- dev-luatex - NTG Mailing Lists

ligatures before dashes not working with LuaLaTeX 1.07.0
by Toni Dietze 28 Oct '18

28 Oct '18

Hello, when I switched from version 1.0.4 (TeX Live 2017/Arch Linux) to 1.07.0 (TeX Live 2018/Arch Linux) of LuaLaTeX, I noticed that before dashes ligatures are not typeset anymore. Consider the following example: \documentclass{article} \begin{document} \Huge kickoff meeting kickoff-meeting rofl copter rofl-copter profi font profi-font \end{document} With LuaLaTeX 1.07.0, only in the non-dashed versions the last two letters of the first words are typeset as ligatures. Before a dash, the letters are typeset without a ligature. With pdfLaTeX 3.14159265- 2.6-1.40.19 (TeX Live 2018/Arch Linux), ligatures are also typeset before dashes. Is this behaviour intended? -- Regards Toni Dietze

1 0

values for `mode` field in tex.nest
by Marcel Krüger 28 Oct '18

28 Oct '18

Hi, according to the documentation, the `mode` field in `tex.nest` is explained as > a number representing the main mode at this level: 0 = no > mode (this happens during \write), 1 = vertical, 127 = hor- > izontal, 253 = display math, -1 = internal vertical, -127 = > restricted horizontal, -253 = inline math This looks as if these values were constant across engine versions, but the real values depend on `max_command_cmd` which changes if new commands are added. Especially the current values are `(-)132` for (restricted) horizontal mode and `(-)263` for (inline) math. So I think the documentation should be either updated to delete these outdated values and instead include a note that you have to query these values yourself or LuaTeX could normalize these values before passing them to Lua. One way to do this could be diff --git a/manual/luatex-tex.tex b/manual/luatex-tex.tex index 7edd561e6..f69906d4c 100644 --- a/manual/luatex-tex.tex +++ b/manual/luatex-tex.tex @@ -943,13 +943,13 @@ The known fields are: \DB key \BC type \BC modes \BC explanation \NC \NR \TB \NC \type{mode} \NC number \NC all \NC a number representing the main mode at this level: - \type {0} = no mode (this happens during \prm {write}), - \type {1} = vertical, - \type {127} = horizontal, - \type {253} = display math, - \type {-1} = internal vertical, - \type {-127} = restricted horizontal, - \type {-253} = inline math \NC \NR + \type {0} = no mode (this happens during \prm {write}), + \type {1} = vertical, + \type {2} = horizontal, + \type {3} = display math, + \type {-1} = internal vertical, + \type {-2} = restricted horizontal, + \type {-3} = inline math \NC \NR \NC \type{modeline} \NC number \NC all \NC source input line where this mode was entered in, negative inside the output routine \NC \NR \NC \type{head} \NC node \NC all \NC the head of the current list \NC \NR diff --git a/source/texk/web2c/luatexdir/lua/ltexlib.c b/source/texk/web2c/luatexdir/lua/ltexlib.c index c3e459b22..c71492644 100644 --- a/source/texk/web2c/luatexdir/lua/ltexlib.c +++ b/source/texk/web2c/luatexdir/lua/ltexlib.c @@ -2355,7 +2355,10 @@ static int lua_nest_getfield(lua_State * L) const char *field = lua_tostring(L, -1); r = *rv; if (lua_key_eq(field,mode)) { - lua_pushinteger(L, r->mode_field); + int mode = r->mode_field; + lua_pushinteger(L, + mode == 0 ? 0 : (mode + max_command_cmd * ((mode > 0) * 2 - 1)) / + (max_command_cmd + 1)); } else if (lua_key_eq(field,head)) { lua_nodelib_push_fast(L, r->head_field); } else if (lua_key_eq(field,tail)) { @@ -2396,7 +2399,8 @@ static int lua_nest_setfield(lua_State * L) r = *rv; if (lua_key_eq(field,mode)) { i = lua_tointeger(L, -1); - r->mode_field = i; + r->mode_field = i == 0 ? 0 : + (max_command_cmd + 1) * i + max_command_cmd * ((i < 0) * 2 - 1); } else if (lua_key_eq(field,head)) { n = check_isnode(L, -1); r->head_field = *n; Best regards Marcel Krüger

1 0

Opening files in safer mode
by Marcel Krüger 25 Oct '18

25 Oct '18

Hi, I think there is a bug in `io.open` in --safer mode: \directlua{ io.open(tex.jobname .. '.tex'):close() } \bye If this document is run with `luatex --safer` it leads to `bad argument #1 to 'gsub' (string expected, got nil)`. The problem is a missing `not` in luatex-core.lua, function luatex_io_open_readonly: The default argument 'r' is used if an explicit argument is given and when no "how" argument is given it tries to interpret the non existing argument. This can be fixed by (I omit the changes to `luatex-core.c` in the diff, they are auto-generated anyway): diff --git a/source/texk/web2c/luatexdir/lua/luatex-core.lua b/source/texk/web2c/luatexdir/lua/luatex-core.lua index f2d55fd99..de7f12006 100644 --- a/source/texk/web2c/luatexdir/lua/luatex-core.lua +++ b/source/texk/web2c/luatexdir/lua/luatex-core.lua @@ -54,7 +54,7 @@ local function luatex_io_open(name,how) end local function luatex_io_open_readonly(name,how) - if how then + if not how then how = 'r' else how = gsub(how,'[^rb]','') Best regards, Marcel Krüger

2 2

Re: [Dev-luatex] Bug in \letcharcode
by Marcel Krüger 23 Oct '18

23 Oct '18

---- On Mon, 22 Oct 2018 22:06:56 +0200 luigi scarso <luigi.scarso(a)gmail.com> wrote ---- > > > On Mon, Oct 22, 2018 at 9:58 PM Marcel Krüger <tex(a)2krueger.de> wrote: > > Thank you for the report. Do you have a sample document ? > On my system at least this is triggered by the following LuaLaTeX document: \documentclass{article} \usepackage{unicode-math} \show\document \letcharcode"305=\relax \show\document \begin{document} \end{document} Beside failing during \begin{document}, you can see that between the `\show\document` lines a occurence of `\ifdim \topskip <1sp` changes to `\ifdim ̅<1sp`. I could not reproduce it with ConTeXt/plain yet.

2 1

Bug for subsetted CFF font output
by Marcel Krüger 23 Oct '18

23 Oct '18

Hi, another bug in LuaTeX: If LuaTeX tries to calculate the size of a CFF table where the second entry in the table is a real number, `pack_real` is called with `work_buffer+i` for a small `i`. This causes aliasing issues because `work_buffer` writes into `work_buffer` using `sprintf`. So the output from `pack_real` modifies `work_buffer` before the digits are read completely, causing LuaTeX to complain about a "invalid character". An example to reproduce the issue: Run the following plainTeX document with the attached version of `LOGO10.otf` (the font was created by reordering the Private table entries and has an invalid checksum, but this is ignored by LuaTeX) \directlua{ font.current(font.define{ name='logo', embedding='subset', encodingbytes=2, format='opentype', filename='LOGO10.otf', characters={ [0x41] = { width = 1000, height = 1000, depth = 1000, index = 0x2, } }, }) } A \bye This can be fixed by adding a separate temporary buffer for `pack_real`: diff --git a/source/texk/web2c/luatexdir/font/writecff.c b/source/texk/web2c/luatexdir/font/writecff.c index cf0751179..1217d1a52 100644 --- a/source/texk/web2c/luatexdir/font/writecff.c +++ b/source/texk/web2c/luatexdir/font/writecff.c @@ -1171,6 +1171,7 @@ static long pack_integer(card8 * dest, long destlen, long value) static long pack_real(card8 * dest, long destlen, double value) { + static char temp_buffer[22]; long e; int i = 0, pos = 2; int res; @@ -1199,20 +1200,18 @@ static long pack_real(card8 * dest, long destlen, double value) e--; } } - res = sprintf(work_buffer, "%1.14g", value); + res = sprintf(temp_buffer, "%1.14g", value); if (res<0) normal_error("cff","invalid conversion"); - if (res>CFF_REAL_MAX_LEN) - res=CFF_REAL_MAX_LEN; for (i = 0; i < res; i++) { unsigned char ch = 0; - if (work_buffer[i] == '\0') { + if (temp_buffer[i] == '\0') { /*tex In fact |res| should prevent this. */ break; - } else if (work_buffer[i] == '.') { + } else if (temp_buffer[i] == '.') { ch = 0x0a; - } else if (work_buffer[i] >= '0' && work_buffer[i] <= '9') { - ch = (unsigned char) (work_buffer[i] - '0'); + } else if (temp_buffer[i] >= '0' && temp_buffer[i] <= '9') { + ch = (unsigned char) (temp_buffer[i] - '0'); } else { normal_error("cff","invalid character"); } @@ -1247,15 +1246,15 @@ static long pack_real(card8 * dest, long destlen, double value) pos++; } if (e != 0) { - sprintf(work_buffer, "%ld", e); + sprintf(temp_buffer, "%ld", e); for (i = 0; i < CFF_REAL_MAX_LEN; i++) { unsigned char ch = 0; - if (work_buffer[i] == '\0') { + if (temp_buffer[i] == '\0') { break; - } else if (work_buffer[i] == '.') { + } else if (temp_buffer[i] == '.') { ch = 0x0a; - } else if (work_buffer[i] >= '0' && work_buffer[i] <= '9') { - ch = (unsigned char) (work_buffer[i] - '0'); + } else if (temp_buffer[i] >= '0' && temp_buffer[i] <= '9') { + ch = (unsigned char) (temp_buffer[i] - '0'); } else { normal_error("cff","invalid character"); }

2 3

MetaPost get_... related problems
by Marcel Krüger 22 Oct '18

22 Oct '18

Hi, I noticed two problems with the mplib `get_...` family of functions: - First there seems to be a LuaTeX specific documentation bug: In the LuaTeX manual these functions are listed as `getboolean`, `getnumeric`, `getstring`and `getpath`, but in the actual implementation the names contain a underscore. - If `get_numeric` is called while two nested loops are in effect in MetaPost, an endless loop is triggered when MetaPost tries to iterate through the current loops. I attached a patch fixing both problems. The second point also applies to the standalone mplib. Best regards Marcel Krüger

2 4

Bug in \letcharcode
by Marcel Krüger 22 Oct '18

22 Oct '18

Hi, in the implementation of \letcharcode LuaTeX calls `set_token_info(cur_cs, cur_cs + cs_token_flag);`. This is weird, because the first argument to `set_token_info` is supposed to be a pointer to a one-word node as returned by `get_avail` and not a `cs`. I've got a document where this caused LuaTeX to overwrite a random token in a different macro every time `\letcharcode` was called. The active token isn't stored anywhere, so this can be fixed by just omitting `set_token_info`. Best regards Marcel Krüger Suggested patch: diff --git a/source/texk/web2c/luatexdir/tex/maincontrol.c b/source/texk/web2c/luatexdir/tex/maincontrol.c index 05e385b27..25acf1ea9 100644 --- a/source/texk/web2c/luatexdir/tex/maincontrol.c +++ b/source/texk/web2c/luatexdir/tex/maincontrol.c @@ -2698,9 +2698,7 @@ void prefixed_command(void) /*tex |letcharcode| */ scan_int(); if (cur_val > 0) { - cur_cs = active_to_cs(cur_val, true); - set_token_info(cur_cs, cur_cs + cs_token_flag); - p = cur_cs; + p = active_to_cs(cur_val, true); do { get_token(); } while (cur_cmd == spacer_cmd);

2 1

Missing opentype font entries in fls file
by Linas Stonys 02 Oct '18

02 Oct '18

Hi, I've noticed that there's no INPUT entries of any modern font (otf, ttf) in .fls file with TL2018. VERSION: This is LuaTeX, Version 1.07.0 (TeX Live 2018/W32TeX) (format=lualatex 2018.6.5) 2 OCT 2018 15:59 EXAMPLE - MYRIAD.TTF I put in local directory: \documentclass{article} \usepackage{fontspec} \setmainfont{lmroman10-regular.otf} \newfontfamily\myrfamily{MYRIAD.TTF} \begin{document} Abc def ghi {\myrfamily used myriad} \end{document} in the log file I see: <./MYRIAD.TTF><d:/texroot/2018/texmf-dist/fonts/opentype/public/lm/lmroman10-regular.otf> TL2016 with this small example acts the same but with larger file and "home made" class file I see INPUT lines but only with TL2016: INPUT C:/Users/lstonys/.texlive2016/texmf-vtex/luatex-cache/generic/fonts/otl/lmroman10-regular.luc INPUT d:/texroot/2016/texmf-dist/fonts/opentype/public/lm/lmroman10-regular.otf INPUT C:/Users/lstonys/.texlive2016/texmf-vtex/luatexja/extra_lmroman10-regular.luc ... INPUT C:/Users/lstonys/.texlive2016/texmf-vtex/luatex-cache/generic/fonts/otl/myriad.luc INPUT ./MYRIAD.TTF INPUT C:/Users/lstonys/.texlive2016/texmf-vtex/luatexja/extra_myriad.luc Could INPUT lines with used fonts appear in future texlive versions? Inputs from fls I use for making proper back up of project used files. Regards, Linas Stonys

1 0

Preserving unknown keys in cached font table
by Marcel Krüger 26 Sep '18

26 Sep '18

Hi, while looking into a [luaotfload bug][bug] caused by the fontloader being confused when LuaTeX overwrites the `parameters` in a cached font table, I wondered why `write_lua_parameters` always creates a new table instead of altering a potentially preexisting table. Changing this should lead to some small performance improvement because Lua no longer has to create a new table every time a cached font table is queried and more importantly it allows additional string keys in `parameters` to be preserved. Best regards Marcel [bug]: https://github.com/u-fischer/luaotfload/issues/3 A patch implementing the suggested change: --- source/texk/web2c/luatexdir/font/luafont.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/source/texk/web2c/luatexdir/font/luafont.c b/source/texk/web2c/luatexdir/font/luafont.c index e64aaffc1..68e74ffc5 100644 --- a/source/texk/web2c/luatexdir/font/luafont.c +++ b/source/texk/web2c/luatexdir/font/luafont.c @@ -349,7 +349,11 @@ static void write_lua_parameters(lua_State * L, int f) { int k; lua_push_string_by_name(L,parameters); - lua_newtable(L); + lua_pushvalue(L, -1); + if (LUA_TNIL == lua_rawget(L, -3)) { + lua_pop(L, 1); + lua_newtable(L); + } for (k = 1; k <= font_params(f); k++) { switch (k) { case slant_code: -- 2.19.0

3 7

Re: [Dev-luatex] luatex 1.08 in TL cannot be built with --with-system-gmp
by Norbert Preining 07 Sep '18

07 Sep '18

Reposting to dev-luatex, seems to be the appropriate place. On Thu, 06 Sep 2018, Norbert Preining wrote: > Hi Luigi, hi all, > > configuring current TL sources with > --with-system-gmp > I get an error: > gcc -DHAVE_CONFIG_H -I. -I../../../texk/web2c -I./w2c -I/usr/include/libpng16 -I/home/norbert/Development/TeX/texlive.git/Build/source/foo/texk -I/home/norbert/Development/TeX/texlive.git/Build/source/texk -I../../../texk/web2c/mplibdir -Wimplicit -Wreturn-type -g -O2 -MT libmplibcore_a-mpmathbinary.o -MD -MP -MF .deps/libmplibcore_a-mpmathbinary.Tpo -c -o libmplibcore_a-mpmathbinary.o `test -f 'mpmathbinary.c' || echo '../../../texk/web2c/'`mpmathbinary.c > In file included from ../../../texk/web2c/mplibdir/mpmathbinary.w:28: > ../../../texk/web2c/mplibdir/mpmathbinary.w:44:10: fatal error: gmp/config.h: No such file or directory > #include <gmp/config.h> > ^~~~~~~~~~~~~~ > compilation terminated. > > > Debian's libgmp-dev does not ship > gmp/config.h > only > gmp.h > > Is it now necessary to build with the TL shipped libgmp? Best Norbert -- PREINING Norbert http://www.preining.info Accelia Inc. + JAIST + TeX Live + Debian Developer GPG: 0x860CDC13 fp: F7D8 A928 26E3 16A1 9FA0 ACF0 6CAC A448 860C DC13

2 2