Dear all,

Thanks to the help of several LyX developers who addressed needed issues
for 2.2.0rc1, we are getting close. I would like to propose Tuesday as
the tentative date to tag and tar rc1. There are still some issues that
are pending but I hope that they can be resolved by then.

Georg is taking care of the translations of math environments, which hopefully
will receive enough attention from translators by Monday.

Günter has proposed layout patches to respond to changes in updated LaTeX class
files at the following two threads:
(acmsiggraph)
https://www.mail-archive.com/search?l=mid&q=ndgr7a%24kuf%241%40ger.gmane.org
(aastex6)
https://www.mail-archive.com/search?l=mid&q=ndj02g%244ht%241%40ger.gmane.org

In addition to the on-going layout discussion, there is still a question of
which Qt version we should release the Windows 2.2.0 binary with.

Shipping with Qt 5.6.0 instead of 5.5.1 would have the following advantages:

1. Uwe has been testing LyX built with 5.6.0 and it works well for him.

2. It is faster for Uwe.

3. It has long-term support.

4. It brings some features, such as better support for HiDPI. See here
for more: https://wiki.qt.io/New_Features_in_Qt_5.6

The disadvantages of using Qt 5.6.0 instead of 5.5.1 are the following:

1. Although Uwe tested and it works for him, no one else has tested. We would
thus lose all Windows-specific testing effort of the beta testers.

2. There is a mysterious crash with the MSVC2015
http://www.lyx.org/trac/ticket/10009 which does not happen with a merged
build. Since we do not understand why a merged build "fixes" the crash we do
not know what other problems might be created by the same cause.

3. We would ignore standard best practice in professional software 
development (which would be to switch compilers at the beginning of a
development cycle, not right before a release).

My own opinion is that we should use 5.5.1, unless there is an important bug
exposed in LyX that 5.6.0 fixes (and that 5.5.1 does not). Using 5.6.0 would
require switching compilers. I personally don't know much about the dangers of
this, but some LyX developers do know a lot about this topic and they are
worried. If someone knows more than I do about a topic, I trust them more than
I do myself.

In any case, Uwe has said that he can provide rc1 with both
Qt5.5.1/MSVC2010 and also a Qt 5.6 version. The only decision is which
should be the official one.

Regarding the Mac installers, Stephan can build with Qt versions 4.8.6,
5.4.2, 5.5.1 and 5.6.0. I think this is great and I'll be interested in
particular with which Qt versions the bugs #9992 and #9985 can be
reproduced. I think there is wide agreement that the official Mac
installer will ship with 5.5.1.

Where should we provide the unofficial installers (i.e. where should
they be stored)?

Should we mention the unofficial installers in the announce email? I am tempted
to say that we should not advertise them at all, and only mention them when a
user reports a problem with the 5.5.1 build so that we can see if the problem
is due to the Qt version. Otherwise the testing effort might be spread across
several installers instead of focused on the official installer.

The following two issues are not blockers. If there is no further action
or opinions regarding them, they will not make it into rc1 (and thus not
into 2.2.0). If someone thinks they are important and should be
addressed for rc1, please let us know.

1. At #10019 Guillaume fixed some ui regressions. The patch seems short
and logical and Stephan has tested it with success on Mac. I would still
like to see if someone can test it on Windows before it is included. If
anyone with Qt knowledge can take a look at the patch, that would also
be nice. I think normally I would not want any .ui changes in at this
point (for the reasons I listed on that ticket), but since they are
fixing regressions it would be nice to have. If someone thinks this is
important and we do not get a Windows user to test, we could potentially
put the patch in and make sure some of our Windows rc1 testers test
specifically this dialog.

2. Enrico has provided a patch regarding separators here:
https://www.mail-archive.com/search?l=mid&q=20160313003332.GA8700%40giove
It is a file format change, I think it would be nice to include because
otherwise users will have to get used to new behavior in 2.2.0 and then
get used to changed behavior again in 2.3.0. I do not understand the
issue though since I rarely use separators.

Thank you very much to those who have done work to discuss and fix the
pressing issues for 2.2.0. I know many such issues were not fun and I
really appreciate your help.

Scott

Attachment: signature.asc
Description: PGP signature

Reply via email to