karl berry

Maybe you can search the linux mint repos for something that provides a file (symlink, probably) named "libglut.so.3", likely in /usr/lib64 or similar. It is widely available so far as I can see from pkgs.org, but mint is not listed there. I'm surprised it was not provided by freeglut3-dev. Maybe just freeglut3? But that should have been installed as a dependency, one would think.

My only comment is that Lua(La)TeX and Xe(La)TeX use entirely different methods to find fonts by name, and Windows is yet another variable, so it's not surprising that a given font might be found by one and not by the other. Unfortunately, I don't know enough about fontconfig or xetex font internals (or cjk or windows) to suggest a workaround. Sorry. I hope someone else reading this can help.

Actually, I think the best fix is for you to set FORCE_SOURCE_DATE=1 as well as SOURCE_DATE_EPOCH. Changing the DVI comment is equivalent to changing \year \month \day, as far as the implementation goes. Unless setting FORCE_SOURCE_DATE is problematic for you for some reason, this would be a far simpler solution than disentangling the two uses in the code.

Thanks Hille. It's because XeTeX's dvi (er, xdv) output includes the usual dvi comment with the timestamp, and xdvipdfmx dutifully passes that along to the Creator: field. It happens with plain TeX too, not latex-specific. (It also happens with the dvi output from any tex, but no one cares about reproducible dvi, apparently.) Anyway, I'll see if I can munge xetex to pay attention to SOURCE_DATE_EPOCH for the dvi comment. Unless someone else gets there first which, of course, would be great :).

Fails to compile using gcc-16

ICU 76 requires C++17 per https://unicode-org.github.io/icu/download/76.html#icu4c-platform-support . Thanks to Norbert, I (r78759) installed the attached change (more or less) in web2c/configure.ac, and updated our copy of ax_cxx_compile_stdcxx.m4 in Build/source/m4.

Thanks for the report, as always. Can you, or anyone, suggest a patch? My minimal C++ knowledge is not up to dealing with this. Also, I wonder if it might be wise to wait until gcc16 is actually released, since it's such bleeding-edge stuff. or are these problems unlikely to change?

I sent Hans the plain TeX example, but I don't think he did anything with it. Feel free to write him yourself if you want, or let's just forget the whole thing. Sorry for the noise. So many more important things to spend time on ...