Re: Plan for 2.2.0rc1

[Peter Kümmel]

Am 27.03.2016 um 21:33 schrieb Uwe Stöhr:

Am 19.03.2016 um 08:18 schrieb Scott Kostyshak:


Uwe, are you OK with shipping the official RC1 with 5.5.1 and MSVC 2010?
 From what I understand, that's the recommendation of Georg and Peter and
it seems like the safest approach.


This is not fine with me. My spare time is limited and I need to minimize the 
time to support LyX. With Qt 5.6 I can rely on a long term support. I already 
benefited from this with Qt 4.8.

I have been busy the last days but meanwhile used LyX 2.2git intensively for 2 
documentation projects with success. I used the Qt 5.6/MSVC2015 build and it is 
is faster than the Qt 5.5.1/MSVC 2010 build. I did not experience any 
regressions between both builds.

So all I need is the patch from Peter in master and then I am ready for LyX 
2.2RC1. Please note that the patch does not destroy anything, I applied it for 
all builds and I cannot see a problem.


I've commited the change.



So once again, please release RC1. I don't see why we should wait any longer.

regards Uwe

Re: Plan for 2.2.0rc1

[Scott Kostyshak]

On Sun, Mar 27, 2016 at 11:55:04PM +0100, Guillaume Munch wrote:

> It's hard to keep track, updates on the situation
> would be helpful.

OK I will send a new update soon. It is difficult for me to decide when
to start a new update thread. For example, we are discussing which Qt
version it is best to compile the official Windows binary with. We have
some discussion already in this thread. If I create a new update thread,
then the discussion will be spread over several threads. Perhaps it
would be better if I started a new thread for the long-running issues
where we are resolving disagreements (like the Qt version issue), and in
the update thread we can focus on discussing whether something is a
blocker or not, rather than the actual issue itself. That might make it
easier for others to follow.

Scott


signature.asc
Description: PGP signature

Re: Plan for 2.2.0rc1

[Guillaume Munch]

Le 27/03/2016 23:14, Scott Kostyshak a écrit :

On Sun, Mar 27, 2016 at 09:33:12PM +0200, Uwe Stöhr wrote:


So once again, please release RC1. I don't see why we should wait any
longer.


We all want to release RC1 as soon as possible. But there are blockers.
Please see the other threads discussing these issues. We are making
progress thanks to some developers working on issues that aren't fun for
them. Hopefully we will be able to release soon.



What are those blockers? It's hard to keep track, updates on the 
situation would be helpful. I have been wondering too why RC1 was not 
out already.

Re: Plan for 2.2.0rc1

[Scott Kostyshak]

On Sun, Mar 27, 2016 at 09:33:12PM +0200, Uwe Stöhr wrote:

> So once again, please release RC1. I don't see why we should wait any
> longer.

We all want to release RC1 as soon as possible. But there are blockers.
Please see the other threads discussing these issues. We are making
progress thanks to some developers working on issues that aren't fun for
them. Hopefully we will be able to release soon.

Scott


signature.asc
Description: PGP signature

Re: Plan for 2.2.0rc1

[Uwe Stöhr]

Am 19.03.2016 um 08:18 schrieb Scott Kostyshak:


Uwe, are you OK with shipping the official RC1 with 5.5.1 and MSVC 2010?
 From what I understand, that's the recommendation of Georg and Peter and
it seems like the safest approach.


This is not fine with me. My spare time is limited and I need to 
minimize the time to support LyX. With Qt 5.6 I can rely on a long term 
support. I already benefited from this with Qt 4.8.


I have been busy the last days but meanwhile used LyX 2.2git intensively 
for 2 documentation projects with success. I used the Qt 5.6/MSVC2015 
build and it is is faster than the Qt 5.5.1/MSVC 2010 build. I did not 
experience any regressions between both builds.


So all I need is the patch from Peter in master and then I am ready for 
LyX 2.2RC1. Please note that the patch does not destroy anything, I 
applied it for all builds and I cannot see a problem.


So once again, please release RC1. I don't see why we should wait any 
longer.


regards Uwe

Re: Plan for 2.2.0rc1

[Scott Kostyshak]

On Wed, Mar 23, 2016 at 03:29:56AM -0400, Scott Kostyshak wrote:
> On Mon, Mar 14, 2016 at 02:19:27AM +0100, Uwe Stöhr wrote:
> > Am 11.03.2016 um 09:48 schrieb Scott Kostyshak:
> 
> > So why can't there be a step by step rule in our development.lyx  that
> > everybody can understand? That would be the decision.
> 
> Good idea. I will work on this. I've spend a few hours today going
> through previous discussions. I'm learning a lot. I will come up with a
> proposal that we can discuss and then put in Development.lyx with the
> hopes of avoiding recurring discussions.

I have started a new thread, titled 'Proposal for a guide on updating
layouts', on lyx-devel:
https://www.mail-archive.com/search?l=mid=20160325063438.GA5517%40cotopaxi

Scott


signature.asc
Description: PGP signature

Re: Plan for 2.2.0rc1

[Scott Kostyshak]

On Mon, Mar 14, 2016 at 02:19:27AM +0100, Uwe Stöhr wrote:
> Am 11.03.2016 um 09:48 schrieb Scott Kostyshak:

> So why can't there be a step by step rule in our development.lyx  that
> everybody can understand? That would be the decision.

Good idea. I will work on this. I've spend a few hours today going
through previous discussions. I'm learning a lot. I will come up with a
proposal that we can discuss and then put in Development.lyx with the
hopes of avoiding recurring discussions.

Scott


signature.asc
Description: PGP signature

Re: Plan for 2.2.0rc1

[Scott Kostyshak]

On Wed, Mar 16, 2016 at 03:08:45PM +1300, Andrew Parsloe wrote:
> On 16/03/2016 10:05 a.m., Georg Baum wrote:
> >I don't know if MSVC 2012 or 2013 would be more reliable than 2015. What I
> >do know is that _any_ compiler switch introduces a risk. If we really
> >release 2.2.0 for windows compiled with a different compiler we throw away
> >a good part of the testing effort of the beta testers. My recommendation
> >would be to use qt 5.5.1 and MSVC 2010 for the official installer, and (if
> >Uwe is willing to invest the extra work) to provide a second one (clearly
> >marked as experimental) using MSVC 2015 and qt 5.6.0.
> 
> >Georg
> I like this proposal. I would be happy to use an experimental MSVC 2015/qt
> 5.6.0 LyX. The testing I've done for the 2.0 alphas and betas has all been
> 'real' work on 'real' documents. I have some big (> 500 pages), complicated
> documents which test LyX without specially trying -- multiple tables,
> figures, equations, footnotes, appendices, master/children, modules, used
> with complicated latex packages like minitoc. I would be happy to use an
> experimental LyX on them. (But I do save regularly and keep multiple copies
> in different places.)

Uwe, are you OK with shipping the official RC1 with 5.5.1 and MSVC 2010?
From what I understand, that's the recommendation of Georg and Peter and
it seems like the safest approach.

Scott


signature.asc
Description: PGP signature

Re: Plan for 2.2.0rc1

[Andrew Parsloe]

On 16/03/2016 10:05 a.m., Georg Baum wrote:
I don't know if MSVC 2012 or 2013 would be more reliable than 2015. 
What I do know is that _any_ compiler switch introduces a risk. If we 
really release 2.2.0 for windows compiled with a different compiler we 
throw away a good part of the testing effort of the beta testers. My 
recommendation would be to use qt 5.5.1 and MSVC 2010 for the official 
installer, and (if Uwe is willing to invest the extra work) to provide 
a second one (clearly marked as experimental) using MSVC 2015 and qt 
5.6.0.


Georg 
I like this proposal. I would be happy to use an experimental MSVC 
2015/qt 5.6.0 LyX. The testing I've done for the 2.0 alphas and betas 
has all been 'real' work on 'real' documents. I have some big (> 500 
pages), complicated documents which test LyX without specially trying -- 
multiple tables, figures, equations, footnotes, appendices, 
master/children, modules, used with complicated latex packages like 
minitoc. I would be happy to use an experimental LyX on them. (But I do 
save regularly and keep multiple copies in different places.)


Andrew

---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus

Re: Plan for 2.2.0rc1

[Georg Baum]

Uwe Stöhr wrote:

> This one of Georg's special posts ;-). I read it now again at least 5
> times and still don't understand what I should do.
> So I should
> 1. rename the existing layout (I guess that is meant with old), but to
> what name?

A name with the version in it: acmsiggraph080.layout

> 2. then I should update a layout with a generic name to the current
> version. But the current version is the one we have in Git and this is
> the old one. What is the one with the generic name? What is a generic
> name?

The generic name is the name without a version in it, in this case 
acmsiggraph.layout

> 3. I should add lyx2lyx code but that would be a fileformat change
> and in the discussion it turned out that I should not use fileformat
> changes.

Doing no file format change was a different proposal, not mine.

IMHO any new layout file is a file format change. The difference to your 
proposed solution is that lyx2lyx would simply change the document class 
from acmsiggraph to acmsiggraph080 instead of modifying the document 
contents.

> So why can't there be a step by step rule in our development.lyx  that
> everybody can understand? That would be the decision.

Yes, eventually we should have a common understanding how new class versions 
should be handled. Up to now we do not have a clear consensus yet.

> Günter maybe you have a clue how to go on. Please take over, you have
> hereby already my +1.

Giving a +1 to a solution that is not yet completely known somehow 
circumvents the intention of the +1 rule ;-/


Georg

Re: Plan for 2.2.0rc1

[Georg Baum]

Uwe Stöhr wrote:

> - we still haven't done anything to fix the 2 layouts that will produce
> uncompilable files (acmsiggraph and agutex). Please make a decision and
> I will apply it that way.. It would be embarassing when we include
> outdated layouts despite that we are aware of this.

These layouts do not produce uncompilable layouts in general. The 
acmsiggraph layout works fine with the latest version of the package unless 
you use one of the removed styles (which you would not do anyway if you 
followed the latest author guidelines).

The AGU interface was changed heavily, and the class authors did therefore 
use a new name of the class: agutex2015.cls vs. agutex.cls. A user can have 
installed both classes in the same LaTeX installation and use them in 
parallel for different documents. Since the class authors decided to use a 
new name for the new class we should do the same for the layout file (we can 
of course mark the old one as obsolete in the header line). I do not see any 
problem with introducing an agutex2015.layout in parallel to the existing 
layout and adjusting the template to the new version for 2.2.0 as long as 
lyx2lyx would not try to convert old files. This should be done by the user 
by changing the document class.


Georg

Re: Plan for 2.2.0rc1

[Georg Baum]

Jean-Marc Lasgouttes wrote:

> Le 14/03/2016 01:21, Uwe Stöhr a écrit :
>>> And I only propose a 2.2.0 release with Qt 5.5.1 and msvc10. We could
>>> not change compilers and Qt versions like socks.
>>
>> But the idea of LyX 2.2 was to use Qt 5.6 because this will be a long
>> term support release.

Yes. Unfortunately at least I did not realize that qt 5.6 needs a compiler 
switch when we initially discussed which qt version we want to have for 
2.2.0. Otherwise I would have warned earlier.

>  > I can build using Qt 5.6 and MSVC 2015 withyout
>> patch and LyX works quite well with it. (I only don't understand why one
>> needs a merged build.)

I guess you mean http://www.lyx.org/trac/ticket/10009? I do not understand 
that either, but I would bet that the merged build does only hide the real 
issue.

> Yes, one year  before release is a good time to update. But it now is
> not good. Actually the best time is just at the start of a development
> round.
> 
> I can understand that MSVC 2010 is too old, but 2015 seems really to
> young to me. What happened to versions 2012 and 2013? Do they stink?

MSVC 2010 is fine in general, it became an inofficial long term support 
release over time. Problems do only arise in conjunction with qt 5.6: It is 
non-trivial to build qt, so we should use the official binary packages, and 
those are no longer offered for MSVC 2010 starting with qt 5.6.

I don't know if MSVC 2012 or 2013 would be more reliable than 2015. What I 
do know is that _any_ compiler switch introduces a risk. If we really 
release 2.2.0 for windows compiled with a different compiler we throw away a 
good part of the testing effort of the beta testers.

My recommendation would be to use qt 5.5.1 and MSVC 2010 for the official 
installer, and (if Uwe is willing to invest the extra work) to provide a 
second one (clearly marked as experimental) using MSVC 2015 and qt 5.6.0.

For 2.3 we will require more C++11 anyway, so we should drop MSVC 2010 
support at the beginning of the development cycle.


Georg

Re: Plan for 2.2.0rc1

[Guillaume Munch]

Le 14/03/2016 20:15, Georg Baum a écrit :

Scott Kostyshak wrote:


Guillaume committed a fix that was tested by Enrico and Uwe. If you have
time to check it, take a look at b3bed292.


Works fine and fast for me (but I did never see the crash, only the
slowlyness).

Very nice, this is exactly what I had in mind but did not now how to
implement.




Thanks. TBH I found this solution a bit with chance: I started with
following this guide
https://doc.qt.io/qt-5/qtwidgets-itemviews-fetchmore-example.html and
then I found out that it worked well even if I fetched a million items
at once.

Re: Plan for 2.2.0rc1

[Georg Baum]

Scott Kostyshak wrote:

> Guillaume committed a fix that was tested by Enrico and Uwe. If you have
> time to check it, take a look at b3bed292.

Works fine and fast for me (but I did never see the crash, only the 
slowlyness).

Very nice, this is exactly what I had in mind but did not now how to 
implement.


Georg

Re: Plan for 2.2.0rc1

[Jean-Marc Lasgouttes]

Le 14/03/2016 01:21, Uwe Stöhr a écrit :

And I only propose a 2.2.0 release with Qt 5.5.1 and msvc10. We could
not change compilers and Qt versions like socks.


But the idea of LyX 2.2 was to use Qt 5.6 because this will be a long
term support release.


I am not sure why this matters to us right now. Qt 5.6 will probably 
become dependable after a few point releases, but we are not there yet. 
I do not care about long term support of something that will be released 
in two days (BTW, the final binaries are available and release date 
should be Wednesday).


> I can build using Qt 5.6 and MSVC 2015 withyout

patch and LyX works quite well with it. (I only don't understand why one
needs a merged build.)



So from my point of view a major LyX release is the perfect time to
switch to new versions. As the Qt people announced there won't be MSVC
2010 support anymore so we cannot rely any longer on MSVC 2010.


Yes, one year  before release is a good time to update. But it now is 
not good. Actually the best time is just at the start of a development 
round.


I can understand that MSVC 2010 is too old, but 2015 seems really to 
young to me. What happened to versions 2012 and 2013? Do they stink?


JMarc

Re: Plan for 2.2.0rc1

[Uwe Stöhr]

Am 11.03.2016 um 09:48 schrieb Scott Kostyshak:


My decision is that if someone gives you a +1 on a patch that is posted
for the two layouts and the lyx2lyx tests pass, then the patch should go
in.


Sorry, but this is no decision. A decision would be what to do, see below.

Why can't we vote and then act? I am tired of discussions without 
decisions. Now I still doesn't know how I should update the layouts:



Note that Georg stated:
https://www.mail-archive.com/search?l=mid=n8i0e7%24pr5%241%40ger.gmane.org
"I would very likely give a +1 to a patch that renames the old layout
file to a different name, updates the one with the generic name to the
current version (as you did it), and a lyx2lyx step that simply changes
the document class to the new name of the old layout file for old
documents. This would even be less work (no real lyx2lyx code needed)."


This one of Georg's special posts ;-). I read it now again at least 5 
times and still don't understand what I should do.

So I should
1. rename the existing layout (I guess that is meant with old), but to 
what name?
2. then I should update a layout with a generic name to the current 
version. But the current version is the one we have in Git and this is 
the old one. What is the one with the generic name? What is a generic name?
3. I should add lyx2lyx code but that would be a fileformat change and 
in the discussion it turned out that I should not use fileformat changes.


So why can't there be a step by step rule in our development.lyx  that 
everybody can understand? That would be the decision.



If I understood correctly, there is support from Kornel and Guenter for
versioned layouts. It seems to me we just need to figure out some of the
details. Guenter provided some points for discussion here:
https://www.mail-archive.com/search?l=mid=n8m1ff%24nnq%241%40ger.gmane.org

Also, I do not see any response to Guenter's patch posted at
https://www.mail-archive.com/search?l=mid=n90eas%24gab%241%40ger.gmane.org


This version is more or less what I proposed. Attached is a patch 
according to Günter's proposal.


Günter maybe you have a clue how to go on. Please take over, you have 
hereby already my +1.


thanks and regards
Uwe
 lib/layouts/acmsiggraph.layout | 123 +-
 lib/templates/ACM-siggraph.lyx | 363 +++--
 2 files changed, 245 insertions(+), 241 deletions(-)

diff --git a/lib/layouts/acmsiggraph.layout b/lib/layouts/acmsiggraph.layout
index 388b1dd..e4fc6bb 100644
--- a/lib/layouts/acmsiggraph.layout
+++ b/lib/layouts/acmsiggraph.layout
@@ -1,27 +1,30 @@
 #% Do not delete he line below; configure depends on this
 #  \DeclareLaTeXClass[acmsiggraph,comment.sty,lineno.sty]{ACM SIGGRAPH}
 #  \DeclareCategory{Articles}
-# ACM SIGGRAPH acmsiggraph textclass definition file.
+# ACM SIGGRAPH textclass definition file.
 # Author : Uwe Stöhr
 #
-# We use acmsiggraph.cls, the Sigplan class file that can be
-# downloaded from
-# http://www.siggraph.org/publications/instructions
+# The required LaTeX class file "acmsiggraph.cls", a LaTex template
+# BibTeX style and template, and the documentation is available at
+# http://www.siggraph.org/learn/instructions-authors
 #
-# ACM style files can be obtained at
-# http://www.acm.org/sigs/pubs/proceed/template.html
+# Versions:
+#  siggraph 0.9:  initial layout
+#  siggraph 0.92: Removed styles "TOG project URL", "TOG video URL", 
+#   "TOG data URL", and "TOG code URL". The corresponding 
+#   LaTeX commands are no longer supported.
 
 Format 59
 
-Columns 1
-Sides   1
+Columns1
+Sides  1
 SecNumDepth3
 TocDepth   3
-DefaultStyleStandard
+DefaultStyle   Standard
 
 ClassOptions
   FontSize 9|10|11|12
-End  
+End
 
 DefaultFont
   Family   Roman
@@ -33,15 +36,15 @@ EndFont
 
 
 Style Standard
-  Category  MainText
+  Category MainText
   Margin   Static
-  LatexTypeParagraph
-  LatexNamedummy
+  LatexTypeParagraph
+  LatexNamedummy
   ParIndentMM
   ParSkip  0.4
   AlignBlock
-  AlignPossibleBlock, Left, Right, Center
-  LabelTypeNo_Label
+  AlignPossibleBlock, Left, Right, Center
+  LabelTypeNo_Label
 End
 
 
@@ -96,39 +99,51 @@ Style "TOG number"
 End
 
 
-Style "TOG article DOI"
+Style "Set_copyright"
CopyStyle   "TOG online ID"
-   LatexName   TOGarticleDOI
-   LabelString "Article DOI:"
+   LatexName   setcopyright
+   LabelString "Copyright type:"
 End
 
 
-Style "TOG project URL"
+Style "Copyright_year"
CopyStyle   "TOG online ID"
-   LatexName   TOGprojectURL
-   KeepEmpty   1
-   LabelString "Project URL:"
+   

Re: Plan for 2.2.0rc1

[Uwe Stöhr]

Am 11.03.2016 um 12:55 schrieb Peter Kümmel:


No, not necessary for builds with msvc10.

And I only propose a 2.2.0 release with Qt 5.5.1 and msvc10. We could
not change compilers and Qt versions like socks.


But the idea of LyX 2.2 was to use Qt 5.6 because this will be a long 
term support release. I can build using Qt 5.6 and MSVC 2015 withyout 
patch and LyX works quite well with it. (I only don't understand why one 
needs a merged build.)


So from my point of view a major LyX release is the perfect time to 
switch to new versions. As the Qt people announced there won't be MSVC 
2010 support anymore so we cannot rely any longer on MSVC 2010.


regards Uwe

Re: Plan for 2.2.0rc1

[Uwe Stöhr]

Am 11.03.2016 um 04:49 schrieb Scott Kostyshak:


Will you be able to also release an installer using Qt 5.5.1 and Qt
4.8.6? This would be nice so that if we get a bug report that we can't
reproduce, we can ask them to try the 5.5.1 installer and that way we
can see whether the bug is due to LyX or to Qt.


OK, I can provide
- Qt 5.6 using MSVC2015
- Qt 5.5.1 using MSVC2010
- Qt 4.8.6 using MSVC2010

regards Uwe

Re: Plan for 2.2.0rc1

[Guenter Milde]

On 2016-03-11, Scott Kostyshak wrote:
> On Thu, Mar 10, 2016 at 10:49:16PM -0500, Scott Kostyshak wrote:
>> On Thu, Mar 10, 2016 at 12:50:25AM +0100, Uwe Stöhr wrote:

>> > - we still haven't done anything to fix the 2 layouts that will produce
>> > uncompilable files (acmsiggraph and agutex). Please make a decision and I
>> > will apply it that way. It would be embarassing when we include outdated
>> > layouts despite that we are aware of this.

>> OK I will take a look at this.

> My decision is that if someone gives you a +1 on a patch that is posted
> for the two layouts and the lyx2lyx tests pass, then the patch should
> go in.

> Note that Georg stated:
> https://www.mail-archive.com/search?l=mid=n8i0e7%24pr5%241%40ger.gmane.org
> "I would very likely give a +1 to a patch that renames the old layout
> file to a different name, updates the one with the generic name to the
> current version (as you did it), and a lyx2lyx step that simply changes
> the document class to the new name of the old layout file for old
> documents. This would even be less work (no real lyx2lyx code needed)."

Sounds reasonable.

In the case of acmsiggraph, we could also just drop the styles corresponding
to LaTeX commands/environments that are no longer supported by the new
documentclass version. (See the post/patch linked below).

> If I understood correctly, there is support from Kornel and Guenter for
> versioned layouts. It seems to me we just need to figure out some of the
> details. Guenter provided some points for discussion here:
> https://www.mail-archive.com/search?l=mid=n8m1ff%24nnq%241%40ger.gmane.org

> Also, I do not see any response to Guenter's patch posted at
> https://www.mail-archive.com/search?l=mid=n90eas%24gab%241%40ger.gmane.org

Günter

Re: Plan for 2.2.0rc1

[Peter Kümmel]

Am 11.03.2016 um 04:49 schrieb Scott Kostyshak:

On Thu, Mar 10, 2016 at 12:50:25AM +0100, Uwe Stöhr wrote:

Am 09.03.2016 um 04:37 schrieb Scott Kostyshak:


8. What am I missing?


- I need for Qt 5.6 this patch from Peter to be applied:
https://github.com/syntheticpp/lyx/commit/2470fb442cb2b04a69b2030f28f1da60221556a7?diff=unified


Peter, do you proppose this for 2.2.0?


No, not necessary for builds with msvc10.

And I only propose a 2.2.0 release with Qt 5.5.1 and msvc10. We could not 
change compilers and Qt versions like socks.




- bug http://www.lyx.org/trac/ticket/10009


I'm glad you figured out a workaround for that.


- we still haven't done anything to fix the 2 layouts that will produce
uncompilable files (acmsiggraph and agutex). Please make a decision and I
will apply it that way.. It would be embarassing when we include outdated
layouts despite that we are aware of this.


OK I will take a look at this.


Uwe and Stephan, which Qt versions will you be able to release RC1 with?


In my opinion the RC should feature Qt 5.6. For Qt 5.6 I need MSVC 2015
Update 2 (no other MSVC version could be used). Qt 5.6 will be released next
week and I assume MSVC 2015 Update 2 too (there is already an RC out). If
http://www.lyx.org/trac/ticket/10009 cannot be solved I will have to try out
MinGW but there are reports that MinGW misses some things necessary for
Windows 10.


Will you be able to also release an installer using Qt 5.5.1 and Qt
4.8.6? This would be nice so that if we get a bug report that we can't
reproduce, we can ask them to try the 5.5.1 installer and that way we
can see whether the bug is due to LyX or to Qt.

Scott

Re: Plan for 2.2.0rc1

[Scott Kostyshak]

On Thu, Mar 10, 2016 at 10:49:16PM -0500, Scott Kostyshak wrote:
> On Thu, Mar 10, 2016 at 12:50:25AM +0100, Uwe Stöhr wrote:

> > - we still haven't done anything to fix the 2 layouts that will produce
> > uncompilable files (acmsiggraph and agutex). Please make a decision and I
> > will apply it that way.. It would be embarassing when we include outdated
> > layouts despite that we are aware of this.
> 
> OK I will take a look at this.

My decision is that if someone gives you a +1 on a patch that is posted
for the two layouts and the lyx2lyx tests pass, then the patch should go
in.

Note that Georg stated:
https://www.mail-archive.com/search?l=mid=n8i0e7%24pr5%241%40ger.gmane.org
"I would very likely give a +1 to a patch that renames the old layout
file to a different name, updates the one with the generic name to the
current version (as you did it), and a lyx2lyx step that simply changes
the document class to the new name of the old layout file for old
documents. This would even be less work (no real lyx2lyx code needed)."

If I understood correctly, there is support from Kornel and Guenter for
versioned layouts. It seems to me we just need to figure out some of the
details. Guenter provided some points for discussion here:
https://www.mail-archive.com/search?l=mid=n8m1ff%24nnq%241%40ger.gmane.org

Also, I do not see any response to Guenter's patch posted at
https://www.mail-archive.com/search?l=mid=n90eas%24gab%241%40ger.gmane.org

Scott


signature.asc
Description: PGP signature

Re: Plan for 2.2.0rc1

[Scott Kostyshak]

On Thu, Mar 10, 2016 at 12:50:25AM +0100, Uwe Stöhr wrote:
> Am 09.03.2016 um 04:37 schrieb Scott Kostyshak:
> 
> >8. What am I missing?
> 
> - I need for Qt 5.6 this patch from Peter to be applied:
> https://github.com/syntheticpp/lyx/commit/2470fb442cb2b04a69b2030f28f1da60221556a7?diff=unified

Peter, do you proppose this for 2.2.0?

> - bug http://www.lyx.org/trac/ticket/10009

I'm glad you figured out a workaround for that.

> - we still haven't done anything to fix the 2 layouts that will produce
> uncompilable files (acmsiggraph and agutex). Please make a decision and I
> will apply it that way.. It would be embarassing when we include outdated
> layouts despite that we are aware of this.

OK I will take a look at this.

> >Uwe and Stephan, which Qt versions will you be able to release RC1 with?
> 
> In my opinion the RC should feature Qt 5.6. For Qt 5.6 I need MSVC 2015
> Update 2 (no other MSVC version could be used). Qt 5.6 will be released next
> week and I assume MSVC 2015 Update 2 too (there is already an RC out). If
> http://www.lyx.org/trac/ticket/10009 cannot be solved I will have to try out
> MinGW but there are reports that MinGW misses some things necessary for
> Windows 10.

Will you be able to also release an installer using Qt 5.5.1 and Qt
4.8.6? This would be nice so that if we get a bug report that we can't
reproduce, we can ask them to try the 5.5.1 installer and that way we
can see whether the bug is due to LyX or to Qt.

Scott


signature.asc
Description: PGP signature

Re: Plan for 2.2.0rc1

[Scott Kostyshak]

On Wed, Mar 09, 2016 at 09:40:38PM +0100, Georg Baum wrote:
> Scott Kostyshak wrote:
> 
> > Dear all,
> > 
> > I think we are getting close to being able to release 2.2.0rc1. We have
> > not discussed whether to release a beta3 or move directly to rc1. If
> > others disagree with moving directly to rc1, please do speak up and
> > let's discuss.
> 
> rc1 is fine.
> 
> > Below are the items to consider fixing for rc1.
> > 
> > 1. Although #9968 is not a fresh regression (there was a problem in
> > 2.1.x), it seems to be worse for 2.2.0dev (although I'm guessing this is
> > due to different Qt versions). Georg has posted a patch which hopefully
> > improves the situation. We just need testing of the patch and we need
> > to decide whether it is good enough for 2.2.0.
> 
> Unfortuntely the patch will not really help, the effect is barely visible.
> 
> I would postpone this. We neither have somebody who can debug the crash on 
> windows, nor somebody who knows QAbstractItemModel better than me stepped up 
> to implement symbol computation on demand (triggered by scrolling through 
> the list). Of course it would be nice to fix at least the crash, but I don't 
> see how that could be done. My previous attempts to explain how to get a 
> backtrace from the MSVC debugger failed.

Guillaume committed a fix that was tested by Enrico and Uwe. If you have
time to check it, take a look at b3bed292.

> > 8. What am I missing?
> 
> There is also the question which compiler to use on windows. In general it 
> is unwise to do all the testing with one compiler and then switch the 
> compiler for the release. IMHO the only justification for switching 
> compilers would be known serious problems with the old one which are fixed 
> by the new one, so that the risk of yet unknown problems would be 
> acceptable.

+1

Scott


signature.asc
Description: PGP signature

Re: Plan for 2.2.0rc1

[Uwe Stöhr]

Am 09.03.2016 um 04:37 schrieb Scott Kostyshak:


8. What am I missing?


- I need for Qt 5.6 this patch from Peter to be applied:
https://github.com/syntheticpp/lyx/commit/2470fb442cb2b04a69b2030f28f1da60221556a7?diff=unified

- bug http://www.lyx.org/trac/ticket/10009

- we still haven't done anything to fix the 2 layouts that will produce 
uncompilable files (acmsiggraph and agutex). Please make a decision and 
I will apply it that way.. It would be embarassing when we include 
outdated layouts despite that we are aware of this.



Uwe and Stephan, which Qt versions will you be able to release RC1 with?


In my opinion the RC should feature Qt 5.6. For Qt 5.6 I need MSVC 2015 
Update 2 (no other MSVC version could be used). Qt 5.6 will be released 
next week and I assume MSVC 2015 Update 2 too (there is already an RC 
out). If http://www.lyx.org/trac/ticket/10009 cannot be solved I will 
have to try out MinGW but there are reports that MinGW misses some 
things necessary for Windows 10.


thanks and regards
Uwe

Re: Plan for 2.2.0rc1

[Georg Baum]

Scott Kostyshak wrote:

> Dear all,
> 
> I think we are getting close to being able to release 2.2.0rc1. We have
> not discussed whether to release a beta3 or move directly to rc1. If
> others disagree with moving directly to rc1, please do speak up and
> let's discuss.

rc1 is fine.

> Below are the items to consider fixing for rc1.
> 
> 1. Although #9968 is not a fresh regression (there was a problem in
> 2.1.x), it seems to be worse for 2.2.0dev (although I'm guessing this is
> due to different Qt versions). Georg has posted a patch which hopefully
> improves the situation. We just need testing of the patch and we need
> to decide whether it is good enough for 2.2.0.

Unfortuntely the patch will not really help, the effect is barely visible.

I would postpone this. We neither have somebody who can debug the crash on 
windows, nor somebody who knows QAbstractItemModel better than me stepped up 
to implement symbol computation on demand (triggered by scrolling through 
the list). Of course it would be nice to fix at least the crash, but I don't 
see how that could be done. My previous attempts to explain how to get a 
backtrace from the MSVC debugger failed.

> 8. What am I missing?

There is also the question which compiler to use on windows. In general it 
is unwise to do all the testing with one compiler and then switch the 
compiler for the release. IMHO the only justification for switching 
compilers would be known serious problems with the old one which are fixed 
by the new one, so that the risk of yet unknown problems would be 
acceptable.


Georg

Re: Plan for 2.2.0rc1

[José Matos]

On Wednesday, March 09, 2016 01:51:26 PM Jean-Marc Lasgouttes wrote:
> Moving to rc1 is OK.

+1
-- 
José Abílio

Re: Plan for 2.2.0rc1

[Stephan Witt]

Am 09.03.2016 um 04:37 schrieb Scott Kostyshak :
> 
> Dear all,
> 
> I think we are getting close to being able to release 2.2.0rc1. We have
> not discussed whether to release a beta3 or move directly to rc1. If
> others disagree with moving directly to rc1, please do speak up and
> let's discuss.
> 
> Below are the items to consider fixing for rc1.
> 
> 1. Although #9968 is not a fresh regression (there was a problem in
> 2.1.x), it seems to be worse for 2.2.0dev (although I'm guessing this is
> due to different Qt versions). Georg has posted a patch which hopefully
> improves the situation. We just need testing of the patch and we need
> to decide whether it is good enough for 2.2.0.

On my Mac LyX is using 100% CPU for about 35 seconds to fill the symbols
grid with and without the patch. Perhaps it’s a little bit faster with 
the patch. It didn’t crash.

> 
> 2. #9979 is not a regression, but would be nice to fix.
> 
> 3. We need a dev experienced in lyx2lyx to complete the patch posted
> here regarding the "multi-page table" change (see also the discussion
> that followed):
> https://www.mail-archive.com/search?l=mid=56B365D0.4040203%40free.fr
> 
> 4. There are a couple of issues that have been difficult to reproduce by
> others, such as #9992 and #9644. It would be nice to see if they have
> been triggered by different Qt versions. This could be done by releasing
> binaries built using a few different Qt versions, such as 4.8.6, 5.5.1,
> and 5.6.0rc1. Perhaps #9971 also falls in this category? Similarly,
> hopefully Lange can confirm that #10002 has indeed been fixed.
> 
> 5. Enrico understands #9990 so hopefully he will have advice on how to
> proceed.
> 
> 6. I'm guessing a fix for #10007 will not be difficult since it is a
> recent regression.
> 
> 7. There is a regression in #9663 that Guillaume and I have discussed
> but we are stuck. Please let us know if someone is interested in taking
> a look to see if they have come across the regression and have a
> reproducible example; or they have an idea at what triggered the
> regression.
> 
> 8. What am I missing?
> 
> Uwe and Stephan, which Qt versions will you be able to release RC1 with?

I’m able to build with 4.8.6, 5.4.2, 5.5.1 and 5.6.0(-rc).

> It would be nice to build using a few different Qt versions so that we
> can see if a few bugs are due to regressions in LyX or regressions in
> Qt. See e.g. item 4 above. We do not have to put all of the installers
> on the ftp in case that could be confusing for users. But it would be
> nice to have them available somewhere (I can put them on my Dropbox and
> share them), so that we can ask someone to test them.
> 
> Uwe and Stephan, similar to above, which Qt version are you thinking
> will be the best to release LyX 2.2.0 with? Qt 5.6.0 is scheduled to be
> released on March 16th, but a further delay would not be surprising.
> Regarding Mac and Windows, which issues (experienced by LyX users) are
> fixed in Qt 5.6.0 that are not fixed with Qt 5.5.1?

To be honest I don’t know the answer for this.

Stephan

Re: Plan for 2.2.0rc1

[Jean-Marc Lasgouttes]

Le 09/03/2016 04:37, Scott Kostyshak a écrit :

Dear all,

I think we are getting close to being able to release 2.2.0rc1. We have
not discussed whether to release a beta3 or move directly to rc1. If
others disagree with moving directly to rc1, please do speak up and
let's discuss.


Moving to rc1 is OK.


6. I'm guessing a fix for #10007 will not be difficult since it is a
recent regression.


I'm looking at it, but the underlying code is fragile.


8. What am I missing?


I will propose a workaround for #9917.

There is also the painting problems in #9971, but I have absolutely no 
idea about it.


JMarc

Plan for 2.2.0rc1

[Scott Kostyshak]

Dear all,

I think we are getting close to being able to release 2.2.0rc1. We have
not discussed whether to release a beta3 or move directly to rc1. If
others disagree with moving directly to rc1, please do speak up and
let's discuss.

Below are the items to consider fixing for rc1.

1. Although #9968 is not a fresh regression (there was a problem in
2.1.x), it seems to be worse for 2.2.0dev (although I'm guessing this is
due to different Qt versions). Georg has posted a patch which hopefully
improves the situation. We just need testing of the patch and we need
to decide whether it is good enough for 2.2.0.

2. #9979 is not a regression, but would be nice to fix.

3. We need a dev experienced in lyx2lyx to complete the patch posted
here regarding the "multi-page table" change (see also the discussion
that followed):
https://www.mail-archive.com/search?l=mid=56B365D0.4040203%40free.fr

4. There are a couple of issues that have been difficult to reproduce by
others, such as #9992 and #9644. It would be nice to see if they have
been triggered by different Qt versions. This could be done by releasing
binaries built using a few different Qt versions, such as 4.8.6, 5.5.1,
and 5.6.0rc1. Perhaps #9971 also falls in this category? Similarly,
hopefully Lange can confirm that #10002 has indeed been fixed.

5. Enrico understands #9990 so hopefully he will have advice on how to
proceed.

6. I'm guessing a fix for #10007 will not be difficult since it is a
recent regression.

7. There is a regression in #9663 that Guillaume and I have discussed
but we are stuck. Please let us know if someone is interested in taking
a look to see if they have come across the regression and have a
reproducible example; or they have an idea at what triggered the
regression.

8. What am I missing?

Uwe and Stephan, which Qt versions will you be able to release RC1 with?
It would be nice to build using a few different Qt versions so that we
can see if a few bugs are due to regressions in LyX or regressions in
Qt. See e.g. item 4 above. We do not have to put all of the installers
on the ftp in case that could be confusing for users. But it would be
nice to have them available somewhere (I can put them on my Dropbox and
share them), so that we can ask someone to test them.

Uwe and Stephan, similar to above, which Qt version are you thinking
will be the best to release LyX 2.2.0 with? Qt 5.6.0 is scheduled to be
released on March 16th, but a further delay would not be surprising.
Regarding Mac and Windows, which issues (experienced by LyX users) are
fixed in Qt 5.6.0 that are not fixed with Qt 5.5.1?

Scott


signature.asc
Description: PGP signature