Hi, Helge! At 2022-03-14T12:29:26+0100, Helge Kreutzmann wrote: > Just to ensure that credit is where credit is due: This was a bug > report by a l10n/i18n team member of Debian and the fix was developed > by Colin Watson, see: > https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=692765 > > Thanks for confirmation and fixing it upstream, so downstreams > (translations) get it automatically. For the sake of neurotic precision, I (and the groff documentation I've been updating over the past 5 years) term the `\&` escape sequence a "non-printing input break", not a "zero-width character" of any sort. The reasons for this are that (1) `\&` affects how the formatter (troff) interprets its _input_, and (2) this escape sequence does not directly produce _anything_ in formatted output. We're familiar with zero-width space characters from HTML and Unicode, but such a concept is not really applicable to *roff formatters. To answer the question you raised in the Debian bug's history... > Why don't you need it for line 1 as well? There is an ! also. The "!" over on the left-hand side of the table has only one space after it, not two. For a potential end-of-sentence character (".", "?", and "!" by default) to be recognized as ending a sentence, it must be followed by a newline or _at least_ two spaces. I have tried to make the formatter's behavior very clear in the groff Texinfo manual[1]. I'd appreciate critique of any ways in which it fails to be. Regards, Branden [1] https://git.savannah.gnu.org/cgit/groff.git/tree/doc/groff.texi#n4775
Attachment:
signature.asc
Description: PGP signature
