Re: Assertion with BATCH_outline-beamer test
Am Thu, 26 Mar 2020 19:05:36 -0400 schrieb Scott Kostyshak : > On Thu, Mar 26, 2020 at 08:08:50PM +0100, Jean-Marc Lasgouttes wrote: > > Le 26/03/2020 à 19:35, Scott Kostyshak a écrit : > > > Note that I made a recent change in a setCursor() call in outline-down > > > so that is a good candidate. > > > > Indeed :) > > Attached is an example file. To reproduce, put the cursor in the first > position of the buffer and press + . For some reason, > outline() was changing the cursor position to 1. I'm not sure if that's > intended behavior, but in any case the assertion is fixed at a higher > level at a1169188. Yes, that's fine. > On a separate issue, there's what I think is undesired behavior in that > outline() should put the frame after the next frame. Instead, it breaks > up the second frame because it wants to put the first before > "References". I'm not sure what is ideal here though, because if you > look at the outliner, you see "frame, frame, references" so you might > expect the frame to go before references after the move. I guess there > needs to be a concept of references being nested in frame and that being > shown in the outliner. > > Scott Not nice, but OTOH easy enough to handle :) Thanks Scott. Kornel pgpQgawLaiAq5.pgp Description: Digitale Signatur von OpenPGP -- lyx-devel mailing list lyx-devel@lists.lyx.org http://lists.lyx.org/mailman/listinfo/lyx-devel
Re: Assertion with BATCH_outline-beamer test
On Thu, Mar 26, 2020 at 08:08:50PM +0100, Jean-Marc Lasgouttes wrote: > Le 26/03/2020 à 19:35, Scott Kostyshak a écrit : > > Note that I made a recent change in a setCursor() call in outline-down > > so that is a good candidate. > > Indeed :) Attached is an example file. To reproduce, put the cursor in the first position of the buffer and press + . For some reason, outline() was changing the cursor position to 1. I'm not sure if that's intended behavior, but in any case the assertion is fixed at a higher level at a1169188. On a separate issue, there's what I think is undesired behavior in that outline() should put the frame after the next frame. Instead, it breaks up the second frame because it wants to put the first before "References". I'm not sure what is ideal here though, because if you look at the outliner, you see "frame, frame, references" so you might expect the frame to go before references after the move. I guess there needs to be a concept of references being nested in frame and that being shown in the outliner. Scott example.23.lyx Description: application/lyx signature.asc Description: PGP signature -- lyx-devel mailing list lyx-devel@lists.lyx.org http://lists.lyx.org/mailman/listinfo/lyx-devel
Re: Assertion with BATCH_outline-beamer test
Le 26/03/2020 à 19:35, Scott Kostyshak a écrit : Note that I made a recent change in a setCursor() call in outline-down so that is a good candidate. Indeed :) JMarc -- lyx-devel mailing list lyx-devel@lists.lyx.org http://lists.lyx.org/mailman/listinfo/lyx-devel
Re: Assertion with BATCH_outline-beamer test
On Thu, Mar 26, 2020 at 06:55:19PM +0100, Jean-Marc Lasgouttes wrote: > Le 26/03/2020 à 18:33, Kornel Benko a écrit : > > > > The command: > > /BUILD/BUILDMint18/BuildLyxGitQt5.9.5local-gcc8.3.0/bin/lyx2.4 '-userdir' > > '/BUILD/BUILDMint18/BuildLyxGitQt5.9.5local-gcc8.3.0/Testing/.lyxbatch' > > '-platform' > > 'offscreen' '-x' 'command-sequence file-open > > beamer_test.lyx;buffer-begin;repeat 150 > > outline-down;repeat 150 outline-up;buffer-export pdflatex;buffer-reload > > dump;lyx-quit' > > > Attached the commands to provoke the assert. > > Looks like Jean-Marc would be happy to fix :) > > If somebody has time, having a more minimalist test would be nice :) I can reproduce. I can try a git bisect. Note that I made a recent change in a setCursor() call in outline-down so that is a good candidate. Scott signature.asc Description: PGP signature -- lyx-devel mailing list lyx-devel@lists.lyx.org http://lists.lyx.org/mailman/listinfo/lyx-devel
Re: Assertion with BATCH_outline-beamer test
Le 26/03/2020 à 18:33, Kornel Benko a écrit : The command: /BUILD/BUILDMint18/BuildLyxGitQt5.9.5local-gcc8.3.0/bin/lyx2.4 '-userdir' '/BUILD/BUILDMint18/BuildLyxGitQt5.9.5local-gcc8.3.0/Testing/.lyxbatch' '-platform' 'offscreen' '-x' 'command-sequence file-open beamer_test.lyx;buffer-begin;repeat 150 outline-down;repeat 150 outline-up;buffer-export pdflatex;buffer-reload dump;lyx-quit' Attached the commands to provoke the assert. Looks like Jean-Marc would be happy to fix :) If somebody has time, having a more minimalist test would be nice :) JMarc -- lyx-devel mailing list lyx-devel@lists.lyx.org http://lists.lyx.org/mailman/listinfo/lyx-devel
Re: Assertion with BATCH_outline-beamer test
Le 26/03/2020 à 18:33, Kornel Benko a écrit : Attached the commands to provoke the assert. Looks like Jean-Marc would be happy to fix :) I am happy to put it on my to-do fifo stack :) JMarc -- lyx-devel mailing list lyx-devel@lists.lyx.org http://lists.lyx.org/mailman/listinfo/lyx-devel
Assertion with BATCH_outline-beamer test
The command: /BUILD/BUILDMint18/BuildLyxGitQt5.9.5local-gcc8.3.0/bin/lyx2.4 '-userdir' '/BUILD/BUILDMint18/BuildLyxGitQt5.9.5local-gcc8.3.0/Testing/.lyxbatch' '-platform' 'offscreen' '-x' 'command-sequence file-open beamer_test.lyx;buffer-begin;repeat 150 outline-down;repeat 150 outline-up;buffer-export pdflatex;buffer-reload dump;lyx-quit' Result: frontends/qt/GuiClipboard.cpp (71): 0 bytes (no QMimeData) CursorSlice.cpp (111): Don't like 1, pos: 1 size: 0 par: 316 support/lassert.cpp (51): ASSERTION false VIOLATED IN /usr2/src/lyx/lyx-git/src/CursorSlice.cpp:112 ( 1) /BUILD/BUILDMint18/BuildLyxGitQt5.9.5local-gcc8.3.0/bin/lyx2.4: lyx::doAssertWithCallstack(bool) ( 2) /BUILD/BUILDMint18/BuildLyxGitQt5.9.5local-gcc8.3.0/bin/lyx2.4: lyx::doAssert(char const*, char const*, long) ( 3) /BUILD/BUILDMint18/BuildLyxGitQt5.9.5local-gcc8.3.0/bin/lyx2.4: lyx::CursorSlice::setPitPos(long, long) ( 4) /BUILD/BUILDMint18/BuildLyxGitQt5.9.5local-gcc8.3.0/bin/lyx2.4: lyx::Text::setCursorIntern(lyx::Cursor&, long, long, bool, bool) ( 5) /BUILD/BUILDMint18/BuildLyxGitQt5.9.5local-gcc8.3.0/bin/lyx2.4: lyx::Text::setCursor(lyx::Cursor&, long, long, bool, bool) ( 6) /BUILD/BUILDMint18/BuildLyxGitQt5.9.5local-gcc8.3.0/bin/lyx2.4: lyx::Text::dispatch(lyx::Cursor&, lyx::FuncRequest&) ... Attached the commands to provoke the assert. Looks like Jean-Marc would be happy to fix :) Kornel batchtest Description: Binary data pgpbuS5hEc0CE.pgp Description: Digitale Signatur von OpenPGP -- lyx-devel mailing list lyx-devel@lists.lyx.org http://lists.lyx.org/mailman/listinfo/lyx-devel
