bug#40951: Weird highlighting

[Kévin Le Gouguec]

積丹尼 Dan Jacobson <jidanni@jidanni.org> writes:

>>>>>> "KLG" == Kévin Le Gouguec <kevin.legouguec@gmail.com> writes:
> KLG> 積丹尼 Dan Jacobson <jidanni@jidanni.org> writes:
>
> KLG> on the other hand I don't see how Gnus could distinguish between a
> KLG> verb added deliberately by the article author, and the same verb
> KLG> added automatically as part of the mail client's citation
> KLG> boilerplate.
>
> Hmmm, well don't boilerplates have some @ and/or < and > etc. in them?

If we focus solely on Gnus's message.el, message-insert-citation-line
can be relied on to use whatever was in the "From" header.  From what I
gather from RFC 5322, that should at least include an "@" sign.

That only covers Gnus's "writes:" though.  Spelunking in lisp/gnus for
other verbs featured in gnus-cite-attribution-suffix, I get the
impression that other MUAs tend to follow suit, but that might be
wishful thinking.

Assuming false negatives (failing to highlight) on articles from
"non-conforming" MUAs are less annoying than false positives
(highlighting part of a line that's not introducing a citation) on
"conforming" ones, I'm still not sure where the right "fix" would
belong.

gnus-cite-attribution-suffix's job is to match just the final verb
before the citation, so I think it's fine the way it is.  There's
probably a way to teach gnus-article-highlight-citation that it
shouldn't apply gnus-cite-attribution-face to the whole line, but I'm
not familiar enough with the code to see how to go about it.