On 8/26/2021 6:23 AM, Lars Ingebrigtsen wrote: > Eli Zaretskii writes: > >> I think we have to revert and wait for the paperwork to run to >> completion. > > OK; now done. Ok, my paperwork is complete. I've attached updated patches that should apply cleanly again (I had to fix the merge to account for changes to NEWS). I also rolled my fix to the term.el tests and Eli's fix to include version info in the new term faces into the second patch. There's one further enhancement that might make sense here: currently, color values are set via a vector of strings for `ansi-color', whereas they're set via faces for `term-mode'. Would it be better to use faces for `ansi-color' too? Perhaps they should even be the *same* faces; is there any reason an Emacs user (or package) would want ANSI colors to be different between `ansi-color' and `term-mode'? If so, maybe each `term-color-FOO' face should inherit from the (hypothetical) corresponding `ansi-color-FOO' face? However, maybe there's a particular reason why `ansi-color' works this way that I'm not aware of. If the current way is best, that's fine by me too. Otherwise, just let me know and I can update the patches to include `ansi-color-FOO' faces.