Re: [patch] support the lightning Unicode character Ux21AF

[Guenter Milde]

On 2012-09-23, Uwe Stöhr wrote:
 Am 13.09.2012 11:18, schrieb Jürgen Spitzmüller:

 you just found some packages to support this character:
 http://wiki.lyx.org/LyX/MissingUnicodeCharacters?action=diff#diff1309586810

 I tested this and the attached patch works fine. Can this go in or is there
 a reason why you did not commit this yet?

 I never bothered with this character. But as Günter says, package clashes
 might be an issue.

 But we already use marvosym and this package is now again under maintenance.

 We can use wasysym instead for this character but this package is not
 under maintenance and that#s why I did not use it. However, we use
 already both packages for many characters so it should not make a
 difference what to use. So OK to commit?

It makes a difference regarding matching symbols or possible clashes between
these two packages. It would be best to choose whichever package is already
loaded or which does not conflict with other required packages.

The | operator seen in e.g. 

   2171: 0x271d \\Cross  bbding|marvosym# LATIN 
CROSS
   
seems to do something in this line already (it is, unfortunately not
documented in the head of the unicodesymbols file).

So, I suggest using marvosym|wasysym as textpreamble entry.

Günter

Re: [LyX 2.0.x] Fix bug #8349: Cannot compile 2.0.x: unresolved external symbol

[Jean-Marc Lasgouttes]

That's amusing. This is a patch from Stephan that I backported, and it 
looks like Stephan did the backport, which does not help communication.


Le 24/09/2012 16:19, Stephan Witt a écrit :

commit 8169347adafeaece14c8b463add8072c6e9a91a8
Author: Stephan Witt sw...@lyx.org
Date:   Sun Sep 23 15:07:41 2012 +0200

 Fix bug #8349: Cannot compile 2.0.x: unresolved external symbol

 correct Messages class implementation for non-NLS


Then there is this loosy no-op commit. Why do we need this junk?

 commit a602f21dc42f6edd5d8ad4719eb85b207373
 Merge: 2512e1f 0ea1944
 Author: Jean-Marc Lasgouttes lasgout...@lyx.org
 Date:   Mon Sep 24 16:01:29 2012 +0200

  Merge branch '2.0.x' of git.lyx.org:lyx into 2.0.x

And then of course there is my own stupidity, which adds another commit. 
It is not a problem since I intended to port this for 2.0.5, but I 
intended to have feedback from Uwe on the other one.



commit 2512e1f085a6ad044d6d0ff68f75390ed1d34c75
Author: Jean-Marc Lasgouttes lasgout...@lyx.org
Date:   Thu Aug 23 15:08:21 2012 +0200

 Tracking correctly available translations (take 2)

 The previous scheme of loading all possible translations and checking
 whether the work is a bit too much brute force and causes problems
 on Mac OS X (documents loaded with the wrong language).

 Now there is an helper static method in Messages class that checks
 whether a readable .mo file exist for the language. There should be an
 API in gettext for doing that, but alas it is not possible.

 As a consequence the method Language::translated() has been removed,
 along with its cache.


Finally is it expected that the commits show in only one e-mail message? 
Is there a reason to do that instead of two e-mails?


JMarc

Re: LyX 2.0.5

[Jean-Marc Lasgouttes]

Le 23/09/2012 23:15, Jean-Marc Lasgouttes a écrit :

This is not changing strings, but I wonder whether I should revert
824fe175 (keep undo information after save as...), after seeing the
crashes found by pavel in bug 8342. The undo errors the commit intended
to fix are fixed by the patch to 8342 anyway.


This is still pending


It seems that there are some link errors in windows after my patch
to to bug 7741. I have not looked at it yet.
http://www.lyx.org/trac/ticket/8349


Fixed in branch now.



I have a second patch I would like to backport to branch: 7a7b9abf1b3f4
It does not change strings. I am waiting for some feedback from Uwe on
the fix to bug 7741 first, because the two touch the same feature (UI
language selection).


Committed by mistake. Let's see the result :) More seriously, my only 
question is that I have written code like:

+   string const filen = locale_dir + / + code
+   + /LC_MESSAGES/PACKAGE.mo;
+   if (FileName(filen).isReadableFile())
+ ...
and I wonder wther this still works under windows. I think I am dealing 
with internal representation, so I can use explicit / for directories, 
right?


Or is it going to break for windows? The symptom would be simple: bad 
list of available translations for GUI language in prefs.


JMarc

Re: [patch] fix warning on branch

[Richard Heck]

On 09/23/2012 04:54 PM, Jean-Marc Lasgouttes wrote:

Le 23/09/12 00:06, Scott Kostyshak a écrit :

How many users *compile* anything :)


True. And out of those who do, how many look at the warnings? :)


Personally, I still think that removing warnings in branch matters.

Yes, I think it's worth doing, but not if it's the kind of thing that 
has any

chance of causing problems. I think that was the issue.

rh

Re: [LyX 2.0.x] Fix bug #8349: Cannot compile 2.0.x: unresolved external symbol

[Richard Heck]

On 09/24/2012 10:28 AM, Jean-Marc Lasgouttes wrote:


Finally is it expected that the commits show in only one e-mail 
message? Is there a reason to do that instead of two e-mails?


It's mostly a matter of re-writing (or somehow updating) the 
email-sending script to do this. I've not had time.


rh

Re: [LyX master] avoid deprecated functions on Mac - use Qt instead

[Pavel Sanda]

Stephan Witt wrote:
 -bool Messages::available() const
 +bool Messages::available(string const  c)
  {
 + (void)c;
   return false;
  }

Why this (void)c; construct? Pavel

Re: [LyX 2.0.x] Fix bug #8349: Cannot compile 2.0.x: unresolved external symbol

[Vincent van Ravesteijn]

Op 24-9-2012 16:28, Jean-Marc Lasgouttes schreef:
That's amusing. This is a patch from Stephan that I backported, and it 
looks like Stephan did the backport, which does not help communication.


Just get used to a more advanced form of communication. It is Stephan's 
code, which you committed. We just need to fix the e-mail send script if 
you really want to



 commit a602f21dc42f6edd5d8ad4719eb85b207373
 Merge: 2512e1f 0ea1944
 Author: Jean-Marc Lasgouttes lasgout...@lyx.org
 Date:   Mon Sep 24 16:01:29 2012 +0200

  Merge branch '2.0.x' of git.lyx.org:lyx into 2.0.x


This is your own stupidity as well. First fetch and merge before you add 
your new commits to the branch. If you already commit something and the 
push fails, then you rebase onto the remote branch. You should merge or 
pull to avoid this no-op commits.




And then of course there is my own stupidity, which adds another 
commit. It is not a problem since I intended to port this for 2.0.5, 
but I intended to have feedback from Uwe on the other one.


Vincent

Re: [LyX master] avoid deprecated functions on Mac - use Qt instead

[Richard Heck]

On 09/24/2012 01:07 PM, Pavel Sanda wrote:

Stephan Witt wrote:

-bool Messages::available() const
+bool Messages::available(string const  c)
  {
+   (void)c;
return false;
  }

Why this (void)c; construct?


Crushes an unused variable warning.

rh

Re: Re: [LyX master] avoid deprecated functions on Mac - use Qt instead

[Kornel Benko]

Am Montag, 24. September 2012 um 14:10:33, schrieb Richard Heck rgh...@lyx.org
 On 09/24/2012 01:07 PM, Pavel Sanda wrote:
  Stephan Witt wrote:
  -bool Messages::available() const
  +bool Messages::available(string const  c)
{
  +  (void)c;
 return false;
}
  Why this (void)c; construct?
 
 Crushes an unused variable warning.
 
 rh

Why not
bool Messages::available(string const  ) { ... }

Kornel

signature.asc
Description: This is a digitally signed message part.

Re: [LyX master] avoid deprecated functions on Mac - use Qt instead

[Richard Heck]

On 09/24/2012 02:44 PM, Kornel Benko wrote:


Am Montag, 24. September 2012 um 14:10:33, schrieb Richard Heck 
rgh...@lyx.org


 On 09/24/2012 01:07 PM, Pavel Sanda wrote:

  Stephan Witt wrote:

  -bool Messages::available() const

  +bool Messages::available(string const  c)

  {

  + (void)c;

  return false;

  }

  Why this (void)c; construct?

 

 Crushes an unused variable warning.



 rh

Why not

bool Messages::available(string const  ) { ... }

Yes, you could also do it that way. He probably wanted to keep the 
signature the same as in the real routine.


rh

Re: Noweb/Literate programming Copier problem

[Pavel Sanda]

Kayvan Sylvan wrote:
 I'm running LyX 2.1.0 SVN on Fedora 17

Are you aware that for long time we use git and SVN is months old codebase?
Pavel

Re: Re: [LyX master] avoid deprecated functions on Mac - use Qt instead

[Pavel Sanda]

Kornel Benko wrote:
 Why not
   bool Messages::available(string const  ) { ... }

exactly ;) p

Re: [patch] LyX bash completion

[Pavel Sanda]

Scott Kostyshak wrote:
 Are these OK to go in to branch and master?

Looks fine if tested ;) Branch is Richard's call.
Pavel

Re: [LyX master] Fix UI for decimal table alignment (bug #8285)

[Pavel Sanda]

Uwe Stöhr wrote:
 I did this now:
 http://www.lyx.org/trac/changeset/237cb51fb15fdaa4951f70e393eea0fdfe6392b3/lyxgit

 I hope this is OK now.

Looks fine. Thanks for your persistence. Pavel

Re: [patch] LyX scrolling issue

[Pavel Sanda]

André Pönitz wrote:
 From the docs:
 
 We suggest only using repaint() if you need an immediate repaint, for
 example during animation. In almost all circumstances update() is
 better, as it permits Qt to optimize for speed and minimize flicker.

The problem is that the speed optimization claim seems to be plain talk ;(
I'm not sure whether the recursive repaint messages we get is worse than
users complaint that scrollbar is unusable...

Pavel

Re: Noweb/Literate programming Copier problem

[Kayvan Sylvan]

Yes. I pulled from git, but the software still identifies itself as 2.1.0
SVN.

On Mon, Sep 24, 2012 at 12:20 PM, Pavel Sanda sa...@lyx.org wrote:

 Kayvan Sylvan wrote:
  I'm running LyX 2.1.0 SVN on Fedora 17

 Are you aware that for long time we use git and SVN is months old codebase?
 Pavel

Re: Noweb/Literate programming Copier problem

[Pavel Sanda]

Kayvan Sylvan wrote:
 Yes. I pulled from git, but the software still identifies itself as 2.1.0
 SVN.

Hummm, I thought we already changed this, will change it now.
Pavel

Re: [LyX master] avoid deprecated functions on Mac - use Qt instead

[Stephan Witt]

Am 24.09.2012 um 21:49 schrieb Pavel Sanda sa...@lyx.org:

 Kornel Benko wrote:
 Why not
  bool Messages::available(string const  ) { ... }
 
 exactly ;) p

I've copied this construct from src/support/Package.cpp

Stephan

Re: [LyX master] avoid deprecated functions on Mac - use Qt instead

[Pavel Sanda]

Stephan Witt wrote:
 I've copied this construct from src/support/Package.cpp

I see, but there it is forced by the ifdef structure. I also wonder
why the (void) btw... Pavel

Re: [LyX master] avoid deprecated functions on Mac - use Qt instead

[Stephan Witt]

Am 24.09.2012 um 23:34 schrieb Pavel Sanda sa...@lyx.org:

 Stephan Witt wrote:
 I've copied this construct from src/support/Package.cpp
 
 I see, but there it is forced by the ifdef structure. I also wonder
 why the (void) btw... Pavel

To avoid other warnings.

The best pattern IMHO is:

bool Messages::available(string const  /* c */)

I can change that.

Stephan

Re: [patch] LyX bash completion

[Richard Heck]

On 09/24/2012 03:54 PM, Pavel Sanda wrote:

Scott Kostyshak wrote:

Are these OK to go in to branch and master?

Looks fine if tested ;) Branch is Richard's call.


I don't know anything about this completion stuff. How do you feel about it?

rh

Re: [patch] LyX bash completion

[Scott Kostyshak]

On Mon, Sep 24, 2012 at 6:49 PM, Richard Heck rgh...@lyx.org wrote:
 On 09/24/2012 03:54 PM, Pavel Sanda wrote:

 Scott Kostyshak wrote:

 Are these OK to go in to branch and master?

 Looks fine if tested ;) Branch is Richard's call.

 I don't know anything about this completion stuff. How do you feel about it?

Not 100% sure and I don't think anyone besides me has tested it. I'll
commit to trunk only.

Would this be a good situation for committing to branch right after
2.0.5 is released?

Thanks,

Scott

Visibility of .lyx~ files as an option in LyX's Open dialog.

[Andrew Parsloe]

The table sorting python script discussed on the users' list overwrites 
the current file on disk: buffer-write; buffer-export lyxtabsort; 
buffer-reload. As Liviu has noted, there is no Undo. But the initial 
buffer-write backs up the pre-sort disk file to .lyx~. That makes easy 
access to the back-up .lyx~ file important. At present the drop-down box 
in the Open dialog shows *.lyx documents and *.lyx13 to *.lyx16 
documents. I think it should also have an option for *.lyx~ documents.


If *.lyx~ files are made visible in the dialog and one is selected to be 
opened, then presumably .lyx~ would be changed to .lyx. If the original 
and back-up files share the same directory (a way I imagine many people 
operate), it would also prompt a message about overwriting.


Andrew

Re: [patch] support the lightning Unicode character Ux21AF

[Guenter Milde]

On 2012-09-23, Uwe Stöhr wrote:
> Am 13.09.2012 11:18, schrieb Jürgen Spitzmüller:

>>> you just found some packages to support this character:
>>> http://wiki.lyx.org/LyX/MissingUnicodeCharacters?action=diff#diff1309586810

>>> I tested this and the attached patch works fine. Can this go in or is there
>>> a reason why you did not commit this yet?

>> I never bothered with this character. But as Günter says, package clashes
>> might be an issue.

> But we already use marvosym and this package is now again under maintenance.

> We can use wasysym instead for this character but this package is not
> under maintenance and that#s why I did not use it. However, we use
> already both packages for many characters so it should not make a
> difference what to use. So OK to commit?

It makes a difference regarding matching symbols or possible clashes between
these two packages. It would be best to choose whichever package is already
loaded or which does not conflict with other required packages.

The "|" operator seen in e.g. 

   2171: 0x271d "\\Cross"  "bbding|marvosym" "" "" "" # LATIN 
CROSS
   
seems to do something in this line already (it is, unfortunately not
documented in the head of the "unicodesymbols" file).

So, I suggest using "marvosym|wasysym" as textpreamble entry.

Günter

Re: [LyX 2.0.x] Fix bug #8349: Cannot compile 2.0.x: unresolved external symbol

[Jean-Marc Lasgouttes]

That's amusing. This is a patch from Stephan that I backported, and it 
looks like Stephan did the backport, which does not help communication.


Le 24/09/2012 16:19, Stephan Witt a écrit :

commit 8169347adafeaece14c8b463add8072c6e9a91a8
Author: Stephan Witt 
Date:   Sun Sep 23 15:07:41 2012 +0200

 Fix bug #8349: Cannot compile 2.0.x: unresolved external symbol

 correct Messages class implementation for non-NLS


Then there is this loosy no-op commit. Why do we need this junk?

> commit a602f21dc42f6edd5d8ad4719eb85b207373
> Merge: 2512e1f 0ea1944
> Author: Jean-Marc Lasgouttes 
> Date:   Mon Sep 24 16:01:29 2012 +0200
>
>  Merge branch '2.0.x' of git.lyx.org:lyx into 2.0.x

And then of course there is my own stupidity, which adds another commit. 
It is not a problem since I intended to port this for 2.0.5, but I 
intended to have feedback from Uwe on the other one.



commit 2512e1f085a6ad044d6d0ff68f75390ed1d34c75
Author: Jean-Marc Lasgouttes 
Date:   Thu Aug 23 15:08:21 2012 +0200

 Tracking correctly available translations (take 2)

 The previous scheme of loading all possible translations and checking
 whether the work is a bit too much "brute force" and causes problems
 on Mac OS X (documents loaded with the wrong language).

 Now there is an helper static method in Messages class that checks
 whether a readable .mo file exist for the language. There should be an
 API in gettext for doing that, but alas it is not possible.

 As a consequence the method Language::translated() has been removed,
 along with its cache.


Finally is it expected that the commits show in only one e-mail message? 
Is there a reason to do that instead of two e-mails?


JMarc

Re: LyX 2.0.5

[Jean-Marc Lasgouttes]

Le 23/09/2012 23:15, Jean-Marc Lasgouttes a écrit :

This is not changing strings, but I wonder whether I should revert
824fe175 (keep undo information after save as...), after seeing the
crashes found by pavel in bug 8342. The undo errors the commit intended
to fix are fixed by the patch to 8342 anyway.


This is still pending


It seems that there are some link errors in windows after my patch
to to bug 7741. I have not looked at it yet.
http://www.lyx.org/trac/ticket/8349


Fixed in branch now.



I have a second patch I would like to backport to branch: 7a7b9abf1b3f4
It does not change strings. I am waiting for some feedback from Uwe on
the fix to bug 7741 first, because the two touch the same feature (UI
language selection).


Committed by mistake. Let's see the result :) More seriously, my only 
question is that I have written code like:

+   string const filen = locale_dir + "/" + code
+   + "/LC_MESSAGES/"PACKAGE".mo";
+   if (FileName(filen).isReadableFile())
+ ...
and I wonder wther this still works under windows. I think I am dealing 
with internal representation, so I can use explicit / for directories, 
right?


Or is it going to break for windows? The symptom would be simple: bad 
list of available translations for GUI language in prefs.


JMarc

Re: [patch] fix warning on branch

[Richard Heck]

On 09/23/2012 04:54 PM, Jean-Marc Lasgouttes wrote:

Le 23/09/12 00:06, Scott Kostyshak a écrit :

How many users *compile* anything :)


True. And out of those who do, how many look at the warnings? :)


Personally, I still think that removing warnings in branch matters.

Yes, I think it's worth doing, but not if it's the kind of thing that 
has any

chance of causing problems. I think that was the issue.

rh

Re: [LyX 2.0.x] Fix bug #8349: Cannot compile 2.0.x: unresolved external symbol

[Richard Heck]

On 09/24/2012 10:28 AM, Jean-Marc Lasgouttes wrote:


Finally is it expected that the commits show in only one e-mail 
message? Is there a reason to do that instead of two e-mails?


It's mostly a matter of re-writing (or somehow updating) the 
email-sending script to do this. I've not had time.


rh

Re: [LyX master] avoid deprecated functions on Mac - use Qt instead

[Pavel Sanda]

Stephan Witt wrote:
> -bool Messages::available() const
> +bool Messages::available(string const & c)
>  {
> + (void)c;
>   return false;
>  }

Why this (void)c; construct? Pavel

Re: [LyX 2.0.x] Fix bug #8349: Cannot compile 2.0.x: unresolved external symbol

[Vincent van Ravesteijn]

Op 24-9-2012 16:28, Jean-Marc Lasgouttes schreef:
That's amusing. This is a patch from Stephan that I backported, and it 
looks like Stephan did the backport, which does not help communication.


Just get used to a more advanced form of communication. It is Stephan's 
code, which you committed. We just need to fix the e-mail send script if 
you really want to



> commit a602f21dc42f6edd5d8ad4719eb85b207373
> Merge: 2512e1f 0ea1944
> Author: Jean-Marc Lasgouttes 
> Date:   Mon Sep 24 16:01:29 2012 +0200
>
>  Merge branch '2.0.x' of git.lyx.org:lyx into 2.0.x


This is your own stupidity as well. First fetch and merge before you add 
your new commits to the branch. If you already commit something and the 
push fails, then you rebase onto the remote branch. You should merge or 
pull to avoid this no-op commits.




And then of course there is my own stupidity, which adds another 
commit. It is not a problem since I intended to port this for 2.0.5, 
but I intended to have feedback from Uwe on the other one.


Vincent

Re: [LyX master] avoid deprecated functions on Mac - use Qt instead

[Richard Heck]

On 09/24/2012 01:07 PM, Pavel Sanda wrote:

Stephan Witt wrote:

-bool Messages::available() const
+bool Messages::available(string const & c)
  {
+   (void)c;
return false;
  }

Why this (void)c; construct?


Crushes an "unused variable" warning.

rh

Re: Re: [LyX master] avoid deprecated functions on Mac - use Qt instead

[Kornel Benko]

Am Montag, 24. September 2012 um 14:10:33, schrieb Richard Heck 
> On 09/24/2012 01:07 PM, Pavel Sanda wrote:
> > Stephan Witt wrote:
> >> -bool Messages::available() const
> >> +bool Messages::available(string const & c)
> >>   {
> >> +  (void)c;
> >>return false;
> >>   }
> > Why this (void)c; construct?
> >
> Crushes an "unused variable" warning.
> 
> rh

Why not
bool Messages::available(string const & ) { ... }

Kornel

signature.asc
Description: This is a digitally signed message part.

Re: [LyX master] avoid deprecated functions on Mac - use Qt instead

[Richard Heck]

On 09/24/2012 02:44 PM, Kornel Benko wrote:


Am Montag, 24. September 2012 um 14:10:33, schrieb Richard Heck 



> On 09/24/2012 01:07 PM, Pavel Sanda wrote:

> > Stephan Witt wrote:

> >> -bool Messages::available() const

> >> +bool Messages::available(string const & c)

> >> {

> >> + (void)c;

> >> return false;

> >> }

> > Why this (void)c; construct?

> >

> Crushes an "unused variable" warning.

>

> rh

Why not

bool Messages::available(string const & ) { ... }

Yes, you could also do it that way. He probably wanted to keep the 
signature the same as in the "real" routine.


rh

Re: Noweb/Literate programming Copier problem

[Pavel Sanda]

Kayvan Sylvan wrote:
> I'm running LyX 2.1.0 SVN on Fedora 17

Are you aware that for long time we use git and SVN is months old codebase?
Pavel

Re: Re: [LyX master] avoid deprecated functions on Mac - use Qt instead

[Pavel Sanda]

Kornel Benko wrote:
> Why not
>   bool Messages::available(string const & ) { ... }

exactly ;) p

Re: [patch] LyX bash completion

[Pavel Sanda]

Scott Kostyshak wrote:
> Are these OK to go in to branch and master?

Looks fine if tested ;) Branch is Richard's call.
Pavel

Re: [LyX master] Fix UI for decimal table alignment (bug #8285)

[Pavel Sanda]

Uwe Stöhr wrote:
> I did this now:
> http://www.lyx.org/trac/changeset/237cb51fb15fdaa4951f70e393eea0fdfe6392b3/lyxgit
>
> I hope this is OK now.

Looks fine. Thanks for your persistence. Pavel

Re: [patch] LyX scrolling issue

[Pavel Sanda]

André Pönitz wrote:
> >From the docs:
> 
> We suggest only using repaint() if you need an immediate repaint, for
> example during animation. In almost all circumstances update() is
> better, as it permits Qt to optimize for speed and minimize flicker.

The problem is that the speed optimization claim seems to be plain talk ;(
I'm not sure whether the recursive repaint messages we get is worse than
users complaint that scrollbar is unusable...

Pavel

Re: Noweb/Literate programming Copier problem

[Kayvan Sylvan]

Yes. I pulled from git, but the software still identifies itself as 2.1.0
SVN.

On Mon, Sep 24, 2012 at 12:20 PM, Pavel Sanda  wrote:

> Kayvan Sylvan wrote:
> > I'm running LyX 2.1.0 SVN on Fedora 17
>
> Are you aware that for long time we use git and SVN is months old codebase?
> Pavel
>

Re: Noweb/Literate programming Copier problem

[Pavel Sanda]

Kayvan Sylvan wrote:
> Yes. I pulled from git, but the software still identifies itself as 2.1.0
> SVN.

Hummm, I thought we already changed this, will change it now.
Pavel

Re: [LyX master] avoid deprecated functions on Mac - use Qt instead

[Stephan Witt]

Am 24.09.2012 um 21:49 schrieb Pavel Sanda :

> Kornel Benko wrote:
>> Why not
>>  bool Messages::available(string const & ) { ... }
> 
> exactly ;) p

I've copied this construct from src/support/Package.cpp

Stephan

Re: [LyX master] avoid deprecated functions on Mac - use Qt instead

[Pavel Sanda]

Stephan Witt wrote:
> I've copied this construct from src/support/Package.cpp

I see, but there it is forced by the ifdef structure. I also wonder
why the (void) btw... Pavel

Re: [LyX master] avoid deprecated functions on Mac - use Qt instead

[Stephan Witt]

Am 24.09.2012 um 23:34 schrieb Pavel Sanda :

> Stephan Witt wrote:
>> I've copied this construct from src/support/Package.cpp
> 
> I see, but there it is forced by the ifdef structure. I also wonder
> why the (void) btw... Pavel

To avoid other warnings.

The best pattern IMHO is:

bool Messages::available(string const & /* c */)

I can change that.

Stephan

Re: [patch] LyX bash completion

[Richard Heck]

On 09/24/2012 03:54 PM, Pavel Sanda wrote:

Scott Kostyshak wrote:

Are these OK to go in to branch and master?

Looks fine if tested ;) Branch is Richard's call.


I don't know anything about this completion stuff. How do you feel about it?

rh

Re: [patch] LyX bash completion

[Scott Kostyshak]

On Mon, Sep 24, 2012 at 6:49 PM, Richard Heck  wrote:
> On 09/24/2012 03:54 PM, Pavel Sanda wrote:
>>
>> Scott Kostyshak wrote:
>>>
>>> Are these OK to go in to branch and master?
>>
>> Looks fine if tested ;) Branch is Richard's call.
>>
> I don't know anything about this completion stuff. How do you feel about it?

Not 100% sure and I don't think anyone besides me has tested it. I'll
commit to trunk only.

Would this be a good situation for committing to branch right after
2.0.5 is released?

Thanks,

Scott

Visibility of .lyx~ files as an option in LyX's Open dialog.

[Andrew Parsloe]

The table sorting python script discussed on the users' list overwrites 
the current file on disk: buffer-write; buffer-export lyxtabsort; 
buffer-reload. As Liviu has noted, there is no Undo. But the initial 
buffer-write backs up the pre-sort disk file to .lyx~. That makes easy 
access to the back-up .lyx~ file important. At present the drop-down box 
in the Open dialog shows *.lyx documents and *.lyx13 to *.lyx16 
documents. I think it should also have an option for *.lyx~ documents.


If *.lyx~ files are made visible in the dialog and one is selected to be 
opened, then presumably .lyx~ would be changed to .lyx. If the original 
and back-up files share the same directory (a way I imagine many people 
operate), it would also prompt a message about overwriting.


Andrew