https://src.fedoraproject.org/rpms/groff/raw/rawhide/f/0006-Revert-upstream-change-of-mapping-special-characters.patch From f4f76740ad4693cb967801e366f1d6410381dd71 Mon Sep 17 00:00:00 2001 From: groff owner Date: Thu, 2 Nov 2023 13:09:37 +0000 Subject: [PATCH] Revert upstream change of mapping special characters for UTF-8 Adopt Debian decision: https://salsa.debian.org/debian/groff/-/commit/d5394c68d70e6c5199b01d2522e094c8fd52e64e --- tmac/man.local | 16 ++++++++++++++++ tmac/mdoc.local | 17 ++++++++++++++++- 2 files changed, 32 insertions(+), 1 deletion(-) --- ./tmac/mdoc.local.orig 2026-02-22 22:09:55.000000000 -0600 +++ ./tmac/mdoc.local 2026-02-28 15:03:00.255800250 -0600 @@ -2,6 +2,23 @@ .\" .\" Put local modifications to groff_mdoc(7)'s behavior here. .\" +.\" Adopting Debian's decision to preserve old mapping for UTF-8 +.\" Debian: As of groff 1.23.0, the "-", "'", "`", "^", and "~" input +.\" characters are mapped to non-Basic Latin code points in output from +.\" the "utf8" device, for consistency with other output devices such as +.\" PDF. Unfortunately in practice this still causes far too much +.\" disruption for innocent users of manual pages who just want to be able +.\" to search for characters normally, so preserve the old behaviour. +.\" Comment this out if you want to make sure that manual pages you're +.\" writing are clear of these problems. +.if '\*[.T]'utf8' \{\ +. char - \- +. char ' \[aq] +. char ` \[ga] +. char ^ \[ha] +. char ~ \[ti] +.\} +.\" .\" Some nroff devices or pager programs may not gracefully handle OSC 8 .\" hyperlink escape sequences that they don't understand. .\" .if n .if !r U .nr U 0 --- ./tmac/man.local.orig 2026-02-22 22:09:55.000000000 -0600 +++ ./tmac/man.local 2026-02-28 15:02:06.547803220 -0600 @@ -2,6 +2,23 @@ .\" .\" Put local modifications to groff_man(7)'s behavior here. .\" +.\" Adopting Debian's decision to preserve old mapping for UTF-8 +.\" Debian: As of groff 1.23.0, the "-", "'", "`", "^", and "~" input +.\" characters are mapped to non-Basic Latin code points in output from +.\" the "utf8" device, for consistency with other output devices such as +.\" PDF. Unfortunately in practice this still causes far too much +.\" disruption for innocent users of manual pages who just want to be able +.\" to search for characters normally, so preserve the old behaviour. +.\" Comment this out if you want to make sure that manual pages you're +.\" writing are clear of these problems. +.if '\*[.T]'utf8' \{\ +. char - \- +. char ' \[aq] +. char ` \[ga] +. char ^ \[ha] +. char ~ \[ti] +.\} +.\" .\" Some nroff devices or pager programs may not gracefully handle OSC 8 .\" hyperlink escape sequences that they don't understand. .\" .if n .if !r U .nr U 0