date:20200815

Am Sat, 15 Aug 2020 08:24:12 +0200
schrieb Jürgen Spitzmüller :

> Am Samstag, den 15.08.2020, 07:57 +0200 schrieb Kornel Benko:
> > Failing:
> > export/export/latex/languages/en-th_utf8_pdf4_systemF  
> This compiles for me (non-TeX fonts, XeTeX).

All latex errors due to export to pdf(xetex)

! Package polyglossia Error: The current latin font TlwgMono(0) does not contai
n the "Thai" script!
...
Missing character: There is no ภ in font [lmroman10-bold]:mapping=tex-text;!
Missing character: There is no า in font [lmroman10-bold]:mapping=tex-text;!
Missing character: There is no ษ in font [lmroman10-bold]:mapping=tex-text;!
Missing character: There is no า in font [lmroman10-bold]:mapping=tex-text;!
and so on

> 
> > export/doc/ru/Math_(dvi|pdf)[345]_.*F  
> 
> Compiles for me as well (DVI, pstopdf)

You missed dvi3 pdf4 pdf5 (according to the regex)

! Package polyglossia Error: The current latin font  does not contain the "Cyri
llic" script!
...
Missing character: There is no Р in font [lmroman12-bold]:mapping=tex-text;!
Missing character: There is no а in font [lmroman12-bold]:mapping=tex-text;!
Missing character: There is no с in font [lmroman12-bold]:mapping=tex-text;!
and so on

Even setting for instance the serif font 'Alegreya Medium' does not help.

> > export/doc/ru/UserGuide_(dvi|pdf)[345]_.*F  
> 
> Ditto.

Same errors as for Math

> > export/examples/ru/Graphics_and_Insets/XY-Figure_(dvi|pdf)[345]_.*F  
> 
> Ditto.

Same errors as for Math

> > export/examples/ru/Graphics_and_Insets/XY-Pic_dvi|pdf)[345]_.*F  
> 
> Ditto.

Again, missing glyphs

> > export/templates/Articles/Hebrew_Article_%28KOMA-
> > Script%29_dvi|pdf)[345]_.*F  
> 
> Ditto.

Does not compile with pdf(luatex)

! Package fontspec Error: The font "DavidCLM" cannot be found.

This may be luatex error, because xetex has no problems.

> Could you give me a recipe and/or the LaTeX errors you get?
> 
> Jürgen

Kornel


pgpBeeKCe8Fep.pgp
Description: Digitale Signatur von OpenPGP
-- 
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: [LyX/master] DocBook: bug fixing with several bibliographic references at once.

Am Samstag, den 01.08.2020, 03:26 +0200 schrieb Thibaut Cuvelier:
> @@ -627,7 +628,7 @@ void InsetCitation::latex(otexstream & os,
> OutputParams const & runparams) const
> 
> // check if we have to do a qualified list
> 
> vector keys =
> getVectorFromString(cleanupWhitespace(key));
> 
> bool const qualified = cs.hasQualifiedList
> 
> -   && (!getParam("pretextlist").empty()
> 
> +   && (!getParam("F").empty()

I take it this went in by accident. In any case it is wrong and also
causes an assertion. I'll revert it.

Thibaut, if there is a rationale behind it, please let me know so we
can figure out the correct way.

Jürgen


signature.asc
Description: This is a digitally signed message part
-- 
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: Crash in export of American_Psychological_Association_%28APA%29,_v._7_

Am Freitag, den 14.08.2020, 10:56 +0200 schrieb Kornel Benko:
> The backtrace attached.
> 
> In the line InsetCommandParams.cpp:612
> LASSERT(info_.hasParam(name), return dummy);

Fixed. This seems to have been caused by accident in fed71fa8c2.

The template still needs to be fixed (also the image is using a local
absolute path). For the bibliography, I suggest using biblatex-
examples.

Jürgen


signature.asc
Description: This is a digitally signed message part
-- 
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: Regression on master: zh_CN tutorial.lyx compiles with error

Am Samstag, den 15.08.2020, 09:18 +0200 schrieb Kornel Benko:
> All latex errors due to export to pdf(xetex)
> 
> ! Package polyglossia Error: The current latin font TlwgMono(0) does
> not contai
> n the "Thai" script!
> ...
> Missing character: There is no ภ in font [lmroman10-
> bold]:mapping=tex-text;!
> Missing character: There is no า in font [lmroman10-
> bold]:mapping=tex-text;!
> Missing character: There is no ษ in font [lmroman10-
> bold]:mapping=tex-text;!
> Missing character: There is no า in font [lmroman10-
> bold]:mapping=tex-text;!
> and so on

Polyglossia bug, fixed in polyglossia devel (1.50dev) at least.
Nothing we can do about here.

> > 
> > > export/doc/ru/Math_(dvi|pdf)[345]_.*F  
> > 
> > Compiles for me as well (DVI, pstopdf)
> 
> You missed dvi3 pdf4 pdf5 (according to the regex)
> 
> ! Package polyglossia Error: The current latin font  does not contain
> the "Cyri
> llic" script!
> ...
> Missing character: There is no Р in font [lmroman12-
> bold]:mapping=tex-text;!
> Missing character: There is no а in font [lmroman12-
> bold]:mapping=tex-text;!
> Missing character: There is no с in font [lmroman12-
> bold]:mapping=tex-text;!
> and so on
> 
> Even setting for instance the serif font 'Alegreya Medium' does not
> help.
> 
> > > export/doc/ru/UserGuide_(dvi|pdf)[345]_.*F  
> > 
> > Ditto.
> 
> Same errors as for Math
> 
> > > export/examples/ru/Graphics_and_Insets/XY-
> Figure_(dvi|pdf)[345]_.*F  
> > 
> > Ditto.
> 
> Same errors as for Math
> 
> > > export/examples/ru/Graphics_and_Insets/XY-Pic_dvi|pdf)[345]_.*F  
> > 
> > Ditto.
> 
> Again, missing glyphs

All these due to the same polyglossia issue.

> 
> > > export/templates/Articles/Hebrew_Article_%28KOMA-
> > > Script%29_dvi|pdf)[345]_.*F  
> > 
> > Ditto.
> 
> Does not compile with pdf(luatex)
> 
> ! Package fontspec Error: The font "DavidCLM" cannot be found.
> 
> This may be luatex error, because xetex has no problems.

LuaTeX and XeTeX look for fonts differently. You need to install
DavidCLM on the OS level (LuaTeX does not look in texmf fonts).

Jürgen

> 
> > Could you give me a recipe and/or the LaTeX errors you get?
> > 
> > Jürgen
> 
> Kornel


signature.asc
Description: This is a digitally signed message part
-- 
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

LyX-devel-list delay

2020-08-15 Thread Stephan Witt

Hi all,

in the past I've noticed there is an substantial delay for delivery of mails
to the mailing list. Again and again I receive answer to emails w/o having
the cited mail in my inbox before. The mails are coming late and in the
wrong order.

This mail had a delay of more 36 hours. Here are the header lines:

Return-Path: 
Received: from whitealder.osuosl.org ([140.211.166.138]) by mx-ha.gmx.net
 (mxgmx114 [212.227.17.5]) with ESMTPS (Nemesis) id 1MtvlM-1kyfQ53jbw-00uJDN
 for ; Sat, 15 Aug 2020 09:09:14 +0200
Received: from localhost (localhost [127.0.0.1])
by whitealder.osuosl.org (Postfix) with ESMTP id C653F88295;
Thu, 13 Aug 2020 19:59:13 + (UTC)
X-Virus-Scanned: amavisd-new at osuosl.org
Received: from whitealder.osuosl.org ([127.0.0.1])
by localhost (.osuosl.org [127.0.0.1]) (amavisd-new, port 10024)
with ESMTP id N2U9GPLTQhpx; Thu, 13 Aug 2020 19:59:12 + (UTC)
Received: from ash.osuosl.org (ash.osuosl.org [140.211.166.34])
by whitealder.osuosl.org (Postfix) with ESMTP id 0D74388091;
Thu, 13 Aug 2020 19:59:12 + (UTC)

Is it only me seeing this or is it a general phenomena?

Stephan

> Am 13.08.2020 um 21:59 schrieb Stephan Witt :
> 
> Am 13.08.2020 um 18:06 schrieb Daniel :
>> 
>> On 2020-08-13 16:51, Richard Kimberly Heck wrote:
>>> On 8/13/20 10:22 AM, Daniel wrote:
 The depth bar looks a bit out of shape (both in stable and master).
 The attached patch fixes this. See the attached screen captures - the
 "before" is taken in stable and the after in master that is the only
 reason why the length differs.
>>> Any idea why this is happening? It would be worth a bisect. Although the
>>> patch may work, it may simply be masking a problem that is coming from
>>> elsewhere.
>> 
>> I can't do a bisect. It takes ages to compile a version because I am just 
>> using the installer script from Stephan which seems to compile everything 
>> from scratch. I couldn't make his proper cmake/Xcode script to work yet.
> 
> With some luck (i.e. not so big structure aka Makefile.am changes) you can do 
> a bisect w/o repeated runs of the installer script.
> 
> The build tree is not removed after package creation on purpose. You may edit 
> a source file and run make install as usual in the build tree. Therefore you 
> can do the bisect that way. You may start the lyx app from command line with 
> .../lyx-build/LyX-2.4.0dev.app if make install succeeds.
> 
> This probably fails for big range of commits because of file additions etc. 
> If that’s the case you have to run the complete installer script to get the 
> Makefiles right.
> 
> Stephan
> 
> PS: The following command line to build and run LyX lasts 3 secs for an empty 
> change on my system.
> 
> (cd /Users/stephan/git/lyx-build/LyX-2.4.0dev.build/src && make install && 
> open /Users/stephan/git/lyx-build/LyX-2.4.0dev.app)
> 
>> But it seems to me that part of the answer is that it wasn't taken into 
>> account that fillRect creates a filled rectangle without the border set by 
>> the pen. It seem pretty clear that this explains the part on the top. And it 
>> has come up elsewhere (https://www.lyx.org/trac/ticket/11925). I guess a 
>> warning about this could be added in the code to the fillRect function. The 
>> lower part I am not so sure.
>> 
>> My fix might look a little indirect because I had to make sure to not come 
>> into conflict with the re-painting of the rows. But I think it's a clear 
>> improvement over the previous state.
>> -- 
>> Daniel

-- 
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: Crash in export of American_Psychological_Association_%28APA%29,_v._7_

Am Sat, 15 Aug 2020 09:28:43 +0200
schrieb Jürgen Spitzmüller :

> Am Freitag, den 14.08.2020, 10:56 +0200 schrieb Kornel Benko:
> > The backtrace attached.
> > 
> > In the line InsetCommandParams.cpp:612
> > LASSERT(info_.hasParam(name), return dummy);  
> 
> Fixed. This seems to have been caused by accident in fed71fa8c2.
> 
> The template still needs to be fixed (also the image is using a local
> absolute path). For the bibliography, I suggest using biblatex-
> examples.
> 
> Jürgen

Thanks, will try to correct.

Kornel


pgploPOkF43Mb.pgp
Description: Digitale Signatur von OpenPGP
-- 
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: Crash in export of American_Psychological_Association_%28APA%29,_v._7_

Am Samstag, den 15.08.2020, 09:28 +0200 schrieb Jürgen Spitzmüller:
> The template still needs to be fixed (also the image is using a local
> absolute path). For the bibliography, I suggest using biblatex-
> examples.

Done.

Jürgen


signature.asc
Description: This is a digitally signed message part
-- 
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: Crash in export of American_Psychological_Association_%28APA%29,_v._7_

Am Sat, 15 Aug 2020 09:53:09 +0200
schrieb Jürgen Spitzmüller :

> Am Samstag, den 15.08.2020, 09:28 +0200 schrieb Jürgen Spitzmüller:
> > The template still needs to be fixed (also the image is using a local
> > absolute path). For the bibliography, I suggest using biblatex-
> > examples.  
> 
> Done.
> 
> Jürgen

You beat me. But there was still a language problem. Will try to make an mwe.

Kornel


mwe.lyx
Description: application/lyx


pgpyPVGJkh3ET.pgp
Description: Digitale Signatur von OpenPGP
-- 
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: Crash in export of American_Psychological_Association_%28APA%29,_v._7_

Am Samstag, den 15.08.2020, 10:26 +0200 schrieb Kornel Benko:
> You beat me. But there was still a language problem. Will try to make
> an mwe.

Yes, I saw that. I'm just about to investigate.

Jürgen


signature.asc
Description: This is a digitally signed message part
-- 
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: [Request/idea] Preamble generalization

2020-08-15 Thread Lorenzo Bertini

> UI-wise, this would probably be better than what we have, but you can
> already do this with Local Layout. Like this:
>
> Format 61
>
> AddToHTMLPreamble
>
> 
> h2.section { font-size: 150%; font-weight: bold; }
> 
>
> EndPremable
>
> You can put anything else in there that you want in the head block, too.
>
> Do feel free to file an enhancement request on the bug tracker.
>
> Riki

Thank you, this solves some problems for me. However now I'm
questioning whether the  section in document->settings
isn't just a shortcut for  in ? In this case I think putting them at the same level is
confusing, and prevented me from ever needing to look deeper in ; I understand not all users would want to learn this but it's
three lines to get a LaTeX or HTML preamble. Ideally there would be a
way to keep the generality of the  section that however
makes it super easy and intuitive to change a preamble.
I leave this decision to you and I'll not file a bug (feature) report
for now, but here's my two cents for keeping things clear and
functional:
1)  section gets removed
2)  section gets an "advanced" tab with the text entry
we have now, and an "easy" tab made by text boxes, buttons and toggles
(for a start, just containing the infamous LaTeX Preamble and HTML
Preamble, then maybe more).
Except for the advantages in clarity on preambles, I think this would
make way easier not only to tinker with layouts but also to learn how
Lyx works and what are the commands. I know this is a bit of an
overhaul, I'm just leaving this here as an idea.
-- 
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: [Request/idea] Preamble generalization

2020-08-15 Thread José Abílio Matos

On Saturday, 15 August 2020 07.52.21 WEST Richard Kimberly Heck wrote:
> I think this is probably very easy. As someone (maybe you) suggested, we
> just need an extra tab to choose between LaTeX and HTML preambles. After
> that, it's just a matter of copying and pasting code (and not making
> copy and paste errors!). Let me know if you have any problems. I'll be
> happy to have a look.
> 
> Riki

I am probably over thinking, that happens a lot :-), but I think that the 
preamble should be made generic. There are several file formats that allow a 
preamble not just latex and html.

I say this because I am thinking that sooner or later it would be nice to 
support directly markdown or structured test. That is a common request and 
currently we suggest using pandoc but just like in the html case we have more 
information than an external converter.

The purpose of this message is just to put things into perspective and not to 
stall any plans.

Regards,
-- 
José Abílio-- 
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: LyX-devel-list delay

2020-08-15 Thread Pavel Sanda

On Sat, Aug 15, 2020 at 09:38:48AM +0200, Stephan Witt wrote:
> Return-Path: 
> Received: from whitealder.osuosl.org ([140.211.166.138]) by mx-ha.gmx.net
>  (mxgmx114 [212.227.17.5]) with ESMTPS (Nemesis) id 1MtvlM-1kyfQ53jbw-00uJDN
>  for ; Sat, 15 Aug 2020 09:09:14 +0200
> Received: from localhost (localhost [127.0.0.1])
>   by whitealder.osuosl.org (Postfix) with ESMTP id C653F88295;
>   Thu, 13 Aug 2020 19:59:13 + (UTC)
> X-Virus-Scanned: amavisd-new at osuosl.org
> Received: from whitealder.osuosl.org ([127.0.0.1])
>   by localhost (.osuosl.org [127.0.0.1]) (amavisd-new, port 10024)
>   with ESMTP id N2U9GPLTQhpx; Thu, 13 Aug 2020 19:59:12 + (UTC)
> Received: from ash.osuosl.org (ash.osuosl.org [140.211.166.34])
>   by whitealder.osuosl.org (Postfix) with ESMTP id 0D74388091;
>   Thu, 13 Aug 2020 19:59:12 + (UTC)
> 
> Is it only me seeing this or is it a general phenomena?

I got this particular email on time, which suggests the bottleneck
is in interaction between mx-ha.gmx.net and whitealder.osuosl.org.

I suffered similar problem when I have subscribed to the list by my
@lyx address and suspected it has something to do with google antispam
solutions, but who knows.
When I subscribed to different address the problem disappeared.

Pavel
-- 
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: Crash in export of American_Psychological_Association_%28APA%29,_v._7_

Am Samstag, den 15.08.2020, 10:32 +0200 schrieb Jürgen Spitzmüller:
> > You beat me. But there was still a language problem. Will try to
> make
> > an mwe.
> 
> Yes, I saw that. I'm just about to investigate.

This should be fixed at 0a2fce358d7.

I tested with all the previously problematic cases, but to be sure,
could you check one more time whether this broke something else?

Thanks
Jürgen


signature.asc
Description: This is a digitally signed message part
-- 
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: Crash in export of American_Psychological_Association_%28APA%29,_v._7_

Am Sat, 15 Aug 2020 12:30:36 +0200
schrieb Jürgen Spitzmüller :

> Am Samstag, den 15.08.2020, 10:32 +0200 schrieb Jürgen Spitzmüller:
> > > You beat me. But there was still a language problem. Will try to  
> > make  
> > > an mwe.  
> > 
> > Yes, I saw that. I'm just about to investigate.  
> 
> This should be fixed at 0a2fce358d7.
> 
> I tested with all the previously problematic cases, but to be sure,
> could you check one more time whether this broke something else?
> 
> Thanks
> Jürgen

Will do.

For the next cases I will open a new thread.

Kornel


pgplL3K2Umx0x.pgp
Description: Digitale Signatur von OpenPGP
-- 
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: Crash in export of American_Psychological_Association_%28APA%29,_v._7_

Am Sat, 15 Aug 2020 12:56:31 +0200
schrieb Kornel Benko :

> Am Sat, 15 Aug 2020 12:30:36 +0200
> schrieb Jürgen Spitzmüller :
> 
> > Am Samstag, den 15.08.2020, 10:32 +0200 schrieb Jürgen Spitzmüller:  
> > > > You beat me. But there was still a language problem. Will try to
> > > make
> > > > an mwe.
> > > 
> > > Yes, I saw that. I'm just about to investigate.
> > 
> > This should be fixed at 0a2fce358d7.
> > 
> > I tested with all the previously problematic cases, but to be sure,
> > could you check one more time whether this broke something else?
> > 
> > Thanks
> > Jürgen  
> 
> Will do.
> 
> For the next cases I will open a new thread.
> 
>   Kornel

Hm, the number of failed is increasing again.
Most of already error-free tests are there again.
Example:
export/doc/es/Math_pdf5_texF
export/doc/es/UserGuide_pdf5_texF
export/doc/fr/Customization_pdf4_systemF
DEFAULTOUTPUT_export/doc/fr/EmbeddedObjects_pdf2
export/doc/fr/Customization_pdf5_systemF

\sum = 68 failed.

Kornel


pgpBd6Krvgjbi.pgp
Description: Digitale Signatur von OpenPGP
-- 
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: Crash in export of American_Psychological_Association_%28APA%29,_v._7_

Am Samstag, den 15.08.2020, 13:58 +0200 schrieb Kornel Benko:
> Hm, the number of failed is increasing again.
> Most of already error-free tests are there again.

No idea how that escaped me. Anyway, better now?

Jürgen


signature.asc
Description: This is a digitally signed message part
-- 
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: Crash in export of American_Psychological_Association_%28APA%29,_v._7_

Am Sat, 15 Aug 2020 15:33:35 +0200
schrieb Jürgen Spitzmüller :

> Am Samstag, den 15.08.2020, 13:58 +0200 schrieb Kornel Benko:
> > Hm, the number of failed is increasing again.
> > Most of already error-free tests are there again.  
> 
> No idea how that escaped me. Anyway, better now?
> 
> Jürgen

Yes, remaining 21 tests attached.

Kornel


LastExportTestsFailed.log.1
Description: Binary data


pgp58KaDpYyFR.pgp
Description: Digitale Signatur von OpenPGP
-- 
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Test entering an endless loop in latex processing


This one is really bad, maybe not even lyx's fault.

TL2020:
export/export/latex/languages/supported-languages_babel_auto-legacy_pdf2

Kornel


pgpAFf5nLn54U.pgp
Description: Digitale Signatur von OpenPGP
-- 
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: Crash in export of American_Psychological_Association_%28APA%29,_v._7_

Am Samstag, den 15.08.2020, 15:49 +0200 schrieb Kornel Benko:
> Yes, remaining 21 tests attached.

> 440:export/export/latex/languages/en-th_utf8_pdf4_systemF

This I cannot reproduce as said.

> 1973:DEFAULTOUTPUT_export/doc/ar/Shortcuts_pdf4_systemF

This is fixed now.

Jürgen


signature.asc
Description: This is a digitally signed message part
-- 
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Allow empty line ... I know

I know LyX is very strict on not allowing empty lines but I think there 
should be one exception: a line that contains only a display formula.


The attached document demonstrates the situation: LyX shows no space 
difference between the text and the display formula. However, the output 
shows such difference. It would be good if LyX makes one aware that one 
has started a new paragraph before a display formula by showing that 
extra empty line created.


Also, the numbering/bullets get moved in front of the display formula in 
LyX. I guess this problem would be solved as well if the empty line 
would be rendered in LyX.


Maybe it would be even good if such an empty line would be highlighted 
with some symbol at the beginning because most of the time the added 
break will not be on purpose. And it can even lead to the infamous LaTeX 
error in LyX when, for example, a description contains only a display 
formula.

--
Daniel
#LyX 2.3 created this file. For more info see http://www.lyx.org/
\lyxformat 544
\begin_document
\begin_header
\save_transient_properties true
\origin unavailable
\textclass article
\begin_preamble
\usepackage{multicol}
\end_preamble
\use_default_options true
\maintain_unincluded_children false
\language english
\language_package default
\inputencoding auto
\fontencoding global
\font_roman "default" "default"
\font_sans "default" "default"
\font_typewriter "default" "default"
\font_math "auto" "auto"
\font_default_family default
\use_non_tex_fonts false
\font_sc false
\font_osf false
\font_sf_scale 100 100
\font_tt_scale 100 100
\use_microtype false
\use_dash_ligatures true
\graphics default
\default_output_format default
\output_sync 0
\bibtex_command default
\index_command default
\paperfontsize default
\spacing single
\use_hyperref true
\pdf_bookmarks true
\pdf_bookmarksnumbered false
\pdf_bookmarksopen false
\pdf_bookmarksopenlevel 1
\pdf_breaklinks false
\pdf_pdfborder false
\pdf_colorlinks false
\pdf_backref false
\pdf_pdfusetitle true
\papersize default
\use_geometry false
\use_package amsmath 1
\use_package amssymb 1
\use_package cancel 1
\use_package esint 1
\use_package mathdots 1
\use_package mathtools 1
\use_package mhchem 1
\use_package stackrel 1
\use_package stmaryrd 1
\use_package undertilde 1
\cite_engine basic
\cite_engine_type default
\biblio_style plain
\use_bibtopic false
\use_indices false
\paperorientation portrait
\suppress_date false
\justification false
\use_refstyle 1
\use_minted 0
\index Index
\shortcut idx
\color #008000
\end_index
\secnumdepth 3
\tocdepth 3
\paragraph_separation indent
\paragraph_indentation default
\is_math_indent 1
\math_indentation default
\math_numbering_side default
\quotes_style english
\dynamic_quotes 0
\papercolumns 1
\papersides 1
\paperpagestyle default
\tracking_changes false
\output_changes false
\html_math_output 0
\html_css_as_file 0
\html_be_strict false
\end_header

\begin_body

\begin_layout Standard
No break after...
 (same spacing in LyX and LaTeX output)
\begin_inset Formula 
\begin{align*}
x
\end{align*}

\end_inset


\end_layout

\begin_layout Standard
Break after...
 (different spacing in LyX and LaTeX output)
\end_layout

\begin_layout Standard
\begin_inset Formula 
\begin{align*}
x
\end{align*}

\end_inset


\end_layout

\begin_layout Standard
With text...
 (same spacing in LyX and LaTeX output)
\end_layout

\begin_layout Enumerate
Text
\begin_inset Formula 
\begin{align*}
x
\end{align*}

\end_inset


\end_layout

\begin_layout Standard
(Unimportant intermezzo for getting an indented paragraph...)
\end_layout

\begin_layout Standard
Without text...
 (different spacing in LyX and LaTeX output)
\end_layout

\begin_layout Enumerate
\begin_inset Formula 
\begin{align*}
x
\end{align*}

\end_inset


\end_layout

\end_body
\end_document
-- 
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: Allow empty line ... I know

2020-08-15 Thread Jean-Marc Lasgouttes

Short answer from my phone : this is not an empty line issue. There is a 
problem reported about space above formula. I have not looked at it yet.

Concerning the bullet case, we could handle it like I did the proof box case.

More on that when I return.

JMarc

Le 15 août 2020 16:50:22 GMT+02:00, Daniel  a écrit :
>I know LyX is very strict on not allowing empty lines but I think there
>
>should be one exception: a line that contains only a display formula.
>
>The attached document demonstrates the situation: LyX shows no space 
>difference between the text and the display formula. However, the
>output 
>shows such difference. It would be good if LyX makes one aware that one
>
>has started a new paragraph before a display formula by showing that 
>extra empty line created.
>
>Also, the numbering/bullets get moved in front of the display formula
>in 
>LyX. I guess this problem would be solved as well if the empty line 
>would be rendered in LyX.
>
>Maybe it would be even good if such an empty line would be highlighted 
>with some symbol at the beginning because most of the time the added 
>break will not be on purpose. And it can even lead to the infamous
>LaTeX 
>error in LyX when, for example, a description contains only a display 
>formula.

-- 
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: Crash in export of American_Psychological_Association_%28APA%29,_v._7_

Am Sat, 15 Aug 2020 16:49:09 +0200
schrieb Jürgen Spitzmüller :

> Am Samstag, den 15.08.2020, 15:49 +0200 schrieb Kornel Benko:
> > Yes, remaining 21 tests attached.  
> 
> > 440:export/export/latex/languages/en-th_utf8_pdf4_systemF  
> 
> This I cannot reproduce as said.

Slipped in unintended

> > 1973:DEFAULTOUTPUT_export/doc/ar/Shortcuts_pdf4_systemF  
> 
> This is fixed now.

Confirmed.

> Jürgen

Thanks for your patience. Sometimes I am insisting too much.

Kornel


pgp2Nd5kh3GH2.pgp
Description: Digitale Signatur von OpenPGP
-- 
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: Crash in export of American_Psychological_Association_%28APA%29,_v._7_

Am Samstag, den 15.08.2020, 16:49 +0200 schrieb Jürgen Spitzmüller:
> Am Samstag, den 15.08.2020, 15:49 +0200 schrieb Kornel Benko:
> > 7509:export/mathmacros/masterOfSpace_dvi
> > 7510:export/mathmacros/masterOfSpace_dvi3_texF
> > 7511:export/mathmacros/masterOfSpace_dvi3_systemF
> > 7512:export/mathmacros/masterOfSpace_pdf
> > 7513:export/mathmacros/masterOfSpace_pdf2
> > 7514:export/mathmacros/masterOfSpace_pdf3
> > 7515:export/mathmacros/masterOfSpace_pdf4_systemF
> > 7516:export/mathmacros/masterOfSpace_pdf5_texF
> > 7517:export/mathmacros/masterOfSpace_pdf5_systemF
> > 7545:export/mathmacros/testcases_environments_dvi
> > 7546:export/mathmacros/testcases_environments_dvi3_texF
> > 7547:export/mathmacros/testcases_environments_dvi3_systemF
> > 7548:export/mathmacros/testcases_environments_pdf
> > 7549:export/mathmacros/testcases_environments_pdf2
> > 7550:export/mathmacros/testcases_environments_pdf3
> > 7551:export/mathmacros/testcases_environments_pdf4_systemF
> > 7552:export/mathmacros/testcases_environments_pdf5_texF
> > 7553:export/mathmacros/testcases_environments_pdf5_systemF

These are all due to math macros. Here I need help (don't know enough
about mathed).

Jürgen


signature.asc
Description: This is a digitally signed message part
-- 
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: Allow empty line ... I know


On 2020-08-15 16:58, Jean-Marc Lasgouttes wrote:

Le 15 août 2020 16:50:22 GMT+02:00, Daniel  a écrit :

I know LyX is very strict on not allowing empty lines but I think there

should be one exception: a line that contains only a display formula.

The attached document demonstrates the situation: LyX shows no space
difference between the text and the display formula. However, the
output
shows such difference. It would be good if LyX makes one aware that one

has started a new paragraph before a display formula by showing that
extra empty line created.

Also, the numbering/bullets get moved in front of the display formula
in
LyX. I guess this problem would be solved as well if the empty line
would be rendered in LyX.

Maybe it would be even good if such an empty line would be highlighted
with some symbol at the beginning because most of the time the added
break will not be on purpose. And it can even lead to the infamous
LaTeX
error in LyX when, for example, a description contains only a display
formula.


Short answer from my phone : this is not an empty line issue. There is a 
problem reported about space above formula. I have not looked at it yet.

Concerning the bullet case, we could handle it like I did the proof box case.

More on that when I return.


Okay, I would have thought that both of these problems are handled (and 
correctly so) if a line (not a paragraph!) before a display formula was 
allowed to be empty. Because that seems exactly what LaTeX is doing.

--
Daniel

--
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: Allow empty line ... I know

2020-08-15 Thread Jean-Marc Lasgouttes

Le 15 août 2020 17:09:45 GMT+02:00, Daniel  a écrit :
>On 2020-08-15 16:58, Jean-Marc Lasgouttes wrote:
>> Short answer from my phone : this is not an empty line issue. There
>is a problem reported about space above formula. I have not looked at
>it yet.
>> 
>> Concerning the bullet case, we could handle it like I did the proof
>box case.
>> 
>> More on that when I return.
>
>Okay, I would have thought that both of these problems are handled (and
>correctly so) if a line (not a paragraph!) before a display formula was
>
>allowed to be empty. Because that seems exactly what LaTeX is doing.
>-- 
>Daniel
>
>-- 
>lyx-devel mailing list
>lyx-devel@lists.lyx.org
>http://lists.lyx.org/mailman/listinfo/lyx-devel

The spacing is created by LaTeX, not by the user, if I understand correctly.

JMarc
-- 
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: Allow empty line ... I know


On 2020-08-15 17:22, Jean-Marc Lasgouttes wrote:

Le 15 août 2020 17:09:45 GMT+02:00, Daniel  a écrit :

On 2020-08-15 16:58, Jean-Marc Lasgouttes wrote:

Short answer from my phone : this is not an empty line issue. There

is a problem reported about space above formula. I have not looked at
it yet.


Concerning the bullet case, we could handle it like I did the proof

box case.


More on that when I return.


Okay, I would have thought that both of these problems are handled (and
correctly so) if a line (not a paragraph!) before a display formula was

allowed to be empty. Because that seems exactly what LaTeX is doing.
--
Daniel

--
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel


The spacing is created by LaTeX, not by the user, if I understand correctly.


In LaTeX the user creates the spacing by adding an extra empty line or 
\item.

--
Daniel

--
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: Allow empty line ... I know

On 8/15/20 11:30 AM, Daniel wrote:
> On 2020-08-15 17:22, Jean-Marc Lasgouttes wrote:
>> Le 15 août 2020 17:09:45 GMT+02:00, Daniel  a écrit :
>>> On 2020-08-15 16:58, Jean-Marc Lasgouttes wrote:
 Short answer from my phone : this is not an empty line issue. There
 is a problem reported about space above formula. I have not looked
 atit yet.

 Concerning the bullet case, we could handle it like I did the proof
 box case.

 More on that when I return.
>>>
>>> Okay, I would have thought that both of these problems are handled (and
>>> correctly so) if a line (not a paragraph!) before a display formula was
>>> allowed to be empty. Because that seems exactly what LaTeX is doing.
>>> -- 
>>> Daniel
>>
>> The spacing is created by LaTeX, not by the user, if I understand
>> correctly.
>
> In LaTeX the user creates the spacing by adding an extra empty line or
> \item.

As JMarc said, I think the problem here is just that LyX doesn't display
the extra space about the display formula. Basically, what we have is a
paragraph break immediately before the displayed formula. So there is
the space for that break in the LaTeX.

I've run into this quite a bit myself so would be happy to see some kind
of indication. Of course, one can turn on the pilcrows

Riki

-- 
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: Test entering an endless loop in latex processing

Am Samstag, den 15.08.2020, 16:02 +0200 schrieb Kornel Benko:
> This one is really bad, maybe not even lyx's fault.
> 
> TL2020:
> export/export/latex/languages/supported-languages_babel_auto-
> legacy_pdf2

This compiles for me (at 3c5c41ddc2d4a4).

Jürgen 


signature.asc
Description: This is a digitally signed message part
-- 
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: Allow empty line ... I know


On 2020-08-15 17:52, Richard Kimberly Heck wrote:

On 8/15/20 11:30 AM, Daniel wrote:

On 2020-08-15 17:22, Jean-Marc Lasgouttes wrote:

Le 15 août 2020 17:09:45 GMT+02:00, Daniel  a écrit :

On 2020-08-15 16:58, Jean-Marc Lasgouttes wrote:

Short answer from my phone : this is not an empty line issue. There
is a problem reported about space above formula. I have not looked
atit yet.

Concerning the bullet case, we could handle it like I did the proof
box case.

More on that when I return.


Okay, I would have thought that both of these problems are handled (and
correctly so) if a line (not a paragraph!) before a display formula was
allowed to be empty. Because that seems exactly what LaTeX is doing.
--
Daniel


The spacing is created by LaTeX, not by the user, if I understand
correctly.


In LaTeX the user creates the spacing by adding an extra empty line or
\item.


As JMarc said, I think the problem here is just that LyX doesn't display
the extra space about the display formula. Basically, what we have is a
paragraph break immediately before the displayed formula. So there is
the space for that break in the LaTeX.


The space does not take care of the bullet/numbering. An empty line 
would take care of both problems. If the line contains a character, then 
LyX does exactly the right thing. So, I thought allowing an empty line 
in this case would make LyX do exactly the right thing. So, LyX would do 
this: remove an empty line only if it is empty by current standards *or 
contains a display formula* (which actually could kind of qualify as 
non-empty since the line continues with a display formula).



I've run into this quite a bit myself so would be happy to see some kind
of indication. Of course, one can turn on the pilcrows


Yes, one can turn on the pilcrows. But since this leads to particular 
problems and is (almost?) never desired, I thought this could be marked 
by default (by a pilcrow?).



--
Daniel

--
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: [Request/idea] Preamble generalization

On 8/15/20 4:41 AM, Lorenzo Bertini wrote:
>> UI-wise, this would probably be better than what we have, but you can
>> already do this with Local Layout. Like this:
>>
>> Format 61
>>
>> AddToHTMLPreamble
>>
>> 
>> h2.section { font-size: 150%; font-weight: bold; }
>> 
>>
>> EndPremable
>>
>> You can put anything else in there that you want in the head block, too.
>>
>> Do feel free to file an enhancement request on the bug tracker.
>>
>> Riki
> Thank you, this solves some problems for me. However now I'm
> questioning whether the  section in document->settings
> isn't just a shortcut for  in  layout>? In this case I think putting them at the same level is
> confusing, and prevented me from ever needing to look deeper in  layout>; I understand not all users would want to learn this but it's
> three lines to get a LaTeX or HTML preamble. Ideally there would be a
> way to keep the generality of the  section that however
> makes it super easy and intuitive to change a preamble.

I can see why it might seem that way, but these are quite different in
principle. The Local Layout box is really intended for layout code, like
you'd find in a module, but which is document-specific (not something
you really want to be able to reuse). E.g., I sometimes need a specific
kind of list for one document, so I put the layout code there. It's
basically a side-effect of the fact that the AddTo...Preamble tags exist
in layout code that one can put preamble info there as well. But I
wouldn't really recommend it if you're going to have very much preamble
code.

There's a long-standing request, actually, for there to be TWO preamble
boxes, whose contents are output at different times (before and after
other code that LyX generates). I'm actually not sure whether the layout
code from Local Layout (or from layout code more generally) is output
before or after the code in the Preamble box. So these may not actually
be equivalent.

> I leave this decision to you and I'll not file a bug (feature) report
> for now, but here's my two cents for keeping things clear and
> functional:
> 1)  section gets removed
> 2)  section gets an "advanced" tab with the text entry
> we have now, and an "easy" tab made by text boxes, buttons and toggles
> (for a start, just containing the infamous LaTeX Preamble and HTML
> Preamble, then maybe more).
> Except for the advantages in clarity on preambles, I think this would
> make way easier not only to tinker with layouts but also to learn how
> Lyx works and what are the commands. I know this is a bit of an
> overhaul, I'm just leaving this here as an idea.

The complete version of this would give us a kind of GUI layout
editor---another longstanding request and something of a holy grail.
It's possible, presumably, but has proven difficult.

Riki

-- 
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: Allow empty line ... I know

On 8/15/20 12:05 PM, Daniel wrote:
> On 2020-08-15 17:52, Richard Kimberly Heck wrote:
>> On 8/15/20 11:30 AM, Daniel wrote:
>>> On 2020-08-15 17:22, Jean-Marc Lasgouttes wrote:
 Le 15 août 2020 17:09:45 GMT+02:00, Daniel  a écrit :
> On 2020-08-15 16:58, Jean-Marc Lasgouttes wrote:
>> Short answer from my phone : this is not an empty line issue. There
>> is a problem reported about space above formula. I have not looked
>> atit yet.
>>
>> Concerning the bullet case, we could handle it like I did the proof
>> box case.
>>
>> More on that when I return.
>
> Okay, I would have thought that both of these problems are handled
> (and
> correctly so) if a line (not a paragraph!) before a display
> formula was
> allowed to be empty. Because that seems exactly what LaTeX is doing.
> -- 
> Daniel

 The spacing is created by LaTeX, not by the user, if I understand
 correctly.
>>>
>>> In LaTeX the user creates the spacing by adding an extra empty line or
>>> \item.
>>
>> As JMarc said, I think the problem here is just that LyX doesn't display
>> the extra space about the display formula. Basically, what we have is a
>> paragraph break immediately before the displayed formula. So there is
>> the space for that break in the LaTeX.
>
> The space does not take care of the bullet/numbering. An empty line
> would take care of both problems. If the line contains a character,
> then LyX does exactly the right thing. So, I thought allowing an empty
> line in this case would make LyX do exactly the right thing. So, LyX
> would do this: remove an empty line only if it is empty by current
> standards *or contains a display formula* (which actually could kind
> of qualify as non-empty since the line continues with a display formula).

The problem is that the following is perfectly valid LaTeX:

test test

$$a=b$$

And, as a general rule, we want to allow the user to write that in LyX
if they want to do so. Or, as people here will sometimes say: Don't try
to outsmart the user. This is a pretty fundamental LyX principle---one
it took me a while to appreciate---and magically removing empty lines
before display formulas violates it. So that's no go.

The problem, as I see it, is that it is very easy to write what's above
when what you wanted was:

test test
$$a = b$$

and LyX doesn't give enough indication of this difference. So we either
need to display more space or, as you suggest, to display a pilcrow in
the former case. The latter is probably not terribly easy to do, though,
since the pilcrow would attach to the paragraph before the one in which
the displayed formula lives. But maybe there's something else we could
display in the space above the displayed equation, just to make it more
obvious.

Riki

-- 
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: [RFC] Missing DocIterator methods

On 8/14/20 4:02 PM, Stephan Witt wrote:
> Hi all,
>
> to implement the correct and Mac conforming solution for navigation through 
> the LyX document with keyboard I need to go forward and backward. Sounds 
> simple and I want to use the DocIterator for that. Nevertheless I missed two 
> methods to use DocIterators for backward iteration. First I’d like to have 
> the opposite test to atEnd() - name it atBegin() to stop the loop at 
> beginning. Second there is the possibility to go forward and skip closed or 
> readonly text insets. But the symmetric method for backward traversal isn’t 
> there.
>
> I’ve attached a patch to implement these methods. Did I overlook something 
> and if not is it the right thing? I found it useful and working and like to 
> add it. Any objections?

No, but I'm not terribly familiar with this code.

Riki


-- 
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: Allow empty line ... I know


On 2020-08-15 18:20, Richard Kimberly Heck wrote:

On 8/15/20 12:05 PM, Daniel wrote:

On 2020-08-15 17:52, Richard Kimberly Heck wrote:

On 8/15/20 11:30 AM, Daniel wrote:

On 2020-08-15 17:22, Jean-Marc Lasgouttes wrote:

Le 15 août 2020 17:09:45 GMT+02:00, Daniel  a écrit :

On 2020-08-15 16:58, Jean-Marc Lasgouttes wrote:

Short answer from my phone : this is not an empty line issue. There
is a problem reported about space above formula. I have not looked
atit yet.

Concerning the bullet case, we could handle it like I did the proof
box case.

More on that when I return.


Okay, I would have thought that both of these problems are handled
(and
correctly so) if a line (not a paragraph!) before a display
formula was
allowed to be empty. Because that seems exactly what LaTeX is doing.
--
Daniel


The spacing is created by LaTeX, not by the user, if I understand
correctly.


In LaTeX the user creates the spacing by adding an extra empty line or
\item.


As JMarc said, I think the problem here is just that LyX doesn't display
the extra space about the display formula. Basically, what we have is a
paragraph break immediately before the displayed formula. So there is
the space for that break in the LaTeX.


The space does not take care of the bullet/numbering. An empty line
would take care of both problems. If the line contains a character,
then LyX does exactly the right thing. So, I thought allowing an empty
line in this case would make LyX do exactly the right thing. So, LyX
would do this: remove an empty line only if it is empty by current
standards *or contains a display formula* (which actually could kind
of qualify as non-empty since the line continues with a display formula).


The problem is that the following is perfectly valid LaTeX:

test test

$$a=b$$

And, as a general rule, we want to allow the user to write that in LyX
if they want to do so. Or, as people here will sometimes say: Don't try
to outsmart the user. This is a pretty fundamental LyX principle---one
it took me a while to appreciate---and magically removing empty lines
before display formulas violates it. So that's no go.

The problem, as I see it, is that it is very easy to write what's above
when what you wanted was:

test test
$$a = b$$

and LyX doesn't give enough indication of this difference. So we either
need to display more space or, as you suggest, to display a pilcrow in
the former case. The latter is probably not terribly easy to do, though,
since the pilcrow would attach to the paragraph before the one in which
the displayed formula lives. But maybe there's something else we could
display in the space above the displayed equation, just to make it more
obvious.


I agree. I suggested to add an empty line in LyX not to remove it in the 
LaTeX output.


The LaTeX spacing seems to behave exactly as if an empty line contained 
a character, i.e. in


test test

$$a=b$$

and

test test
x
$$a=b$$

the spacing between test test and the display formula is exactly the same.

The same with

\item

$$a=b$$

and

\item x

$$a=b$$

That's why I thought an empty line would be the solution.

And adding a special character to the empty line might be easier too. I 
take back my pilcrow suggestion though because this would be an empty 
line *without* a paragraph break. Maybe a right pointing arrow or so 
might be more suitable.

--
Daniel

--
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: Allow empty line ... I know


On 2020-08-15 18:39, Daniel wrote:

On 2020-08-15 18:20, Richard Kimberly Heck wrote:

On 8/15/20 12:05 PM, Daniel wrote:

On 2020-08-15 17:52, Richard Kimberly Heck wrote:

On 8/15/20 11:30 AM, Daniel wrote:

On 2020-08-15 17:22, Jean-Marc Lasgouttes wrote:
Le 15 août 2020 17:09:45 GMT+02:00, Daniel  a 
écrit :

On 2020-08-15 16:58, Jean-Marc Lasgouttes wrote:

Short answer from my phone : this is not an empty line issue. There
is a problem reported about space above formula. I have not looked
atit yet.

Concerning the bullet case, we could handle it like I did the proof
box case.

More on that when I return.


Okay, I would have thought that both of these problems are handled
(and
correctly so) if a line (not a paragraph!) before a display
formula was
allowed to be empty. Because that seems exactly what LaTeX is doing.
--
Daniel


The spacing is created by LaTeX, not by the user, if I understand
correctly.


In LaTeX the user creates the spacing by adding an extra empty line or
\item.


As JMarc said, I think the problem here is just that LyX doesn't 
display

the extra space about the display formula. Basically, what we have is a
paragraph break immediately before the displayed formula. So there is
the space for that break in the LaTeX.


The space does not take care of the bullet/numbering. An empty line
would take care of both problems. If the line contains a character,
then LyX does exactly the right thing. So, I thought allowing an empty
line in this case would make LyX do exactly the right thing. So, LyX
would do this: remove an empty line only if it is empty by current
standards *or contains a display formula* (which actually could kind
of qualify as non-empty since the line continues with a display 
formula).


The problem is that the following is perfectly valid LaTeX:

test test

$$a=b$$

And, as a general rule, we want to allow the user to write that in LyX
if they want to do so. Or, as people here will sometimes say: Don't try
to outsmart the user. This is a pretty fundamental LyX principle---one
it took me a while to appreciate---and magically removing empty lines
before display formulas violates it. So that's no go.

The problem, as I see it, is that it is very easy to write what's above
when what you wanted was:

test test
$$a = b$$

and LyX doesn't give enough indication of this difference. So we either
need to display more space or, as you suggest, to display a pilcrow in
the former case. The latter is probably not terribly easy to do, though,
since the pilcrow would attach to the paragraph before the one in which
the displayed formula lives. But maybe there's something else we could
display in the space above the displayed equation, just to make it more
obvious.


I agree. I suggested to add an empty line in LyX not to remove it in the 
LaTeX output.


The LaTeX spacing seems to behave exactly as if an empty line contained 
a character, i.e. in


test test

$$a=b$$

and

test test
x
$$a=b$$

the spacing between test test and the display formula is exactly the same.

The same with

\item

$$a=b$$

and

\item x

$$a=b$$

That's why I thought an empty line would be the solution.

And adding a special character to the empty line might be easier too. I 
take back my pilcrow suggestion though because this would be an empty 
line *without* a paragraph break. Maybe a right pointing arrow or so 
might be more suitable.


Okay, I just noticed another argument for the empty line: If there is an 
bullet/numbering, it would be odd to show it in the space without the 
possibility to click right after it to add text to it. Instead one would 
have to click before the display formula...

--
Daniel

--
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: Crash in export of American_Psychological_Association_%28APA%29,_v._7_

Am Samstag, den 15.08.2020, 15:49 +0200 schrieb Kornel Benko:
> Am Sat, 15 Aug 2020 15:33:35 +0200
> > 402:DEFAULTOUTPUT_export/export/latex/CJK/zh_CN-en-de-el-ru_language-
> > default_pdf2

This should be fixed now as well. Again, this has potential to break
other things, so another cycle of testing would be highly appreciated.

Jürgen


signature.asc
Description: This is a digitally signed message part
-- 
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: [LyX/master] Consider file encoding for modules in reconfigure

2020-08-15 Thread José Abílio Matos

On Saturday, 15 August 2020 17.23.55 WEST JosÃ© Matos wrote:
> commit a8094051c1ae9c546c76bb0d3300d83e6cdbadef
> Author: José Matos 
> Date:   Sat Aug 15 17:49:23 2020 +0100
> 
> Consider file encoding for modules in reconfigure
> 
> If the modules are not in utf8 then we warn and skip that file
> like it happens for layout files.
> 
> It would be nice in both cases to have a warn in the gui and not only in
> the config.log

Riki this should be considered for 2.3.x. :-)

-- 
José Abílio


-- 
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: Test entering an endless loop in latex processing

Am Sat, 15 Aug 2020 17:58:18 +0200
schrieb Jürgen Spitzmüller :

> Am Samstag, den 15.08.2020, 16:02 +0200 schrieb Kornel Benko:
> > This one is really bad, maybe not even lyx's fault.
> > 
> > TL2020:
> > export/export/latex/languages/supported-languages_babel_auto-
> > legacy_pdf2  
> 
> This compiles for me (at 3c5c41ddc2d4a4).
> 
> Jürgen 

If I make a fresh userdir
 $ cd build-dir
 $ rm -rf Testing/.lyx
 $ mkdir Testing/.lyx
 $ ctest -R 
export/export/latex/languages/supported-languages_babel_auto-legacy_pdf2
Test project /BUILD/BUILDMint18/BuildLyxGitQt5.9.5local-gcc8.4.0
Start 485: 
export/export/latex/languages/supported-languages_babel_auto-legacy_pdf2
1/1 Test #485:
export/export/latex/languages/supported-languages_babel_auto-legacy_pdf2 ...   
Passed
8.40 sec

then the test passes.

BUT ... already with the next call to
 $ ctest -R 
export/export/latex/languages/supported-languages_babel_auto-legacy_pdf2

and I can wait, and wait, ...

Kornel


pgpXSGJVvVD0s.pgp
Description: Digitale Signatur von OpenPGP
-- 
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: Crash in export of American_Psychological_Association_%28APA%29,_v._7_

Am Sat, 15 Aug 2020 19:22:49 +0200
schrieb Jürgen Spitzmüller :

> Am Samstag, den 15.08.2020, 15:49 +0200 schrieb Kornel Benko:
> > Am Sat, 15 Aug 2020 15:33:35 +0200  
> > > 402:DEFAULTOUTPUT_export/export/latex/CJK/zh_CN-en-de-el-ru_language-
> > > default_pdf2  
> 
> This should be fixed now as well. Again, this has potential to break
> other things, so another cycle of testing would be highly appreciated.
> 
> Jürgen

Confirmed.
But the full test needs some time.

Kornel


pgp5keGuJ8Ms2.pgp
Description: Digitale Signatur von OpenPGP
-- 
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: [LyX/master] Consider file encoding for modules in reconfigure

On 8/15/20 1:36 PM, José Abílio Matos wrote:
> On Saturday, 15 August 2020 17.23.55 WEST JosÃ© Matos wrote:
>> commit a8094051c1ae9c546c76bb0d3300d83e6cdbadef
>> Author: José Matos 
>> Date:   Sat Aug 15 17:49:23 2020 +0100
>>
>> Consider file encoding for modules in reconfigure
>>
>> If the modules are not in utf8 then we warn and skip that file
>> like it happens for layout files.
>>
>> It would be nice in both cases to have a warn in the gui and not only in
>> the config.log
> Riki this should be considered for 2.3.x. :-)

Thank you, go ahead.

Riki


-- 
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: [Request/idea] Preamble generalization

On 2020-08-15 18:09, Richard Kimberly Heck wrote:

On 8/15/20 4:41 AM, Lorenzo Bertini wrote:

UI-wise, this would probably be better than what we have, but you can
already do this with Local Layout. Like this:

Format 61

AddToHTMLPreamble

h2.section { font-size: 150%; font-weight: bold; }

EndPremable

You can put anything else in there that you want in the head block, too.

Do feel free to file an enhancement request on the bug tracker.

Riki

Thank you, this solves some problems for me. However now I'm
questioning whether the section in document->settings
isn't just a shortcut for in ? In this case I think putting them at the same level is
confusing, and prevented me from ever needing to look deeper in ; I understand not all users would want to learn this but it's
three lines to get a LaTeX or HTML preamble. Ideally there would be a
way to keep the generality of the section that however
makes it super easy and intuitive to change a preamble.

I can see why it might seem that way, but these are quite different in
principle. The Local Layout box is really intended for layout code, like
you'd find in a module, but which is document-specific (not something
you really want to be able to reuse). E.g., I sometimes need a specific
kind of list for one document, so I put the layout code there. It's
basically a side-effect of the fact that the AddTo...Preamble tags exist
in layout code that one can put preamble info there as well. But I
wouldn't really recommend it if you're going to have very much preamble
code.

There's a long-standing request, actually, for there to be TWO preamble
boxes, whose contents are output at different times (before and after
other code that LyX generates). I'm actually not sure whether the layout
code from Local Layout (or from layout code more generally) is output
before or after the code in the Preamble box. So these may not actually
be equivalent.

I leave this decision to you and I'll not file a bug (feature) report
for now, but here's my two cents for keeping things clear and
functional:
1) section gets removed
2) section gets an "advanced" tab with the text entry
we have now, and an "easy" tab made by text boxes, buttons and toggles
(for a start, just containing the infamous LaTeX Preamble and HTML
Preamble, then maybe more).
Except for the advantages in clarity on preambles, I think this would
make way easier not only to tinker with layouts but also to learn how
Lyx works and what are the commands. I know this is a bit of an
overhaul, I'm just leaving this here as an idea.

The complete version of this would give us a kind of GUI layout
editor---another longstanding request and something of a holy grail.
It's possible, presumably, but has proven difficult.

I guess you are not referring to the general preamble stuff here, right?
By GUI layout editor, you mean something like the style widgets of other
word processors, or? There you have a list of all styles and the ability
to edit them and add new ones. For example, one could create new
enumerations via a dialog to set the numbering style. So, this has been
tried? Is it possible to see some of the attempts or difficulties
encountered?

--
Daniel

--
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: Crash in export of American_Psychological_Association_%28APA%29,_v._7_

Am Sat, 15 Aug 2020 20:11:08 +0200
schrieb Kornel Benko :

> Am Sat, 15 Aug 2020 19:22:49 +0200
> schrieb Jürgen Spitzmüller :
> 
> > Am Samstag, den 15.08.2020, 15:49 +0200 schrieb Kornel Benko:  
> > > Am Sat, 15 Aug 2020 15:33:35 +0200
> > > > 402:DEFAULTOUTPUT_export/export/latex/CJK/zh_CN-en-de-el-ru_language-
> > > > default_pdf2
> > 
> > This should be fixed now as well. Again, this has potential to break
> > other things, so another cycle of testing would be highly appreciated.
> > 
> > Jürgen  
> 
> Confirmed.
> But the full test needs some time.
> 
>   Kornel

Hm, you was right.

Kornel


LastExportTestsFailed.log.1.tmp
Description: Binary data


pgpV05eqlp3Ax.pgp
Description: Digitale Signatur von OpenPGP
-- 
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: [RFC] Missing DocIterator methods

2020-08-15 Thread Jean-Marc Lasgouttes

Le 15 août 2020 18:21:18 GMT+02:00, Richard Kimberly Heck  a 
écrit :
>On 8/14/20 4:02 PM, Stephan Witt wrote:
>> Hi all,
>>
>> to implement the correct and Mac conforming solution for navigation
>through the LyX document with keyboard I need to go forward and
>backward. Sounds simple and I want to use the DocIterator for that.
>Nevertheless I missed two methods to use DocIterators for backward
>iteration. First I’d like to have the opposite test to atEnd() - name
>it atBegin() to stop the loop at beginning. Second there is the
>possibility to go forward and skip closed or readonly text insets. But
>the symmetric method for backward traversal isn’t there.
>>
>> I’ve attached a patch to implement these methods. Did I overlook
>something and if not is it the right thing? I found it useful and
>working and like to add it. Any objections?
>
>No, but I'm not terribly familiar with this code.
>
>Riki
>
>
>-- 
>lyx-devel mailing list
>lyx-devel@lists.lyx.org
>http://lists.lyx.org/mailman/listinfo/lyx-devel

I cannot look at the patch but the principle is fine.

JMarc
-- 
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: [LyX/master] DocBook: bug fixing with several bibliographic references at once.

2020-08-15 Thread Thibaut Cuvelier

On Sat, 15 Aug 2020 at 09:26, Jürgen Spitzmüller  wrote:

> Am Samstag, den 01.08.2020, 03:26 +0200 schrieb Thibaut Cuvelier:
> > @@ -627,7 +628,7 @@ void InsetCitation::latex(otexstream & os,
> > OutputParams const & runparams) const
> >
> > // check if we have to do a qualified list
> >
> > vector keys =
> > getVectorFromString(cleanupWhitespace(key));
> >
> > bool const qualified = cs.hasQualifiedList
> >
> > -   && (!getParam("pretextlist").empty()
> >
> > +   && (!getParam("F").empty()
>
> I take it this went in by accident. In any case it is wrong and also
> causes an assertion. I'll revert it.
>
> Thibaut, if there is a rationale behind it, please let me know so we
> can figure out the correct way.
>

 It may be due to an attempt to solve a problem when many references are
present in a single inset, but testing your code shows no regression on
bibliography_precooked_aastex
.
I suppose there was no real reason to correct it in the first place.
-- 
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: [LyX/master] DocBook: bug fixing with several bibliographic references at once.

Am Sat, 15 Aug 2020 22:59:19 +0200
schrieb Thibaut Cuvelier :

> On Sat, 15 Aug 2020 at 09:26, Jürgen Spitzmüller  wrote:
> 
> > Am Samstag, den 01.08.2020, 03:26 +0200 schrieb Thibaut Cuvelier:
> > > @@ -627,7 +628,7 @@ void InsetCitation::latex(otexstream & os,
> > > OutputParams const & runparams) const
> > >
> > > // check if we have to do a qualified list
> > >
> > > vector keys =
> > > getVectorFromString(cleanupWhitespace(key));
> > >
> > > bool const qualified = cs.hasQualifiedList
> > >
> > > -   && (!getParam("pretextlist").empty()
> > >
> > > +   && (!getParam("F").empty()
> >
> > I take it this went in by accident. In any case it is wrong and also
> > causes an assertion. I'll revert it.
> >
> > Thibaut, if there is a rationale behind it, please let me know so we
> > can figure out the correct way.
> >
> 
>  It may be due to an attempt to solve a problem when many references are
> present in a single inset, but testing your code shows no regression on
> bibliography_precooked_aastex
> .
> I suppose there was no real reason to correct it in the first place.

It led to a crash of a testcase. (Exporting
American_Psychological_Association_%28APA%29,_v._7.lyx)

Kornel


pgpHXem6m5yog.pgp
Description: Digitale Signatur von OpenPGP
-- 
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Re: Allow empty line ... I know