GNU bug report logs - #77539
31.0.50; nnatom: missing content in articles.

Previous Next

Full log

View this message in rfc822 format

From: Daniel Semyonov <daniel <at> dsemy.com>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: 77539 <at> debbugs.gnu.org, fernandodemorais.jf <at> gmail.com
Subject: bug#77539: 31.0.50; nnatom: missing content in articles.
Date: Mon, 21 Jul 2025 16:05:54 +0300

>>>>> Daniel Semyonov writes:

>>>>> Daniel Semyonov writes:
>>>>> Eli Zaretskii writes:
    >>> Ping! Ping! Ping! Daniel, please respond.

    >> Sorry, I'm very sick for a few weeks now (getting better hopefully),
    >> this issue is specific to nnatom, I did test but need to do more
    >> research, I think I'll be able to do this soon.
    >> Did not expect to be away for so long, sorry again, I'll make sure to
    >> write soon whether or not I made progress

    > Not sure how no one noticed this for so long, but nnatom/nnfeed don't
    > take into account the fact that MIME messages can only contain ASCII
    > text (in the case of Gnus, it seems non-ASCII text is simply not shown
    > when an article is displayed); Non-ASCII text must encoded in one of a
    > two ways specific to MIME to be displayed correctly.

    > However, I made some simple changes to encode the body of articles using
    > 'quoted-printable' encoding to test this, and yet non-ASCII characters
    > still seem to be missing, even though looking at the raw article I can
    > see it was encoded correctly and all the appropriate MIME headers are
    > set correctly (I even tested with an HTML email with identical content
    > where non-ASCII characters are visible through Gnus, and it seems the
    > raw articles are nearly identical, barring some headers specific to the
    > email unrelated to encoding).

    > I'm honestly not sure how to proceed from here, I don't know what could
    > be causing this difference at this point.

    > From a cursory look at how nnrss handles this, it seems that encoding
    > the body is handled automatically there, as the message is structured
    > with MML and then converted to MIME using 'mml-to-mime' (which also
    > encodes the message body); if I don't figure it out soon I'll guess I'll
    > convert nnatom to use this method too.

Just wanted to give an update, I'm still dealing with health issues so
it's taking me a while, but I've decided to convert nnfeed and nnatom to
also use MML; this unfortunately requires changing a lot of the
existing code.

Daniel

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.