Re: test of math previews

[Enrico Forestieri]

On Thu, Jun 25, 2015 at 10:16:21AM +0200, Enrico Forestieri wrote:
 
 The third patch is the simpler one proposed by Guillaume. I have already
 applied the better patch to master but there is a glitch with it.
 Namely, if a preview snippet produces a blank image, its metrics are
 not invalidated and thus the image would not be visible on the LyX
 screen (actually, it is replaced by a small red square). I still have
 to take a look at it, so I think that for the time being the simpler
 patch will do.

Actually, it seems doing the right thing, as things such as $\quad$ or
$\relax$ are left as is, while $\raisebox{6mm}{}$ or $\raisebox{-6mm}{}$
produce a small vertical strip with the required characteristics (as
revealed by selecting the images in LyX).
So, I think the better patch can also be applied.

-- 
Enrico

Re: [LyX/master] Fix setting of pref hide scrollbar in fullscreen

[Scott Kostyshak]

On Thu, Jun 25, 2015 at 02:42:40PM +0200, Scott Kostyshak wrote:
 commit 5e7cbc8f2c141889950628d0b81fcdb148fa278b
 Author: Scott Kostyshak skost...@lyx.org
 Date:   Thu Jun 25 08:21:46 2015 -0400
 
 Fix setting of pref hide scrollbar in fullscreen
 
 The symptom of the bug is that the preference could be incorrectly
 displayed (based on the value of the statusbar preference).
 
 This seems to have been a copy/paste mistake introduced in 214f7ed2.

OK for branch?

Scott

Re: Unable to enter custom negative spaces

[Kornel Benko]

Am Donnerstag, 25. Juni 2015 um 10:04:47, schrieb Scott Kostyshak 
skost...@lyx.org
 On Thu, Jun 25, 2015 at 07:01:06AM -0700, Kornel Benko wrote:
  The dialog in master does not allow negative values anymore.
  Neither for horizontal, nor vertical space.
 
 I can reproduce. Possibly related to 9e166088.
 
 Scott

Good guess. Bisecting leads to it too.
9e16608867f72e417b04267d080fed28802e92a4 is the first bad commit
commit 9e16608867f72e417b04267d080fed28802e92a4
Author: Juergen Spitzmueller sp...@lyx.org
Date:   Wed May 20 19:12:23 2015 +0200

Fix problems with immediate apply and length validators

Fixes bugs #7022 and #7599.

:04 04 6b9a5994435f4a86e54f26f53258254a3e907f23 
4cfd99cbc33ae00afdf5eb20273fddf8908ed229 M  src

Kornel

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

Unable to enter custom negative spaces

[Kornel Benko]

The dialog in master does not allow negative values anymore.
Neither for horizontal, nor vertical space.

Entering the values with external editor works and is also correctly displayed 
afterwards.

Kornel


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

Re: Unable to enter custom negative spaces

[Scott Kostyshak]

On Thu, Jun 25, 2015 at 07:01:06AM -0700, Kornel Benko wrote:
 The dialog in master does not allow negative values anymore.
 Neither for horizontal, nor vertical space.

I can reproduce. Possibly related to 9e166088.

Scott

Re: test of math previews

[Enrico Forestieri]

On Thu, Jun 25, 2015 at 01:44:15PM +0200, Enrico Forestieri wrote:
 On Thu, Jun 25, 2015 at 10:16:21AM +0200, Enrico Forestieri wrote:
  
  The third patch is the simpler one proposed by Guillaume. I have already
  applied the better patch to master but there is a glitch with it.
  Namely, if a preview snippet produces a blank image, its metrics are
  not invalidated and thus the image would not be visible on the LyX
  screen (actually, it is replaced by a small red square). I still have
  to take a look at it, so I think that for the time being the simpler
  patch will do.
 
 Actually, it seems doing the right thing, as things such as $\quad$ or
 $\relax$ are left as is, while $\raisebox{6mm}{}$ or $\raisebox{-6mm}{}$
 produce a small vertical strip with the required characteristics (as
 revealed by selecting the images in LyX).

I had a closer look and this behavior is independent of the patch.
Only when using dvipng blank images can be detected reliably, so what
described above occurs only when using the legacy route for generating
images.

 So, I think the better patch can also be applied.

More rightly, as in stable the legacy route is always used as a fallback.

-- 
Enrico

Re: test of math previews

[Richard Heck]

On 06/25/2015 02:13 PM, Enrico Forestieri wrote:

On Thu, Jun 25, 2015 at 01:44:15PM +0200, Enrico Forestieri wrote:

On Thu, Jun 25, 2015 at 10:16:21AM +0200, Enrico Forestieri wrote:

The third patch is the simpler one proposed by Guillaume. I have already
applied the better patch to master but there is a glitch with it.
Namely, if a preview snippet produces a blank image, its metrics are
not invalidated and thus the image would not be visible on the LyX
screen (actually, it is replaced by a small red square). I still have
to take a look at it, so I think that for the time being the simpler
patch will do.

Actually, it seems doing the right thing, as things such as $\quad$ or
$\relax$ are left as is, while $\raisebox{6mm}{}$ or $\raisebox{-6mm}{}$
produce a small vertical strip with the required characteristics (as
revealed by selecting the images in LyX).

I had a closer look and this behavior is independent of the patch.
Only when using dvipng blank images can be detected reliably, so what
described above occurs only when using the legacy route for generating
images.


So, I think the better patch can also be applied.

More rightly, as in stable the legacy route is always used as a fallback.


OK with me then.

This more or less completes this work? (For which thanks very much, to 
you and Guillaume.)


Richard

Re: test of math previews

[Enrico Forestieri]

On Thu, Jun 25, 2015 at 02:16:19PM -0400, Richard Heck wrote:
 On 06/25/2015 02:13 PM, Enrico Forestieri wrote:
 On Thu, Jun 25, 2015 at 01:44:15PM +0200, Enrico Forestieri wrote:
 On Thu, Jun 25, 2015 at 10:16:21AM +0200, Enrico Forestieri wrote:
 The third patch is the simpler one proposed by Guillaume. I have already
 applied the better patch to master but there is a glitch with it.
 Namely, if a preview snippet produces a blank image, its metrics are
 not invalidated and thus the image would not be visible on the LyX
 screen (actually, it is replaced by a small red square). I still have
 to take a look at it, so I think that for the time being the simpler
 patch will do.
 Actually, it seems doing the right thing, as things such as $\quad$ or
 $\relax$ are left as is, while $\raisebox{6mm}{}$ or $\raisebox{-6mm}{}$
 produce a small vertical strip with the required characteristics (as
 revealed by selecting the images in LyX).
 I had a closer look and this behavior is independent of the patch.
 Only when using dvipng blank images can be detected reliably, so what
 described above occurs only when using the legacy route for generating
 images.
 
 So, I think the better patch can also be applied.
 More rightly, as in stable the legacy route is always used as a fallback.
 
 OK with me then.
 
 This more or less completes this work?

I think so. The most important glitches were identified and solved.

 (For which thanks very much, to you
 and Guillaume.)

The credit goes mainly to Guillaume for his accurate testing.

-- 
Enrico

Re: test of math previews

[Richard Heck]

On 06/25/2015 02:58 PM, Enrico Forestieri wrote:

On Thu, Jun 25, 2015 at 02:16:19PM -0400, Richard Heck wrote:

This more or less completes this work?

I think so. The most important glitches were identified and solved.


(For which thanks very much, to you
and Guillaume.)

The credit goes mainly to Guillaume for his accurate testing.


Somebody had to write the code

rh

Re: test of math previews

[Guillaume M-M]

Le 25/06/2015 23:45, Richard Heck a écrit :

On 06/25/2015 02:58 PM, Enrico Forestieri wrote:

On Thu, Jun 25, 2015 at 02:16:19PM -0400, Richard Heck wrote:

This more or less completes this work?

I think so. The most important glitches were identified and solved.


(For which thanks very much, to you
and Guillaume.)

The credit goes mainly to Guillaume for his accurate testing.


Somebody had to write the code

rh




This is also what I believe. Thank you!

By the way will f3b03699 be backported as well? It seems that segfaults 
are almost guaranteed for people with previews on during Ctrl+C...

Re: test of math previews

[Enrico Forestieri]

On Wed, Jun 24, 2015 at 09:36:50PM -0400, Richard Heck wrote:
 On 06/24/2015 08:49 PM, Guillaume M-M wrote:
 Le 25/06/2015 00:44, Enrico Forestieri a écrit :
 On Mon, Jun 22, 2015 at 11:32:17PM +0100, Guillaume M-M wrote:
 
 3) Previews have the incorrect size with layouts that offer font sizes
 other
 than 1[0-2]pt (lyx-preview-fontsize.lyx). The legacy method offers a
 function that retrieves the actual font size and calculates the actual
 dpi.
 Attached is a patch (extract_resolution.diff) that imports this
 function in
 the new script.
 
 Thank you, this is much better than trying to catch all possible cases.
 I have applied the patch to master at 0751f96d.
 
 
 Thank you for applying it.
 
 Here's a trivial patch for stable, if you are not ready to backport
 0751f96d. Both were tested in stable.
 
 This is fine with me, as is the original patch, as Enrico sees fit.

Richard, I am attaching 3 patches for stable. The first one fix crashes
occurring when zooming after latex completely failed to generate an
output and when copying a macro with instant preview on. This last
crash also occurs in 2.1.3.

The second patch assures that a lonely math inset is previewed.
The third patch is the simpler one proposed by Guillaume. I have already
applied the better patch to master but there is a glitch with it.
Namely, if a preview snippet produces a blank image, its metrics are
not invalidated and thus the image would not be visible on the LyX
screen (actually, it is replaced by a small red square). I still have
to take a look at it, so I think that for the time being the simpler
patch will do.

-- 
Enrico
diff --git a/src/graphics/PreviewLoader.cpp b/src/graphics/PreviewLoader.cpp
index eb25805..e20f1ad 100644
--- a/src/graphics/PreviewLoader.cpp
+++ b/src/graphics/PreviewLoader.cpp
@@ -709,6 +709,7 @@ void PreviewLoader::Impl::finishedGenerating(pid_t pid, int 
retval)
if (git == in_progress_.end()) {
lyxerr  PreviewLoader::finishedGenerating(): unable to find 
data for PID   pid  endl;
+   finished_generating_ = true;
return;
}
 
@@ -717,8 +718,11 @@ void PreviewLoader::Impl::finishedGenerating(pid_t pid, 
int retval)
LYXERR(Debug::GRAPHICS, PreviewLoader::finishedInProgress(
 retval  ): processing   status
  for   command);
-   if (retval  0)
+   if (retval  0) {
+   in_progress_.erase(git);
+   finished_generating_ = true;
return;
+   }
 
// Read the metrics file, if it exists
vectordouble ascent_fractions(git-second.snippets.size());
diff --git a/src/mathed/MacroTable.h b/src/mathed/MacroTable.h
index 3bd04ea..030e6f7 100644
--- a/src/mathed/MacroTable.h
+++ b/src/mathed/MacroTable.h
@@ -69,6 +69,8 @@ public:
char const * MathMLtype() const;
///
void setSymbol(latexkeys const * sym) { sym_ = sym; }
+   ///
+   DocIterator const  pos() { return pos_; }
 
/// lock while being drawn to avoid recursions
int lock() const { return ++lockCount_; }
diff --git a/src/mathed/MathMacro.cpp b/src/mathed/MathMacro.cpp
index 06cfd5c..a25c5db 100644
--- a/src/mathed/MathMacro.cpp
+++ b/src/mathed/MathMacro.cpp
@@ -176,6 +176,14 @@ void MathMacro::assign(MathMacro const  that)
if (p)
p-setOwner(this);
}
+   if (macro_  lyxrc.preview == LyXRC::PREVIEW_ON) {
+   // We need to update macro_ by ourselves because in this case
+   // MathData::metrics() is not called when selecting a math inset
+   DocIterator const  pos = macroBackup_.pos();
+   macro_ = pos.buffer() ? pos.buffer()-getMacro(name(), pos) : 0;
+   if (!macro_)
+   macro_ = macroBackup_;
+   }
 }
 
 
diff --git a/src/mathed/MathMacro.cpp b/src/mathed/MathMacro.cpp
index 06cfd5c..9744366 100644
--- a/src/mathed/MathMacro.cpp
+++ b/src/mathed/MathMacro.cpp
@@ -604,6 +612,20 @@ bool MathMacro::validName() const
 
 void MathMacro::validate(LaTeXFeatures  features) const
 {
+   // Immediately after a document is loaded, in some cases the MacroData
+   // of the global macros defined in the lib/symbols file may still not
+   // be known to the macro machinery because it will be set only after
+   // the first call to updateMacros(). This is not a problem unless
+   // instant preview is on for math, in which case we will be missing
+   // the corresponding requirements.
+   // In this case, we get the required info from the global macro table.
+   if (requires_.empty()  !macro_) {
+   // Update requires for known global macros.
+   MacroData const * data = MacroTable::globalMacros().get(name());
+   if (data  !data-requires().empty())
+   features.require(data-requires());
+   }
+
 

Re: test of math previews

[Enrico Forestieri]

On Wed, Jun 24, 2015 at 09:36:50PM -0400, Richard Heck wrote:
> On 06/24/2015 08:49 PM, Guillaume M-M wrote:
> >Le 25/06/2015 00:44, Enrico Forestieri a écrit :
> >>On Mon, Jun 22, 2015 at 11:32:17PM +0100, Guillaume M-M wrote:
> >>>
> >>>3) Previews have the incorrect size with layouts that offer font sizes
> >>>other
> >>>than 1[0-2]pt (lyx-preview-fontsize.lyx). The legacy method offers a
> >>>function that retrieves the actual font size and calculates the actual
> >>>dpi.
> >>>Attached is a patch (extract_resolution.diff) that imports this
> >>>function in
> >>>the new script.
> >>
> >>Thank you, this is much better than trying to catch all possible cases.
> >>I have applied the patch to master at 0751f96d.
> >>
> >
> >Thank you for applying it.
> >
> >Here's a trivial patch for stable, if you are not ready to backport
> >0751f96d. Both were tested in stable.
> 
> This is fine with me, as is the original patch, as Enrico sees fit.

Richard, I am attaching 3 patches for stable. The first one fix crashes
occurring when zooming after latex completely failed to generate an
output and when copying a macro with instant preview on. This last
crash also occurs in 2.1.3.

The second patch assures that a lonely math inset is previewed.
The third patch is the simpler one proposed by Guillaume. I have already
applied the better patch to master but there is a glitch with it.
Namely, if a preview snippet produces a blank image, its metrics are
not invalidated and thus the image would not be visible on the LyX
screen (actually, it is replaced by a small red square). I still have
to take a look at it, so I think that for the time being the simpler
patch will do.

-- 
Enrico
diff --git a/src/graphics/PreviewLoader.cpp b/src/graphics/PreviewLoader.cpp
index eb25805..e20f1ad 100644
--- a/src/graphics/PreviewLoader.cpp
+++ b/src/graphics/PreviewLoader.cpp
@@ -709,6 +709,7 @@ void PreviewLoader::Impl::finishedGenerating(pid_t pid, int 
retval)
if (git == in_progress_.end()) {
lyxerr << "PreviewLoader::finishedGenerating(): unable to find "
"data for PID " << pid << endl;
+   finished_generating_ = true;
return;
}
 
@@ -717,8 +718,11 @@ void PreviewLoader::Impl::finishedGenerating(pid_t pid, 
int retval)
LYXERR(Debug::GRAPHICS, "PreviewLoader::finishedInProgress("
<< retval << "): processing " << status
<< " for " << command);
-   if (retval > 0)
+   if (retval > 0) {
+   in_progress_.erase(git);
+   finished_generating_ = true;
return;
+   }
 
// Read the metrics file, if it exists
vector ascent_fractions(git->second.snippets.size());
diff --git a/src/mathed/MacroTable.h b/src/mathed/MacroTable.h
index 3bd04ea..030e6f7 100644
--- a/src/mathed/MacroTable.h
+++ b/src/mathed/MacroTable.h
@@ -69,6 +69,8 @@ public:
char const * MathMLtype() const;
///
void setSymbol(latexkeys const * sym) { sym_ = sym; }
+   ///
+   DocIterator const & pos() { return pos_; }
 
/// lock while being drawn to avoid recursions
int lock() const { return ++lockCount_; }
diff --git a/src/mathed/MathMacro.cpp b/src/mathed/MathMacro.cpp
index 06cfd5c..a25c5db 100644
--- a/src/mathed/MathMacro.cpp
+++ b/src/mathed/MathMacro.cpp
@@ -176,6 +176,14 @@ void MathMacro::assign(MathMacro const & that)
if (p)
p->setOwner(this);
}
+   if (macro_ && lyxrc.preview == LyXRC::PREVIEW_ON) {
+   // We need to update macro_ by ourselves because in this case
+   // MathData::metrics() is not called when selecting a math inset
+   DocIterator const & pos = macroBackup_.pos();
+   macro_ = pos.buffer() ? pos.buffer()->getMacro(name(), pos) : 0;
+   if (!macro_)
+   macro_ = _;
+   }
 }
 
 
diff --git a/src/mathed/MathMacro.cpp b/src/mathed/MathMacro.cpp
index 06cfd5c..9744366 100644
--- a/src/mathed/MathMacro.cpp
+++ b/src/mathed/MathMacro.cpp
@@ -604,6 +612,20 @@ bool MathMacro::validName() const
 
 void MathMacro::validate(LaTeXFeatures & features) const
 {
+   // Immediately after a document is loaded, in some cases the MacroData
+   // of the global macros defined in the lib/symbols file may still not
+   // be known to the macro machinery because it will be set only after
+   // the first call to updateMacros(). This is not a problem unless
+   // instant preview is on for math, in which case we will be missing
+   // the corresponding requirements.
+   // In this case, we get the required info from the global macro table.
+   if (requires_.empty() && !macro_) {
+   // Update requires for known global macros.
+   MacroData const * data = MacroTable::globalMacros().get(name());
+   if (data && 

Re: test of math previews

[Enrico Forestieri]

On Thu, Jun 25, 2015 at 10:16:21AM +0200, Enrico Forestieri wrote:
> 
> The third patch is the simpler one proposed by Guillaume. I have already
> applied the better patch to master but there is a glitch with it.
> Namely, if a preview snippet produces a blank image, its metrics are
> not invalidated and thus the image would not be visible on the LyX
> screen (actually, it is replaced by a small red square). I still have
> to take a look at it, so I think that for the time being the simpler
> patch will do.

Actually, it seems doing the right thing, as things such as $\quad$ or
$\relax$ are left as is, while $\raisebox{6mm}{}$ or $\raisebox{-6mm}{}$
produce a small vertical strip with the required characteristics (as
revealed by selecting the images in LyX).
So, I think the better patch can also be applied.

-- 
Enrico

Re: [LyX/master] Fix setting of pref "hide scrollbar in fullscreen"

[Scott Kostyshak]

On Thu, Jun 25, 2015 at 02:42:40PM +0200, Scott Kostyshak wrote:
> commit 5e7cbc8f2c141889950628d0b81fcdb148fa278b
> Author: Scott Kostyshak 
> Date:   Thu Jun 25 08:21:46 2015 -0400
> 
> Fix setting of pref "hide scrollbar in fullscreen"
> 
> The symptom of the bug is that the preference could be incorrectly
> displayed (based on the value of the statusbar preference).
> 
> This seems to have been a copy/paste mistake introduced in 214f7ed2.

OK for branch?

Scott

Unable to enter custom negative spaces

[Kornel Benko]

The dialog in master does not allow negative values anymore.
Neither for horizontal, nor vertical space.

Entering the values with external editor works and is also correctly displayed 
afterwards.

Kornel


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

Re: Unable to enter custom negative spaces

[Scott Kostyshak]

On Thu, Jun 25, 2015 at 07:01:06AM -0700, Kornel Benko wrote:
> The dialog in master does not allow negative values anymore.
> Neither for horizontal, nor vertical space.

I can reproduce. Possibly related to 9e166088.

Scott

Re: Unable to enter custom negative spaces

[Kornel Benko]

Am Donnerstag, 25. Juni 2015 um 10:04:47, schrieb Scott Kostyshak 

> On Thu, Jun 25, 2015 at 07:01:06AM -0700, Kornel Benko wrote:
> > The dialog in master does not allow negative values anymore.
> > Neither for horizontal, nor vertical space.
> 
> I can reproduce. Possibly related to 9e166088.
> 
> Scott

Good guess. Bisecting leads to it too.
9e16608867f72e417b04267d080fed28802e92a4 is the first bad commit
commit 9e16608867f72e417b04267d080fed28802e92a4
Author: Juergen Spitzmueller 
Date:   Wed May 20 19:12:23 2015 +0200

Fix problems with immediate apply and length validators

Fixes bugs #7022 and #7599.

:04 04 6b9a5994435f4a86e54f26f53258254a3e907f23 
4cfd99cbc33ae00afdf5eb20273fddf8908ed229 M  src

Kornel

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

Re: test of math previews

[Enrico Forestieri]

On Thu, Jun 25, 2015 at 01:44:15PM +0200, Enrico Forestieri wrote:
> On Thu, Jun 25, 2015 at 10:16:21AM +0200, Enrico Forestieri wrote:
> > 
> > The third patch is the simpler one proposed by Guillaume. I have already
> > applied the better patch to master but there is a glitch with it.
> > Namely, if a preview snippet produces a blank image, its metrics are
> > not invalidated and thus the image would not be visible on the LyX
> > screen (actually, it is replaced by a small red square). I still have
> > to take a look at it, so I think that for the time being the simpler
> > patch will do.
> 
> Actually, it seems doing the right thing, as things such as $\quad$ or
> $\relax$ are left as is, while $\raisebox{6mm}{}$ or $\raisebox{-6mm}{}$
> produce a small vertical strip with the required characteristics (as
> revealed by selecting the images in LyX).

I had a closer look and this behavior is independent of the patch.
Only when using dvipng blank images can be detected reliably, so what
described above occurs only when using the legacy route for generating
images.

> So, I think the better patch can also be applied.

More rightly, as in stable the legacy route is always used as a fallback.

-- 
Enrico

Re: test of math previews

[Richard Heck]

On 06/25/2015 02:13 PM, Enrico Forestieri wrote:

On Thu, Jun 25, 2015 at 01:44:15PM +0200, Enrico Forestieri wrote:

On Thu, Jun 25, 2015 at 10:16:21AM +0200, Enrico Forestieri wrote:

The third patch is the simpler one proposed by Guillaume. I have already
applied the better patch to master but there is a glitch with it.
Namely, if a preview snippet produces a blank image, its metrics are
not invalidated and thus the image would not be visible on the LyX
screen (actually, it is replaced by a small red square). I still have
to take a look at it, so I think that for the time being the simpler
patch will do.

Actually, it seems doing the right thing, as things such as $\quad$ or
$\relax$ are left as is, while $\raisebox{6mm}{}$ or $\raisebox{-6mm}{}$
produce a small vertical strip with the required characteristics (as
revealed by selecting the images in LyX).

I had a closer look and this behavior is independent of the patch.
Only when using dvipng blank images can be detected reliably, so what
described above occurs only when using the legacy route for generating
images.


So, I think the better patch can also be applied.

More rightly, as in stable the legacy route is always used as a fallback.


OK with me then.

This more or less completes this work? (For which thanks very much, to 
you and Guillaume.)


Richard

Re: test of math previews

[Enrico Forestieri]

On Thu, Jun 25, 2015 at 02:16:19PM -0400, Richard Heck wrote:
> On 06/25/2015 02:13 PM, Enrico Forestieri wrote:
> >On Thu, Jun 25, 2015 at 01:44:15PM +0200, Enrico Forestieri wrote:
> >>On Thu, Jun 25, 2015 at 10:16:21AM +0200, Enrico Forestieri wrote:
> >>>The third patch is the simpler one proposed by Guillaume. I have already
> >>>applied the better patch to master but there is a glitch with it.
> >>>Namely, if a preview snippet produces a blank image, its metrics are
> >>>not invalidated and thus the image would not be visible on the LyX
> >>>screen (actually, it is replaced by a small red square). I still have
> >>>to take a look at it, so I think that for the time being the simpler
> >>>patch will do.
> >>Actually, it seems doing the right thing, as things such as $\quad$ or
> >>$\relax$ are left as is, while $\raisebox{6mm}{}$ or $\raisebox{-6mm}{}$
> >>produce a small vertical strip with the required characteristics (as
> >>revealed by selecting the images in LyX).
> >I had a closer look and this behavior is independent of the patch.
> >Only when using dvipng blank images can be detected reliably, so what
> >described above occurs only when using the legacy route for generating
> >images.
> >
> >>So, I think the better patch can also be applied.
> >More rightly, as in stable the legacy route is always used as a fallback.
> 
> OK with me then.
> 
> This more or less completes this work?

I think so. The most important glitches were identified and solved.

> (For which thanks very much, to you
> and Guillaume.)

The credit goes mainly to Guillaume for his accurate testing.

-- 
Enrico

Re: test of math previews

[Richard Heck]

On 06/25/2015 02:58 PM, Enrico Forestieri wrote:

On Thu, Jun 25, 2015 at 02:16:19PM -0400, Richard Heck wrote:

This more or less completes this work?

I think so. The most important glitches were identified and solved.


(For which thanks very much, to you
and Guillaume.)

The credit goes mainly to Guillaume for his accurate testing.


Somebody had to write the code

rh

Re: test of math previews

[Guillaume M-M]

Le 25/06/2015 23:45, Richard Heck a écrit :

On 06/25/2015 02:58 PM, Enrico Forestieri wrote:

On Thu, Jun 25, 2015 at 02:16:19PM -0400, Richard Heck wrote:

This more or less completes this work?

I think so. The most important glitches were identified and solved.


(For which thanks very much, to you
and Guillaume.)

The credit goes mainly to Guillaume for his accurate testing.


Somebody had to write the code

rh




This is also what I believe. Thank you!

By the way will f3b03699 be backported as well? It seems that segfaults 
are almost guaranteed for people with previews on during Ctrl+C...