Re: Any rc1 blockers?

[Jürgen Spitzmüller]

Am Montag, den 18.09.2017, 11:41 +0200 schrieb Jürgen Spitzmüller:
> > What is not reasonable is that we insist in disabling aspell and
> > enchant what hunspell is used. What is the point? (I know, it is
> > separate from your effort).
> 
> I do not find this reasonable either.

Fixed in master and 2.3.x.

Jürgen

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

Re: Any rc1 blockers?

[Jürgen Spitzmüller]

2017-09-18 13:14 GMT+02:00 Kornel Benko :

> > CMakers, does -DLYX_EXTERNAL_HUNSPELL=OFF -DLYX_USE_ASPELL=ON
> > -DLYX_USE_ENCHANT=ON give you all three spellers as option?
>
> Yes, I see all three spellers as option.
> To be sure I don't use external HUNSPELL:
> # ldd /usr/local/bin/lyx2.4 | egrep -i hun
> ==> Exit 1
> (otherwise should have seen "/usr/lib/x86_64-linux-gnu/libhunspell.so")
>

Thanks, Kornel. Then there is even less reason for the automake behavior.

Jürgen

Re: Any rc1 blockers?

[Kornel Benko]

Am Montag, 18. September 2017 um 12:14:30, schrieb Jürgen Spitzmüller 

> 2017-09-18 11:53 GMT+02:00 Jean-Marc Lasgouttes :
> 
> > What could make it useful is that you could decide to enforce the use of
> > the new C++ API (if we ever get it to work). If the system hunspell is not
> > good enough, one can fall back on the bundled one.
> >
> 
> Yes, right. I did not mean, however, that I don't find the internal
> hunspell unreasonable, but the fact that it disables the other spellers.
> The point probably is that automake is nowadays only used by people on
> machnies that have hunspell installed anyway. Which does not mean that it
> shouldn't be possible to use it.
> 
> CMakers, does -DLYX_EXTERNAL_HUNSPELL=OFF -DLYX_USE_ASPELL=ON
> -DLYX_USE_ENCHANT=ON give you all three spellers as option?

Yes, I see all three spellers as option.
To be sure I don't use external HUNSPELL:
# ldd /usr/local/bin/lyx2.4 | egrep -i hun
==> Exit 1
(otherwise should have seen "/usr/lib/x86_64-linux-gnu/libhunspell.so")

> Jürgen
> 
> 
> >
> > JMarc

Kornel

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

Re: Any rc1 blockers?

[Jean-Marc Lasgouttes]

Le 18/09/2017 à 12:12, Stephan Witt a écrit :

I’m using a self-compiled hunspell with explicit given include path. Please, 
don’t break this for RC1.


I do not think we are going to make big changes like that for rc1. We 
just want to update the built-in one.


JMarc

Re: Any rc1 blockers?

[Jürgen Spitzmüller]

2017-09-18 12:12 GMT+02:00 Stephan Witt :

> I’m using a self-compiled hunspell with explicit given include path.
> Please, don’t break this for RC1.
>

I will commit to master only first and ask CMakers to check before I
backport.

Jürgen


>
> Stephan

Re: Any rc1 blockers?

[Jürgen Spitzmüller]

2017-09-18 11:53 GMT+02:00 Jean-Marc Lasgouttes :

> What could make it useful is that you could decide to enforce the use of
> the new C++ API (if we ever get it to work). If the system hunspell is not
> good enough, one can fall back on the bundled one.
>

Yes, right. I did not mean, however, that I don't find the internal
hunspell unreasonable, but the fact that it disables the other spellers.
The point probably is that automake is nowadays only used by people on
machnies that have hunspell installed anyway. Which does not mean that it
shouldn't be possible to use it.

CMakers, does -DLYX_EXTERNAL_HUNSPELL=OFF -DLYX_USE_ASPELL=ON
-DLYX_USE_ENCHANT=ON give you all three spellers as option?

Jürgen


>
> JMarc
>

Re: Any rc1 blockers?

[Stephan Witt]

Am 18.09.2017 um 11:53 schrieb Jean-Marc Lasgouttes :
> 
> Le 18/09/2017 à 11:41, Jürgen Spitzmüller a écrit :
>> I do not find this reasonable either. But it seems that nobody ever used 
>> internal hunspell with automake anyway (otherwise, we would have known 
>> earlier that it did not work).
> 
> What could make it useful is that you could decide to enforce the use of the 
> new C++ API (if we ever get it to work). If the system hunspell is not good 
> enough, one can fall back on the bundled one.

I’m using a self-compiled hunspell with explicit given include path. Please, 
don’t break this for RC1.

Stephan 

Re: Any rc1 blockers?

[Jean-Marc Lasgouttes]

Le 18/09/2017 à 11:41, Jürgen Spitzmüller a écrit :
I do not find this reasonable either. But it seems that nobody ever used 
internal hunspell with automake anyway (otherwise, we would have known 
earlier that it did not work).


What could make it useful is that you could decide to enforce the use of 
the new C++ API (if we ever get it to work). If the system hunspell is 
not good enough, one can fall back on the bundled one.


JMarc

Re: Any rc1 blockers?

[Jürgen Spitzmüller]

2017-09-18 11:34 GMT+02:00 Jean-Marc Lasgouttes :

> What is not reasonable is that we insist in disabling aspell and enchant
> what hunspell is used. What is the point? (I know, it is separate from your
> effort).
>

I do not find this reasonable either. But it seems that nobody ever used
internal hunspell with automake anyway (otherwise, we would have known
earlier that it did not work).

Jürgen


>
> JMarc
>

Re: Any rc1 blockers?

[Jean-Marc Lasgouttes]

Le 17/09/2017 à 18:28, Jürgen Spitzmüller a écrit :

Are you still trying? I have successfully updated to 1.6.2 here (more
or less the same subset, apart from one pair of files that has been
dropped from hunspell meanwhile, dictmgr.{cpp,h}).


I did not find time to progress on this front. Please commit what you 
came up with.



Builds and works fine at least with automake. However, I had to do
tweak spell.m4 in order to make internal hunspell functional at all
with automake (USE_HUNSPELL was not defined with internal hunspell):



This looks reasonable. What is not reasonable is that we insist in 
disabling aspell and enchant what hunspell is used. What is the point? 
(I know, it is separate from your effort).


JMarc

Re: Any rc1 blockers?

[Jürgen Spitzmüller]

Am Dienstag, den 12.09.2017, 12:20 +0200 schrieb Jean-Marc Lasgouttes:
> Le 12/09/2017 à 11:41, Jean-Marc Lasgouttes a écrit :
> > Le 12/09/2017 à 10:59, Jürgen Spitzmüller a écrit :
> > > Hunspell is also outdated.
> > 
> > I'll have a go at it.
> 
> Not so easy after all. What we have is a trimmed down version of
> 1.3.3, 
> but I am not sure of what we want to trim down.

Are you still trying? I have successfully updated to 1.6.2 here (more
or less the same subset, apart from one pair of files that has been
dropped from hunspell meanwhile, dictmgr.{cpp,h}).

Builds and works fine at least with automake. However, I had to do
tweak spell.m4 in order to make internal hunspell functional at all
with automake (USE_HUNSPELL was not defined with internal hunspell):

diff --git a/config/spell.m4 b/config/spell.m4
index c733dab2b4..fdac0b1f07 100644
--- a/config/spell.m4
+++ b/config/spell.m4
@@ -97,6 +97,7 @@ dnl the user wanted to use the included hunspell, so
do not check for the other
lyx_use_aspell=false
lyx_use_enchant=false
lyx_use_hunspell=true
+   AC_DEFINE(USE_HUNSPELL, 1, [Define as 1 to use the
hunspell library])
lyx_flags="$lyx_flags use-hunspell"
else
CHECK_WITH_ASPELL

Jürgen

> 
> JMarc

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

Re: Any rc1 blockers?

[Uwe Stöhr]

El 12.09.2017 a las 18:49, Scott Kostyshak escribió:


Thanks. Uwe, please go ahead.


Done.

regards Uwe

Re: Any rc1 blockers?

[Scott Kostyshak]

On Tue, Sep 12, 2017 at 10:59:00AM +0200, Jürgen Spitzmüller wrote:
> 2017-09-12 10:15 GMT+02:00 Jean-Marc Lasgouttes :
> 
> > Le 11/09/2017 à 23:34, Uwe Stöhr a écrit :
> >
> >> Concerning libiconv 1.15, there are not yet opinions.
> >>
> >
> > Updating looks like a good idea IMO.
> >
> > We could also consider updating boost to 1.65.1 (we have 1.62 right now).
> >
> 
> Hunspell is also outdated.

As for boost and hunspell, I trust your judgments. I'll leave it up to
you to decide whether the changes are worth the risk.

Scott


signature.asc
Description: PGP signature

Re: Any rc1 blockers?

[Scott Kostyshak]

On Tue, Sep 12, 2017 at 10:15:25AM +0200, Jean-Marc Lasgouttes wrote:
> Le 11/09/2017 à 23:34, Uwe Stöhr a écrit :
> > Concerning libiconv 1.15, there are not yet opinions.
> 
> Updating looks like a good idea IMO.

Thanks. Uwe, please go ahead.

Scott


signature.asc
Description: PGP signature

Re: Any rc1 blockers?

[Jürgen Spitzmüller]

2017-09-12 13:09 GMT+02:00 Jürgen Spitzmüller :

> 2017-09-12 12:20 GMT+02:00 Jean-Marc Lasgouttes :
>
>> Le 12/09/2017 à 11:41, Jean-Marc Lasgouttes a écrit :
>> Not so easy after all. What we have is a trimmed down version of 1.3.3,
>> but I am not sure of what we want to trim down.
>>
>
> Just the relevant READMEs, licenses and src/ (without hunspell2, which is
> the forthcoming version that won't work anyway with LyX yet).
>

and msvc/config.h > win_api/config.h

Jürgen


>
> Jürgen
>
>
>>
>> JMarc
>>
>
>

Re: Any rc1 blockers?

[Jürgen Spitzmüller]

2017-09-12 12:20 GMT+02:00 Jean-Marc Lasgouttes :

> Le 12/09/2017 à 11:41, Jean-Marc Lasgouttes a écrit :
> Not so easy after all. What we have is a trimmed down version of 1.3.3,
> but I am not sure of what we want to trim down.
>

Just the relevant READMEs, licenses and src/ (without hunspell2, which is
the forthcoming version that won't work anyway with LyX yet).

Jürgen


>
> JMarc
>

Re: Any rc1 blockers?

[Jean-Marc Lasgouttes]

Le 12/09/2017 à 11:41, Jean-Marc Lasgouttes a écrit :

Le 12/09/2017 à 10:59, Jürgen Spitzmüller a écrit :

Hunspell is also outdated.


I'll have a go at it.


Not so easy after all. What we have is a trimmed down version of 1.3.3, 
but I am not sure of what we want to trim down.


JMarc

Re: Any rc1 blockers?

[Jean-Marc Lasgouttes]

Le 12/09/2017 à 10:59, Jürgen Spitzmüller a écrit :

Hunspell is also outdated.


I'll have a go at it.

JMarc

Re: Any rc1 blockers?

[Jürgen Spitzmüller]

2017-09-12 10:15 GMT+02:00 Jean-Marc Lasgouttes :

> Le 11/09/2017 à 23:34, Uwe Stöhr a écrit :
>
>> Concerning libiconv 1.15, there are not yet opinions.
>>
>
> Updating looks like a good idea IMO.
>
> We could also consider updating boost to 1.65.1 (we have 1.62 right now).
>

Hunspell is also outdated.

Jürgen


>
> JMarc
>

Re: Any rc1 blockers?

[Jean-Marc Lasgouttes]

Le 11/09/2017 à 23:34, Uwe Stöhr a écrit :

Concerning libiconv 1.15, there are not yet opinions.


Updating looks like a good idea IMO.

We could also consider updating boost to 1.65.1 (we have 1.62 right now).

JMarc

Re: Any rc1 blockers?

[Scott Kostyshak]

On Mon, Sep 11, 2017 at 11:34:34PM +0200, Uwe Stöhr wrote:
> El 10.09.2017 a las 01:38, Uwe Stöhr escribió:
> 
> > Just the patch for
> > http://www.lyx.org/trac/ticket/10679
> > should go in before.
> 
> Jürgen found now a fix.

Nice.

> Meanwhile I testes LyX 2.3 a lot the last 2 days. It is stable for my needs.

Good to know.

> Concerning libiconv 1.15, there are not yet opinions.

Let's wait until Friday. If no opinion by then, feel free to put it in.

> Concerning lyx2lyx I guess we will get more bug reports first after RC1 is
> out and people try to convert their real life documents.

Yeah that's a good point.

Scott


signature.asc
Description: PGP signature

Re: Any rc1 blockers?

[Uwe Stöhr]

El 10.09.2017 a las 01:38, Uwe Stöhr escribió:


Just the patch for
http://www.lyx.org/trac/ticket/10679
should go in before.


Jürgen found now a fix.

Meanwhile I testes LyX 2.3 a lot the last 2 days. It is stable for my needs.

Concerning libiconv 1.15, there are not yet opinions.

Concerning lyx2lyx I guess we will get more bug reports first after RC1 
is out and people try to convert their real life documents.


regards Uwe

Re: Any rc1 blockers?

[Scott Kostyshak]

On Fri, Sep 08, 2017 at 07:54:02AM +0200, Stephan Witt wrote:
> Am 08.09.2017 um 04:30 schrieb Scott Kostyshak :
> > 
> > Dear all,
> > 
> > I think most would agree that the 2.3.x branch is quite stable, and I'm
> > satisfied that 2.3.0beta1 has received a significant amount of testing.
> > I think we should start thinking about rc1. If anyone disagrees and
> > prefers to release a beta2 first instead, please let me know now.
> > 
> > Assuming everyone is fine with moving towards rc1, are there any issues
> > that you view as rc1 blockers?
> 
> Yes, the missing SVG resp. SVGZ converter for the inset-icon instances
> in splash.lyx and other help documents on Mac is not solved yet.

Thanks, Stephan. I agree we should fix that.

Scott


signature.asc
Description: PGP signature

Re: Any rc1 blockers?

[Scott Kostyshak]

On Sun, Sep 10, 2017 at 01:38:26AM +0200, Uwe Stöhr wrote:
> El 08.09.2017 a las 04:30, Scott Kostyshak escribió:
> 
> > I think we should start thinking about rc1.
> 
> Fine with me. Just the patch for
> http://www.lyx.org/trac/ticket/10679
> should go in before. Then the new 2.3.0 feature of supporting SVG+text files
> will also work under Windows.

Sounds good. Thanks to you and Jürgen for making progress on that.

Scott


signature.asc
Description: PGP signature

Re: Any rc1 blockers?

[Uwe Stöhr]

El 08.09.2017 a las 04:30, Scott Kostyshak escribió:


I think we should start thinking about rc1.


Fine with me. Just the patch for
http://www.lyx.org/trac/ticket/10679
should go in before. Then the new 2.3.0 feature of supporting SVG+text 
files will also work under Windows.


regards Uwe

Re: Any rc1 blockers?

[Joel Kulesza]

On Thu, Sep 7, 2017 at 11:54 PM, Stephan Witt  wrote:

> Am 08.09.2017 um 04:30 schrieb Scott Kostyshak :
> >
> > Dear all,
> >
> > I think most would agree that the 2.3.x branch is quite stable, and I'm
> > satisfied that 2.3.0beta1 has received a significant amount of testing.
> > I think we should start thinking about rc1. If anyone disagrees and
> > prefers to release a beta2 first instead, please let me know now.
> >
> > Assuming everyone is fine with moving towards rc1, are there any issues
> > that you view as rc1 blockers?
>
> Yes, the missing SVG resp. SVGZ converter for the inset-icon instances
> in splash.lyx and other help documents on Mac is not solved yet.


Regarding Mac, are there any thoughts on
https://www.lyx.org/trac/ticket/10662 and/or
https://www.lyx.org/trac/ticket/10671 (the more important of these two)?

Re: Any rc1 blockers?

[Stephan Witt]

Am 08.09.2017 um 04:30 schrieb Scott Kostyshak :
> 
> Dear all,
> 
> I think most would agree that the 2.3.x branch is quite stable, and I'm
> satisfied that 2.3.0beta1 has received a significant amount of testing.
> I think we should start thinking about rc1. If anyone disagrees and
> prefers to release a beta2 first instead, please let me know now.
> 
> Assuming everyone is fine with moving towards rc1, are there any issues
> that you view as rc1 blockers?

Yes, the missing SVG resp. SVGZ converter for the inset-icon instances
in splash.lyx and other help documents on Mac is not solved yet.

Stephan

> I would like to send an email to translators soon to start preparing for
> the rc1 release.
> 
> Thanks,
> 
> Scott