misc/pinfo presents Unicode symbols as garbage even on UTF-8 terminals with ru_RU.UTF-8 locale. How to reproduce. Find an info file with Unicode Characters (e.g. U+2018 and U+2019), there are many such files. For example libidn2.info has numerous "LEFT SINGLE QUOTATION MARK" and "LEFT SINGLE QUOTATION MARK" glyphs. View them with pinfo and see that those multibyte characters are presented as garbage: M-b~@~T ~@~X
print/texinfo seems to just skip (ignore) the non-ascii characters in the same documents. I don't know which is worse.
A commit references this bug: Author: bapt Date: Fri Sep 13 09:19:02 UTC 2019 New revision: 511918 URL: https://svnweb.freebsd.org/changeset/ports/511918 Log: Ensure the port is linked to libncursesw to properly support wildchar. While here make it use USES=ncurses to be able to properly choose between base ncurses and the port version PR: 240553 Reported by: Victor Sudakov <vas@mpeks.tomsk.su> Changes: head/misc/pinfo/Makefile head/misc/pinfo/files/ head/misc/pinfo/files/patch-macros_curses.m4
A commit references this bug: Author: bapt Date: Fri Sep 13 09:40:43 UTC 2019 New revision: 511920 URL: https://svnweb.freebsd.org/changeset/ports/511920 Log: Fix texinfo playing with unicode variables Chose ncurses (tinfo in this case) with widechar support first. PR: 240553 Submitted by: Victor Sudakov <vas@mpeks.tomsk.su> Changes: head/print/texinfo/Makefile
Can you try both changes I have made? with locales set to fr_FR.UTF-8 I can properly read unicode characters with both pinfo and info commands (after my patches)
(In reply to Baptiste Daroussin from comment #4) > Can you try both changes I have made? Both misc/pinfo and print/texinfo are fine now, at least in the ru_RU.UTF-8 locale. Thank you!!
^Triage: Assign to committer that resolved