From spitz at lyx.org Mon Nov 1 08:56:05 2021 From: spitz at lyx.org (=?ISO-8859-1?Q?J=FCrgen_Spitzm=FCller?=) Date: Mon, 01 Nov 2021 09:56:05 +0100 Subject: Footnote size is larger on master (\normalsize is prepended) In-Reply-To: <20211031171101.a32end5eh45mx5x2@poole> References: <20211031171101.a32end5eh45mx5x2@poole> Message-ID: <6164e06e3490b4e31f8baa4a383278f198520ff1.camel@lyx.org> Am Sonntag, dem 31.10.2021 um 13:11 -0400 schrieb Scott Kostyshak: > See the attached file. On master, \normalsize is included inside the > footnote, which is different from 2.3.x. To fix this, one needs? (1.) to set inheritFont() in InsetFootlike to true (i.e., back to default) in order to prevent unnecessary font switches and (2.) revert b37929fd to correct the workarea font in footnotes (which otherwise inherits the font of the context, e.g. section) JMarc, what do you think about (2.)? It was part of your series to address #12204. J?rgen -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 488 bytes Desc: This is a digitally signed message part URL: From spitz at lyx.org Mon Nov 1 09:29:33 2021 From: spitz at lyx.org (=?ISO-8859-1?Q?J=FCrgen_Spitzm=FCller?=) Date: Mon, 01 Nov 2021 10:29:33 +0100 Subject: Footnote size is larger on master (\normalsize is prepended) In-Reply-To: <6164e06e3490b4e31f8baa4a383278f198520ff1.camel@lyx.org> References: <20211031171101.a32end5eh45mx5x2@poole> <6164e06e3490b4e31f8baa4a383278f198520ff1.camel@lyx.org> Message-ID: <83950bd68c4552be972f950e3453f5eaf56b0de4.camel@lyx.org> Am Montag, dem 01.11.2021 um 09:56 +0100 schrieb J?rgen Spitzm?ller: > (2.) revert b37929fd to correct the workarea font in footnotes (which > otherwise inherits the font of the context, e.g. section) > > JMarc, what do you think about (2.)? It was part of your series to > address #12204. On another look I tend to think that, rather than overwriting metrics and draw in individual insets, we would need a method similar to inheritFont() which only applies to screen font (I thought that resetsFonEdit() is that, but it isn't). J?rgen -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 488 bytes Desc: This is a digitally signed message part URL: From spitz at lyx.org Mon Nov 1 10:21:49 2021 From: spitz at lyx.org (=?ISO-8859-1?Q?J=FCrgen_Spitzm=FCller?=) Date: Mon, 01 Nov 2021 11:21:49 +0100 Subject: Footnote size is larger on master (\normalsize is prepended) In-Reply-To: <83950bd68c4552be972f950e3453f5eaf56b0de4.camel@lyx.org> References: <20211031171101.a32end5eh45mx5x2@poole> <6164e06e3490b4e31f8baa4a383278f198520ff1.camel@lyx.org> <83950bd68c4552be972f950e3453f5eaf56b0de4.camel@lyx.org> Message-ID: Am Montag, dem 01.11.2021 um 10:29 +0100 schrieb J?rgen Spitzm?ller: > On another look I tend to think that, rather than overwriting metrics > and draw in individual insets, we would need a method similar to > inheritFont() which only applies to screen font (I thought that > resetsFonEdit() is that, but it isn't). Something like the attached. I figure, in the long run we also need a layout tag. J?rgen -------------- next part -------------- A non-text attachment was scrubbed... Name: x.diff Type: text/x-patch Size: 2530 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 488 bytes Desc: This is a digitally signed message part URL: From yuriy.skalko at gmail.com Mon Nov 1 13:52:02 2021 From: yuriy.skalko at gmail.com (Yuriy Skalko) Date: Mon, 1 Nov 2021 15:52:02 +0200 Subject: Marking inverted branch insets In-Reply-To: <94b4c7d4-926a-a8e2-acf0-c7318539b04f@lyx.org> References: <94b4c7d4-926a-a8e2-acf0-c7318539b04f@lyx.org> Message-ID: <35ef4fb7-5020-e889-57cb-c448a8bbf39c@gmail.com> > Sounds reasonable to me. > > Riki Riki, is it OK to have this also in stable? Yuriy From sanda at lyx.org Mon Nov 1 13:15:49 2021 From: sanda at lyx.org (Pavel Sanda) Date: Mon, 1 Nov 2021 14:15:49 +0100 Subject: master crash (tocwidget code) Message-ID: <20211101131549.GA13671@jabberwock.ucw.cz> Hi, I get the crash in majority of cases when using the following recipy: 1. launch lyx & open introduction manual 2. open new window via file -> new window 3. view->hidden->introduction 4. kaboom /usr/include/c++/10/debug/vector:434: In function: std::__debug::vector<_Tp, _Allocator>::const_reference std::__debug::vector<_Tp, _Allocator>::operator[](std::__debug::vector<_Tp, _Allocator>::size_type) const [with _Tp = lyx::TocItem; _Allocator = std::allocator; std::__debug::vector<_Tp, _Allocator>::const_reference = const lyx::TocItem&; std::__debug::vector<_Tp, _Allocator>::size_type = long unsigned int] Error: attempt to subscript container with out-of-bounds index 0, but container only holds 0 elements. Objects involved in the operation: sequence "this" @ 0x0x555557f3b0d0 { type = std::__debug::vector >; } Thread 1 "lyx" received signal SIGABRT, Aborted. __GI_raise (sig=sig at entry=6) at ../sysdeps/unix/sysv/linux/raise.c:50 50 ../sysdeps/unix/sysv/linux/raise.c: No such file or directory. (gdb) bt #0 __GI_raise (sig=sig at entry=6) at ../sysdeps/unix/sysv/linux/raise.c:50 #1 0x00007ffff6aa4537 in __GI_abort () at abort.c:79 #2 0x00007ffff694eeb1 in ?? () from /usr/lib/x86_64-linux-gnu/libstdc++.so.6 #3 0x0000555556100051 in lyx::frontend::TocModel::tocItem(QModelIndex const&) const () #4 0x0000555556102cbc in lyx::frontend::TocModels::currentItem(QString const&, QModelIndex const&) const () #5 0x00005555563080c4 in lyx::frontend::TocWidget::filterContents() () #6 0x000055555630856f in lyx::frontend::TocWidget::finishUpdateView() () #7 0x000055555630947b in lyx::frontend::TocWidget::qt_static_metacall(QObject*, QMetaObject::Call, int, void**) () bisect leads to: commit 4d7f4762a1af2bbe4d2051bd86e45c35361f0cfc Author: Juergen Spitzmueller Date: Sun Mar 15 10:46:35 2020 +0100 Outliner: Add filter combo for non-output items Addresses #11442, #10786 Looking at the backtrace and commit it seems that + TocItem const & item = + gui_view_.tocModels().currentItem(current_type_, index); triggers the crash. Juergen, can you reproduce? Pavel From spitz at lyx.org Mon Nov 1 16:01:34 2021 From: spitz at lyx.org (=?ISO-8859-1?Q?J=FCrgen_Spitzm=FCller?=) Date: Mon, 01 Nov 2021 17:01:34 +0100 Subject: master crash (tocwidget code) In-Reply-To: <20211101131549.GA13671@jabberwock.ucw.cz> References: <20211101131549.GA13671@jabberwock.ucw.cz> Message-ID: <7f44ca07ff7c3d8844639fa96a86e10a8eee75e5.camel@lyx.org> Am Montag, dem 01.11.2021 um 14:15 +0100 schrieb Pavel Sanda: > Looking at the backtrace and commit it seems that > +???????? TocItem const & item = > +???????????????? gui_view_.tocModels().currentItem(current_type_, > index); > triggers the crash. > > Juergen, can you reproduce? Yes. I am not sure what goes on here exactly, but the toc seems empty after the window switch. I don't think the code above is to blame, it just reveals the problem. The attached patch prevents the crash, but I am not sure whether it doesn't hide the actual problem. Do we need to update the respective toc somewhere? But where? J?rgen -------------- next part -------------- A non-text attachment was scrubbed... Name: y.diff Type: text/x-patch Size: 868 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 488 bytes Desc: This is a digitally signed message part URL: From skostysh at lyx.org Mon Nov 1 19:27:36 2021 From: skostysh at lyx.org (Scott Kostyshak) Date: Mon, 1 Nov 2021 15:27:36 -0400 Subject: Footnote size is larger on master (\normalsize is prepended) In-Reply-To: References: <20211031171101.a32end5eh45mx5x2@poole> <6164e06e3490b4e31f8baa4a383278f198520ff1.camel@lyx.org> <83950bd68c4552be972f950e3453f5eaf56b0de4.camel@lyx.org> Message-ID: <20211101192736.pz3iney6umsbgo6n@poole> On Mon, Nov 01, 2021 at 11:21:49AM +0100, J?rgen Spitzm?ller wrote: > Am Montag, dem 01.11.2021 um 10:29 +0100 schrieb J?rgen Spitzm?ller: > > On another look I tend to think that, rather than overwriting metrics > > and draw in individual insets, we would need a method similar to > > inheritFont() which only applies to screen font (I thought that > > resetsFonEdit() is that, but it isn't). > > Something like the attached. I figure, in the long run we also need a > layout tag. Fixes the original issue from what I can see, although I know nothing about the code. I'm not sure if the following is a related issue. See attached. It's hard to see, but the in-line footnote number itself is now bold on master. Note the "\bfseries" in the difference in TeX code produced below. master with your patch: To set the bracket size for the second line correctly, the first line is ended with \textbf{\textbackslash right.}\index{Commands!R!right@\textbackslash right} and the second line with \textbf{\textbackslash left.}{\bfseries\footnote{for more about \textbf{\textbackslash left} and \textbf{\textbackslash right} 2.3.x: To set the bracket size for the second line correctly, the first line is ended with \textbf{\textbackslash right.}\index{Commands!R!right@\textbackslash right} and the second line with \textbf{\textbackslash left.}\footnote{for more about \textbf{\textbackslash left} and \textbf{\textbackslash right} Scott -------------- next part -------------- A non-text attachment was scrubbed... Name: screenshot1635794790.png Type: image/png Size: 93267 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 488 bytes Desc: not available URL: From sanda at lyx.org Mon Nov 1 21:51:24 2021 From: sanda at lyx.org (Pavel Sanda) Date: Mon, 1 Nov 2021 22:51:24 +0100 Subject: master crash (tocwidget code) In-Reply-To: <7f44ca07ff7c3d8844639fa96a86e10a8eee75e5.camel@lyx.org> References: <20211101131549.GA13671@jabberwock.ucw.cz> <7f44ca07ff7c3d8844639fa96a86e10a8eee75e5.camel@lyx.org> Message-ID: <20211101215124.GA26903@jabberwock.ucw.cz> On Mon, Nov 01, 2021 at 05:01:34PM +0100, J?rgen Spitzm?ller wrote: > Am Montag, dem 01.11.2021 um 14:15 +0100 schrieb Pavel Sanda: > > Looking at the backtrace and commit it seems that > > +???????? TocItem const & item = > > +???????????????? gui_view_.tocModels().currentItem(current_type_, > > index); > > triggers the crash. > > > > Juergen, can you reproduce? > > Yes. I am not sure what goes on here exactly, but the toc seems empty > after the window switch. I don't think the code above is to blame, it > just reveals the problem. > > The attached patch prevents the crash, but I am not sure whether it > doesn't hide the actual problem. Do we need to update the respective > toc somewhere? But where? I don't know the code in question either, but I can try to look later this week. Pavel From tcuvelier at lyx.org Mon Nov 1 23:55:54 2021 From: tcuvelier at lyx.org (Thibaut Cuvelier) Date: Tue, 2 Nov 2021 00:55:54 +0100 Subject: A couple of lilypond docbook tests failing In-Reply-To: <20211031170245.6otzlkzvdateromh@poole> References: <20211031170245.6otzlkzvdateromh@poole> Message-ID: Hi Scott, What version of LilyPond are you using? I am indeed getting errors at least with the first document, but not at all the same ones. The error you get should have been solved by e22f52e7, at least on Windows (os.chdir solved it for me in some cases). You could chime in https://gitlab.com/lilypond/lilypond/-/issues/6187 to indicate you have the same issue on Linux. Here is the error I get (only the LilyPond part of the output): >> Given arguments: >> LilyPond: present. >> LilyPond callable as: C:\Program Files (x86)\LilyPond\usr\bin/lilypond-book. >> LilyPond path: C:\Program Files (x86)\LilyPond\usr\bin. >> Input file: C:/Users/Thibaut/AppData/Local/Temp/lyx_tmpdir.qugZiVZlvBWx/lyx_tmpbuf0/LilyPond_Book.xml. >> Output file: D:/LyX/lyx-unstable/lib/examples/Modules/LilyPond_Book.xml. >> Input folder: C:/Users/Thibaut/AppData/Local/Temp/lyx_tmpdir.qugZiVZlvBWx/lyx_tmpbuf0. >> Output folder: D:/LyX/lyx-unstable/lib/examples/Modules. >> The input file needs a LilyPond pass and LilyPond is available. >> Rewriting C:/Users/Thibaut/AppData/Local/Temp/lyx_tmpdir.qugZiVZlvBWx/lyx_tmpbuf0/LilyPond_Book.xml >> as C:/Users/Thibaut/AppData/Local/Temp/lyx_tmpdir.qugZiVZlvBWx/lyx_tmpbuf0/LilyPond_Book.lyxml. >> Running LilyPond. lilypond-book (GNU LilyPond) 2.23.3 Reading `C:\Users\Thibaut\AppData\Local\Temp\lyx_tmpdir.qugZiVZlvBWx\lyx_tmpbuf0\LilyPond_Book.lyxml' Dissecting... Writing snippets... Processing... Running lilypond... Processing `C:/Users/Thibaut/AppData/Local/Temp/lyx_tmpdir.qugZiVZlvBWx/lyx_tmpbuf0/21/ lily-0a4ea353.ly' Parsing... C:/Users/Thibaut/AppData/Local/Temp/lyx_tmpdir.qugZiVZlvBWx/lyx_tmpbuf0/21/lily-0a4ea353.ly:127:17: error: undefined cha racter or shorthand: & \new PianoStaff << % RH Staff \new Staff { \clef treble \key aes \major \time 2/4 %\set Staff.finger ingOrientations = #'(up) %\set Score.currentBarNumber = #9 \partial 8 <aes''-4 c''>8\staccato | %\s et doubleSlurs = ##t <bes''-5 des''-2>4( <aes''-4 c''>8\staccato) \noBeam c''8-5\staccato\pp | \unset doubleSlurs bes'8..( aes'32 g'8)\staccato f'8\staccato | e'4-2 r8 %\once \override S cript #'script-priority = #-100 \afterGrace f'8(\trill^\markup { \finger "3-2" } { e'16[ f'16] } | g'8..- 3 f'32 e'8-1)\staccato d'8-2\staccato | c'4 r4 } % LH Staff \new Staff { \key aes \major \cle f treble %\override Fingering #'direction = #down %\set Staff.fingeringOrientations = #'(down) \partial 8 &l t;aes' aes>8\staccato %\set doubleSlurs = ##t <g'-2 ees'-3>4( <aes' aes>8)\staccato \noBeam \c C:/Users/Thibaut/AppData/Local/Temp/lyx_tmpdir.qugZiVZlvBWx/lyx_tmpbuf0/21/lily-0a4ea353.ly:127:17: error: syntax error, unexpected STRING I have reported this issue to LilyPond, because there is nothing LyX could do to prevent it: https://gitlab.com/lilypond/lilypond/-/issues/6204. On Sun, 31 Oct 2021 at 18:02, Scott Kostyshak wrote: > The following tests are failing for me: > > The following tests FAILED: > 4010 - export/examples/Modules/LilyPond_Book_docbook5 (Failed) > 5165 - > INVERTED.DOCBOOK_export/examples/fr/Modules/Linguistics_docbook5 (Failed) > 5557 - export/examples/ja/Modules/LilyPond_Book_docbook5 (Failed) > > When I open lib/examples/Modules/LilyPond_Book.lyx and export to docbook5, > I get the following output in the terminal: > > Command '/usr/bin/lilypond --formats=ps,png -dbackend=eps -I "./" -I > "/tmp/lyx_tmpdir.xdBexxcIwFHp/lyx_tmpbuf0" -deps-box-padding=3.000000 > -dread-file-list -dno-strip-output-dir > "/tmp/lyx_tmpdir.xdBexxcIwFHp/lyx_tmpbuf0/ > snippet-names-8f8c627f87a88430f6904f4119093d6b.ly"' returned non-zero > exit status 1. > python: can't open file > '/tmp/lyx_tmpdir.xdBexxcIwFHp/lyx_tmpbuf0/lilypond-book': [Errno 2] No such > file or directory > >> Error from LilyPond. The successive calls were: > Traceback (most recent call last): > File "/home/scott/lyxbuilds/master/repo/lib/scripts/docbook_copy.py", > line 174, in > DocBookCopier(sys.argv).copy() > File "/home/scott/lyxbuilds/master/repo/lib/scripts/docbook_copy.py", > line 161, in copy > self.call_lilypond() > File "/home/scott/lyxbuilds/master/repo/lib/scripts/docbook_copy.py", > line 128, in call_lilypond > print('>> (' + i + ') Error from trying ' + str(cmd) + ':') > TypeError: can only concatenate str (not "int") to str > support/Systemcall.cpp (291): Systemcall: 'python > "/home/scott/lyxbuilds/master/repo/lib/scripts/docbook_copy.py" > lilypond-book "/tmp/lyx_tmpdir.xdBexxcIwFHp/lyx_tmpbuf0/LilyPond_Book.xml" > "/home/scott/lyxbuilds/master/repo/lib/examples/Modules/LilyPond_Book.xml"' > finished with exit code 1 > Error: Couldn't copy file > > Scott > -------------- next part -------------- An HTML attachment was scrubbed... URL: From skostysh at lyx.org Tue Nov 2 01:13:04 2021 From: skostysh at lyx.org (Scott Kostyshak) Date: Mon, 1 Nov 2021 21:13:04 -0400 Subject: A couple of lilypond docbook tests failing In-Reply-To: References: <20211031170245.6otzlkzvdateromh@poole> Message-ID: <20211102011304.dmrcr7gjpxxwpd4o@poole> On Tue, Nov 02, 2021 at 12:55:54AM +0100, Thibaut Cuvelier wrote: > Hi Scott, > > What version of LilyPond are you using? Probably an old one: $ lilypond -v GNU LilyPond 2.22.0 > I am indeed getting errors at least > with the first document, but not at all the same ones. The error you get > should have been solved by e22f52e7, at least on Windows (os.chdir solved > it for me in some cases). Strange, I did the tests on current master. > You could chime in > https://gitlab.com/lilypond/lilypond/-/issues/6187 to indicate you have the > same issue on Linux. Thanks for making the upstream bug reports. I subscribed to the issue. Scott > Here is the error I get (only the LilyPond part of the output): > > >> Given arguments: > >> LilyPond: present. > >> LilyPond callable as: C:\Program Files > (x86)\LilyPond\usr\bin/lilypond-book. > >> LilyPond path: C:\Program Files (x86)\LilyPond\usr\bin. > >> Input file: > C:/Users/Thibaut/AppData/Local/Temp/lyx_tmpdir.qugZiVZlvBWx/lyx_tmpbuf0/LilyPond_Book.xml. > >> Output file: D:/LyX/lyx-unstable/lib/examples/Modules/LilyPond_Book.xml. > >> Input folder: > C:/Users/Thibaut/AppData/Local/Temp/lyx_tmpdir.qugZiVZlvBWx/lyx_tmpbuf0. > >> Output folder: D:/LyX/lyx-unstable/lib/examples/Modules. > >> The input file needs a LilyPond pass and LilyPond is available. > >> Rewriting > C:/Users/Thibaut/AppData/Local/Temp/lyx_tmpdir.qugZiVZlvBWx/lyx_tmpbuf0/LilyPond_Book.xml > >> as > C:/Users/Thibaut/AppData/Local/Temp/lyx_tmpdir.qugZiVZlvBWx/lyx_tmpbuf0/LilyPond_Book.lyxml. > >> Running LilyPond. > lilypond-book (GNU LilyPond) 2.23.3 > Reading > `C:\Users\Thibaut\AppData\Local\Temp\lyx_tmpdir.qugZiVZlvBWx\lyx_tmpbuf0\LilyPond_Book.lyxml' > Dissecting... > Writing snippets... > Processing... > Running lilypond... > Processing > `C:/Users/Thibaut/AppData/Local/Temp/lyx_tmpdir.qugZiVZlvBWx/lyx_tmpbuf0/21/ > lily-0a4ea353.ly' > Parsing... > C:/Users/Thibaut/AppData/Local/Temp/lyx_tmpdir.qugZiVZlvBWx/lyx_tmpbuf0/21/lily-0a4ea353.ly:127:17: > error: undefined cha > racter or shorthand: & > \new PianoStaff > << % RH Staff \new Staff { \clef treble \key > aes \major \time 2/4 %\set Staff.finger > ingOrientations = #'(up) %\set Score.currentBarNumber = #9 \partial 8 > <aes''-4 c''>8\staccato | %\s > et doubleSlurs = ##t <bes''-5 des''-2>4( <aes''-4 > c''>8\staccato) \noBeam c''8-5\staccato\pp > | \unset doubleSlurs bes'8..( aes'32 g'8)\staccato > f'8\staccato | e'4-2 r8 %\once \override S > cript #'script-priority = #-100 \afterGrace f'8(\trill^\markup { > \finger "3-2" } { e'16[ f'16] } | g'8..- > 3 f'32 e'8-1)\staccato d'8-2\staccato | c'4 r4 } % LH > Staff \new Staff { \key aes \major \cle > f treble %\override Fingering #'direction = #down %\set > Staff.fingeringOrientations = #'(down) \partial 8 &l > t;aes' aes>8\staccato %\set doubleSlurs = ##t <g'-2 > ees'-3>4( <aes' aes>8)\staccato \noBeam > \c > C:/Users/Thibaut/AppData/Local/Temp/lyx_tmpdir.qugZiVZlvBWx/lyx_tmpbuf0/21/lily-0a4ea353.ly:127:17: > error: syntax error, > unexpected STRING > > I have reported this issue to LilyPond, because there is nothing LyX could > do to prevent it: https://gitlab.com/lilypond/lilypond/-/issues/6204. > > On Sun, 31 Oct 2021 at 18:02, Scott Kostyshak wrote: > > > The following tests are failing for me: > > > > The following tests FAILED: > > 4010 - export/examples/Modules/LilyPond_Book_docbook5 (Failed) > > 5165 - > > INVERTED.DOCBOOK_export/examples/fr/Modules/Linguistics_docbook5 (Failed) > > 5557 - export/examples/ja/Modules/LilyPond_Book_docbook5 (Failed) > > > > When I open lib/examples/Modules/LilyPond_Book.lyx and export to docbook5, > > I get the following output in the terminal: > > > > Command '/usr/bin/lilypond --formats=ps,png -dbackend=eps -I "./" -I > > "/tmp/lyx_tmpdir.xdBexxcIwFHp/lyx_tmpbuf0" -deps-box-padding=3.000000 > > -dread-file-list -dno-strip-output-dir > > "/tmp/lyx_tmpdir.xdBexxcIwFHp/lyx_tmpbuf0/ > > snippet-names-8f8c627f87a88430f6904f4119093d6b.ly"' returned non-zero > > exit status 1. > > python: can't open file > > '/tmp/lyx_tmpdir.xdBexxcIwFHp/lyx_tmpbuf0/lilypond-book': [Errno 2] No such > > file or directory > > >> Error from LilyPond. The successive calls were: > > Traceback (most recent call last): > > File "/home/scott/lyxbuilds/master/repo/lib/scripts/docbook_copy.py", > > line 174, in > > DocBookCopier(sys.argv).copy() > > File "/home/scott/lyxbuilds/master/repo/lib/scripts/docbook_copy.py", > > line 161, in copy > > self.call_lilypond() > > File "/home/scott/lyxbuilds/master/repo/lib/scripts/docbook_copy.py", > > line 128, in call_lilypond > > print('>> (' + i + ') Error from trying ' + str(cmd) + ':') > > TypeError: can only concatenate str (not "int") to str > > support/Systemcall.cpp (291): Systemcall: 'python > > "/home/scott/lyxbuilds/master/repo/lib/scripts/docbook_copy.py" > > lilypond-book "/tmp/lyx_tmpdir.xdBexxcIwFHp/lyx_tmpbuf0/LilyPond_Book.xml" > > "/home/scott/lyxbuilds/master/repo/lib/examples/Modules/LilyPond_Book.xml"' > > finished with exit code 1 > > Error: Couldn't copy file > > > > Scott > > > -- > lyx-devel mailing list > lyx-devel at lists.lyx.org > http://lists.lyx.org/mailman/listinfo/lyx-devel -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 488 bytes Desc: not available URL: From spitz at lyx.org Tue Nov 2 07:08:05 2021 From: spitz at lyx.org (=?ISO-8859-1?Q?J=FCrgen_Spitzm=FCller?=) Date: Tue, 02 Nov 2021 08:08:05 +0100 Subject: master crash (tocwidget code) In-Reply-To: <20211101215124.GA26903@jabberwock.ucw.cz> References: <20211101131549.GA13671@jabberwock.ucw.cz> <7f44ca07ff7c3d8844639fa96a86e10a8eee75e5.camel@lyx.org> <20211101215124.GA26903@jabberwock.ucw.cz> Message-ID: Am Montag, dem 01.11.2021 um 22:51 +0100 schrieb Pavel Sanda: > I don't know the code in question either, but I can try to look later > this week. OTOH we test for empty toc_ in several other places of the TocModel, so the check the patch adds might be reasonable (maybe the toc is not yet re-created at that point). J?rgen -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 488 bytes Desc: This is a digitally signed message part URL: From spitz at lyx.org Tue Nov 2 07:33:47 2021 From: spitz at lyx.org (=?ISO-8859-1?Q?J=FCrgen_Spitzm=FCller?=) Date: Tue, 02 Nov 2021 08:33:47 +0100 Subject: Footnote size is larger on master (\normalsize is prepended) In-Reply-To: <20211101192736.pz3iney6umsbgo6n@poole> References: <20211031171101.a32end5eh45mx5x2@poole> <6164e06e3490b4e31f8baa4a383278f198520ff1.camel@lyx.org> <83950bd68c4552be972f950e3453f5eaf56b0de4.camel@lyx.org> <20211101192736.pz3iney6umsbgo6n@poole> Message-ID: <2e25e908d8079d866c1afd43356f88bdeb1e117d.camel@lyx.org> Am Montag, dem 01.11.2021 um 15:27 -0400 schrieb Scott Kostyshak: > Fixes the original issue from what I can see, although I know nothing > about the code. I'm not sure if the following is a related issue. See > attached. It's hard to see, but the in-line footnote number itself is > now bold on master. Note the "\bfseries" in the difference in TeX > code > produced below. > > master with your patch: > > ? To set the bracket size for the second line correctly, the first > line > ? is ended with \textbf{\textbackslash > right.}\index{Commands!R!right@\textbackslash right} > ? and the second line with \textbf{\textbackslash > left.}{\bfseries\footnote{for more about \textbf{\textbackslash left} > and \textbf{\textbackslash right} > > 2.3.x: > > ? To set the bracket size for the second line correctly, the first > line > ? is ended with \textbf{\textbackslash > right.}\index{Commands!R!right@\textbackslash right} > ? and the second line with \textbf{\textbackslash left.}\footnote{for > more about \textbf{\textbackslash left} and \textbf{\textbackslash > right} I'd need to see the LyX file, but if the footnote itself is set bold there, master is correct and stable is wrong. This is something we fixed in master. J?rgen -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 488 bytes Desc: This is a digitally signed message part URL: From lasgouttes at lyx.org Tue Nov 2 13:08:29 2021 From: lasgouttes at lyx.org (Jean-Marc Lasgouttes) Date: Tue, 2 Nov 2021 14:08:29 +0100 Subject: Footnote size is larger on master (\normalsize is prepended) In-Reply-To: <83950bd68c4552be972f950e3453f5eaf56b0de4.camel@lyx.org> References: <20211031171101.a32end5eh45mx5x2@poole> <6164e06e3490b4e31f8baa4a383278f198520ff1.camel@lyx.org> <83950bd68c4552be972f950e3453f5eaf56b0de4.camel@lyx.org> Message-ID: Le 01/11/2021 ? 10:29, J?rgen Spitzm?ller a ?crit?: > Am Montag, dem 01.11.2021 um 09:56 +0100 schrieb J?rgen Spitzm?ller: >> (2.) revert b37929fd to correct the workarea font in footnotes (which >> otherwise inherits the font of the context, e.g. section) >> >> JMarc, what do you think about (2.)? It was part of your series to >> address #12204. > > On another look I tend to think that, rather than overwriting metrics > and draw in individual insets, we would need a method similar to > inheritFont() which only applies to screen font (I thought that > resetsFonEdit() is that, but it isn't). I do not see the needed for treating the screen differently from the lyx file. I'll try to propose something, but currently I do not have much free time. JMarc From spitz at lyx.org Tue Nov 2 13:45:14 2021 From: spitz at lyx.org (=?ISO-8859-1?Q?J=FCrgen_Spitzm=FCller?=) Date: Tue, 02 Nov 2021 14:45:14 +0100 Subject: Footnote size is larger on master (\normalsize is prepended) In-Reply-To: References: <20211031171101.a32end5eh45mx5x2@poole> <6164e06e3490b4e31f8baa4a383278f198520ff1.camel@lyx.org> <83950bd68c4552be972f950e3453f5eaf56b0de4.camel@lyx.org> Message-ID: <1a9052954ffb47fb103c2209d133ed1cf85f89d8.camel@lyx.org> Am Dienstag, dem 02.11.2021 um 14:08 +0100 schrieb Jean-Marc Lasgouttes: > I do not see the needed for treating the screen differently from the > lyx file.? My point was to differentiate inheritance of font properties on screen from inheritance in the TeX output (nothing to do with the LyX file). Some contexts (footnote is a case in point) require us to use different font on screen but not change the font in output (as this is done by the contexts themselves). So footnote should inheritFont() but not inheritScreenFont(). This is not possible with the current mechanism (and it only worked since font metrics were hardcoded in InsetFootlike before you removed that hardcoding). > I'll try to propose something, but currently I do not have much > free time. OK. J?rgen -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 488 bytes Desc: This is a digitally signed message part URL: From lasgouttes at lyx.org Tue Nov 2 14:40:00 2021 From: lasgouttes at lyx.org (Jean-Marc Lasgouttes) Date: Tue, 2 Nov 2021 15:40:00 +0100 Subject: Footnote size is larger on master (\normalsize is prepended) In-Reply-To: <1a9052954ffb47fb103c2209d133ed1cf85f89d8.camel@lyx.org> References: <20211031171101.a32end5eh45mx5x2@poole> <6164e06e3490b4e31f8baa4a383278f198520ff1.camel@lyx.org> <83950bd68c4552be972f950e3453f5eaf56b0de4.camel@lyx.org> <1a9052954ffb47fb103c2209d133ed1cf85f89d8.camel@lyx.org> Message-ID: <871ec28b-fc97-8bbd-aaa8-04a76a669558@lyx.org> Le 02/11/2021 ? 14:45, J?rgen Spitzm?ller a ?crit?: > Am Dienstag, dem 02.11.2021 um 14:08 +0100 schrieb Jean-Marc > Lasgouttes: >> I do not see the needed for treating the screen differently from the >> lyx file. > > My point was to differentiate inheritance of font properties on screen > from inheritance in the TeX output (nothing to do with the LyX file). > Some contexts (footnote is a case in point) require us to use different > font on screen but not change the font in output (as this is done by > the contexts themselves). So footnote should inheritFont() but not > inheritScreenFont(). This is not possible with the current mechanism > (and it only worked since font metrics were hardcoded in InsetFootlike > before you removed that hardcoding). Why do you want to inherit anything for the footnote? OK, it turns out that using \normalfont (as \footnotetext does) resets everything but the font size (I believe), but footnote does reset the font completely. And I think that it makes sense that it does that. So the font at the point where the footnote is is not relevant semantically. Since we have had this discussion already, so I guess the point is that you disagree with me. Otherwise, I think I know how we could coerce lyx2lyx to convert from resetfont to inheritfont. JMarc From spitz at lyx.org Tue Nov 2 14:52:43 2021 From: spitz at lyx.org (=?ISO-8859-1?Q?J=FCrgen_Spitzm=FCller?=) Date: Tue, 02 Nov 2021 15:52:43 +0100 Subject: Footnote size is larger on master (\normalsize is prepended) In-Reply-To: <871ec28b-fc97-8bbd-aaa8-04a76a669558@lyx.org> References: <20211031171101.a32end5eh45mx5x2@poole> <6164e06e3490b4e31f8baa4a383278f198520ff1.camel@lyx.org> <83950bd68c4552be972f950e3453f5eaf56b0de4.camel@lyx.org> <1a9052954ffb47fb103c2209d133ed1cf85f89d8.camel@lyx.org> <871ec28b-fc97-8bbd-aaa8-04a76a669558@lyx.org> Message-ID: Am Dienstag, dem 02.11.2021 um 15:40 +0100 schrieb Jean-Marc Lasgouttes: > Why do you want to inherit anything for the footnote? OK, it turns > out that using \normalfont (as \footnotetext does) resets everything > but the font size (I believe), but footnote does reset the font > completely. With inheritFont() == false, we end up in unnecessary font switches at footnote start, e.g. \footnote{\normalsize text ...} in section headings (see Scott's example), and I want to avoid that. With inheritFont() == true, these switches are gone (as it should be), but the font in the workarea is wrong. resetsFontEdit() (layout's ResetsFont) does not help here, it does something different (don't ask me, what). If we are stuck with the current choices, I'd rather prefer wrong workarea font that cluttered LaTeX output. J?rgen -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 488 bytes Desc: This is a digitally signed message part URL: From spitz at lyx.org Tue Nov 2 15:00:27 2021 From: spitz at lyx.org (=?ISO-8859-1?Q?J=FCrgen_Spitzm=FCller?=) Date: Tue, 02 Nov 2021 16:00:27 +0100 Subject: Footnote size is larger on master (\normalsize is prepended) In-Reply-To: References: <20211031171101.a32end5eh45mx5x2@poole> <6164e06e3490b4e31f8baa4a383278f198520ff1.camel@lyx.org> <83950bd68c4552be972f950e3453f5eaf56b0de4.camel@lyx.org> <1a9052954ffb47fb103c2209d133ed1cf85f89d8.camel@lyx.org> <871ec28b-fc97-8bbd-aaa8-04a76a669558@lyx.org> Message-ID: Am Dienstag, dem 02.11.2021 um 15:52 +0100 schrieb J?rgen Spitzm?ller: > With inheritFont() == false, we end up in unnecessary font switches > at footnote start, e.g. \footnote{\normalsize text ...} in section > headings (see Scott's example), and I want to avoid that. And note that this is not only cosmetic, but also wrong, as this resets the footnote text size from \footnotesize to \normalsize. See attached example and compare the output (both TeX and PDF) in stable and master. J?rgen -------------- next part -------------- A non-text attachment was scrubbed... Name: footnote-size-is-different.23.lyx Type: application/x-lyx Size: 3644 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 488 bytes Desc: This is a digitally signed message part URL: From spitz at lyx.org Tue Nov 2 16:11:34 2021 From: spitz at lyx.org (=?ISO-8859-1?Q?J=FCrgen_Spitzm=FCller?=) Date: Tue, 02 Nov 2021 17:11:34 +0100 Subject: Qt6 crash Message-ID: I just notices that, when compiling master against Qt 6.2.0, LyX crashes reliably at document compilation (somewhere in QFutureWatcher). Can anyone confirm or is this something on my end? Thanks, J?rgen -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 488 bytes Desc: This is a digitally signed message part URL: From sanda at lyx.org Tue Nov 2 21:07:20 2021 From: sanda at lyx.org (Pavel Sanda) Date: Tue, 2 Nov 2021 22:07:20 +0100 Subject: master crash (tocwidget code) In-Reply-To: References: <20211101131549.GA13671@jabberwock.ucw.cz> <7f44ca07ff7c3d8844639fa96a86e10a8eee75e5.camel@lyx.org> <20211101215124.GA26903@jabberwock.ucw.cz> Message-ID: <20211102210720.GA30980@jabberwock.ucw.cz> On Tue, Nov 02, 2021 at 08:08:05AM +0100, J?rgen Spitzm?ller wrote: > Am Montag, dem 01.11.2021 um 22:51 +0100 schrieb Pavel Sanda: > > I don't know the code in question either, but I can try to look later > > this week. > > OTOH we test for empty toc_ in several other places of the TocModel, so > the check the patch adds might be reasonable Yes. > (maybe the toc is not yet re-created at that point). I believe something else is going on. When you add something like +++ b/src/frontends/qt/TocModel.cpp @@ -327,6 +330,7 @@ TocItem const TocModels::currentItem(QString const & type, } LASSERT(index.model() == it.value()->model(), return TocItem()); + lyxerr<toc_->empty() << "\n"; return it.value()->tocItem(index); } You can see steady flow of identical lines, e.g. tableofcontents 0x55a7b8d94af8 0x55a7b9c931f0 0 associated with the first window. Then you create second window open new view of the same file and get steady flow of second type e.g. tableofcontents 0x55a7ba5e5438 0x55a7ba817910 0 until the last line before crash tableofcontents 0x55a7b8d94af8 0x55a7b9c931f0 1 which is again pointing to structure associated with window 1 but now empty. So instead of missing initialization we likely destroy the content of first structure meanwhile. We call various reset() and clear() routines many many times, and its unclear to me why we repeatedlt call the whole machinery. My time for this issue is spent for today... Pavel From skostysh at lyx.org Wed Nov 3 01:43:54 2021 From: skostysh at lyx.org (Scott Kostyshak) Date: Tue, 2 Nov 2021 21:43:54 -0400 Subject: Footnote size is larger on master (\normalsize is prepended) In-Reply-To: <2e25e908d8079d866c1afd43356f88bdeb1e117d.camel@lyx.org> References: <20211031171101.a32end5eh45mx5x2@poole> <6164e06e3490b4e31f8baa4a383278f198520ff1.camel@lyx.org> <83950bd68c4552be972f950e3453f5eaf56b0de4.camel@lyx.org> <20211101192736.pz3iney6umsbgo6n@poole> <2e25e908d8079d866c1afd43356f88bdeb1e117d.camel@lyx.org> Message-ID: <20211103014354.fj5l3m42bxks5psy@poole> On Tue, Nov 02, 2021 at 08:33:47AM +0100, J?rgen Spitzm?ller wrote: > Am Montag, dem 01.11.2021 um 15:27 -0400 schrieb Scott Kostyshak: > > Fixes the original issue from what I can see, although I know nothing > > about the code. I'm not sure if the following is a related issue. See > > attached. It's hard to see, but the in-line footnote number itself is > > now bold on master. Note the "\bfseries" in the difference in TeX > > code > > produced below. > > > > master with your patch: > > > > ? To set the bracket size for the second line correctly, the first > > line > > ? is ended with \textbf{\textbackslash > > right.}\index{Commands!R!right@\textbackslash right} > > ? and the second line with \textbf{\textbackslash > > left.}{\bfseries\footnote{for more about \textbf{\textbackslash left} > > and \textbf{\textbackslash right} > > > > 2.3.x: > > > > ? To set the bracket size for the second line correctly, the first > > line > > ? is ended with \textbf{\textbackslash > > right.}\index{Commands!R!right@\textbackslash right} > > ? and the second line with \textbf{\textbackslash left.}\footnote{for > > more about \textbf{\textbackslash left} and \textbf{\textbackslash > > right} > > I'd need to see the LyX file, but if the footnote itself is set bold > there, master is correct and stable is wrong. This is something we > fixed in master. Thanks for taking a look. I see, so master behavior is correct. I think in most cases the intent is to not have it bold, i.e., I might look into removing the bold in the .lyx files if I am motivated. Scott -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 488 bytes Desc: not available URL: From st.witt at gmx.net Wed Nov 3 02:30:37 2021 From: st.witt at gmx.net (Stephan Witt) Date: Wed, 3 Nov 2021 03:30:37 +0100 Subject: Qt6 crash In-Reply-To: References: Message-ID: Am 02.11.2021 um 17:11 schrieb J?rgen Spitzm?ller : > > I just notices that, when compiling master against Qt 6.2.0, LyX > crashes reliably at document compilation (somewhere in QFutureWatcher). > > Can anyone confirm or is this something on my end? I?ve built master with Qt 6.2.0 on Mac and it doesn?t crash. I?m compiling the UG with pdflatex. Stephan From spitz at lyx.org Wed Nov 3 07:12:55 2021 From: spitz at lyx.org (=?UTF-8?B?SsO8cmdlbiBTcGl0em3DvGxsZXI=?=) Date: Wed, 3 Nov 2021 08:12:55 +0100 Subject: Footnote size is larger on master (\normalsize is prepended) In-Reply-To: <20211103014354.fj5l3m42bxks5psy@poole> References: <20211031171101.a32end5eh45mx5x2@poole> <6164e06e3490b4e31f8baa4a383278f198520ff1.camel@lyx.org> <83950bd68c4552be972f950e3453f5eaf56b0de4.camel@lyx.org> <20211101192736.pz3iney6umsbgo6n@poole> <2e25e908d8079d866c1afd43356f88bdeb1e117d.camel@lyx.org> <20211103014354.fj5l3m42bxks5psy@poole> Message-ID: Scott Kostyshak schrieb am Mi., 3. Nov. 2021, 02:44: > > Thanks for taking a look. I see, so master behavior is correct. I think > in most cases the intent is to not have it bold, i.e., I might look into > removing the bold in the .lyx files if I am motivated. > I think a non-bold footnote symbol in otherwise bold context looks odd. J?rgen > Scott > -- > lyx-devel mailing list > lyx-devel at lists.lyx.org > http://lists.lyx.org/mailman/listinfo/lyx-devel > -------------- next part -------------- An HTML attachment was scrubbed... URL: From spitz at lyx.org Wed Nov 3 07:14:43 2021 From: spitz at lyx.org (=?UTF-8?B?SsO8cmdlbiBTcGl0em3DvGxsZXI=?=) Date: Wed, 3 Nov 2021 08:14:43 +0100 Subject: master crash (tocwidget code) In-Reply-To: <20211102210720.GA30980@jabberwock.ucw.cz> References: <20211101131549.GA13671@jabberwock.ucw.cz> <7f44ca07ff7c3d8844639fa96a86e10a8eee75e5.camel@lyx.org> <20211101215124.GA26903@jabberwock.ucw.cz> <20211102210720.GA30980@jabberwock.ucw.cz> Message-ID: Am Di., 2. Nov. 2021 um 22:08 Uhr schrieb Pavel Sanda : > I believe something else is going on. > > When you add something like > +++ b/src/frontends/qt/TocModel.cpp > @@ -327,6 +330,7 @@ TocItem const TocModels::currentItem(QString const & > type, > } > LASSERT(index.model() == it.value()->model(), return TocItem()); > > + lyxerr< it.value()->toc_->empty() << "\n"; > return it.value()->tocItem(index); > } > > > You can see steady flow of identical lines, e.g. > tableofcontents 0x55a7b8d94af8 0x55a7b9c931f0 0 > > associated with the first window. Then you create second window > open new view of the same file and get steady flow of second type e.g. > > tableofcontents 0x55a7ba5e5438 0x55a7ba817910 0 > > until the last line before crash > tableofcontents 0x55a7b8d94af8 0x55a7b9c931f0 1 > > which is again pointing to structure associated with window 1 > but now empty. So instead of missing initialization we likely > destroy the content of first structure meanwhile. > Thanks, that's interesting. > We call various reset() and clear() routines many many times, > and its unclear to me why we repeatedlt call the whole machinery. > Maybe this code (TocWidget::updateView()) is to blame? if (!gui_view_.documentBufferView()) { tocTV->setModel(nullptr); depthSL->setMaximum(0); depthSL->setValue(0); setEnabled(false); return; } J?rgen -------------- next part -------------- An HTML attachment was scrubbed... URL: From spitz at lyx.org Wed Nov 3 07:28:59 2021 From: spitz at lyx.org (=?UTF-8?B?SsO8cmdlbiBTcGl0em3DvGxsZXI=?=) Date: Wed, 3 Nov 2021 08:28:59 +0100 Subject: master crash (tocwidget code) In-Reply-To: References: <20211101131549.GA13671@jabberwock.ucw.cz> <7f44ca07ff7c3d8844639fa96a86e10a8eee75e5.camel@lyx.org> <20211101215124.GA26903@jabberwock.ucw.cz> <20211102210720.GA30980@jabberwock.ucw.cz> Message-ID: Am Mi., 3. Nov. 2021 um 08:14 Uhr schrieb J?rgen Spitzm?ller : > Maybe this code (TocWidget::updateView()) is to blame? > > if (!gui_view_.documentBufferView()) { > > tocTV->setModel(nullptr); > > depthSL->setMaximum(0); > > depthSL->setValue(0); > > setEnabled(false); > > return; > > } > > Commenting this out (except for the return) at least cures the crash. Why do we need to reset the model in this case? J?rgen -------------- next part -------------- An HTML attachment was scrubbed... URL: From yuriy.skalko at gmail.com Wed Nov 3 07:54:20 2021 From: yuriy.skalko at gmail.com (Yuriy Skalko) Date: Wed, 3 Nov 2021 09:54:20 +0200 Subject: Latex run after toggling search pane Message-ID: Hello, In current master I'm getting automatic run of latex after toggling Find and replace button on the toolbar. Bisect gives this commit: bca1b63d89e27b. J?rgen, can you check this? Yuriy From spitz at lyx.org Wed Nov 3 11:35:59 2021 From: spitz at lyx.org (=?UTF-8?B?SsO8cmdlbiBTcGl0em3DvGxsZXI=?=) Date: Wed, 3 Nov 2021 12:35:59 +0100 Subject: Latex run after toggling search pane In-Reply-To: References: Message-ID: Am Mi., 3. Nov. 2021 um 08:54 Uhr schrieb Yuriy Skalko < yuriy.skalko at gmail.com>: > Hello, > > In current master I'm getting automatic run of latex after toggling Find > and replace button on the toolbar. > > Bisect gives this commit: bca1b63d89e27b. J?rgen, can you check this? > Yes, this is due to an utter stupidity of mine. Fixed now. J?rgen > > > Yuriy > > > -- > lyx-devel mailing list > lyx-devel at lists.lyx.org > http://lists.lyx.org/mailman/listinfo/lyx-devel > -------------- next part -------------- An HTML attachment was scrubbed... URL: From sanda at lyx.org Wed Nov 3 12:41:37 2021 From: sanda at lyx.org (Pavel Sanda) Date: Wed, 3 Nov 2021 13:41:37 +0100 Subject: master crash (tocwidget code) In-Reply-To: References: <20211101131549.GA13671@jabberwock.ucw.cz> <7f44ca07ff7c3d8844639fa96a86e10a8eee75e5.camel@lyx.org> <20211101215124.GA26903@jabberwock.ucw.cz> <20211102210720.GA30980@jabberwock.ucw.cz> Message-ID: <20211103124137.GA4228@jabberwock.ucw.cz> On Wed, Nov 03, 2021 at 08:28:59AM +0100, J?rgen Spitzm?ller wrote: > > Maybe this code (TocWidget::updateView()) is to blame? > > > > if (!gui_view_.documentBufferView()) { > > > > tocTV->setModel(nullptr); > > > > depthSL->setMaximum(0); > > > > depthSL->setValue(0); > > > > setEnabled(false); > > > > return; > > > > } > > > > > Commenting this out (except for the return) at least cures the crash. Why > do we need to reset the model in this case? I can't reproduce this and do get the crash even when commenting the block statements except return. Also I checked that gui_view_ is correctly different when calling for the new window so I would have hard time to understand what cancels toc_ content. Pavel From kornel at lyx.org Wed Nov 3 12:53:42 2021 From: kornel at lyx.org (Kornel Benko) Date: Wed, 3 Nov 2021 13:53:42 +0100 Subject: master crash (tocwidget code) In-Reply-To: <20211103124137.GA4228@jabberwock.ucw.cz> References: <20211101131549.GA13671@jabberwock.ucw.cz> <7f44ca07ff7c3d8844639fa96a86e10a8eee75e5.camel@lyx.org> <20211101215124.GA26903@jabberwock.ucw.cz> <20211102210720.GA30980@jabberwock.ucw.cz> <20211103124137.GA4228@jabberwock.ucw.cz> Message-ID: <20211103135342.2f511026@admin1-desktop> Am Wed, 3 Nov 2021 13:41:37 +0100 schrieb Pavel Sanda : > On Wed, Nov 03, 2021 at 08:28:59AM +0100, J?rgen Spitzm?ller wrote: > > > Maybe this code (TocWidget::updateView()) is to blame? > > > > > > if (!gui_view_.documentBufferView()) { > > > > > > tocTV->setModel(nullptr); > > > > > > depthSL->setMaximum(0); > > > > > > depthSL->setValue(0); > > > > > > setEnabled(false); > > > > > > return; > > > > > > } > > > > > > > > Commenting this out (except for the return) at least cures the crash. Why > > do we need to reset the model in this case? > > I can't reproduce this and do get the crash even when commenting the block > statements except return. > > Also I checked that gui_view_ is correctly different when calling for the new > window so I would have hard time to understand what cancels toc_ content. > > > Pavel Not getting the crash here with latest master. Kornel -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 488 bytes Desc: Digitale Signatur von OpenPGP URL: From spitz at lyx.org Wed Nov 3 12:59:15 2021 From: spitz at lyx.org (=?UTF-8?B?SsO8cmdlbiBTcGl0em3DvGxsZXI=?=) Date: Wed, 3 Nov 2021 13:59:15 +0100 Subject: master crash (tocwidget code) In-Reply-To: <20211103124137.GA4228@jabberwock.ucw.cz> References: <20211101131549.GA13671@jabberwock.ucw.cz> <7f44ca07ff7c3d8844639fa96a86e10a8eee75e5.camel@lyx.org> <20211101215124.GA26903@jabberwock.ucw.cz> <20211102210720.GA30980@jabberwock.ucw.cz> <20211103124137.GA4228@jabberwock.ucw.cz> Message-ID: Am Mi., 3. Nov. 2021 um 13:42 Uhr schrieb Pavel Sanda : > I can't reproduce this and do get the crash even when commenting the block > statements except return. > Right. Maybe it didn't crash for me by coincidence. J?rgen -------------- next part -------------- An HTML attachment was scrubbed... URL: From sanda at lyx.org Wed Nov 3 14:16:46 2021 From: sanda at lyx.org (Pavel Sanda) Date: Wed, 3 Nov 2021 15:16:46 +0100 Subject: master crash (tocwidget code) In-Reply-To: <20211102210720.GA30980@jabberwock.ucw.cz> References: <20211101131549.GA13671@jabberwock.ucw.cz> <7f44ca07ff7c3d8844639fa96a86e10a8eee75e5.camel@lyx.org> <20211101215124.GA26903@jabberwock.ucw.cz> <20211102210720.GA30980@jabberwock.ucw.cz> Message-ID: <20211103141646.GB4228@jabberwock.ucw.cz> On Tue, Nov 02, 2021 at 10:07:20PM +0100, Pavel Sanda wrote: > We call various reset() and clear() routines many many times, > and its unclear to me why we repeatedlt call the whole machinery. >From what I can see in the code the route to touching toc_ is: Guiview::structureChanged() -> TocModels::reset -> TocModel::reset TocModels::clear -> TocModel::clear and structureChanged() get's called way more often than in the case of document change, for example in the case of just getting focus on the lyx window or in the case of opening empty new window (ironically the document itself is in readonly mode). So the likely scenario is that the moment new window is created the toc_ in the old one gets emptied via the route above, the we load the old document in new window which in chain of events also triggers filterContents in the old window and you're done. I am afraid unless we want to revamp the whole toc update machinery we just shouldn't by default that expect toc_ is nonempty. Pavel From spitz at lyx.org Wed Nov 3 14:59:25 2021 From: spitz at lyx.org (=?UTF-8?B?SsO8cmdlbiBTcGl0em3DvGxsZXI=?=) Date: Wed, 3 Nov 2021 15:59:25 +0100 Subject: master crash (tocwidget code) In-Reply-To: <20211103141646.GB4228@jabberwock.ucw.cz> References: <20211101131549.GA13671@jabberwock.ucw.cz> <7f44ca07ff7c3d8844639fa96a86e10a8eee75e5.camel@lyx.org> <20211101215124.GA26903@jabberwock.ucw.cz> <20211102210720.GA30980@jabberwock.ucw.cz> <20211103141646.GB4228@jabberwock.ucw.cz> Message-ID: Am Mi., 3. Nov. 2021 um 15:17 Uhr schrieb Pavel Sanda : > I am afraid unless we want to revamp the whole toc update machinery we just > shouldn't by default that expect toc_ is nonempty. > This means that my patch which checks for toc_ emptiness should go in, right? J?rgen > > Pavel > -- > lyx-devel mailing list > lyx-devel at lists.lyx.org > http://lists.lyx.org/mailman/listinfo/lyx-devel > -------------- next part -------------- An HTML attachment was scrubbed... URL: From sanda at lyx.org Wed Nov 3 15:43:04 2021 From: sanda at lyx.org (Pavel Sanda) Date: Wed, 3 Nov 2021 16:43:04 +0100 Subject: master crash (tocwidget code) In-Reply-To: References: <20211101131549.GA13671@jabberwock.ucw.cz> <7f44ca07ff7c3d8844639fa96a86e10a8eee75e5.camel@lyx.org> <20211101215124.GA26903@jabberwock.ucw.cz> <20211102210720.GA30980@jabberwock.ucw.cz> <20211103141646.GB4228@jabberwock.ucw.cz> Message-ID: <20211103154304.GA26579@jabberwock.ucw.cz> On Wed, Nov 03, 2021 at 03:59:25PM +0100, J?rgen Spitzm?ller wrote: > Am Mi., 3. Nov. 2021 um 15:17 Uhr schrieb Pavel Sanda : > > > I am afraid unless we want to revamp the whole toc update machinery we just > > shouldn't by default that expect toc_ is nonempty. > > > > This means that my patch which checks for toc_ emptiness should go in, > right? Well, nothing against this particular patch, but the problem is that if we accept that models_.find(type) might sucesfully return hashes which are actually bogus, then all other models_.find occurences might be in the same danger. So yes, this particular crash is gone, but other corner cases might be lurking around. The comments in various TocModel classess/attributes are such that I struggle to understand what they are supposed to represent and what should I expect or not. Pavel From spitz at lyx.org Wed Nov 3 16:18:44 2021 From: spitz at lyx.org (=?ISO-8859-1?Q?J=FCrgen_Spitzm=FCller?=) Date: Wed, 03 Nov 2021 17:18:44 +0100 Subject: master crash (tocwidget code) In-Reply-To: <20211103154304.GA26579@jabberwock.ucw.cz> References: <20211101131549.GA13671@jabberwock.ucw.cz> <7f44ca07ff7c3d8844639fa96a86e10a8eee75e5.camel@lyx.org> <20211101215124.GA26903@jabberwock.ucw.cz> <20211102210720.GA30980@jabberwock.ucw.cz> <20211103141646.GB4228@jabberwock.ucw.cz> <20211103154304.GA26579@jabberwock.ucw.cz> Message-ID: <3232207c12fd082b937f9e2b94884907354e672f.camel@lyx.org> Am Mittwoch, dem 03.11.2021 um 16:43 +0100 schrieb Pavel Sanda: > Well, nothing against this particular patch, but the problem is that > if we accept that models_.find(type) might sucesfully return hashes > which are actually bogus, then all other models_.find occurences > might be in the same danger. So yes, this particular crash is gone, > but other corner cases might be lurking around. So what do you propose instead? J?rgen -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 488 bytes Desc: This is a digitally signed message part URL: From spitz at lyx.org Wed Nov 3 16:44:44 2021 From: spitz at lyx.org (=?ISO-8859-1?Q?J=FCrgen_Spitzm=FCller?=) Date: Wed, 03 Nov 2021 17:44:44 +0100 Subject: Qt6 crash In-Reply-To: References: Message-ID: Am Mittwoch, dem 03.11.2021 um 03:30 +0100 schrieb Stephan Witt: > I?ve built master with Qt 6.2.0 on Mac and it doesn?t crash. I?m > compiling the UG with pdflatex. Thanks. Retried with a clean rebuild. Still crashing (after the PDF has been created): ( 1) src/lyx: lyx::frontend::Alert::doError(std::__cxx11::basic_string, std::allocator > const&, std::__cxx11::basic_string, std::allocator > const&, bool) ( 2) src/lyx: std::_Function_handler, std::allocator > const>, std::reference_wrapper, std::allocator > const>, std::reference_wrapper))(std::__cxx11::basic_string, std::allocator > const&, std::__cxx11::basic_string, std::allocator > const&, bool)> >::_M_invoke(std::_Any_data const&) ( 3) src/lyx: lyx::frontend::InGuiThread::synchronousFunctionCall() ( 4) src/lyx: lyx::frontend::IntoGuiThreadMover::callInGuiThread() ( 5) src/lyx: lyx::frontend::Alert::error(std::__cxx11::basic_string, std::allocator > const&, std::__cxx11::basic_string, std::allocator > const&, bool) ( 6) src/lyx: src/lyx() [0x9d5f33] ( 7) /lib64/libc.so.6: /lib64/libc.so.6(+0x427a0) [0x7fa36873e7a0] ( 8) /lib64/libstdc: /lib64/libstdc++.so.6(_ZSt18_Rb_tree_decrementPKSt18_Rb_tree_node_base+ 0x23) [0x7fa368ac8003] ( 9) /lib64/libQt6Core.so.6: QtPrivate::ResultStoreBase::resultAt(int) const ( 10) src/lyx: lyx::frontend::GuiView::processingThreadFinished() ( 11) src/lyx: src/lyx() [0xf74af0] ( 12) /lib64/libQt6Core.so.6: /lib64/libQt6Core.so.6(+0x1bd97f) [0x7fa368e1c97f] ( 13) /lib64/libQt6Core.so.6: QFutureWatcherBase::event(QEvent*) ( 14) /lib64/libQt6Widgets.so.6: QApplicationPrivate::notify_helper(QObject*, QEvent*) ( 15) src/lyx: lyx::frontend::GuiApplication::notify(QObject*, QEvent*) ( 16) /lib64/libQt6Core.so.6: QCoreApplication::notifyInternal2(QObject*, QEvent*) ( 17) /lib64/libQt6Core.so.6: QCoreApplicationPrivate::sendPostedEvents(QObject*, int, QThreadData*) ( 18) /lib64/libQt6Core.so.6: /lib64/libQt6Core.so.6(+0x3568c3) [0x7fa368fb58c3] ( 19) /lib64/libglib-2.0.so.0: /lib64/libglib- 2.0.so.0(g_main_context_dispatch+0x16f) [0x7fa368426d5f] ( 20) /lib64/libglib-2.0.so.0: /lib64/libglib-2.0.so.0(+0x570e8) [0x7fa3684270e8] ( 21) /lib64/libglib-2.0.so.0: /lib64/libglib- 2.0.so.0(g_main_context_iteration+0x2f) [0x7fa36842719f] ( 22) /lib64/libQt6Core.so.6: QEventDispatcherGlib::processEvents(QFlags) ( 23) /lib64/libQt6Core.so.6: QEventLoop::exec(QFlags) ( 24) /lib64/libQt6Core.so.6: QCoreApplication::exec() ( 25) src/lyx: lyx::frontend::GuiApplication::exec() ( 26) src/lyx: lyx::LyX::exec(int&, char**) ( 27) src/lyx: src/lyx(main+0x45) [0x76d32b] ( 28) /lib64/libc.so.6: /lib64/libc.so.6(+0x2d540) [0x7fa368729540] ( 29) /lib64/libc.so.6: /lib64/libc.so.6(__libc_start_main+0x7c) [0x7fa3687295ec] ( 30) src/lyx: src/lyx(_start+0x25) [0x76d225] J?rgen -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 488 bytes Desc: This is a digitally signed message part URL: From skostysh at lyx.org Wed Nov 3 16:52:58 2021 From: skostysh at lyx.org (Scott Kostyshak) Date: Wed, 3 Nov 2021 12:52:58 -0400 Subject: Footnote size is larger on master (\normalsize is prepended) In-Reply-To: References: <20211031171101.a32end5eh45mx5x2@poole> <6164e06e3490b4e31f8baa4a383278f198520ff1.camel@lyx.org> <83950bd68c4552be972f950e3453f5eaf56b0de4.camel@lyx.org> <20211101192736.pz3iney6umsbgo6n@poole> <2e25e908d8079d866c1afd43356f88bdeb1e117d.camel@lyx.org> <20211103014354.fj5l3m42bxks5psy@poole> Message-ID: <20211103165258.bm7dmjydewmedyh3@poole> On Wed, Nov 03, 2021 at 08:12:55AM +0100, J?rgen Spitzm?ller wrote: > Scott Kostyshak schrieb am Mi., 3. Nov. 2021, 02:44: > > > > > Thanks for taking a look. I see, so master behavior is correct. I think > > in most cases the intent is to not have it bold, i.e., I might look into > > removing the bold in the .lyx files if I am motivated. > > > > I think a non-bold footnote symbol in otherwise bold context looks odd. Good to know. I'll skip it then. Scott -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 488 bytes Desc: not available URL: From spitz at lyx.org Wed Nov 3 17:03:16 2021 From: spitz at lyx.org (=?ISO-8859-1?Q?J=FCrgen_Spitzm=FCller?=) Date: Wed, 03 Nov 2021 18:03:16 +0100 Subject: Qt6 crash In-Reply-To: References: Message-ID: <2900378c065ef2ddf84f8f1f21a6306770bc75bd.camel@lyx.org> Am Mittwoch, dem 03.11.2021 um 17:44 +0100 schrieb J?rgen Spitzm?ller: > Am Mittwoch, dem 03.11.2021 um 03:30 +0100 schrieb Stephan Witt: > > I?ve built master with Qt 6.2.0 on Mac and it doesn?t crash. I?m > > compiling the UG with pdflatex. > > Thanks. Retried with a clean rebuild. Still crashing (after the PDF > has been created): Recompiling with --enable-build-type=rel (rather than --enable- maintainer-mode) it doesn't crash, so it seems the inclusion of debug symbols is relevant. J?rgen -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 488 bytes Desc: This is a digitally signed message part URL: From forenr at lyx.org Wed Nov 3 19:43:45 2021 From: forenr at lyx.org (Enrico Forestieri) Date: Wed, 3 Nov 2021 20:43:45 +0100 Subject: Qt6 crash In-Reply-To: <2900378c065ef2ddf84f8f1f21a6306770bc75bd.camel@lyx.org> References: <2900378c065ef2ddf84f8f1f21a6306770bc75bd.camel@lyx.org> Message-ID: On Wed, Nov 03, 2021 at 06:03:16PM +0100, J?rgen Spitzm?ller wrote: > Am Mittwoch, dem 03.11.2021 um 17:44 +0100 schrieb J?rgen Spitzm?ller: > > Am Mittwoch, dem 03.11.2021 um 03:30 +0100 schrieb Stephan Witt: > > > I?ve built master with Qt 6.2.0 on Mac and it doesn?t crash. I?m > > > compiling the UG with pdflatex. > > > > Thanks. Retried with a clean rebuild. Still crashing (after the PDF > > has been created): > > Recompiling with --enable-build-type=rel (rather than --enable- > maintainer-mode) it doesn't crash, so it seems the inclusion of debug > symbols is relevant. What if you simply use --disable-stdlib-debug ? -- Enrico From lasgouttes at lyx.org Wed Nov 3 20:33:06 2021 From: lasgouttes at lyx.org (Jean-Marc Lasgouttes) Date: Wed, 3 Nov 2021 21:33:06 +0100 Subject: Footnote size is larger on master (\normalsize is prepended) In-Reply-To: References: <20211031171101.a32end5eh45mx5x2@poole> <6164e06e3490b4e31f8baa4a383278f198520ff1.camel@lyx.org> <83950bd68c4552be972f950e3453f5eaf56b0de4.camel@lyx.org> <1a9052954ffb47fb103c2209d133ed1cf85f89d8.camel@lyx.org> <871ec28b-fc97-8bbd-aaa8-04a76a669558@lyx.org> Message-ID: Le 02/11/2021 ? 15:52, J?rgen Spitzm?ller a ?crit?: > With inheritFont() == false, we end up in unnecessary font switches at > footnote start, e.g. \footnote{\normalsize text ...} in section > headings (see Scott's example), and I want to avoid that. Agreed. However it seems that the code 1/ looks at the outside font, sees that it is \normalsize 2/ look at footnote font and sees that it is \footnotesize 3/ decides to override the font size to do as if a footnote should plainly inherit the outside font. There is _no_ reason in my opinion why the font outside the inset should be considered at all when inheritFont() is false. It is however difficult to follow the code in Paragraph::latex: // If we are inside an non inheritFont() inset, the real outerfont is local_font Font const real_outerfont = (!inInset().inheritFont() && runparams.local_font != nullptr) ? Font(runparams.local_font->fontInfo()) : outerfont; I think I know at some time what local_font was, but I forgot. I see that it is set in dozens of places in the code (which murmurs "fragile code" in my ear), but I am not sure why. AFAICS, this is more a language thing than a font thing (but I am even not sure of this). OK, there is a comment about this variable: /** the font at the point where the inset is */ Font const * local_font = nullptr; If the comment is correct, then the code is wrong IMO. JMarc From ajparsloe at gmail.com Wed Nov 3 20:56:01 2021 From: ajparsloe at gmail.com (Andrew Parsloe) Date: Thu, 4 Nov 2021 09:56:01 +1300 Subject: False spacing if a skip is emphasized Message-ID: LyX 2.4 alpha 3 on windows 10. If a vertical skip between two standard paragraphs is emphasized (or bolded or nouned or ...) it produces a false linespace in the pdf within the second paragraph. The effect doesn't show in the LyX window. It's easy to emphasize a skip if the first word of the second paragraph is emphasized and the cursor is at the start of that paragraph when the skip is inserted. The code preview pane shows: This is a standard paragraph. \emph{\bigskip{} }This is a second standard paragraph. This is a second standard paragraph. This is a second standard paragraph. This is a second standard paragraph. This is a second standard paragraph. This is a second standard paragraph. This is a second standard paragraph. I've attached the pdf, where the skip has been inserted between the first and second lines of the the second paragraph. The two paragraphs also have slightly different indents in the pdf. Unemphasizing the skip solves the issues, but since the LyX window shows nothing wrong, it can easily slip by unnoticed. Andrew -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: skipspacing.pdf Type: application/pdf Size: 27394 bytes Desc: not available URL: From sanda at lyx.org Wed Nov 3 21:33:30 2021 From: sanda at lyx.org (Pavel Sanda) Date: Wed, 3 Nov 2021 22:33:30 +0100 Subject: master crash (tocwidget code) In-Reply-To: <3232207c12fd082b937f9e2b94884907354e672f.camel@lyx.org> References: <20211101131549.GA13671@jabberwock.ucw.cz> <7f44ca07ff7c3d8844639fa96a86e10a8eee75e5.camel@lyx.org> <20211101215124.GA26903@jabberwock.ucw.cz> <20211102210720.GA30980@jabberwock.ucw.cz> <20211103141646.GB4228@jabberwock.ucw.cz> <20211103154304.GA26579@jabberwock.ucw.cz> <3232207c12fd082b937f9e2b94884907354e672f.camel@lyx.org> Message-ID: <20211103213330.GA5449@ucw.cz> On Wed, Nov 03, 2021 at 05:18:44PM +0100, J?rgen Spitzm?ller wrote: > Am Mittwoch, dem 03.11.2021 um 16:43 +0100 schrieb Pavel Sanda: > > Well, nothing against this particular patch, but the problem is that > > if we accept that models_.find(type) might sucesfully return hashes > > which are actually bogus, then all other models_.find occurences > > might be in the same danger. So yes, this particular crash is gone, > > but other corner cases might be lurking around. > > So what do you propose instead? I would like to play a little more to understand the logic between resetting and seting up the toc_. We clear it all the time and yet we do not see the crashes, so there must be some mechanism of filling it up immediately which fails now. But it will take some time, time resources quite limited ATM. Pavel From spitz at lyx.org Thu Nov 4 07:48:16 2021 From: spitz at lyx.org (=?UTF-8?B?SsO8cmdlbiBTcGl0em3DvGxsZXI=?=) Date: Thu, 4 Nov 2021 08:48:16 +0100 Subject: Qt6 crash In-Reply-To: References: <2900378c065ef2ddf84f8f1f21a6306770bc75bd.camel@lyx.org> Message-ID: Am Mi., 3. Nov. 2021 um 20:43 Uhr schrieb Enrico Forestieri : > What if you simply use --disable-stdlib-debug ? > This helps as well. BTW I have reproduced the issue on a different machine (both running OpenSuse Tumbleweed). J?rgen -------------- next part -------------- An HTML attachment was scrubbed... URL: From spitz at lyx.org Thu Nov 4 08:15:39 2021 From: spitz at lyx.org (=?UTF-8?B?SsO8cmdlbiBTcGl0em3DvGxsZXI=?=) Date: Thu, 4 Nov 2021 09:15:39 +0100 Subject: False spacing if a skip is emphasized In-Reply-To: References: Message-ID: Am Mi., 3. Nov. 2021 um 21:56 Uhr schrieb Andrew Parsloe < ajparsloe at gmail.com>: > LyX 2.4 alpha 3 on windows 10. > > If a vertical skip between two standard paragraphs is emphasized (or > bolded or nouned or ...) it produces a false linespace in the pdf within > the second paragraph. The effect doesn't show in the LyX window. It's easy > to emphasize a skip if the first word of the second paragraph is emphasized > and the cursor is at the start of that paragraph when the skip is inserted. > The code preview pane shows: > This is a standard paragraph. > > > \emph{\bigskip{} > > }This is a second standard paragraph. This is a second standard paragraph. > > This is a second standard paragraph. This is a second standard paragraph. > > This is a second standard paragraph. This is a second standard paragraph. > > This is a second standard paragraph. > Also in stable and has probably been there since always. VSkip must come at the very beginning of paragraphs (before hmode is entered). This does not only apply to font setting, but to everything else; hence a bug which is hard to tackle. J?rgen -------------- next part -------------- An HTML attachment was scrubbed... URL: From lasgouttes at lyx.org Thu Nov 4 23:27:20 2021 From: lasgouttes at lyx.org (Jean-Marc Lasgouttes) Date: Fri, 5 Nov 2021 00:27:20 +0100 Subject: Qt6 crash In-Reply-To: References: <2900378c065ef2ddf84f8f1f21a6306770bc75bd.camel@lyx.org> Message-ID: <0bc49e3c-0123-1d7b-cd0a-4e110080ac80@lyx.org> Le 04/11/2021 ? 08:48, J?rgen Spitzm?ller a ?crit?: > Am Mi., 3. Nov. 2021 um 20:43?Uhr schrieb Enrico Forestieri > >: > > What if you simply use --disable-stdlib-debug ? > > > This helps as well. Is there some message from stdlib-debug before the backtrace? Or is it a crash related to object layout incompatibility? One possibility is a missing #include "config.h" JMarc From skostysh at lyx.org Fri Nov 5 04:10:20 2021 From: skostysh at lyx.org (Scott Kostyshak) Date: Fri, 5 Nov 2021 00:10:20 -0400 Subject: SIGSEGV on master when exporting MergedManuals.lyx in GUI In-Reply-To: <20210414164504.x43cvcjf2iitsuhj@tallinn> References: <20210413161724.rbbcph3eok4yym3l@tallinn> <3535838.MHq7AAxBmi@griffin> <20210414164504.x43cvcjf2iitsuhj@tallinn> Message-ID: <20211105041020.kpszs6ypsixqgcwa@poole> On Wed, Apr 14, 2021 at 12:45:04PM -0400, Scott Kostyshak wrote: > On Tue, Apr 13, 2021 at 07:53:11PM +0100, Jos? Ab?lio Matos wrote: > > On Tuesday, April 13, 2021 5:17:24 PM WEST Scott Kostyshak wrote: > > > I get the following terminal messages followed by a SIGSEGV when opening and > > > viewing MergedManuals.lyx in the GUI: > > > > > > QSocketNotifier: Invalid socket 16 and type 'Read', disabling... > > > > > > Does anyone else see this on master? > > > > > > Scott > > > > For me it is socket 10. :-) > > > > LyX does not crash but it enters an infinite cycle (read that after 5 minutes > > it continues to output that line) sending that line to stderr. > > I am not sure if I am patient enough to wait for infinity before recovering > > LyX's control. :-) > > Thanks for confirming, Jos?. I can still reproduce the infinite cycle (but not the original SIGSEGV) on master. Scott -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 488 bytes Desc: not available URL: From spitz at lyx.org Fri Nov 5 07:10:00 2021 From: spitz at lyx.org (=?ISO-8859-1?Q?J=FCrgen_Spitzm=FCller?=) Date: Fri, 05 Nov 2021 08:10:00 +0100 Subject: Qt6 crash In-Reply-To: <0bc49e3c-0123-1d7b-cd0a-4e110080ac80@lyx.org> References: <2900378c065ef2ddf84f8f1f21a6306770bc75bd.camel@lyx.org> <0bc49e3c-0123-1d7b-cd0a-4e110080ac80@lyx.org> Message-ID: <3b2ef9ff10b7771bbd023e01d28f94bfc0afd6a5.camel@lyx.org> Am Freitag, dem 05.11.2021 um 00:27 +0100 schrieb Jean-Marc Lasgouttes: > Is there some message from stdlib-debug before the backtrace?? No. > Or is it a crash related to object layout incompatibility? One > possibility is a missing #include "config.h" I have no clue. J?rgen > > JMarc -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 488 bytes Desc: This is a digitally signed message part URL: From spitz at lyx.org Fri Nov 5 08:22:00 2021 From: spitz at lyx.org (=?ISO-8859-1?Q?J=FCrgen_Spitzm=FCller?=) Date: Fri, 05 Nov 2021 09:22:00 +0100 Subject: Qt6 crash In-Reply-To: <3b2ef9ff10b7771bbd023e01d28f94bfc0afd6a5.camel@lyx.org> References: <2900378c065ef2ddf84f8f1f21a6306770bc75bd.camel@lyx.org> <0bc49e3c-0123-1d7b-cd0a-4e110080ac80@lyx.org> <3b2ef9ff10b7771bbd023e01d28f94bfc0afd6a5.camel@lyx.org> Message-ID: <1042b448930c16fdad18952fbb65994890d27595.camel@lyx.org> Am Freitag, dem 05.11.2021 um 08:10 +0100 schrieb J?rgen Spitzm?ller: > Am Freitag, dem 05.11.2021 um 00:27 +0100 schrieb Jean-Marc > Lasgouttes: > > Is there some message from stdlib-debug before the backtrace?? > > No. > > > Or is it a crash related to object layout incompatibility? One > > possibility is a missing #include "config.h" > > I have no clue. FWIW it crashes at GuiView.cpp:901 Buffer::ExportStatus const status = watcher->result(); J?rgen -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 488 bytes Desc: This is a digitally signed message part URL: From skostysh at lyx.org Sat Nov 6 01:18:05 2021 From: skostysh at lyx.org (Scott Kostyshak) Date: Fri, 5 Nov 2021 21:18:05 -0400 Subject: Link error when compiling with -fsanitize=undefined In-Reply-To: <20210922190848.42ebf968@admin1-desktop> References: <20210919114519.3sdpro3auz3p264t@poole> <20210919140033.2b57fe1f@admin1-desktop> <20210919181521.mwwwjvttpzjinqdi@poole> <20210919203716.184d1881@admin1-desktop> <20210920001915.gzojhqabgyjgatnx@poole> <20210920114837.6902e5aa@admin1-desktop> <20210920183014.750b45ea@admin1-desktop> <20210921145234.yf55hklka7jljg3j@poole> <20210922190848.42ebf968@admin1-desktop> Message-ID: <20211106011805.v32v6pe2yvklh3ua@poole> On Wed, Sep 22, 2021 at 07:08:48PM +0200, Kornel Benko wrote: > Am Wed, 22 Sep 2021 10:28:44 +0200 > schrieb Jean-Marc Lasgouttes : > > > Le 21/09/2021 ? 16:52, Scott Kostyshak a ?crit?: > > >> Some debugging lead to following observation: > > >> Each inserted branch add a color to the list. The original enum defined colors (about > > >> 102), last Color_ignore. > > >> So each branch will now get colors 103, 104, ..., and so on, so that the original > > >> enum (probably used as 'char') cannot contain the value beyond 127). > > >> > > >> Expanding the colors in src/ColorCode.h as in attached, removes the sanitize > > >> messages. > > > > > > Nice, that makes sense. 500 seems to be a reasonable ceiling. I suppose an > > > alternative would be to have the code stop assigning new colors past the max. That > > > is, I imagine they could just assign the max color value to multiple insets? > > > > But why the 127? Is it that the underlying type of the enum is a signed > > char? What does sizeof(ColorCode) return? > > That is not important from my POV. Sanitize is checking also for portability, so if > there is one compiler which would use 'char' for this enum, that would justify the > message. Just my 2 cents. > > > If this is the case, using > > enum ColorCode : int { > > should help. But I am surprised because the underlying type should be > > int (unless -fshort-enums is used): > > https://stackoverflow.com/questions/1122096/what-is-the-underlying-type-of-a-c-enum > > 'int' type may be too big (think on 64-bit (or more) intergers). > > > > I didn't understand how JMarc's suggestion to change the type to int would solve the > > > (non) issue. > > > > I was under the impression that the code would check that the enum value > > was one of the declared values. But this is not the case, since a number > > of synthetic values between 100 and 127 seem to be valid. > > > > I also wonder how you triggered it initially. Is it that you really had > > 25+ branches, or is it that the code "leaks" color code and creates too > > many of them? > > > > JMarc JMarc any objection to Kornel's patch? Scott -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 488 bytes Desc: not available URL: From jeanpierre.chretien at free.fr Mon Nov 8 10:03:25 2021 From: jeanpierre.chretien at free.fr (=?UTF-8?Q?Jean-Pierre_Chr=c3=a9tien?=) Date: Mon, 8 Nov 2021 11:03:25 +0100 Subject: Upgrade from buster to bullseye Message-ID: <5b369feb-541f-c155-9ae9-4d34f9ad35a2@free.fr> Dear developers I upgraded my Debian yesterday and I tested compilation and execution of master and 2.3.7dev. Compilation and display of the UserGuide run flawlessly on master. On 2.3.x, I first checked the fr.po file status and when I ran the make lyx-2.3.7dev.pot-update in the po dir of the build directory, I got in the following error: /bin/sh: 1: /usr/bin/python: not found make: *** [Makefile:774 : qt4_l10n.pot] Erreur 127 So I created a link: sudo ln -s /usr/bin/python3 /usr/bin/python and the problem was solved. The I compiled lyx-2.3.7dev all right with QT5 (QT4 fails). However, when compiling the pdf of the UserGuide, none of the svgz icons could be converted, so the padf dispaly failed with a pdf image not found. Here is first of the error messages: convert-im6.q16: attempt to perform an operation not allowed by the security policy `PDF' @ error/constitute.c/IsCoderAuthorized/421. /usr/local/share/lyx-2.3.7dev/scripts/convertDefault.py ERROR Execution of "convert" failed. support/Systemcall.cpp (276): Systemcall: 'python -tt "/usr/local/share/lyx-2.3.7dev/scripts/convertDefault.py" svgz "/tmp/lyx_tmpdir.ISfQxXhtSZQG/lyx_tmpbuf0/0_usr_local_share_lyx-2_3_7dev_images_buffer-view.svgz" pdf "/tmp/lyx_tmpdir.ISfQxXhtSZQG/lyx_tmpbuf0/0_usr_local_share_lyx-2_3_7dev_images_buffer-view.pdf"' finished with exit code 1 -- Jean-Pierre From lasgouttes at lyx.org Mon Nov 8 10:20:26 2021 From: lasgouttes at lyx.org (Jean-Marc Lasgouttes) Date: Mon, 8 Nov 2021 11:20:26 +0100 Subject: Upgrade from buster to bullseye In-Reply-To: <5b369feb-541f-c155-9ae9-4d34f9ad35a2@free.fr> References: <5b369feb-541f-c155-9ae9-4d34f9ad35a2@free.fr> Message-ID: <40b2d547-2b41-ff1d-775a-5e786a82725b@lyx.org> Le 08/11/2021 ? 11:03, Jean-Pierre Chr?tien a ?crit?: > Dear developers > > I upgraded my Debian yesterday and I tested compilation and execution of > master and 2.3.7dev. > > Compilation and display of the UserGuide run flawlessly on master. > > On 2.3.x, I first checked the fr.po file status and when I ran the make > lyx-2.3.7dev.pot-update in the po dir of the build directory, I got in > the following error: > > /bin/sh: 1: /usr/bin/python: not found > make: *** [Makefile:774 : qt4_l10n.pot] Erreur 127 This is for Jos?, right? > > So I created a link: > > sudo ln -s /usr/bin/python3 /usr/bin/python > > and the problem was solved. > > The I compiled lyx-2.3.7dev all right with QT5 (QT4 fails). > > However, when compiling the pdf of the UserGuide, none of the svgz icons > could be converted, so the padf dispaly failed with a pdf image not > found. Here is first of the error messages: > > convert-im6.q16: attempt to perform an operation not allowed by the > security policy `PDF' @ error/constitute.c/IsCoderAuthorized/421. > /usr/local/share/lyx-2.3.7dev/scripts/convertDefault.py ERROR > Execution of "convert" failed. Do you have rsvg-convert intalled (in librsvg2-bin package here). This "convert" situation is really horrible, I do not understand that there is no workaround yet. JMarc From st.witt at gmx.net Mon Nov 8 10:35:05 2021 From: st.witt at gmx.net (Stephan Witt) Date: Mon, 8 Nov 2021 11:35:05 +0100 Subject: Upgrade from buster to bullseye In-Reply-To: <40b2d547-2b41-ff1d-775a-5e786a82725b@lyx.org> References: <5b369feb-541f-c155-9ae9-4d34f9ad35a2@free.fr> <40b2d547-2b41-ff1d-775a-5e786a82725b@lyx.org> Message-ID: Am 08.11.2021 um 11:20 schrieb Jean-Marc Lasgouttes : > > Le 08/11/2021 ? 11:03, Jean-Pierre Chr?tien a ?crit : >> Dear developers >> I upgraded my Debian yesterday and I tested compilation and execution of master and 2.3.7dev. >> Compilation and display of the UserGuide run flawlessly on master. >> On 2.3.x, I first checked the fr.po file status and when I ran the make lyx-2.3.7dev.pot-update in the po dir of the build directory, I got in the following error: >> /bin/sh: 1: /usr/bin/python: not found >> make: *** [Makefile:774 : qt4_l10n.pot] Erreur 127 > > This is for Jos?, right? > >> So I created a link: >> sudo ln -s /usr/bin/python3 /usr/bin/python >> and the problem was solved. >> The I compiled lyx-2.3.7dev all right with QT5 (QT4 fails). >> However, when compiling the pdf of the UserGuide, none of the svgz icons could be converted, so the padf dispaly failed with a pdf image not found. Here is first of the error messages: >> convert-im6.q16: attempt to perform an operation not allowed by the security policy `PDF' @ error/constitute.c/IsCoderAuthorized/421. >> /usr/local/share/lyx-2.3.7dev/scripts/convertDefault.py ERROR >> Execution of "convert" failed. > > Do you have rsvg-convert intalled (in librsvg2-bin package here). > > This "convert" situation is really horrible, I do not understand that there is no workaround yet. There is a fallback converter utility lyxconvert. (src/convert/lyxconvert.cpp) It uses the Qt-library to convert from SVG to PDF et.al. But it has the drawback to have limited function and doesn?t understand the full SVG standard. I?m not sure how useful it is to improve the situation. Stephan From lasgouttes at lyx.org Mon Nov 8 10:58:13 2021 From: lasgouttes at lyx.org (Jean-Marc Lasgouttes) Date: Mon, 8 Nov 2021 11:58:13 +0100 Subject: Upgrade from buster to bullseye In-Reply-To: <40b2d547-2b41-ff1d-775a-5e786a82725b@lyx.org> References: <5b369feb-541f-c155-9ae9-4d34f9ad35a2@free.fr> <40b2d547-2b41-ff1d-775a-5e786a82725b@lyx.org> Message-ID: <77b643ff-136b-0e74-5052-77b90c3ffb98@lyx.org> Le 08/11/2021 ? 11:20, Jean-Marc Lasgouttes a ?crit?: >> However, when compiling the pdf of the UserGuide, none of the svgz >> icons could be converted, so the padf dispaly failed with a pdf image >> not found. Here is first of the error messages: >> >> convert-im6.q16: attempt to perform an operation not allowed by the >> security policy `PDF' @ error/constitute.c/IsCoderAuthorized/421. >> /usr/local/share/lyx-2.3.7dev/scripts/convertDefault.py ERROR >> Execution of "convert" failed. There is something I actually do not understand: this convert error triggers when converting a PDF file to something else. What is the path that created a pdf file from svg? Can we avoid it? JMarc From cornelisbb at gmail.com Mon Nov 8 11:10:20 2021 From: cornelisbb at gmail.com (Cor Blom) Date: Mon, 8 Nov 2021 12:10:20 +0100 Subject: Upgrade from buster to bullseye In-Reply-To: <77b643ff-136b-0e74-5052-77b90c3ffb98@lyx.org> References: <5b369feb-541f-c155-9ae9-4d34f9ad35a2@free.fr> <40b2d547-2b41-ff1d-775a-5e786a82725b@lyx.org> <77b643ff-136b-0e74-5052-77b90c3ffb98@lyx.org> Message-ID: <5d5feccd-132e-cf66-753e-a4b07ff8dbd5@gmail.com> Op 08-11-2021 om 11:58 schreef Jean-Marc Lasgouttes: > Le 08/11/2021 ? 11:20, Jean-Marc Lasgouttes a ?crit?: >>> However, when compiling the pdf of the UserGuide, none of the svgz >>> icons could be converted, so the padf dispaly failed with a pdf image >>> not found. Here is first of the error messages: >>> >>> convert-im6.q16: attempt to perform an operation not allowed by the >>> security policy `PDF' @ error/constitute.c/IsCoderAuthorized/421. >>> /usr/local/share/lyx-2.3.7dev/scripts/convertDefault.py ERROR >>> Execution of "convert" failed. > > There is something I actually do not understand: this convert error > triggers when converting a PDF file to something else. What is the path > that created a pdf file from svg? Can we avoid it? > > JMarc The words "security policy" and "PDF" reminds me of a security policy in openSUSE, which disables imagemagick working with PS/PDF by default. A additional package must be installed to enable this. No idea if this is the same, but who knows... Cor From lasgouttes at lyx.org Mon Nov 8 11:16:37 2021 From: lasgouttes at lyx.org (Jean-Marc Lasgouttes) Date: Mon, 8 Nov 2021 12:16:37 +0100 Subject: Upgrade from buster to bullseye In-Reply-To: <5d5feccd-132e-cf66-753e-a4b07ff8dbd5@gmail.com> References: <5b369feb-541f-c155-9ae9-4d34f9ad35a2@free.fr> <40b2d547-2b41-ff1d-775a-5e786a82725b@lyx.org> <77b643ff-136b-0e74-5052-77b90c3ffb98@lyx.org> <5d5feccd-132e-cf66-753e-a4b07ff8dbd5@gmail.com> Message-ID: <9f905cf5-72c0-6b47-0ea9-a991550f5567@lyx.org> Le 08/11/2021 ? 12:10, Cor Blom a ?crit?: > The words "security policy" and "PDF" reminds me of a security policy in > openSUSE, which disables imagemagick working with PS/PDF? by default. A > additional package must be installed to enable this. > > No idea if this is the same, but who knows... Yes, I think so. There has been a lot of work in ghostscript to address this, but I do not know whether this is enough. JMarc From kornel at lyx.org Mon Nov 8 13:12:22 2021 From: kornel at lyx.org (Kornel Benko) Date: Mon, 8 Nov 2021 14:12:22 +0100 Subject: Upgrade from buster to bullseye In-Reply-To: <9f905cf5-72c0-6b47-0ea9-a991550f5567@lyx.org> References: <5b369feb-541f-c155-9ae9-4d34f9ad35a2@free.fr> <40b2d547-2b41-ff1d-775a-5e786a82725b@lyx.org> <77b643ff-136b-0e74-5052-77b90c3ffb98@lyx.org> <5d5feccd-132e-cf66-753e-a4b07ff8dbd5@gmail.com> <9f905cf5-72c0-6b47-0ea9-a991550f5567@lyx.org> Message-ID: <20211108141222.1bd6dbb6@admin1-desktop> Am Mon, 8 Nov 2021 12:16:37 +0100 schrieb Jean-Marc Lasgouttes : > Le 08/11/2021 ? 12:10, Cor Blom a ?crit?: > > The words "security policy" and "PDF" reminds me of a security policy in > > openSUSE, which disables imagemagick working with PS/PDF? by default. A > > additional package must be installed to enable this. > > > > No idea if this is the same, but who knows... > > Yes, I think so. There has been a lot of work in ghostscript to address > this, but I do not know whether this is enough. > > JMarc > For ubuntu/debian it was sufficient to modify the imagemagic policy. Like: $ cp /etc/ImageMagick-6/policy.xml ~/.config/ImageMagick/. and add/modify policies for PS, PDF and EPS in ~/.config/ImageMagick/policy.xml to Kornel -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 488 bytes Desc: Digitale Signatur von OpenPGP URL: From lasgouttes at lyx.org Mon Nov 8 13:33:31 2021 From: lasgouttes at lyx.org (Jean-Marc Lasgouttes) Date: Mon, 8 Nov 2021 14:33:31 +0100 Subject: Upgrade from buster to bullseye In-Reply-To: <20211108141222.1bd6dbb6@admin1-desktop> References: <5b369feb-541f-c155-9ae9-4d34f9ad35a2@free.fr> <40b2d547-2b41-ff1d-775a-5e786a82725b@lyx.org> <77b643ff-136b-0e74-5052-77b90c3ffb98@lyx.org> <5d5feccd-132e-cf66-753e-a4b07ff8dbd5@gmail.com> <9f905cf5-72c0-6b47-0ea9-a991550f5567@lyx.org> <20211108141222.1bd6dbb6@admin1-desktop> Message-ID: Le 08/11/2021 ? 14:12, Kornel Benko a ?crit?: > Am Mon, 8 Nov 2021 12:16:37 +0100 > schrieb Jean-Marc Lasgouttes : > >> Le 08/11/2021 ? 12:10, Cor Blom a ?crit?: >>> The words "security policy" and "PDF" reminds me of a security policy in >>> openSUSE, which disables imagemagick working with PS/PDF? by default. A >>> additional package must be installed to enable this. >>> >>> No idea if this is the same, but who knows... >> >> Yes, I think so. There has been a lot of work in ghostscript to address >> this, but I do not know whether this is enough. >> >> JMarc >> > > For ubuntu/debian it was sufficient to modify the imagemagic policy. > Like: > $ cp /etc/ImageMagick-6/policy.xml ~/.config/ImageMagick/. > and add/modify policies for PS, PDF and EPS in ~/.config/ImageMagick/policy.xml to > > > Sure, but it would be nice if our users could use LyX without editing system configuration and without risking their security from other sources... JMarc From jeanpierre.chretien at free.fr Mon Nov 8 15:01:36 2021 From: jeanpierre.chretien at free.fr (=?UTF-8?Q?Jean-Pierre_Chr=c3=a9tien?=) Date: Mon, 8 Nov 2021 16:01:36 +0100 Subject: Upgrade from buster to bullseye In-Reply-To: <40b2d547-2b41-ff1d-775a-5e786a82725b@lyx.org> References: <5b369feb-541f-c155-9ae9-4d34f9ad35a2@free.fr> <40b2d547-2b41-ff1d-775a-5e786a82725b@lyx.org> Message-ID: <2d94aa2f-9b1e-ea93-1b68-156d49cf43a1@free.fr> Le 08/11/2021 ? 11:20, Jean-Marc Lasgouttes a ?crit?: > Le 08/11/2021 ? 11:03, Jean-Pierre Chr?tien a ?crit?: >> Dear developers >> >> I upgraded my Debian yesterday and I tested compilation and execution of >> master and 2.3.7dev. >> >> Compilation and display of the UserGuide run flawlessly on master. >> >> On 2.3.x, I first checked the fr.po file status and when I ran the make >> lyx-2.3.7dev.pot-update in the po dir of the build directory, I got in the >> following error: >> >> /bin/sh: 1: /usr/bin/python: not found >> make: *** [Makefile:774 : qt4_l10n.pot] Erreur 127 > > This is for Jos?, right? I guess so. > >> >> So I created a link: >> >> sudo ln -s /usr/bin/python3 /usr/bin/python >> >> and the problem was solved. >> >> The I compiled lyx-2.3.7dev all right with QT5 (QT4 fails). >> >> However, when compiling the pdf of the UserGuide, none of the svgz icons could >> be converted, so the padf dispaly failed with a pdf image not found. Here is >> first of the error messages: >> >> convert-im6.q16: attempt to perform an operation not allowed by the security >> policy `PDF' @ error/constitute.c/IsCoderAuthorized/421. >> /usr/local/share/lyx-2.3.7dev/scripts/convertDefault.py ERROR >> Execution of "convert" failed. > > Do you have rsvg-convert intalled (in librsvg2-bin package here). No I hadn't. So I installed librsvg2-bin, reconfigured LyX and it seems that the situation is worse : the conversion does not even seem to take place. > > This "convert" situation is really horrible, I do not understand that there is > no workaround yet. As the problem seems to be gone with master, I guess that such a workaround won't be needed in the (near?) future. But why is the behavior different between 2.3.7 and 2.4.0? -- Jean-Pierre From skostysh at lyx.org Mon Nov 8 16:48:16 2021 From: skostysh at lyx.org (Scott Kostyshak) Date: Mon, 8 Nov 2021 11:48:16 -0500 Subject: Upgrade from buster to bullseye In-Reply-To: <2d94aa2f-9b1e-ea93-1b68-156d49cf43a1@free.fr> References: <5b369feb-541f-c155-9ae9-4d34f9ad35a2@free.fr> <40b2d547-2b41-ff1d-775a-5e786a82725b@lyx.org> <2d94aa2f-9b1e-ea93-1b68-156d49cf43a1@free.fr> Message-ID: <20211108164816.jkhnpnwxsr4dblk4@poole> On Mon, Nov 08, 2021 at 04:01:36PM +0100, Jean-Pierre Chr?tien wrote: > Le 08/11/2021 ? 11:20, Jean-Marc Lasgouttes a ?crit?: > > Le 08/11/2021 ? 11:03, Jean-Pierre Chr?tien a ?crit?: > > > Dear developers > > > > > > I upgraded my Debian yesterday and I tested compilation and > > > execution of master and 2.3.7dev. > > > > > > Compilation and display of the UserGuide run flawlessly on master. > > > > > > On 2.3.x, I first checked the fr.po file status and when I ran the > > > make lyx-2.3.7dev.pot-update in the po dir of the build directory, I > > > got in the following error: > > > > > > /bin/sh: 1: /usr/bin/python: not found > > > make: *** [Makefile:774 : qt4_l10n.pot] Erreur 127 > > > > This is for Jos?, right? > > I guess so. Jos? made a couple of proposals a few months ago. See here: https://www.mail-archive.com/search?l=mid&q=2061214.Mh6RI2rZIc%40griffin Scott -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 488 bytes Desc: not available URL: From skostysh at lyx.org Mon Nov 8 16:52:17 2021 From: skostysh at lyx.org (Scott Kostyshak) Date: Mon, 8 Nov 2021 11:52:17 -0500 Subject: I extended the expiration date of my PGP key Message-ID: <20211108165217.6sbtt5hupbjmwrjv@poole> My PGP key recently expired. I extended it. If you are checking my signatures, to update my key I think you need to run the following command: gpg --keyserver pgp.mit.edu --recv-keys B6470BEB Sorry for the bother. Scott -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 488 bytes Desc: not available URL: From jeanpierre.chretien at free.fr Mon Nov 8 17:40:35 2021 From: jeanpierre.chretien at free.fr (=?UTF-8?Q?Jean-Pierre_Chr=c3=a9tien?=) Date: Mon, 8 Nov 2021 18:40:35 +0100 Subject: Upgrade from buster to bullseye In-Reply-To: <20211108141222.1bd6dbb6@admin1-desktop> References: <5b369feb-541f-c155-9ae9-4d34f9ad35a2@free.fr> <40b2d547-2b41-ff1d-775a-5e786a82725b@lyx.org> <77b643ff-136b-0e74-5052-77b90c3ffb98@lyx.org> <5d5feccd-132e-cf66-753e-a4b07ff8dbd5@gmail.com> <9f905cf5-72c0-6b47-0ea9-a991550f5567@lyx.org> <20211108141222.1bd6dbb6@admin1-desktop> Message-ID: Le 08/11/2021 ? 14:12, Kornel Benko a ?crit?: >> > > For ubuntu/debian it was sufficient to modify the imagemagic policy. > Like: > $ cp /etc/ImageMagick-6/policy.xml ~/.config/ImageMagick/. > and add/modify policies for PS, PDF and EPS in ~/.config/ImageMagick/policy.xml to > > > > I removen rsvh-conbert, reconfigured, created and edited ~/.config/ImageMagick/policy.xml and now, same behavior as with rsvg-convert, latex does not find the files. I'm lost. What is broken now ? I will recompile lyx-2.3.7dev from scratch. -- Jean-Pierre From sanda at lyx.org Mon Nov 8 20:57:37 2021 From: sanda at lyx.org (Pavel Sanda) Date: Mon, 8 Nov 2021 21:57:37 +0100 Subject: Upgrade from buster to bullseye In-Reply-To: <5b369feb-541f-c155-9ae9-4d34f9ad35a2@free.fr> References: <5b369feb-541f-c155-9ae9-4d34f9ad35a2@free.fr> Message-ID: <20211108205737.GA4582@jabberwock.ucw.cz> On Mon, Nov 08, 2021 at 11:03:25AM +0100, Jean-Pierre Chr?tien wrote: > On 2.3.x, I first checked the fr.po file status and when I ran the make > lyx-2.3.7dev.pot-update in the po dir of the build directory, I got in the > following error: > > /bin/sh: 1: /usr/bin/python: not found > make: *** [Makefile:774 : qt4_l10n.pot] Erreur 127 > > So I created a link: > > sudo ln -s /usr/bin/python3 /usr/bin/python > > and the problem was solved. This is actually not correct way how to solve this under bullseye. "python" is gone for good and scripts should use either python2 or python3 binary. If you insist on using "python" you should install specific package (IIRC "python") via: apt install python > However, when compiling the pdf of the UserGuide, none of the svgz icons > could be converted, so the padf dispaly failed with a pdf image not found. > Here is first of the error messages: > > convert-im6.q16: attempt to perform an operation not allowed by the security > policy `PDF' @ error/constitute.c/IsCoderAuthorized/421. > /usr/local/share/lyx-2.3.7dev/scripts/convertDefault.py ERROR > Execution of "convert" failed. > support/Systemcall.cpp (276): Systemcall: 'python -tt > "/usr/local/share/lyx-2.3.7dev/scripts/convertDefault.py" svgz "/tmp/lyx_tmpdir.ISfQxXhtSZQG/lyx_tmpbuf0/0_usr_local_share_lyx-2_3_7dev_images_buffer-view.svgz" > pdf "/tmp/lyx_tmpdir.ISfQxXhtSZQG/lyx_tmpbuf0/0_usr_local_share_lyx-2_3_7dev_images_buffer-view.pdf"' > finished with exit code 1 Yes, this is default imagemagick policy on debian now. First introduced in buster to overcome ghostscritp security issues, then reverted in testing (forthcoming bullseye) and then reintroduced again. I tried to explain the situation in couple debian bugs but no one seem to care. The situation is very unfortunate, because this default policy ban on postscript conversions can not be overriden by user setting and if you don't have root access on the machine you work, you can't use postscript anymore. We could perhaps improve error handling to catch this explain situation in some error dialog to users... Pavel From sanda at lyx.org Mon Nov 8 20:59:59 2021 From: sanda at lyx.org (Pavel Sanda) Date: Mon, 8 Nov 2021 21:59:59 +0100 Subject: Upgrade from buster to bullseye In-Reply-To: References: <5b369feb-541f-c155-9ae9-4d34f9ad35a2@free.fr> <40b2d547-2b41-ff1d-775a-5e786a82725b@lyx.org> <77b643ff-136b-0e74-5052-77b90c3ffb98@lyx.org> <5d5feccd-132e-cf66-753e-a4b07ff8dbd5@gmail.com> <9f905cf5-72c0-6b47-0ea9-a991550f5567@lyx.org> <20211108141222.1bd6dbb6@admin1-desktop> Message-ID: <20211108205959.GB4582@jabberwock.ucw.cz> On Mon, Nov 08, 2021 at 06:40:35PM +0100, Jean-Pierre Chr?tien wrote: > >For ubuntu/debian it was sufficient to modify the imagemagic policy. > >Like: > > $ cp /etc/ImageMagick-6/policy.xml ~/.config/ImageMagick/. > >and add/modify policies for PS, PDF and EPS in ~/.config/ImageMagick/policy.xml to > > > > > > > > > I removen rsvh-conbert, reconfigured, created and edited > ~/.config/ImageMagick/policy.xml You need to directly edit /etc/ImageMagick-6/policy.xml Pavel From jamatos at lyx.org Tue Nov 9 01:16:45 2021 From: jamatos at lyx.org (=?ISO-8859-1?Q?Jos=E9_Ab=EDlio?= Matos) Date: Tue, 09 Nov 2021 01:16:45 +0000 Subject: Upgrade from buster to bullseye In-Reply-To: <20211108205737.GA4582@jabberwock.ucw.cz> References: <5b369feb-541f-c155-9ae9-4d34f9ad35a2@free.fr> <20211108205737.GA4582@jabberwock.ucw.cz> Message-ID: <4664809.GXAFRqVoOG@griffin> On Monday, 8 November 2021 20.57.37 WET Pavel Sanda wrote: > This is actually not correct way how to solve this under bullseye. > "python" is gone for good and scripts should use either python2 > or python3 binary. > If you insist on using "python" you should install specific package (IIRC > "python") via: apt install python Now that python2 is gone (the last supported versions were released last year and no new one will be released) it is safe to use python again since it refers to python3. :-) For the moment that is the easiest solution. The more complete solution is the one that Scott mentioned because we already search for python in a more general way, Even python3 is a symlink to some python version. As an example in my computer running Fedora 35 I have: $ ls -l /usr/bin/python lrwxrwxrwx. 1 root root 9 Oct 4 23:36 /usr/bin/python -> ./python3 $ ls -l /usr/bin/python3 lrwxrwxrwx. 1 root root 10 Oct 4 23:10 /usr/bin/python3 -> python3.10 The previous case where python worked relied on python referring to python2. -- Jos? Ab?lio -------------- next part -------------- An HTML attachment was scrubbed... URL: From jeanpierre.chretien at free.fr Tue Nov 9 10:07:26 2021 From: jeanpierre.chretien at free.fr (=?UTF-8?Q?Jean-Pierre_Chr=c3=a9tien?=) Date: Tue, 9 Nov 2021 11:07:26 +0100 Subject: Upgrade from buster to bullseye In-Reply-To: <20211108205737.GA4582@jabberwock.ucw.cz> References: <5b369feb-541f-c155-9ae9-4d34f9ad35a2@free.fr> <20211108205737.GA4582@jabberwock.ucw.cz> Message-ID: <09cd32b4-63ee-5eb7-a073-53788777507b@free.fr> Le 08/11/2021 ? 21:57, Pavel Sanda a ?crit?: > On Mon, Nov 08, 2021 at 11:03:25AM +0100, Jean-Pierre Chr?tien wrote: >> On 2.3.x, I first checked the fr.po file status and when I ran the make >> lyx-2.3.7dev.pot-update in the po dir of the build directory, I got in the >> following error: >> >> /bin/sh: 1: /usr/bin/python: not found >> make: *** [Makefile:774 : qt4_l10n.pot] Erreur 127 >> >> So I created a link: >> >> sudo ln -s /usr/bin/python3 /usr/bin/python >> >> and the problem was solved. > > This is actually not correct way how to solve this under bullseye. > "python" is gone for good and scripts should use either python2 > or python3 binary. I was quite aware that my workaround was not the correct solution. However, as make lyx-2.3.7dev.pot-update fails because of missing python while make lyx-2.4.0dev.pot-update succeeds, is it possible to cherry-pick the python management from master ? > If you insist on using "python" you should install specific package (IIRC "python") via: > apt install python It is not me who insists, it is lyx-2.3.7dev :-) -- Jean-Pierre From jeanpierre.chretien at free.fr Tue Nov 9 10:28:30 2021 From: jeanpierre.chretien at free.fr (=?UTF-8?Q?Jean-Pierre_Chr=c3=a9tien?=) Date: Tue, 9 Nov 2021 11:28:30 +0100 Subject: Upgrade from buster to bullseye In-Reply-To: <20211108205737.GA4582@jabberwock.ucw.cz> References: <5b369feb-541f-c155-9ae9-4d34f9ad35a2@free.fr> <20211108205737.GA4582@jabberwock.ucw.cz> Message-ID: Le 08/11/2021 ? 21:57, Pavel Sanda a ?crit?: > > Yes, this is default imagemagick policy on debian now. > First introduced in buster to overcome ghostscritp security issues, then reverted > in testing (forthcoming bullseye) and then reintroduced again. I tried to explain > the situation in couple debian bugs but no one seem to care. I did not get these messages with buster, maybe I upgraded from etch after the reversion to testing. > > The situation is very unfortunate, because this default policy ban on > postscript conversions can not be overriden by user setting and if you > don't have root access on the machine you work, you can't use > postscript anymore. I edited /etc/ImageMagick-6/policy.xml without success, here is the log of the first file to convert, extracted from lyx-2.3.7dev -dbg graphics insets/InsetGraphics.cpp (119): findTargetFormat: PDF mode insets/InsetGraphics.cpp (742): we have: from svgz to pdf6 insets/InsetGraphics.cpp (748): the orig file is: /usr/local/share/lyx-2.3.7dev/images/buffer-new.svgz insets/InsetGraphics.cpp (794): The original file is /usr/local/share/lyx-2.3.7dev/images/buffer-new.svgz A copy has been made and convert is to be called with: file to convert = /tmp/lyx_tmpdir.ukkrIUtySeWW/lyx_tmpbuf0/0_usr_local_share_lyx-2_3_7dev_images_buffer-new.svgz from svgz to pdf6 insets/InsetGraphics.cpp (924): InsetGraphics::latex outputting: \includegraphics[width=1em]{0_usr_local_share_lyx-2_3_7dev_images_buffer-new.pdf} With 2.4.0dev (wich works all right), I get this : insets/InsetGraphics.cpp (117): findTargetFormat: PDF mode insets/InsetGraphics.cpp (735): we have: from svgz to pdf6 insets/InsetGraphics.cpp (741): the orig file is: /usr/local/share/lyx-2.4.0dev/images/buffer-new.svgz insets/InsetGraphics.cpp (787): The original file is /usr/local/share/lyx-2.4.0dev/images/buffer-new.svgz A copy has been made and convert is to be called with: file to convert = /tmp/lyx_tmpdir.HrwLplbmyPtA/lyx_tmpbuf0/export_buffer-new_svgz_c835ff7b1cb15ad019ad9820ad3e422677e2fb0ffb807224b41310571151427e.svgz from svgz to pdf6 insets/InsetGraphics.cpp (916): InsetGraphics::latex outputting: \includegraphics[width=1em]{export_buffer-new_svgz_c835ff7b1cb15ad019ad9820ad3e422677e2fb0ffb807224b41310571151427e.pdf} How can I display the conversion process from svgz to pdf6 itself? -- Jean-Pierre From sanda at lyx.org Tue Nov 9 11:08:24 2021 From: sanda at lyx.org (Pavel Sanda) Date: Tue, 9 Nov 2021 12:08:24 +0100 Subject: Upgrade from buster to bullseye In-Reply-To: <4664809.GXAFRqVoOG@griffin> References: <5b369feb-541f-c155-9ae9-4d34f9ad35a2@free.fr> <20211108205737.GA4582@jabberwock.ucw.cz> <4664809.GXAFRqVoOG@griffin> Message-ID: <20211109110824.GA710@jabberwock.ucw.cz> On Tue, Nov 09, 2021 at 01:16:45AM +0000, Jos? Ab?lio Matos wrote: > On Monday, 8 November 2021 20.57.37 WET Pavel Sanda wrote: > > This is actually not correct way how to solve this under bullseye. > > "python" is gone for good and scripts should use either python2 > > or python3 binary. > > If you insist on using "python" you should install specific package (IIRC > > "python") via: apt install python > > Now that python2 is gone (the last supported versions were released last year > and no new one will be released) it is safe to use python again since it > refers to python3. :-) Sure, my instructions were debian specific, because no "python" exists in vanilla system anymore. Pavel From sanda at lyx.org Tue Nov 9 11:22:03 2021 From: sanda at lyx.org (Pavel Sanda) Date: Tue, 9 Nov 2021 12:22:03 +0100 Subject: Upgrade from buster to bullseye In-Reply-To: References: <5b369feb-541f-c155-9ae9-4d34f9ad35a2@free.fr> <20211108205737.GA4582@jabberwock.ucw.cz> Message-ID: <20211109112203.GA4094@ucw.cz> On Tue, Nov 09, 2021 at 11:28:30AM +0100, Jean-Pierre Chr?tien wrote: > I edited /etc/ImageMagick-6/policy.xml without success, here is the log of > the first file to convert, extracted from lyx-2.3.7dev -dbg graphics Can show the current content of that config? > How can I display the conversion process from svgz to pdf6 itself? You need to enable "files" debug level. Pavel From jeanpierre.chretien at free.fr Wed Nov 10 13:07:03 2021 From: jeanpierre.chretien at free.fr (=?UTF-8?Q?Jean-Pierre_Chr=c3=a9tien?=) Date: Wed, 10 Nov 2021 14:07:03 +0100 Subject: Upgrade from buster to bullseye In-Reply-To: <20211109112203.GA4094@ucw.cz> References: <5b369feb-541f-c155-9ae9-4d34f9ad35a2@free.fr> <20211108205737.GA4582@jabberwock.ucw.cz> <20211109112203.GA4094@ucw.cz> Message-ID: Le 09/11/2021 ? 12:22, Pavel Sanda a ?crit?: > On Tue, Nov 09, 2021 at 11:28:30AM +0100, Jean-Pierre Chr?tien wrote: >> I edited /etc/ImageMagick-6/policy.xml without success, here is the log of >> the first file to convert, extracted from lyx-2.3.7dev -dbg graphics > > Can show the current content of that config? File attached. This is the original bullseye file in which I changed lines 93, 96, 97. > >> How can I display the conversion process from svgz to pdf6 itself? > > You need to enable "files" debug level. This not very informative, in fact I would like to see the conversion command to be able to test it from the console. -- Jean-Pierre -------------- next part -------------- A non-text attachment was scrubbed... Name: policy.xml Type: text/xml Size: 4694 bytes Desc: not available URL: From lasgouttes at lyx.org Wed Nov 10 13:32:52 2021 From: lasgouttes at lyx.org (Jean-Marc Lasgouttes) Date: Wed, 10 Nov 2021 14:32:52 +0100 Subject: Upgrade from buster to bullseye In-Reply-To: References: <5b369feb-541f-c155-9ae9-4d34f9ad35a2@free.fr> <20211108205737.GA4582@jabberwock.ucw.cz> <20211109112203.GA4094@ucw.cz> Message-ID: <2d844af5-993b-6c70-ad50-d4b0b6a5cb0c@lyx.org> Le 10/11/2021 ? 14:07, Jean-Pierre Chr?tien a ?crit?: >> You need to enable "files" debug level. > > This not very informative, in fact I would like to see the conversion > command to be able to test it from the console. I should (from Converter::convert, l. 661): LYXERR(Debug::FILES, "Calling " << command); What do you have as output? JMarc From skostysh at lyx.org Thu Nov 11 03:07:27 2021 From: skostysh at lyx.org (Scott Kostyshak) Date: Wed, 10 Nov 2021 22:07:27 -0500 Subject: LabelCounter with Beamer frame? Message-ID: <20211111030727.wxbx66pyxeidzotv@poole> I would find it helpful to display the Frame number in LyX when working with Beamer presentations. Attached is an attempt that doesn't work and screenshot showing that it doesn't work. I'm clueless about layouts. Any advice? Scott -------------- next part -------------- A non-text attachment was scrubbed... Name: 0001-Failed-attempt-to-get-beamer-frame-numbers.patch Type: text/x-diff Size: 1084 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: screenshot1636598041.png Type: image/png Size: 43379 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 488 bytes Desc: not available URL: From xracoonx at gmx.de Thu Nov 11 07:38:53 2021 From: xracoonx at gmx.de (Daniel) Date: Thu, 11 Nov 2021 08:38:53 +0100 Subject: LabelCounter with Beamer frame? In-Reply-To: <20211111030727.wxbx66pyxeidzotv@poole> References: <20211111030727.wxbx66pyxeidzotv@poole> Message-ID: On 2021-11-11 04:07, Scott Kostyshak wrote: > I would find it helpful to display the Frame number in LyX when working > with Beamer presentations. > > Attached is an attempt that doesn't work and screenshot showing that it > doesn't work. > > I'm clueless about layouts. Any advice? > > Scott Adding this to your Document > Local Layout works for me: Counter frame End Style Frame LabelString "Frame \theframe" LabelCounter frame TocLevel 3 End For some reason, if one keeps the TocLevel to 4, the label "Frame ..." disappears. But, I guess, as long as you don't use Subsubsection in your presentation, it should be no problem? Daniel From xracoonx at gmx.de Thu Nov 11 07:44:30 2021 From: xracoonx at gmx.de (Daniel) Date: Thu, 11 Nov 2021 08:44:30 +0100 Subject: LabelCounter with Beamer frame? In-Reply-To: References: <20211111030727.wxbx66pyxeidzotv@poole> Message-ID: On 2021-11-11 08:38, Daniel wrote: > On 2021-11-11 04:07, Scott Kostyshak wrote: >> I would find it helpful to display the Frame number in LyX when working >> with Beamer presentations. >> >> Attached is an attempt that doesn't work and screenshot showing that it >> doesn't work. >> >> I'm clueless about layouts. Any advice? >> >> Scott > > Adding this to your Document > Local Layout works for me: > > > Counter frame > End > > Style Frame > ? LabelString????? "Frame \theframe" > ? LabelCounter???? frame > ? TocLevel???????? 3 > End > > > For some reason, if one keeps the TocLevel to 4, the label "Frame ..." > disappears. But, I guess, as long as you don't use Subsubsection in your > presentation, it should be no problem? > > Daniel > Actually, it seems the local layout is read after the non-local layout is read. So you have to either tweak the other frame layouts (PlainFrame etc.) accordingly, or create a local copy of beamer.layout and tweak the Frame layout directly there. Numbering the frames seems to be good idea in general. Maybe it could become the default? (Then the TocLevel stuff would have to be fixed somehow though.) Daniel From jeanpierre.chretien at free.fr Thu Nov 11 10:52:23 2021 From: jeanpierre.chretien at free.fr (=?UTF-8?Q?Jean-Pierre_Chr=c3=a9tien?=) Date: Thu, 11 Nov 2021 11:52:23 +0100 Subject: Upgrade from buster to bullseye In-Reply-To: <2d844af5-993b-6c70-ad50-d4b0b6a5cb0c@lyx.org> References: <5b369feb-541f-c155-9ae9-4d34f9ad35a2@free.fr> <20211108205737.GA4582@jabberwock.ucw.cz> <20211109112203.GA4094@ucw.cz> <2d844af5-993b-6c70-ad50-d4b0b6a5cb0c@lyx.org> Message-ID: Le 10/11/2021 ? 14:32, Jean-Marc Lasgouttes a ?crit?: > Le 10/11/2021 ? 14:07, Jean-Pierre Chr?tien a ?crit?: >>> You need to enable "files" debug level. >> >> This not very informative, in fact I would like to see the conversion command >> to be able to test it from the console. > > I should (from Converter::convert, l. 661): > ????LYXERR(Debug::FILES, "Calling " << command); > > What do you have as output? No Converter.cpp reference, so the message is not triggered. All references are to ConverterCache.cpp, so I guess that I should clean the cache, but I do not remember how... -- Jean-Pierre From jeanpierre.chretien at free.fr Thu Nov 11 11:02:32 2021 From: jeanpierre.chretien at free.fr (=?UTF-8?Q?Jean-Pierre_Chr=c3=a9tien?=) Date: Thu, 11 Nov 2021 12:02:32 +0100 Subject: Upgrade from buster to bullseye In-Reply-To: References: <5b369feb-541f-c155-9ae9-4d34f9ad35a2@free.fr> <20211108205737.GA4582@jabberwock.ucw.cz> <20211109112203.GA4094@ucw.cz> <2d844af5-993b-6c70-ad50-d4b0b6a5cb0c@lyx.org> Message-ID: Le 11/11/2021 ? 11:52, Jean-Pierre Chr?tien a ?crit?: > Le 10/11/2021 ? 14:32, Jean-Marc Lasgouttes a ?crit?: >> Le 10/11/2021 ? 14:07, Jean-Pierre Chr?tien a ?crit?: >>>> You need to enable "files" debug level. >>> >>> This not very informative, in fact I would like to see the conversion command >>> to be able to test it from the console. >> >> I should (from Converter::convert, l. 661): >> ?????LYXERR(Debug::FILES, "Calling " << command); >> >> What do you have as output? > > No Converter.cpp reference, so the message is not triggered. All references are > to ConverterCache.cpp, so I guess that I should clean the cache, but I do not > remember how... > I found a cache directory in my personal LyX dir, so I removed the contents ans bingo! compilation of the UserGuide works flawlessly. ConverterCache.cpp (401): not in cache. Looking for python v2.x or 3.x ... Examining /opt/texlive/2021/bin/x86_64-linux/pythontex Examining /usr/bin/python2 Found Python 2.7.18 Converter.cpp (430): No converter defined! I use convertDefault.py: /usr/bin/python2 -tt "/usr/local/share/lyx-2.3.7dev/scripts/convertDefault.py" svgz "/tmp/lyx_tmpdir.XrMxvxhYnibI/lyx_tmpbuf0/0_usr_local_share_lyx-2_3_7dev_images_buffer-new.svgz" pdf "/tmp/lyx_tmpdir.XrMxvxhYnibI/lyx_tmpbuf0/0_usr_local_share_lyx-2_3_7dev_images_buffer-new.pdf" ConverterCache.cpp (270): /usr/local/share/lyx-2.3.7dev/images/buffer-new.svgz pdf6 /tmp/lyx_tmpdir.XrMxvxhYnibI/lyx_tmpbuf0/0_usr_local_share_lyx-2_3_7dev_images_buffer-new.pdf Editing policy.html was the trick. Thanks to all of you who helped. -- Jean-Pierre From lasgouttes at lyx.org Thu Nov 11 12:23:46 2021 From: lasgouttes at lyx.org (Jean-Marc Lasgouttes) Date: Thu, 11 Nov 2021 13:23:46 +0100 Subject: Upgrade from buster to bullseye In-Reply-To: References: <5b369feb-541f-c155-9ae9-4d34f9ad35a2@free.fr> <20211108205737.GA4582@jabberwock.ucw.cz> <20211109112203.GA4094@ucw.cz> <2d844af5-993b-6c70-ad50-d4b0b6a5cb0c@lyx.org> Message-ID: <98bc0ec1-408a-0812-b605-218040b9e0e5@lyx.org> Le 11/11/2021 ? 12:02, Jean-Pierre Chr?tien a ?crit?: > Converter.cpp (430): No converter defined! I use convertDefault.py: > ????/usr/bin/python2 -tt > "/usr/local/share/lyx-2.3.7dev/scripts/convertDefault.py" svgz > "/tmp/lyx_tmpdir.XrMxvxhYnibI/lyx_tmpbuf0/0_usr_local_share_lyx-2_3_7dev_images_buffer-new.svgz" > pdf > "/tmp/lyx_tmpdir.XrMxvxhYnibI/lyx_tmpbuf0/0_usr_local_share_lyx-2_3_7dev_images_buffer-new.pdf" > > ConverterCache.cpp (270): > /usr/local/share/lyx-2.3.7dev/images/buffer-new.svgz pdf6 > /tmp/lyx_tmpdir.XrMxvxhYnibI/lyx_tmpbuf0/0_usr_local_share_lyx-2_3_7dev_images_buffer-new.pdf > > > > Editing policy.html was the trick. > Thanks to all of you who helped. Do you have rsvg-convert installed? Is it used in this situation? JMarc From skostysh at lyx.org Thu Nov 11 14:24:57 2021 From: skostysh at lyx.org (Scott Kostyshak) Date: Thu, 11 Nov 2021 09:24:57 -0500 Subject: LabelCounter with Beamer frame? In-Reply-To: References: <20211111030727.wxbx66pyxeidzotv@poole> Message-ID: <20211111142457.hb3ymkubc6on2hkl@poole> On Thu, Nov 11, 2021 at 08:44:30AM +0100, Daniel wrote: > On 2021-11-11 08:38, Daniel wrote: > > On 2021-11-11 04:07, Scott Kostyshak wrote: > > > I would find it helpful to display the Frame number in LyX when working > > > with Beamer presentations. > > > > > > Attached is an attempt that doesn't work and screenshot showing that it > > > doesn't work. > > > > > > I'm clueless about layouts. Any advice? > > > > > > Scott > > > > Adding this to your Document > Local Layout works for me: > > > > > > Counter frame > > End > > > > Style Frame > > ? LabelString????? "Frame \theframe" > > ? LabelCounter???? frame > > ? TocLevel???????? 3 > > End > > > > > > For some reason, if one keeps the TocLevel to 4, the label "Frame ..." > > disappears. But, I guess, as long as you don't use Subsubsection in your > > presentation, it should be no problem? > > > > Daniel > > > > Actually, it seems the local layout is read after the non-local layout is > read. So you have to either tweak the other frame layouts (PlainFrame etc.) > accordingly, or create a local copy of beamer.layout and tweak the Frame > layout directly there. This works great. Thank you! I had to set 'InitialValue' to 2 since the title takes up a frame for me. It's really nice to have the numbers. I find it especially helpful when I print out a copy and make edits and then want to input those hardcopy edits to the file. Navigating by slide numbers is a bit faster than searching for titles or (for frames without a title) content text. Thanks for the help, Daniel! > Numbering the frames seems to be good idea in general. Maybe it could become > the default? (Then the TocLevel stuff would have to be fixed somehow > though.) I think that might be nice. Indeed we'd have to smooth a few things out. If anyone is interested in working on this feature, maybe we could first ask on lyx-users what opinions are. Scott -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 488 bytes Desc: not available URL: From martinrd3d at gmail.com Thu Nov 11 16:12:57 2021 From: martinrd3d at gmail.com (Doug Martin) Date: Thu, 11 Nov 2021 08:12:57 -0800 Subject: LabelCounter with Beamer frame? In-Reply-To: <20211111142457.hb3ymkubc6on2hkl@poole> References: <20211111030727.wxbx66pyxeidzotv@poole> <20211111142457.hb3ymkubc6on2hkl@poole> Message-ID: All, I wouldn't be able to work on this feature, but I am using Beamer all the time now, and it would be great to have it. Doug On Thu, Nov 11, 2021 at 6:25 AM Scott Kostyshak wrote: > On Thu, Nov 11, 2021 at 08:44:30AM +0100, Daniel wrote: > > On 2021-11-11 08:38, Daniel wrote: > > > On 2021-11-11 04:07, Scott Kostyshak wrote: > > > > I would find it helpful to display the Frame number in LyX when > working > > > > with Beamer presentations. > > > > > > > > Attached is an attempt that doesn't work and screenshot showing that > it > > > > doesn't work. > > > > > > > > I'm clueless about layouts. Any advice? > > > > > > > > Scott > > > > > > Adding this to your Document > Local Layout works for me: > > > > > > > > > Counter frame > > > End > > > > > > Style Frame > > > LabelString "Frame \theframe" > > > LabelCounter frame > > > TocLevel 3 > > > End > > > > > > > > > For some reason, if one keeps the TocLevel to 4, the label "Frame ..." > > > disappears. But, I guess, as long as you don't use Subsubsection in > your > > > presentation, it should be no problem? > > > > > > Daniel > > > > > > > Actually, it seems the local layout is read after the non-local layout is > > read. So you have to either tweak the other frame layouts (PlainFrame > etc.) > > accordingly, or create a local copy of beamer.layout and tweak the Frame > > layout directly there. > > This works great. Thank you! I had to set 'InitialValue' to 2 since the > title takes up a frame for me. It's really nice to have the numbers. I > find it especially helpful when I print out a copy and make edits and > then want to input those hardcopy edits to the file. Navigating by slide > numbers is a bit faster than searching for titles or (for frames without > a title) content text. > > Thanks for the help, Daniel! > > > Numbering the frames seems to be good idea in general. Maybe it could > become > > the default? (Then the TocLevel stuff would have to be fixed somehow > > though.) > > I think that might be nice. Indeed we'd have to smooth a few things out. > If anyone is interested in working on this feature, maybe we could first > ask on lyx-users what opinions are. > > Scott > -- > lyx-devel mailing list > lyx-devel at lists.lyx.org > http://lists.lyx.org/mailman/listinfo/lyx-devel > -- R. Douglas Martin Professor Emeritus in Applied Mathematics and Statistics University of Washington -------------- next part -------------- An HTML attachment was scrubbed... URL: From jeanpierre.chretien at free.fr Thu Nov 11 16:15:33 2021 From: jeanpierre.chretien at free.fr (=?UTF-8?Q?Jean-Pierre_Chr=c3=a9tien?=) Date: Thu, 11 Nov 2021 17:15:33 +0100 Subject: Upgrade from buster to bullseye In-Reply-To: <98bc0ec1-408a-0812-b605-218040b9e0e5@lyx.org> References: <5b369feb-541f-c155-9ae9-4d34f9ad35a2@free.fr> <20211108205737.GA4582@jabberwock.ucw.cz> <20211109112203.GA4094@ucw.cz> <2d844af5-993b-6c70-ad50-d4b0b6a5cb0c@lyx.org> <98bc0ec1-408a-0812-b605-218040b9e0e5@lyx.org> Message-ID: <556b36a5-75bf-9c82-d266-11a94b84e9f2@free.fr> Le 11/11/2021 ? 13:23, Jean-Marc Lasgouttes a >> >> Editing policy.html was the trick. >> Thanks to all of you who helped. > > Do you have rsvg-convert installed? Is it used in this situation? I installed librsvg2-bin, reconfigured lyx-2.3.7dev, removed files from the cache and again, compilation of the UserGuide went flawlessly with calls to rsvg. The rsvg path is actually easier on Debian that editing the ImageMagick policy file. I never use Debian install of LyX because it requires to install Texlive as well (and I install it separately from tug.org), but is librsvg2-bin in the Lyx dependencies ? -- Jean-Pierre From jeanpierre.chretien at free.fr Thu Nov 11 16:41:43 2021 From: jeanpierre.chretien at free.fr (JP) Date: Thu, 11 Nov 2021 17:41:43 +0100 Subject: Upgrade from buster to bullseye In-Reply-To: <556b36a5-75bf-9c82-d266-11a94b84e9f2@free.fr> References: <5b369feb-541f-c155-9ae9-4d34f9ad35a2@free.fr> <20211108205737.GA4582@jabberwock.ucw.cz> <20211109112203.GA4094@ucw.cz> <2d844af5-993b-6c70-ad50-d4b0b6a5cb0c@lyx.org> <98bc0ec1-408a-0812-b605-218040b9e0e5@lyx.org> <556b36a5-75bf-9c82-d266-11a94b84e9f2@free.fr> Message-ID: <17d0fe05d70.27d4.e01bbce198d34a7a74a23cfe2737f717@free.fr> Le 11 novembre 2021 17:15:47 Jean-Pierre Chr?tien a ?crit > > I never use Debian install of LyX because it requires to install Texlive as > well > (and I install it separately from tug.org), but is librsvg2-bin in the Lyx > dependencies ? I checked on Debian packages: it is only suggested (while ImageMagick is recommended). -- Jean-Pierre -------------- next part -------------- An HTML attachment was scrubbed... URL: From jkennan at gmail.com Thu Nov 11 17:14:23 2021 From: jkennan at gmail.com (john kennan) Date: Thu, 11 Nov 2021 11:14:23 -0600 Subject: compiling LyX Message-ID: Hi I'd like to be able to compile LyX (on Mac OS Big Sur). I tried following the instructions on https://wiki.lyx.org/Devel/Git Got as far as this: (base) jk at jkmac-3 ~ % git config --global user.name jkennan (base) jk at jkmac-3 ~ % git config --global user.email jkennan at gmail.com (base) jk at jkmac-3 ~ %git checkout master fatal: not a git repository (or any of the parent directories): .git John Kennan -------------- next part -------------- An HTML attachment was scrubbed... URL: From skostysh at lyx.org Thu Nov 11 17:26:28 2021 From: skostysh at lyx.org (Scott Kostyshak) Date: Thu, 11 Nov 2021 12:26:28 -0500 Subject: LabelCounter with Beamer frame? In-Reply-To: References: <20211111030727.wxbx66pyxeidzotv@poole> <20211111142457.hb3ymkubc6on2hkl@poole> Message-ID: <20211111172628.ptxe7wisq3qtjyht@poole> On Thu, Nov 11, 2021 at 08:12:57AM -0800, Doug Martin wrote: > All, > > I wouldn't be able to work on this feature, but I am using Beamer all the > time now, > and it would be great to have it. Good to know, Doug. Is there a particular use case where you would find it useful? Scott -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 488 bytes Desc: not available URL: From skostysh at lyx.org Thu Nov 11 17:39:25 2021 From: skostysh at lyx.org (Scott Kostyshak) Date: Thu, 11 Nov 2021 12:39:25 -0500 Subject: compiling LyX In-Reply-To: References: Message-ID: <20211111173925.cbxo4h4qak4hfuix@poole> On Thu, Nov 11, 2021 at 11:14:23AM -0600, john kennan wrote: > Hi > > I'd like to be able to compile LyX (on Mac OS Big Sur). > > I tried following the instructions on https://wiki.lyx.org/Devel/Git > Got as far as this: > > (base) jk at jkmac-3 ~ % git config --global user.name jkennan > (base) jk at jkmac-3 ~ % git config --global user.email jkennan at gmail.com > (base) jk at jkmac-3 ~ %git checkout master > fatal: not a git repository (or any of the parent directories): .git > > John Kennan Hi John, First you need to clone: git clone "git://git.lyx.org/lyx" That will create a directory called "lyx" wherever you run that command. After cloning, then you need to hope that Stephan can help you :). I think he has a build script for compiling LyX on macOS. Scott -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 488 bytes Desc: not available URL: From jkennan at gmail.com Thu Nov 11 18:00:45 2021 From: jkennan at gmail.com (john kennan) Date: Thu, 11 Nov 2021 12:00:45 -0600 Subject: compiling LyX In-Reply-To: References: Message-ID: I don't know how to reply to messages on the list (I'm already subscribed) Edited the wiki (https://wiki.lyx.org/Devel/Git) to say that one must change to the lyx directory after the first couple of steps John On Thu, Nov 11, 2021 at 11:14 AM john kennan wrote: > Hi > > I'd like to be able to compile LyX (on Mac OS Big Sur). > > I tried following the instructions on https://wiki.lyx.org/Devel/Git > Got as far as this: > > (base) jk at jkmac-3 ~ % git config --global user.name jkennan > (base) jk at jkmac-3 ~ % git config --global user.email jkennan at gmail.com > (base) jk at jkmac-3 ~ %git checkout master > fatal: not a git repository (or any of the parent directories): .git > > John Kennan > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From martinrd3d at gmail.com Thu Nov 11 18:08:27 2021 From: martinrd3d at gmail.com (Doug Martin) Date: Thu, 11 Nov 2021 10:08:27 -0800 Subject: LabelCounter with Beamer frame? In-Reply-To: <20211111172628.ptxe7wisq3qtjyht@poole> References: <20211111030727.wxbx66pyxeidzotv@poole> <20211111142457.hb3ymkubc6on2hkl@poole> <20211111172628.ptxe7wisq3qtjyht@poole> Message-ID: I would just use it all the time, as hany reference information. And for example, if I am reviewing the entire compiled pdf slide deck, and see some slides that I want to move to a different order, I could find it easily in the LYX file. By the way, I almost replied to the long thread END OF FRAME once .. , started by Ehud Kaplan, if only to say that I had shared much of the frustration in getting started, and remembering what to do upon returning to LYX Beamer after not using it for a while. Back in the trench again with it, I think a good user guide would go a long way to helping many people. On Thu, Nov 11, 2021 at 9:26 AM Scott Kostyshak wrote: > On Thu, Nov 11, 2021 at 08:12:57AM -0800, Doug Martin wrote: > > All, > > > > I wouldn't be able to work on this feature, but I am using Beamer all the > > time now, > > and it would be great to have it. > > Good to know, Doug. Is there a particular use case where you would find it > useful? > > Scott > -- > lyx-devel mailing list > lyx-devel at lists.lyx.org > http://lists.lyx.org/mailman/listinfo/lyx-devel > -- R. Douglas Martin Professor Emeritus in Applied Mathematics and Statistics University of Washington -------------- next part -------------- An HTML attachment was scrubbed... URL: From skostysh at lyx.org Thu Nov 11 19:43:13 2021 From: skostysh at lyx.org (Scott Kostyshak) Date: Thu, 11 Nov 2021 14:43:13 -0500 Subject: LabelCounter with Beamer frame? In-Reply-To: References: <20211111030727.wxbx66pyxeidzotv@poole> <20211111142457.hb3ymkubc6on2hkl@poole> <20211111172628.ptxe7wisq3qtjyht@poole> Message-ID: <20211111194313.a4iztqzb5nojtfmi@poole> On Thu, Nov 11, 2021 at 10:08:27AM -0800, Doug Martin wrote: > I would just use it all the time, as hany reference information. > And for example, if I am reviewing the entire compiled pdf slide deck, and > see some > slides that I want to move to a different order, I could find it easily in > the LYX file. Good to know. That's similar to what I am planning to use it for. > By the way, I almost replied to the long thread END OF FRAME once .. , > started by Ehud Kaplan, > if only to say that I had shared much of the frustration in getting > started, and remembering what to do > upon returning to LYX Beamer after not using it for a while. Back in the > trench again with it, I think > a good user guide would go a long way to helping many people. Makes sense. My opinion is that the hard work under the hood has been done (I know J?rgen did a lot to improve consistency and features for beamer). I think we just need to smooth out the user experience a bit. In any case, I don't see this happening until after LyX 2.4.0 but I'm confident we can improve things in the long run. Scott -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 488 bytes Desc: not available URL: From lasgouttes at lyx.org Thu Nov 11 21:37:09 2021 From: lasgouttes at lyx.org (Jean-Marc Lasgouttes) Date: Thu, 11 Nov 2021 22:37:09 +0100 Subject: compiling LyX In-Reply-To: References: Message-ID: Le 11/11/2021 ? 19:00, john kennan a ?crit?: > I don't know how to reply to messages on the list (I'm already subscribed) Hello John, What is the problem? That you answer to sender instead? What are you using for your emails? JMarc From jkennan at gmail.com Thu Nov 11 22:30:13 2021 From: jkennan at gmail.com (john kennan) Date: Thu, 11 Nov 2021 16:30:13 -0600 Subject: Fwd: compiling LyX In-Reply-To: References: Message-ID: I'm using Gmail, looking at the Mail Archive. I can send messages to the list, but I see no way to connect what I'm sending with what I'm replying to. John ---------- Forwarded message --------- From: john kennan Date: Thu, Nov 11, 2021 at 12:00 PM Subject: Re: compiling LyX To: LyX Developers I don't know how to reply to messages on the list (I'm already subscribed) Edited the wiki (https://wiki.lyx.org/Devel/Git) to say that one must change to the lyx directory after the first couple of steps John On Thu, Nov 11, 2021 at 11:14 AM john kennan wrote: > Hi > > I'd like to be able to compile LyX (on Mac OS Big Sur). > > I tried following the instructions on https://wiki.lyx.org/Devel/Git > Got as far as this: > > (base) jk at jkmac-3 ~ % git config --global user.name jkennan > (base) jk at jkmac-3 ~ % git config --global user.email jkennan at gmail.com > (base) jk at jkmac-3 ~ %git checkout master > fatal: not a git repository (or any of the parent directories): .git > > John Kennan > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From lasgouttes at lyx.org Thu Nov 11 22:33:18 2021 From: lasgouttes at lyx.org (Jean-Marc Lasgouttes) Date: Thu, 11 Nov 2021 23:33:18 +0100 Subject: Fwd: compiling LyX In-Reply-To: References: Message-ID: <23f0e2bf-2dfc-2237-67c7-48e5c69f1fe1@lyx.org> Le 11/11/2021 ? 23:30, john kennan a ?crit?: > I'm using Gmail, looking at the Mail Archive. I can send messages to the > list, but I see no way to connect what I'm sending with what I'm > replying to. There is something I do not understand. You are subscribed, but read lyx-devel through the mail archive? Did I get this right? If I did, I have no idea of how to make things easier for you. JMarc From xracoonx at gmx.de Sun Nov 14 08:05:21 2021 From: xracoonx at gmx.de (Daniel) Date: Sun, 14 Nov 2021 09:05:21 +0100 Subject: LabelCounter with Beamer frame? In-Reply-To: <20211111142457.hb3ymkubc6on2hkl@poole> References: <20211111030727.wxbx66pyxeidzotv@poole> <20211111142457.hb3ymkubc6on2hkl@poole> Message-ID: On 2021-11-11 15:24, Scott Kostyshak wrote: > This works great. Thank you! I had to set 'InitialValue' to 2 since the > title takes up a frame for me. I guess a more ideal solution would be to make the title also increase the counter by one since sometimes one might have a title frame and sometimes not (or a custom one consisting of a normal frame). However, I did not manage to get the counter to step up with the Title style *without* adding a label. For example, I tried Style Title LabelType Static LabelString "" LabelCounter frame End But instead of an empty LabelString, LyX seems to treat it the same as if it hadn't been defined and adds just the number in front. Maybe this is a bug? Shouldn't there be a difference between an empty string and no string set? In general, it would be good if one could let counters increase without defining a special label which presupposes too much that one has understood the inner workings of counters. Or at least there should be a better documentation for how it does (and does not) work. I guess Scott having failed is pretty good evidence that something is not great here. Daniel From skostysh at lyx.org Sun Nov 14 14:17:42 2021 From: skostysh at lyx.org (Scott Kostyshak) Date: Sun, 14 Nov 2021 09:17:42 -0500 Subject: LabelCounter with Beamer frame? In-Reply-To: References: <20211111030727.wxbx66pyxeidzotv@poole> <20211111142457.hb3ymkubc6on2hkl@poole> Message-ID: <20211114141742.rg35rgqkxrp5m4w4@poole> On Sun, Nov 14, 2021 at 09:05:21AM +0100, Daniel wrote: > On 2021-11-11 15:24, Scott Kostyshak wrote: > > This works great. Thank you! I had to set 'InitialValue' to 2 since the > > title takes up a frame for me. > > I guess a more ideal solution would be to make the title also increase the > counter by one since sometimes one might have a title frame and sometimes > not (or a custom one consisting of a normal frame). However, I did not > manage to get the counter to step up with the Title style *without* adding a > label. For example, I tried > > > Style Title > LabelType Static > LabelString "" > LabelCounter frame > End Good idea, that is a better approach. > But instead of an empty LabelString, LyX seems to treat it the same as if it > hadn't been defined and adds just the number in front. Maybe this is a bug? > Shouldn't there be a difference between an empty string and no string set? > In general, it would be good if one could let counters increase without > defining a special label which presupposes too much that one has understood > the inner workings of counters. Or at least there should be a better > documentation for how it does (and does not) work. I guess Scott having > failed is pretty good evidence that something is not great here. Things are definitely complex. A lot of it is that I just don't have much familiarity with layout tags. In any case, thanks for the help! Scott -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 488 bytes Desc: not available URL: From lasgouttes at lyx.org Wed Nov 17 13:44:54 2021 From: lasgouttes at lyx.org (Jean-Marc Lasgouttes) Date: Wed, 17 Nov 2021 14:44:54 +0100 Subject: LabelCounter with Beamer frame? In-Reply-To: References: <20211111030727.wxbx66pyxeidzotv@poole> <20211111142457.hb3ymkubc6on2hkl@poole> Message-ID: <7800bc53-058b-4419-a2ca-63c79bae4b7c@lyx.org> Le 14/11/2021 ? 09:05, Daniel a ?crit?: > On 2021-11-11 15:24, Scott Kostyshak wrote: >> This works great. Thank you! I had to set 'InitialValue' to 2 since the >> title takes up a frame for me. > > I guess a more ideal solution would be to make the title also increase > the counter by one since sometimes one might have a title frame and > sometimes not (or a custom one consisting of a normal frame). However, I > did not manage to get the counter to step up with the Title style > *without* adding a label. For example, I tried > > > Style Title > ? LabelType??????? Static > ? LabelString????? "" > ? LabelCounter???? frame > End It would not be difficult in Buffer::Impl::setLabel to handle the case where LabelType==NoLabel and there is a LabelCounter specified. All is needed is to step the counter if it exists. JMarc From xracoonx at gmx.de Wed Nov 17 19:43:27 2021 From: xracoonx at gmx.de (Daniel) Date: Wed, 17 Nov 2021 20:43:27 +0100 Subject: LabelCounter with Beamer frame? In-Reply-To: <7800bc53-058b-4419-a2ca-63c79bae4b7c@lyx.org> References: <20211111030727.wxbx66pyxeidzotv@poole> <20211111142457.hb3ymkubc6on2hkl@poole> <7800bc53-058b-4419-a2ca-63c79bae4b7c@lyx.org> Message-ID: On 2021-11-17 14:44, Jean-Marc Lasgouttes wrote: > Le 14/11/2021 ? 09:05, Daniel a ?crit?: >> On 2021-11-11 15:24, Scott Kostyshak wrote: >>> This works great. Thank you! I had to set 'InitialValue' to 2 since the >>> title takes up a frame for me. >> >> I guess a more ideal solution would be to make the title also increase >> the counter by one since sometimes one might have a title frame and >> sometimes not (or a custom one consisting of a normal frame). However, >> I did not manage to get the counter to step up with the Title style >> *without* adding a label. For example, I tried >> >> >> Style Title >> ?? LabelType??????? Static >> ?? LabelString????? "" >> ?? LabelCounter???? frame >> End > > It would not be difficult in Buffer::Impl::setLabel to handle the case > where LabelType==NoLabel and there is a LabelCounter specified. All is > needed is to step the counter if it exists. How about checking for whether, in case of an environment, this is the first paragraph in the sequence? Similar to what is done in the case of the other labels, all in all something like: docstring const & lcounter = layout.counter; if (!lcounter.empty() && counters.hasCounter(lcounter) && (layout.latextype != LATEX_ENVIRONMENT || it.text()->isFirstInSequence(it.pit()))) {) counters.step(lcounter, utype); } Daniel From skostysh at lyx.org Wed Nov 24 02:35:17 2021 From: skostysh at lyx.org (Scott Kostyshak) Date: Tue, 23 Nov 2021 21:35:17 -0500 Subject: [LyX/master] Meet per-document spelling dictionaries (fixes #86 [sic!]) In-Reply-To: <20210821134021.73i3u2repkcsqqr7@poole> References: <20210306155212.803DC280262@lyx.lyx.org> <03c2466a-00dc-5782-52b6-a8cfe2e8918c@lyx.org> <20210306172549.5p3t4fbqj2ivckom@tallinn> <20210819200418.47dbppay6sknd7ni@poole> <20210821134021.73i3u2repkcsqqr7@poole> Message-ID: <20211124023517.7vxctqz34qaceakf@poole> On Sat, Aug 21, 2021 at 09:40:21AM -0400, Scott Kostyshak wrote: > On Sat, Aug 21, 2021 at 03:20:05PM +0200, J?rgen Spitzm?ller wrote: > > Am Donnerstag, dem 19.08.2021 um 16:04 -0400 schrieb Scott Kostyshak: > > > I've just tested it and it is working well. I will enjoy using this > > > feature for my personal use. A few comments, although I understand if > > > you don't have time these days: > > > > Yes, and I fear this will be true for quite a while. > > > > > > > > 1. When I do "Ignore all", a subsequent occurrence is still found as > > > ?? misspelled. For example, in "hello blahblahxx blahblahxx" if I put > > > ?? "Ignore All" for the first occurrence of "blahblahxx", the > > > spellcheck > > > ?? still goes to the second occurrence of "blahblahxx". Is this > > > ?? expected? I expected "Ignore all" to imply "Skip all" but perhaps > > > ?? I've misunderstood? > > > > No, you're right. We need to re-check the whole buffer. Done now. > > Please check. > > Thanks for the quick fixes. Tested and works well! I think I found another instance of this issue. To reproduce, open the attached file and do spellcheck. "abcxyz" is misspelled, but if you click on "ignore all" it goes to the next instance of that word. I think the issue is that both instances are in an inset (in this case the footnote inset). If the footnote is dissolved, there is no issue. Scott -------------- next part -------------- A non-text attachment was scrubbed... Name: ignore-all-example.23.lyx Type: application/x-lyx Size: 1955 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 488 bytes Desc: not available URL: From nospam at lisse.NA Thu Nov 25 12:24:32 2021 From: nospam at lisse.NA (Dr Eberhard Lisse) Date: Thu, 25 Nov 2021 14:24:32 +0200 Subject: 2.3.6.2 compiled as Universal or M1 App? Message-ID: Hi, I got me my new MacBook Air M1 BtO last night and LyX 2.3.6.2 works well, as I had run softwareupdate --install-rosetta already, for another package. I now wonder if someone has a 2.6.3.2 production-ready Universal or M1 build that (s)he could part with so I can see how it work? Please contact me off list (as below) if it is not of general interest. I see however that is is an Intel app and I would like to have either an Universal or Silicone build to play with. greetings, el -- To email me replace 'nospam' with 'el' From spitz at lyx.org Thu Nov 25 16:06:43 2021 From: spitz at lyx.org (=?ISO-8859-1?Q?J=FCrgen_Spitzm=FCller?=) Date: Thu, 25 Nov 2021 17:06:43 +0100 Subject: [LyX/master] Meet per-document spelling dictionaries (fixes #86 [sic!]) In-Reply-To: <20211124023517.7vxctqz34qaceakf@poole> References: <20210306155212.803DC280262@lyx.lyx.org> <03c2466a-00dc-5782-52b6-a8cfe2e8918c@lyx.org> <20210306172549.5p3t4fbqj2ivckom@tallinn> <20210819200418.47dbppay6sknd7ni@poole> <20210821134021.73i3u2repkcsqqr7@poole> <20211124023517.7vxctqz34qaceakf@poole> Message-ID: Am Dienstag, dem 23.11.2021 um 21:35 -0500 schrieb Scott Kostyshak: > I think I found another instance of this issue. To reproduce, open > the attached file and do spellcheck. "abcxyz" is misspelled, but if > you click on "ignore all" it goes to the next instance of that word. > > I think the issue is that both instances are in an inset (in this > case the footnote inset). If the footnote is dissolved, there is no > issue. Right. Should be fixed as of 5a6498c3d3bfe0c. J?rgen > > Scott -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 488 bytes Desc: This is a digitally signed message part URL: From skostysh at lyx.org Thu Nov 25 18:55:46 2021 From: skostysh at lyx.org (Scott Kostyshak) Date: Thu, 25 Nov 2021 13:55:46 -0500 Subject: [LyX/master] Meet per-document spelling dictionaries (fixes #86 [sic!]) In-Reply-To: References: <20210306155212.803DC280262@lyx.lyx.org> <03c2466a-00dc-5782-52b6-a8cfe2e8918c@lyx.org> <20210306172549.5p3t4fbqj2ivckom@tallinn> <20210819200418.47dbppay6sknd7ni@poole> <20210821134021.73i3u2repkcsqqr7@poole> <20211124023517.7vxctqz34qaceakf@poole> Message-ID: <20211125185546.uku3e6mp37habprh@poole> On Thu, Nov 25, 2021 at 05:06:43PM +0100, J?rgen Spitzm?ller wrote: > Am Dienstag, dem 23.11.2021 um 21:35 -0500 schrieb Scott Kostyshak: > > I think I found another instance of this issue. To reproduce, open > > the attached file and do spellcheck. "abcxyz" is misspelled, but if > > you click on "ignore all" it goes to the next instance of that word. > > > > I think the issue is that both instances are in an inset (in this > > case the footnote inset). If the footnote is dissolved, there is no > > issue. > > Right. Should be fixed as of 5a6498c3d3bfe0c. Works well, thank you! Scott -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 488 bytes Desc: not available URL: From xracoonx at gmx.de Fri Nov 26 15:03:59 2021 From: xracoonx at gmx.de (Daniel) Date: Fri, 26 Nov 2021 16:03:59 +0100 Subject: Lists don't toggle and show Default icon Message-ID: For me in current master, list toolbar buttons stopped toggling and show only the Default layout icon. Daniel From lasgouttes at lyx.org Fri Nov 26 16:16:31 2021 From: lasgouttes at lyx.org (Jean-Marc Lasgouttes) Date: Fri, 26 Nov 2021 17:16:31 +0100 Subject: Lists don't toggle and show Default icon In-Reply-To: References: Message-ID: <2880dd7d-1655-29b1-5835-cfc7c75ff57a@lyx.org> Le 26/11/2021 ? 16:03, Daniel a ?crit?: > For me in current master, list toolbar buttons stopped toggling and show > only the Default layout icon. Do you have your own stdtoolbar.inc file? When toggling was implemented, it was decided that transparently upgrading the toolbars was not worth it. The new code looks like: BidiItem "Numbered list" "layout-toggle Enumerate" BidiItem "Itemized list" "layout-toggle Itemize" BidiItem "Labeled List" "layout-toggle Labeling" BidiItem "Description" "layout-toggle Description" JMarc From st.witt at gmx.net Fri Nov 26 20:09:14 2021 From: st.witt at gmx.net (Stephan Witt) Date: Fri, 26 Nov 2021 21:09:14 +0100 Subject: [LyX/master] Fix a number of issues that were stopping compilation with MSVC 19. In-Reply-To: <20200504232437.386EB280218@lyx.lyx.org> References: <20200504232437.386EB280218@lyx.lyx.org> Message-ID: Am 05.05.2020 um 01:24 schrieb Richard Kimberly Heck : > > commit c506f304bc522ea91ad5a7e97cd4e3c7d54376b0 > Author: Richard Kimberly Heck > Date: Mon May 4 19:41:18 2020 -0400 > > Fix a number of issues that were stopping compilation with MSVC 19. > > Patch from Thibaut Cuvelier, modified slightly by me (mostly for style). > --- > lib/generate_contributions.py | 8 +++++ > src/BiblioInfo.cpp | 12 ++----- > src/BranchList.cpp | 7 +--- > src/Format.cpp | 47 +++++++++++++++++---------- > src/Format.h | 2 + > src/Lexer.cpp | 28 +++++++--------- > src/TextClass.cpp | 64 +++++++++++++++++++----------------- > src/TextClass.h | 6 +++- > src/frontends/qt/GuiDocument.cpp | 1 - > src/frontends/tests/biblio.cpp | 2 +- > src/insets/InsetCommandParams.cpp | 2 +- > src/lyxfind.cpp | 2 +- > 12 files changed, 99 insertions(+), 82 deletions(-) Hi Riki, one part of this patch is required to compile 2.3.x on latest Mac. I?ve attached the change to InsetCommandParams::getFirstNonOptParam. I don?t know if it is safe. But it would be nice to be able to build LyX 2.3.x for Apple M1 architecture too. Would it be an option to backport this part of the change? If there is another solution I?d like to test it. BR, Stephan -------------- next part -------------- A non-text attachment was scrubbed... Name: InsetCommandParams.patch Type: application/octet-stream Size: 563 bytes Desc: not available URL: From xracoonx at gmx.de Sat Nov 27 06:28:43 2021 From: xracoonx at gmx.de (Daniel) Date: Sat, 27 Nov 2021 07:28:43 +0100 Subject: Lists don't toggle and show Default icon In-Reply-To: <2880dd7d-1655-29b1-5835-cfc7c75ff57a@lyx.org> References: <2880dd7d-1655-29b1-5835-cfc7c75ff57a@lyx.org> Message-ID: On 2021-11-26 17:16, Jean-Marc Lasgouttes wrote: > Le 26/11/2021 ? 16:03, Daniel a ?crit?: >> For me in current master, list toolbar buttons stopped toggling and >> show only the Default layout icon. > > Do you have your own stdtoolbar.inc file? When toggling was implemented, > it was decided that transparently upgrading the toolbars was not worth it. > > The new code looks like: > ????BidiItem "Numbered list" "layout-toggle Enumerate" > ????BidiItem "Itemized list" "layout-toggle Itemize" > ????BidiItem "Labeled List" "layout-toggle Labeling" > ????BidiItem "Description" "layout-toggle Description" > > JMarc Strange, I thought I checked for custom toolbar.inc. But indeed there is one. Sorry for the noise. I am not fully sure what "transparently upgrading" is but are you sure that "breaking" (the icons) with old toolbar.inc files is intended? Is it to let the user know that they need to change something? Daniel From st.witt at gmx.net Sat Nov 27 08:47:36 2021 From: st.witt at gmx.net (Stephan Witt) Date: Sat, 27 Nov 2021 09:47:36 +0100 Subject: Failure to make dist on Mac Message-ID: Hi all, I?ve tried to make a dist package to test it on one other machine. 1. Step - configure 2. Step - make dist $ (cd lyx-build/LyX-2.3.6dev.build;make dist) /Applications/Xcode.app/Contents/Developer/usr/bin/make dist-bzip2 dist-gzip am__post_remove_distdir='@:' /Applications/Xcode.app/Contents/Developer/usr/bin/make distdir-am if test -d "LyX-2.3-2.3.6dev"; then find "LyX-2.3-2.3.6dev" -type d ! -perm -200 -exec chmod u+w {} ';' && rm -rf "LyX-2.3-2.3.6dev" || { sleep 5 && rm -rf "LyX-2.3-2.3.6dev"; }; else :; fi test -d "LyX-2.3-2.3.6dev" || mkdir "LyX-2.3-2.3.6dev" (cd 3rdparty && /Applications/Xcode.app/Contents/Developer/usr/bin/make top_distdir=../LyX-2.3-2.3.6dev distdir=../LyX-2.3-2.3.6dev/3rdparty \ am__remove_distdir=: am__skip_length_check=: am__skip_mode_fix=: distdir) /Applications/Xcode.app/Contents/Developer/usr/bin/make distdir-am (cd boost && /Applications/Xcode.app/Contents/Developer/usr/bin/make top_distdir=../../LyX-2.3-2.3.6dev distdir=../../LyX-2.3-2.3.6dev/3rdparty/boost \ am__remove_distdir=: am__skip_length_check=: am__skip_mode_fix=: distdir) /Applications/Xcode.app/Contents/Developer/usr/bin/make distdir-am (cd hunspell && /Applications/Xcode.app/Contents/Developer/usr/bin/make top_distdir=../../LyX-2.3-2.3.6dev distdir=../../LyX-2.3-2.3.6dev/3rdparty/hunspell \ am__remove_distdir=: am__skip_length_check=: am__skip_mode_fix=: distdir) /Applications/Xcode.app/Contents/Developer/usr/bin/make distdir-am make[6]: *** No rule to make target `1.6.2/src/hunspell/affentry.cxx', needed by `distdir-am'. Stop. What am I doing wrong? Is hunspell 1.6.2 required to put somewhere? Is it required to configure LyX with the matching hunspell included? BR, Stephan From sanda at lyx.org Sat Nov 27 17:36:21 2021 From: sanda at lyx.org (Pavel Sanda) Date: Sat, 27 Nov 2021 18:36:21 +0100 Subject: Failure to make dist on Mac In-Reply-To: References: Message-ID: <20211127173621.GA6845@jabberwock.ucw.cz> On Sat, Nov 27, 2021 at 09:47:36AM +0100, Stephan Witt wrote: > 2. Step - make dist > > $ (cd lyx-build/LyX-2.3.6dev.build;make dist) > /Applications/Xcode.app/Contents/Developer/usr/bin/make dist-bzip2 dist-gzip am__post_remove_distdir='@:' > /Applications/Xcode.app/Contents/Developer/usr/bin/make distdir-am > if test -d "LyX-2.3-2.3.6dev"; then find "LyX-2.3-2.3.6dev" -type d ! -perm -200 -exec chmod u+w {} ';' && rm -rf "LyX-2.3-2.3.6dev" || { sleep 5 && rm -rf "LyX-2.3-2.3.6dev"; }; else :; fi > test -d "LyX-2.3-2.3.6dev" || mkdir "LyX-2.3-2.3.6dev" > (cd 3rdparty && /Applications/Xcode.app/Contents/Developer/usr/bin/make top_distdir=../LyX-2.3-2.3.6dev distdir=../LyX-2.3-2.3.6dev/3rdparty \ > am__remove_distdir=: am__skip_length_check=: am__skip_mode_fix=: distdir) > /Applications/Xcode.app/Contents/Developer/usr/bin/make distdir-am > (cd boost && /Applications/Xcode.app/Contents/Developer/usr/bin/make top_distdir=../../LyX-2.3-2.3.6dev distdir=../../LyX-2.3-2.3.6dev/3rdparty/boost \ > am__remove_distdir=: am__skip_length_check=: am__skip_mode_fix=: distdir) > /Applications/Xcode.app/Contents/Developer/usr/bin/make distdir-am > (cd hunspell && /Applications/Xcode.app/Contents/Developer/usr/bin/make top_distdir=../../LyX-2.3-2.3.6dev distdir=../../LyX-2.3-2.3.6dev/3rdparty/hunspell \ > am__remove_distdir=: am__skip_length_check=: am__skip_mode_fix=: distdir) > /Applications/Xcode.app/Contents/Developer/usr/bin/make distdir-am > make[6]: *** No rule to make target `1.6.2/src/hunspell/affentry.cxx', needed by `distdir-am'. Stop. > > What am I doing wrong? Is hunspell 1.6.2 required to put somewhere? Is it required to configure LyX with the matching hunspell included? Seems to work on the current branch (2.3.7dev). Pavel From st.witt at gmx.net Sun Nov 28 09:31:23 2021 From: st.witt at gmx.net (Stephan Witt) Date: Sun, 28 Nov 2021 10:31:23 +0100 Subject: Failure to make dist on Mac In-Reply-To: <20211127173621.GA6845@jabberwock.ucw.cz> References: <20211127173621.GA6845@jabberwock.ucw.cz> Message-ID: Am 27.11.2021 um 18:36 schrieb Pavel Sanda : > > On Sat, Nov 27, 2021 at 09:47:36AM +0100, Stephan Witt wrote: >> 2. Step - make dist >> >> $ (cd lyx-build/LyX-2.3.6dev.build;make dist) >> /Applications/Xcode.app/Contents/Developer/usr/bin/make dist-bzip2 dist-gzip am__post_remove_distdir='@:' >> /Applications/Xcode.app/Contents/Developer/usr/bin/make distdir-am >> if test -d "LyX-2.3-2.3.6dev"; then find "LyX-2.3-2.3.6dev" -type d ! -perm -200 -exec chmod u+w {} ';' && rm -rf "LyX-2.3-2.3.6dev" || { sleep 5 && rm -rf "LyX-2.3-2.3.6dev"; }; else :; fi >> test -d "LyX-2.3-2.3.6dev" || mkdir "LyX-2.3-2.3.6dev" >> (cd 3rdparty && /Applications/Xcode.app/Contents/Developer/usr/bin/make top_distdir=../LyX-2.3-2.3.6dev distdir=../LyX-2.3-2.3.6dev/3rdparty \ >> am__remove_distdir=: am__skip_length_check=: am__skip_mode_fix=: distdir) >> /Applications/Xcode.app/Contents/Developer/usr/bin/make distdir-am >> (cd boost && /Applications/Xcode.app/Contents/Developer/usr/bin/make top_distdir=../../LyX-2.3-2.3.6dev distdir=../../LyX-2.3-2.3.6dev/3rdparty/boost \ >> am__remove_distdir=: am__skip_length_check=: am__skip_mode_fix=: distdir) >> /Applications/Xcode.app/Contents/Developer/usr/bin/make distdir-am >> (cd hunspell && /Applications/Xcode.app/Contents/Developer/usr/bin/make top_distdir=../../LyX-2.3-2.3.6dev distdir=../../LyX-2.3-2.3.6dev/3rdparty/hunspell \ >> am__remove_distdir=: am__skip_length_check=: am__skip_mode_fix=: distdir) >> /Applications/Xcode.app/Contents/Developer/usr/bin/make distdir-am >> make[6]: *** No rule to make target `1.6.2/src/hunspell/affentry.cxx', needed by `distdir-am'. Stop. >> >> What am I doing wrong? Is hunspell 1.6.2 required to put somewhere? Is it required to configure LyX with the matching hunspell included? > > Seems to work on the current branch (2.3.7dev). Yes, it works. I used a mix of some old build directory and current branch checkout. I missed the wrong version number in path name. Thank you for checking. Stephan From st.witt at gmx.net Sun Nov 28 13:14:00 2021 From: st.witt at gmx.net (Stephan Witt) Date: Sun, 28 Nov 2021 14:14:00 +0100 Subject: [LyX/master] Fix a number of issues that were stopping compilation with MSVC 19. In-Reply-To: References: <20200504232437.386EB280218@lyx.lyx.org> Message-ID: Am 26.11.2021 um 21:09 schrieb Stephan Witt : > > Am 05.05.2020 um 01:24 schrieb Richard Kimberly Heck : >> >> commit c506f304bc522ea91ad5a7e97cd4e3c7d54376b0 >> Author: Richard Kimberly Heck >> Date: Mon May 4 19:41:18 2020 -0400 >> >> Fix a number of issues that were stopping compilation with MSVC 19. >> >> Patch from Thibaut Cuvelier, modified slightly by me (mostly for style). >> --- >> lib/generate_contributions.py | 8 +++++ >> src/BiblioInfo.cpp | 12 ++----- >> src/BranchList.cpp | 7 +--- >> src/Format.cpp | 47 +++++++++++++++++---------- >> src/Format.h | 2 + >> src/Lexer.cpp | 28 +++++++--------- >> src/TextClass.cpp | 64 +++++++++++++++++++----------------- >> src/TextClass.h | 6 +++- >> src/frontends/qt/GuiDocument.cpp | 1 - >> src/frontends/tests/biblio.cpp | 2 +- >> src/insets/InsetCommandParams.cpp | 2 +- >> src/lyxfind.cpp | 2 +- >> 12 files changed, 99 insertions(+), 82 deletions(-) > > Hi Riki, > > one part of this patch is required to compile 2.3.x on latest Mac. > I?ve attached the change to InsetCommandParams::getFirstNonOptParam. > I don?t know if it is safe. But it would be nice to be able to build > LyX 2.3.x for Apple M1 architecture too. > > Would it be an option to backport this part of the change? > > If there is another solution I?d like to test it. > > BR, Stephan > > Hi all, I?d like to add some details and a solution to the issue. With Apple clang and -std=c++14 I?m getting a warning but it compiles: CXX insets/InsetCommandParams.o lyx-2.3.x/src/insets/InsetCommandParams.cpp:596:9: warning: 'mem_fun_ref' is deprecated [-Wdeprecated-declarations] not1(mem_fun_ref(&ParamInfo::ParamData::isOptional))); With Apple clang and -std=c++17 I?m getting an error: CXX insets/InsetCommandParams.o lyx-2.3.x/src/insets/InsetCommandParams.cpp:596:6: error: use of undeclared identifier 'mem_fun_ref' not1(mem_fun_ref(&ParamInfo::ParamData::isOptional))); So the solution is simple. One cannot compile LyX-2.3.x with the C++17 - so one should?t try it. :) Now I can build LyX 2.3.x for Apple M1 architecture with Apple clang 12.0.0 too. BR, Stephan