Michael Schmitt wrote:
> enclosed please find some new (?) bug reports.
> Is QPreambleDialogBase.ui old cruft that should be removed???
Yes.
--
Angus
Hello,
enclosed please find some new (?) bug reports. I will enter them into
bugzilla at the weekend.
Regards, Michael
1. New document
2. Activate change tracking
3. Add empty table
4. Open table dialog
On Jul 21, 2005, at 12:18 PM, Angus Leeming wrote:
Bennett Helm wrote:
Ok ... with a few tweaks to the ./configure line of the script (Mac OS
X needs "--enable-optimization=-Os --disable-concept-checks" in
addition to the other stuff), it works.
The trouble is, it produces a .tar.gz file that,
Jean-Marc Lasgouttes <[EMAIL PROTECTED]> writes:
>> "Lars" == Lars Gullik Bjønnes <[EMAIL PROTECTED]> writes:
>
| Lars> I think a log of dependencies on [] has crept, so changing to
| ^^^lot?
right.
| Lars> std::list might be simple... but with a semi huge patch.
>
| So not f
Bennett Helm wrote:
> Ok ... with a few tweaks to the ./configure line of the script (Mac OS
> X needs "--enable-optimization=-Os --disable-concept-checks" in
> addition to the other stuff), it works.
>
> The trouble is, it produces a .tar.gz file that, at 3.3 MB, is too big
> to upload to the wik
On Thu, Jul 21, 2005 at 06:00:21PM +0200, Jean-Marc Lasgouttes wrote:
> I can somewhat make it work by typing "a=5" selecting and then C-m,
> but it seems to create two math insets. And I thought we were supposed
> to add the $ $ around the string.
Ah, the $ $ was causing
Tokens: [$,3] <#>
VAN[=
On Thu, Jul 21, 2005 at 05:56:57PM +0200, Jean-Marc Lasgouttes wrote:
> John> There's a nasty comment saying I'm not allowed to add data
> John> members
>
> Indeed :) What kind of support do you want to obtain anyway? You do
> not plan to support change tracking _inside_ insets, I guess?
Sure I
> "John" == John Levon <[EMAIL PROTECTED]> writes:
John> Is it intentional that I can no longer select latex and press
John> control-M to get mathed to parse it?
I can somewhat make it work by typing "a=5" selecting and then C-m,
but it seems to create two math insets. And I thought we were s
Is it intentional that I can no longer select latex and press control-M
to get mathed to parse it?
regards
john
> "John" == John Levon <[EMAIL PROTECTED]> writes:
John> On Thu, Jul 21, 2005 at 05:31:33PM +0200, Jean-Marc Lasgouttes
John> wrote:
>> Why don't you add the editable() stuff directly to InsetBase? Both
>> math and normal insets derive from that these days.
John> There's a nasty comment sayin
On Thu, Jul 21, 2005 at 05:31:33PM +0200, Jean-Marc Lasgouttes wrote:
> Why don't you add the editable() stuff directly to InsetBase? Both
> math and normal insets derive from that these days.
There's a nasty comment saying I'm not allowed to add data members
regards
john
> "Lars" == Lars Gullik Bjønnes <[EMAIL PROTECTED]> writes:
Lars> I think a log of dependencies on [] has crept, so changing to
^^^lot?
Lars> std::list might be simple... but with a semi huge patch.
So not for 1.4.0, right?
JMarc
> "John" == John Levon <[EMAIL PROTECTED]> writes:
John> This version includes basic support for making math formulae
John> non-editable. Andre, please look it over
Why don't you add the editable() stuff directly to InsetBase? Both
math and normal insets derive from that these days.
JMarc
Angus Leeming <[EMAIL PROTECTED]> writes:
| Lars Gullik Bjønnes wrote:
>> I cannot remember the speed arguemtn, but more the arguemtn for [] and
>> the use of a fixed offset.
>
| Talking of speed, your fingers are too fast for your brain ;-)
unsyncronized fingers
| Hmmm. Does the code now assume
... Bringing this over from the user's list ...
On Jul 21, 2005, at 10:04 AM, Angus Leeming wrote:
Bennett, this is just the executable. You'll need a bunch of support
stuff
too (layout files, lyx2lyx to go from format 243 back to 221).
If you grab the make_tex2lyx_dist.sh script from
http://
Hmm, it stil claims to be the interface between lyx and mathed, but this
isn't true. Andre, it looks like I have to add an erased_ bool to every
math inset and populate it? Or can I get the outer inset somehow and
make an erased() find that one?
john
This version includes basic support for making math formulae
non-editable. Andre, please look it over
thanks
john
Index: lyxfunc.C
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/lyxfunc.C,v
retrieving revision 1.663
diff -u -a -
Jean-Marc Lasgouttes wrote:
> Angus> Anyway, following the code path, stripping out the noise...
> [...]
> Thanks for the nice analysis. I really have a hard time following the
> paths in this code.
I think that the trick is to always start in controllers/Dialog.C.
ControlSpellchecker and QSpellc
> "Jose'" == Jose' Matos <[EMAIL PROTECTED]> writes:
Jose'> On Thursday 21 July 2005 14:32, Jean-Marc Lasgouttes wrote:
>> It is easier to follow somebody who knows all the corridors of a
>> big house that to find your way out later.
Jose'> Funny, I thought that you meant to say "instead of
On Thursday 21 July 2005 14:32, Jean-Marc Lasgouttes wrote:
> It is easier to follow somebody who knows all the corridors of a big
> house that to find your way out later.
Funny, I thought that you meant to say "instead of going up and down
through all those stairs". ;-)
> JMarc
--
José Abíl
> "Angus" == Angus Leeming <[EMAIL PROTECTED]> writes:
Angus> Jean-Marc Lasgouttes wrote: I'll make some time to investigate.
Angus> Will probably take me a couple of days. OK?
>> No problem really. I am so lost, that any help anytime is fine.
Angus> Note that I don't know this code at all b
Lars Gullik Bjønnes wrote:
> I cannot remember the speed arguemtn, but more the arguemtn for [] and
> the use of a fixed offset.
Talking of speed, your fingers are too fast for your brain ;-)
Hmmm. Does the code now assume implicitly that the data is stored
contiguously? Ie, would you be making a
Angus Leeming <[EMAIL PROTECTED]> writes:
| Lars Gullik Bjønnes wrote:
>> ... Actually... since we are not _really_ working on the front it
>> won't be much faster. std::list otoh would be fast.
>
| In this case, maybe. But André moved to a vector to store the paragraphs
| because his profiling te
Lars Gullik Bjønnes wrote:
> ... Actually... since we are not _really_ working on the front it
> won't be much faster. std::list otoh would be fast.
In this case, maybe. But André moved to a vector to store the paragraphs
because his profiling tests showed it to be quicker in most common
operation
On Thu, Jul 21, 2005 at 10:31:05AM +0200, Lars Gullik Bj?nnes wrote:
> | So? How does either point justify you unilaterally breaking the freeze that
> | you yourself imposed?
>
> Because this has been decided a long time ago. (And no, I don't tell
> you of every decision I make)
So perhaps you s
Jean-Marc Lasgouttes wrote:
> Angus> I'll make some time to investigate. Will probably take me a
> Angus> couple of days. OK?
>
> No problem really. I am so lost, that any help anytime is fine.
Note that I don't know this code at all but it has been broken from day
one. Somehow the Spellchecker d
Jean-Marc Lasgouttes <[EMAIL PROTECTED]> writes:
>> "Lars" == Lars Gullik Bjønnes <[EMAIL PROTECTED]> writes:
>
| Lars> When using change tracking whole paragraphs will afaik very
| Lars> selcom be deleted.
>
| Lars> I'll look at bit on this one.
>
| I leave it to you, then.
One simple soluti
> "Angus" == Angus Leeming <[EMAIL PROTECTED]> writes:
Angus> I'll make some time to investigate. Will probably take me a
Angus> couple of days. OK?
No problem really. I am so lost, that any help anytime is fine.
JMarc
Jean-Marc Lasgouttes wrote:
So the real problem is at step 3/. How can a dialog which is not shown
on screen and which has been hidden according to MVC machinery still
be on Qt list of dialogs?
I'll make some time to investigate. Will probably take me a couple of days. OK?
Angus
> "Angus" == Angus Leeming <[EMAIL PROTECTED]> writes:
Angus> It looks to me like this signal is entirely within the Qt event
Angus> loop machinery, no? No wonder it's driving you nuts.
Indeed :)
>> #0 lyx::frontend::ControlSpellchecker::showSummary (this=0x88337a8)
>> at src/frontends/contr
http://bugzilla.lyx.org/show_bug.cgi?id=1938
Patch attached.
However, there are still some problems:
1. if you change a fixed width cell with paragraph breaks to a
non-fixed-width cell, the paragraphs are not remerged.
2. In the case of newlines, LyX even asserts if the cursor is in lastpos
> "Bennett" == Bennett Helm <[EMAIL PROTECTED]> writes:
>> Would it be a good idea to decide that, when the \server_pipe value
>> is set to a relative path, then it should be understood as relative
>> to user_dir?
Bennett> I'm not sure of the issues here: why would I be likely to put
Bennett>
> "Lars" == Lars Gullik Bjønnes <[EMAIL PROTECTED]> writes:
Lars> When using change tracking whole paragraphs will afaik very
Lars> selcom be deleted.
Lars> I'll look at bit on this one.
I leave it to you, then.
JMarc
Jean-Marc Lasgouttes <[EMAIL PROTECTED]> writes:
>> "Lars" == Lars Gullik Bjønnes <[EMAIL PROTECTED]> writes:
>
| Lars> At the very least I'd like an explict check on change tracking.
>
| Yes, this is what I alluded to in my message.
>
| Also, the deletion in case of change tracking would be f
> "Lars" == Lars Gullik Bjønnes <[EMAIL PROTECTED]> writes:
Lars> At the very least I'd like an explict check on change tracking.
Yes, this is what I alluded to in my message.
Also, the deletion in case of change tracking would be faster if we
loop was reversed (less data copying), but I do
> "John" == John Levon <[EMAIL PROTECTED]> writes:
John> On Thu, Jul 21, 2005 at 12:40:42AM +0200, Jean-Marc Lasgouttes
John> wrote:
>> What about this patch, then? It is like the old code, except that
>> it deletes stuff by chunks. It is still slower than Lars' solution,
>> so I guess we coul
Angus Leeming <[EMAIL PROTECTED]> writes:
| So? How does either point justify you unilaterally breaking the freeze that
| you yourself imposed?
Because this has been decided a long time ago. (And no, I don't tell
you of every decision I make)
| You're planning to make a *huge* change to our code
Lars Gullik Bjønnes wrote:
> | A freeze is a freeze, Lars. Live by the same rules you expect others to
> | adhere to.
>
> I have had this planned for a long time, and at least I tell you up
> front.
So? How does either point justify you unilaterally breaking the freeze that
you yourself imposed?
Jean-Marc Lasgouttes <[EMAIL PROTECTED]> writes:
| John Levon wrote:
>> Sure. When we do erase(0, size()), that uses the paragraph_pimpl erase()
>> function, and that updates the change table. For old text, it does not
>> delete the chars.
>> So, we only remove the paragraph is its size is now 0 -
Angus Leeming <[EMAIL PROTECTED]> writes:
| Lars Gullik Bjønnes wrote:
>> This is just a heads up. Boost is planning release of 1.33.0 at Aug 1.
>> I am going to upgrade 1.4cvs to use it.
>
| Why? Will it fix any problems? Err, No. Do we need any of its new features?
| Err, No again.
>
| How does
40 matches
Mail list logo
