[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
bug#36573: 26.1; bufferpos-to-filepos assumes coding-system-eol-type giv
From: |
Eli Zaretskii |
Subject: |
bug#36573: 26.1; bufferpos-to-filepos assumes coding-system-eol-type gives integer |
Date: |
Thu, 12 Aug 2021 11:21:07 +0300 |
> From: Lars Ingebrigtsen <larsi@gnus.org>
> Date: Wed, 11 Aug 2021 20:42:08 +0200
> Cc: 36573@debbugs.gnu.org
>
> > I think it probably still makes sense to add a note to the manual for
> > bufferpos-to-filepos and filepos-to-bufferpos to point out that they
> > don't work if the line ending convention is unknown.
>
> Well, manual does say that the coding system has to be known (which
> implies that the line ending convention should also be known)...
>
> > For bonus points, maybe it makes sense to add something to fail
> > earlier (and more understandably!) in the info reading code. I'm not
> > sure exactly where that would be, though.
>
> I think those functions probably shouldn't signal an error in this
> situation if asked to do an approximate calculation (but just guess at a
> one-byte ending), perhaps. But give a better error message when asked
> to do an exact calculation.
I think, since these cases are very rare, it should be okay to assume
the -unix EOL type when this happens, and document that if this is the
case, the result could be an approximation even if the exact result
was requested.