On Wed, May 16 2012, Jani Nikula wrote: > Tomi Ollila <tomi.ollila at iki.fi> writes: >> IT h\344ppens. >> >> Attempting to reply to this email should expose the problem. >> >> \304\344li\366 \344l\344 ly\366, \366\366li\344 l\344ikkyy.
(\nnn text above hand-(query)-replaced) > > The problem bisected to f6c170fabca8f39e74705e3813504137811bf162 > ("emacs: Correctly quote non-text/plain parts in reply"). The commit > reverts cleanly, replying without it now. > > Tomi, it would be great if you could post your message as a patch to the > test suite. Haa, It doesn't matter which is the original encoding of the message; notmuch reply id:20120515194455.B7AD5100646 at guru.guru-group.fi where notmuch show --format=raw ^^^ outputs (among other lines): Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable and notmuch reply id:"878vgsbprq.fsf at nikula.org" where notmuch show --format=raw ^^^ outputs (among other lines): Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: base64 produce correct reply content, both in utf-8. So it is the emacs side which breaks replies. I'll see what kind of test cases I can create... > > BR, > Jani. Tomi