Hi, Here is uniParse.tcl’s output for the “current” UnicodeData.txt
(https://www.unicode.org/Public/UNIDATA/UnicodeData.txt) I made no changes except for fixing the inf0 typo in the static output part. Generating this was indeed dead easy, but whether it works correctly I can't test. Someone could try overwriting the slnudata.c and build a private luatex executable to run tests, I assume. Best wishes, Taco
tclUniData.c
Description: Binary data
uniParse.tcl
Description: Tcl script
> On 30 Apr 2026, at 15:11, Vítek Starý Novotný <[email protected]> wrote: > > On Thu, Apr 30, 2026 at 02:39:21PM +0200, luigi scarso wrote: >> On Thu, 30 Apr 2026 at 13:46, Werner LEMBERG <[email protected]> wrote: >>> >>> Are there plans to update this library? >> >> Definitely yes, if there is a bug that can only be fixed by updating the >> library, but I have no bug reports at the moment. > > The main issue lies in functions such as `find()`, `gfind()`, and > `gsub()`, which support character classes like `%a` (covering Unicode > classes Lu, Ll, Lt, Lm, and Lo) and `%n` (covering Nd, Nl, and No), > among others. Since Unicode 3.1, these classes have expanded > significantly to include many additional code points that `slnunicode` > is unable to match. > > Best, > Vitek — Taco Hoekwater E: [email protected] genderfluid (all pronouns)
