Arun Isaac <arunisaac@systemreboot.net> writes:

> Thanks for the bug report!
>
>>   From: =?UTF-8?Q?Ludovic_Court=E8s?= <ludo@gnu.org>
>
> This MIME encoded word is incorrectly encoded. UTF-8 is declared as the
> encoding, while actually ISO-8859-1 has been used.

Right.. OK :)

>> Similarly to the recent changes, specifying 'substitute as the third
>> argument to bytevector->string at least avoids crashing in this case.
>
> Indeed, the substitute conversion strategy is the only correct way to
> proceed. I've made this change and pushed to master.
>
> https://git.systemreboot.net/guile-email/commit/?id=9d82904011516530b6ef1bcd53cef220db485e7a
>
> I'm ok with having another bugfix release, but I think we should wait a
> week or two in case more bugs are found. WDYT?

Thanks Arun :)

So I've tried to parse all the mbox files for the guix-commits mailing
list with this change, and I haven't found any new problems.

But yeah, I'm in no rush to get this fixed. I was looking ahead at
loading old mbox files in to the Guix Data Service to load information
about old revisions, but I'm not ready to do that on the processing side
in the Guix Data Service.

Thanks again,

Chris