- dev-luatex - NTG Mailing Lists

Make the late_lua node accessible from the executed code
by Marcel Fabian Krüger 28 Feb '22

28 Feb '22

Hi, one issue I regularly encounter with late_lua whatits is that they don't allow the executed code to determine which node is currently executed. Especially this means that attributes and properties can't be reliably accessed and the code can't access related nodes next to the late_lua node. This can be worked around by storing a closure in the data field which captures a reference to the node, but this breaks in hard to debug ways when the node is copied and the copied node has the same code but a different identify. Therefore it would be great if there would be a way to access the currently evaluated late_lua whatsit node. I would have suggested passing this as an argument to the executed function, but that could break compatbility and it would appear odd due to LuaMetaTeX using the second argument of lua functions is different ways. So the attached suggested implemention instead adds a separate function which allows to access this node: node.current_latelua (or node.direct.current_latelua) Best, Marcel

2 2

Weird behavior of \exceptionpenalty
by Marcel Fabian Krüger 25 Feb '22

25 Feb '22

Hi, in an discussion on TeX StackExchange \exceptionpenalty came up and it's current behavior seems a bit odd and is at odds with the documentation: According to the manual: > Because we introduced penalties in discretionary nodes, the exception syntax now > also can take a penalty specification. The value between square brackets is a multiplier for > \exceptionpenalty. Here we have set it to 10000 so effectively we get > 30000 in the example. and then shows the example \hyphenation{z{a-}{-b}{z}{a-}{-b}{z}[3]{a-}{-b}{z}[1]{a-}{-b}{z}z} As the text says, it could be expected that this causes a penalty of 30000. But the currently implemented behavior is different: It assigns a penalty of 10000 to the second discretionary. The reason is that the number in square brackets is only used as a multiplier if exception_penalty is higher than 100000 (sic). For any value of \exceptionpenalty lower or equal to 100000, the number is ignored and the value of \exceptionpenalty is used as is. I'm not sure if this is just a bug (maybe 100000 was supposed to be a upper limit and not a lower limit?) or if this is intended to be used as a marker which later gets replaced by some callback and therefore 100000 was intentionally used as a unrealistic penalty value, but I think that at least the documentation should get fixed. I'm happy to provide a patch for this if you tell me what to implement. I basically see the following options: * Just change the manual to describe the implemented behavior. I think that this would be sad since it makes the feature much less useful since having multiples of badness higher than 100000 seems to have few applications, but this could be picked up by callbacks ansd would be the most compatible option. * Decide that The documentation is more important than the implemented behavior and fix the code to implement the documentation, removing the 100000 criterium (or making it an upper limit instead of a lower one) If 100000 becomes an upper limit, it would also be possible to substract 100000 from exception_penalty before using it to allow emulating the old behavior of using a fixed exception_penalty. (Of course, the opposite: Keep the current code but reduce \exceptionpenalty by 100000 first in the multiplication case could also be done, but IMO that would be a weird interface.) * Add a new flag to select between the new and old behavior or add a new bit to \hyphenationmode. Best, Marcel

2 1

Is \toksapp assignment?
by Petr Olsak 07 Feb '22

07 Feb '22

Hello, If I try: \def\test{test} \afterassignment\test \advance\count0 by2 then \test is processed immediately after the \advance command. But if I do: \afterassignment\test \toksapp 0{xx} \the\toks0 then \test waits to another assignment than \toksapp. It means, that \toksapp <toks> (and friends \gtoksapp etc.) is not assignment. Is it an intention or a bug? Petr Olšák

4 6

What is the current lfs version?
by Jérôme LAURENS 01 Feb '22

01 Feb '22

We can obtain the current lfs version by running luatex '\directlua{tex.write(lfs._VERSION)}\bye' && open texput.pdf Then we must remove texput.pdf and texput.log. This is a bit cumbersome. It would be great if this information could be in the credits: luatex --credits luajittex --credits BTW, is lfs 1.8 planned in forthcoming? (1.7 does not play with windows very well)

3 2

Re: [Dev-luatex] bug#48064: texlive-* packages fail to build non-deterministically
by Ludovic Courtès 31 Dec '21

31 Dec '21

Hello, While investigating luatex crashes in the TeX Live 2020 package of GNU Guix¹, we identified the following heap corruption reported by Valgrind (this is on GNU/Linux, with glibc 2.33): --8<---------------cut here---------------start------------->8--- sh-5.0$ ~ludo/.guix-profile/bin/valgrind --extra-debuginfo-path=/gnu/store/f933bvd6ab6l2lg6xl6k1a6jwvcls0z6-glibc-2.33-debug/lib/debug "luatex" "-interaction=nonstopmode" "-output-directory=build" "&luatex" "amsbsy.dtx" ==28531== Memcheck, a memory error detector ==28531== Copyright (C) 2002-2017, and GNU GPL'd, by Julian Seward et al. ==28531== Using Valgrind-3.16.1 and LibVEX; rerun with -h for copyright info ==28531== Command: luatex -interaction=nonstopmode -output-directory=build &luatex amsbsy.dtx ==28531== This is LuaTeX, Version 1.12.0 (TeX Live 2020) restricted system commands enabled. ==28531== Invalid write of size 8 ==28531== at 0x485C691: lua_pushlstring (in /gnu/store/w20xxg8p0wksbrxxvj3y46fngvr3954w-texlive-bin-20200406/lib/libtexlua53.so.5.3.5) ==28531== by 0x568E03: load_hyphenation (in /gnu/store/w20xxg8p0wksbrxxvj3y46fngvr3954w-texlive-bin-20200406/bin/luatex) ==28531== by 0x56B41C: undump_language_data (in /gnu/store/w20xxg8p0wksbrxxvj3y46fngvr3954w-texlive-bin-20200406/bin/luatex) ==28531== by 0x4DFB9F: load_fmt_file (in /gnu/store/w20xxg8p0wksbrxxvj3y46fngvr3954w-texlive-bin-20200406/bin/luatex) ==28531== by 0x4EF0ED: main_body (in /gnu/store/w20xxg8p0wksbrxxvj3y46fngvr3954w-texlive-bin-20200406/bin/luatex) ==28531== by 0x45118D: main (in /gnu/store/w20xxg8p0wksbrxxvj3y46fngvr3954w-texlive-bin-20200406/bin/luatex) ==28531== Address 0xae14170 is 0 bytes after a block of size 1,168 alloc'd ==28531== at 0x483EBE1: realloc (in /gnu/store/jlmh0jbgr6zn4iyvhfbvxps8pykzj5ry-valgrind-3.16.1/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==28531== by 0x46695D: ??? (in /gnu/store/w20xxg8p0wksbrxxvj3y46fngvr3954w-texlive-bin-20200406/bin/luatex) ==28531== by 0x486D932: ??? (in /gnu/store/w20xxg8p0wksbrxxvj3y46fngvr3954w-texlive-bin-20200406/lib/libtexlua53.so.5.3.5) ==28531== by 0x48660F2: ??? (in /gnu/store/w20xxg8p0wksbrxxvj3y46fngvr3954w-texlive-bin-20200406/lib/libtexlua53.so.5.3.5) ==28531== by 0x4868BE7: ??? (in /gnu/store/w20xxg8p0wksbrxxvj3y46fngvr3954w-texlive-bin-20200406/lib/libtexlua53.so.5.3.5) ==28531== by 0x4868FCF: ??? (in /gnu/store/w20xxg8p0wksbrxxvj3y46fngvr3954w-texlive-bin-20200406/lib/libtexlua53.so.5.3.5) ==28531== by 0x486988B: ??? (in /gnu/store/w20xxg8p0wksbrxxvj3y46fngvr3954w-texlive-bin-20200406/lib/libtexlua53.so.5.3.5) ==28531== by 0x485C6BB: lua_pushlstring (in /gnu/store/w20xxg8p0wksbrxxvj3y46fngvr3954w-texlive-bin-20200406/lib/libtexlua53.so.5.3.5) ==28531== by 0x568E03: load_hyphenation (in /gnu/store/w20xxg8p0wksbrxxvj3y46fngvr3954w-texlive-bin-20200406/bin/luatex) ==28531== by 0x56B41C: undump_language_data (in /gnu/store/w20xxg8p0wksbrxxvj3y46fngvr3954w-texlive-bin-20200406/bin/luatex) ==28531== by 0x4DFB9F: load_fmt_file (in /gnu/store/w20xxg8p0wksbrxxvj3y46fngvr3954w-texlive-bin-20200406/bin/luatex) ==28531== by 0x4EF0ED: main_body (in /gnu/store/w20xxg8p0wksbrxxvj3y46fngvr3954w-texlive-bin-20200406/bin/luatex) ==28531== ==28531== Invalid write of size 4 ==28531== at 0x485C6A2: lua_pushlstring (in /gnu/store/w20xxg8p0wksbrxxvj3y46fngvr3954w-texlive-bin-20200406/lib/libtexlua53.so.5.3.5) ==28531== by 0x568E03: load_hyphenation (in /gnu/store/w20xxg8p0wksbrxxvj3y46fngvr3954w-texlive-bin-20200406/bin/luatex) ==28531== by 0x56B41C: undump_language_data (in /gnu/store/w20xxg8p0wksbrxxvj3y46fngvr3954w-texlive-bin-20200406/bin/luatex) ==28531== by 0x4DFB9F: load_fmt_file (in /gnu/store/w20xxg8p0wksbrxxvj3y46fngvr3954w-texlive-bin-20200406/bin/luatex) ==28531== by 0x4EF0ED: main_body (in /gnu/store/w20xxg8p0wksbrxxvj3y46fngvr3954w-texlive-bin-20200406/bin/luatex) ==28531== by 0x45118D: main (in /gnu/store/w20xxg8p0wksbrxxvj3y46fngvr3954w-texlive-bin-20200406/bin/luatex) ==28531== Address 0xae14178 is 8 bytes after a block of size 1,168 alloc'd ==28531== at 0x483EBE1: realloc (in /gnu/store/jlmh0jbgr6zn4iyvhfbvxps8pykzj5ry-valgrind-3.16.1/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==28531== by 0x46695D: ??? (in /gnu/store/w20xxg8p0wksbrxxvj3y46fngvr3954w-texlive-bin-20200406/bin/luatex) ==28531== by 0x486D932: ??? (in /gnu/store/w20xxg8p0wksbrxxvj3y46fngvr3954w-texlive-bin-20200406/lib/libtexlua53.so.5.3.5) ==28531== by 0x48660F2: ??? (in /gnu/store/w20xxg8p0wksbrxxvj3y46fngvr3954w-texlive-bin-20200406/lib/libtexlua53.so.5.3.5) ==28531== by 0x4868BE7: ??? (in /gnu/store/w20xxg8p0wksbrxxvj3y46fngvr3954w-texlive-bin-20200406/lib/libtexlua53.so.5.3.5) ==28531== by 0x4868FCF: ??? (in /gnu/store/w20xxg8p0wksbrxxvj3y46fngvr3954w-texlive-bin-20200406/lib/libtexlua53.so.5.3.5) ==28531== by 0x486988B: ??? (in /gnu/store/w20xxg8p0wksbrxxvj3y46fngvr3954w-texlive-bin-20200406/lib/libtexlua53.so.5.3.5) ==28531== by 0x485C6BB: lua_pushlstring (in /gnu/store/w20xxg8p0wksbrxxvj3y46fngvr3954w-texlive-bin-20200406/lib/libtexlua53.so.5.3.5) ==28531== by 0x568E03: load_hyphenation (in /gnu/store/w20xxg8p0wksbrxxvj3y46fngvr3954w-texlive-bin-20200406/bin/luatex) ==28531== by 0x56B41C: undump_language_data (in /gnu/store/w20xxg8p0wksbrxxvj3y46fngvr3954w-texlive-bin-20200406/bin/luatex) ==28531== by 0x4DFB9F: load_fmt_file (in /gnu/store/w20xxg8p0wksbrxxvj3y46fngvr3954w-texlive-bin-20200406/bin/luatex) ==28531== by 0x4EF0ED: main_body (in /gnu/store/w20xxg8p0wksbrxxvj3y46fngvr3954w-texlive-bin-20200406/bin/luatex) ==28531== [...] valgrind: m_mallocfree.c:305 (get_bszB_as_is): Assertion 'bszB_lo == bszB_hi' failed. valgrind: Heap block lo/hi size mismatch: lo = 1232, hi = 68. This is probably caused by your program erroneously writing past the end of a heap block and corrupting heap metadata. If you fix any invalid writes reported by Memcheck, this assertion failure will probably go away. Please try that before reporting this as a bug. --8<---------------cut here---------------end--------------->8--- Does that ring a bell? Is there a chance this problem was fixed in the meantime? Thanks in advance, Ludovic. ¹ https://issues.guix.gnu.org/48064

3 4

Erroneous error and help messages
by Hironobu Yamashita 28 Dec '21

28 Dec '21

Hi, The following MWE shows two LuaTeX-specific issues: * It does not print the primitive name "\write" * Tiny typo: double apostrophes in the word "didn''t" ===== \nonstopmode \outer\def\F{hoge} \immediate\write-1{\F}% => looks wrong % ! Forbidden control sequence found while scanning text of \ . \message{\F}% => OK % ! Forbidden control sequence found while scanning text of \message. \par\spacefactor\relax% The log file contains % "I'll just pretend that you didn''t ask for it." \bye ===== Best, Hironobu Yamashita

2 3

os.uname on 64 bit Windows
by Reinhard Kotucha 22 Dec '21

22 Dec '21

Hi, on a 64 bit Windows machine os.uname() erroneously returns machine i686 On the same machine I get under Linux machine x86_64 with both, x86_64-linux and i386-linux binaries. Thus I assume that the keyword 'machine' refers to the hardware, not to how the binaries were compiled (32 vs. 64 bit). The complete output of for k,v in pairs(os.uname()) do print(k,v) end is release build 9200 nodename R804 version 6.02 machine i686 sysname Windows 8 Regards, Reinhard -- ------------------------------------------------------------------ Reinhard Kotucha Phone: +49-511-3373112 Marschnerstr. 25 D-30167 Hannover mailto:reinhard.kotucha@web.de ------------------------------------------------------------------

3 8

luatex svn git sync
by Norbert Preining 21 Dec '21

21 Dec '21

Hi all, I am tending to disable the svn-git mirroring I am doing of luatex from svn to github, since now since considerable time the svn server seems to be broken: > Can't create session: Unable to connect to a repository at URL 'https://serveur-svn.lri.fr/svn/modhel/luatex': Unexpected server error 500 'Proxy Error' on > '/svn/modhel/luatex' at /usr/share/perl5/Git/SVN.pm line 148. Any comments? And btw, is it really still necessary to use this rather unconventional development space ...? All the best Norbert -- PREINING Norbert https://www.preining.info Fujitsu Research + IFMGA Guide + TU Wien + TeX Live GPG: 0x860CDC13 fp: F7D8 A928 26E3 16A1 9FA0 ACF0 6CAC A448 860C DC13

6 11

Wrong placement using \displaylimits and unicode-math
by Jakub Kaczor 08 Dec '21

08 Dec '21

Hello there, there were some reports regarding the placement of the limits in integrals a few years ago <https://tug.org/pipermail/luatex/2015-March/005076.html>. It was with regard to display math and `\limits`. They seem to work fine for me, but I spotted some weird behavior using `\displaylimits` *in in-line math*. In my opinion, the limits of integration are placed way too much on the right. The expected behavior is to have the same placement as with `\nolimits` in in-line math, and as with `\limits` in display math. The problem doesn't occur with XeLaTeX and it seems to be independent of a font. Here is a minimal working example comparing different options: ``` latex %! TeX program = lualatex \documentclass{article} \usepackage{unicode-math} \begin{document} $\int\limits_a^b f(x)\,\mathrm dx$ $\int\nolimits_a^b f(x)\,\mathrm dx$ $\int\displaylimits_a^b f(x)\,\mathrm dx$ \[\int\limits_a^b f(x)\,\mathrm dx\] \[\int\nolimits_a^b f(x)\,\mathrm dx\] \[\int\displaylimits_a^b f(x)\,\mathrm dx\] \end{document} ```` Here, the third line after `\begin{document}` is the faulty one. Best regards Jakub Kaczor

1 0

luahbtex Mac M1 memory
by Oliver Marugg 08 Dec '21

08 Dec '21

Hi Yesterday I experienced some strange memory issues on mac mini running maco 12.0.1 (monterey) with luabhbtex 1.13.2 (installed via texlive 2021 w. latest updates via tlmgr) and texstudio 4.1.1 as frontend. I read about memory leaks with m1 CPUS. I found out that the process luahbtex run with 33GB virtual memory permanently (without having a job to run) and used 99-100% CPU load, CPU was running hot. Instead of investigating, I killed the process and memory as well CPU load went back to normal. What could I do for you next time I experience such an effect? Many thanks. Ollie Marugg

1 0