt graphics
dialog needs an update, that removes the redundant buttons and intput fields.
I would like to leave that up to you.

Indeed I'm removing stuff and reinterpreting the left-over stuff.
Backwards compatibility may be broken in some odd cases.

But that Format Id is not in 1.2.x, was it? If am right, there has not been yet
a release with the Format Id in the lyx file. So we still can modify the format in
1.3.0cvs, without changing the Format Id number.
After the release of 1.3.0, a format change should also imply push up of the Format Id.

Good point why it has to be in each graphics inset! Couldn't it be moved
to the header of the lyx file, to have this Format Id only once in the lyx file?

Regards,
Rob.




Re: [PrePatch] New Graphics dialog for Xforms; Help needed!

2002-08-18 Thread Dekel Tsur

On Sun, Aug 18, 2002 at 04:50:51PM +0100, John Levon wrote:
> 
> That's what I mean - why ? I admit I am too lazy to look at the patch
> but I thought Rob was just removing stuff.

Even if he was only removing stuff, it should be  considered as a format
change.
But we already have the change size_kind->size_type.



Re: [PrePatch] New Graphics dialog for Xforms; Help needed!

2002-08-18 Thread John Levon

On Mon, Aug 19, 2002 at 01:18:30AM +0900, Rob Lahaye wrote:

> graphics
> dialog needs an update, that removes the redundant buttons and intput 
> fields.
> I would like to leave that up to you.

Sure.

> Indeed I'm removing stuff and reinterpreting the left-over stuff.
> Backwards compatibility may be broken in some odd cases.

Right.

> Good point why it has to be in each graphics inset! Couldn't it be moved
> to the header of the lyx file, to have this Format Id only once in the lyx 
> file?

We should bump the file minor version. Also I didn't check but do you
patch development/FORMAT with your changes (removed, added and changed)
?

I wonder if it would be useful to indicate whether lyx2lyx does/needs to
handle each entry in FORMAT

regards
john

-- 
"Someone turn off the good idea tap; we're drowning here!"
- Rusty Russell



Re: lyx-devel src/: BufferView_pimpl.C text.C ChangeLog

2002-08-18 Thread Dekel Tsur

On Sun, Aug 18, 2002 at 07:15:26PM +, [EMAIL PROTECTED] wrote:
> CVSROOT:  /usr/local/lyx/cvsroot
> Module name:  lyx-devel
> Repository:   lyx-devel/src/
> Changes by:   [EMAIL PROTECTED]  02/08/18 19:15:26
> 
> Modified files:
>   lyx-devel/src/: BufferView_pimpl.C text.C ChangeLog 
> 
> Log message:
>   RTL fixes

It should also be applied to 1.2.1cvs.
A pactch is attached.



patch.gz
Description: GNU Zip compressed data


Re: LFUNs, insets and the outside world

2002-08-18 Thread Lars Gullik Bjønnes

Angus Leeming <[EMAIL PROTECTED]> writes:

| On Friday 16 August 2002 9:09 am, Andre Poenitz wrote:
| > On Fri, Aug 16, 2002 at 09:50:53AM +0200, Lars Gullik Bjønnes wrote:
| > > | Actually I think this implements _your_ idea of "every interaction
| > > | between frontend and core should go through an LFUN" rather well.
| > >
| > > This is not how I read it... I read it more that all LFUNS go through
| > > dialogs...
| >
| > Ok, than let's wait until Angus is back. I think he wants just to call
| > dispatch() from a dialog's apply() instead of hard-wiring the connection to
| > the inset by a Inset* member. [And I certainly like _this_ idea.]
| 
| That's exactly what I want to do. The frontends should call dispatch() and 
| then its up to the core to decide what to do with the data.
| 
| Lars, I can't see how you've reached your conclusion. It's exactly upside 
| down to what I mean!

the LFUN_OPEN, LFUN_CLOSE ting...

-- 
Lgb



Re: reducing compile times

2002-08-18 Thread Lars Gullik Bjønnes

Angus Leeming <[EMAIL PROTECTED]> writes:

| On Friday 16 August 2002 2:19 pm, Andre Poenitz wrote:
| > On Fri, Aug 16, 2002 at 01:57:32PM +0100, Angus Leeming wrote:
| > > So we would lose lazy construction.
| >
| > What was this good for anyway? To reduce startup time?
| 
| Conceptual elegance? Reducing compile times? Lars knows.

Just back from a wonderfull weekend at the west coast of Norway; what
does Lars know?

-- 
Lgb



[PATCH] New Xforms Graphics Dialog

2002-08-18 Thread Rob Lahaye


Hi,

Attached is the patch that implements the new layout for the graphics dialog
with xforms. Two files in the Qt directory have been modified, only to keep
Qt compiling; the Qt Graphics dialog itself needs an update similar to the
xforms version of the dialog. John, can you give that a shot?

Few remarks:

1) Now a new graphics inset will always open the dialog in the file-tab,
irrespectively of what tab was active before.

2) Only non-default values for the inset keywords are saved to the lyx file.
See development/FORMAT for some more info.
Do we need to push up the format version?

3) What is this strange setting of defaultUnit in FormGraphics:
if (lyxrc.default_papersize < 3) defaultUnit = "in";
(? what is this magic "3" ?).

4) insetgraphics.C::InsetGraphics::readFigInset()
I commented out the keywords that do not exist anymore, but I
don't know if I have done the right thing here (and in general
in this file). Things seem to work here at my side, though.

5) Please test and check the backward compatibility for "reasonable" cases.

...and apply.

Cheers,
Rob.



NewGraphicsDialog.diff.gz
Description: application/gzip


fix for 570

2002-08-18 Thread Andre Poenitz


This applies to 1.2.1cvs, I'll have something similar in my 1.3.0 tree
which should get commited sometimes.

Jean-Marc, I think this is safe for 1.2.x.

Andre'


-- 
Those who desire to give up Freedom in order to gain Security,
will not have, nor do they deserve, either one. (T. Jefferson)


Index: mathed/math_hullinset.C
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/mathed/math_hullinset.C,v
retrieving revision 1.22.2.1
diff -u -p -r1.22.2.1 math_hullinset.C
--- mathed/math_hullinset.C 31 May 2002 10:38:48 -  1.22.2.1
+++ mathed/math_hullinset.C 19 Aug 2002 05:56:03 -
@@ -442,6 +442,8 @@ void MathHullInset::appendRow()
 
 void MathHullInset::delRow(row_type row)
 {
+   if (nrows() <= 1)
+   return;
MathGridInset::delRow(row);
nonum_.erase(nonum_.begin() + row);
label_.erase(label_.begin() + row);
@@ -480,6 +482,8 @@ void MathHullInset::addCol(col_type col)
 
 void MathHullInset::delCol(col_type col)
 {
+   if (ncols() <= 1)
+   return;
switch (getType()) {
case LM_OT_ALIGNAT:
case LM_OT_XALIGNAT:



Re: reducing compile times

2002-08-18 Thread Andre Poenitz

On Mon, Aug 19, 2002 at 12:04:47AM +0200, Lars Gullik Bjønnes wrote:
> Angus Leeming <[EMAIL PROTECTED]> writes:
> 
> | On Friday 16 August 2002 2:19 pm, Andre Poenitz wrote:
> | > On Fri, Aug 16, 2002 at 01:57:32PM +0100, Angus Leeming wrote:
> | > > So we would lose lazy construction.
> | >
> | > What was this good for anyway? To reduce startup time?
> | 
> | Conceptual elegance? Reducing compile times? Lars knows.
> 
> Just back from a wonderfull weekend at the west coast of Norway; what
> does Lars know?

People say Lars should know what lazy dialog construction is good for.
My guess was better startup times, Angus thought of "conceptual elegance".

Andre'

-- 
Those who desire to give up Freedom in order to gain Security,
will not have, nor do they deserve, either one. (T. Jefferson)