bug#54423: 29.0.50; gnus-fetch-original-field returns nil in digest

[Sam Steingold]

> * Lars Ingebrigtsen <ynefv@tahf.bet> [2022-03-17 11:59:43 +0100]:
>
> Sam Steingold <sds@gnu.org> writes:
>
>> In the group gmane.comp.security.risks on server nntp:news.gmane.io open
>> an article and type C-d (gnus-summary-enter-digest-group).
>> Then enter a message from the summary buffer (e.g., SPC)
>> and now (gnus-fetch-original-field "Newsgroups") returns nil instead of
>> "gmane.comp.security.risks" as it should.
>> The variable gnus-original-article-buffer evaluates to
>> " *Original Article gmane.comp.security.risks*"
>> and that buffer starts like this:
>
> I think there's a misunderstanding here -- the "original" that the
> function is fetching is the unformatted "original" version of the thing
> that's displayed in the *Article* buffer.  It has nothing to do with how
> the article was sourced (which was via a digest in this case).

Nevertheless, the '''unformatted "original" version of the thing
that's displayed in the *Article* buffer''' is the
`gnus-original-article-buffer' which points to buffer
" *Original Article gmane.comp.security.risks*"
which _does_ contain all the needed goodies,
_BUT_ `gnus-fetch-original-field' does _NOT_ return them.

IOW, `mail-fetch-field',
called by `message-fetch-field',
called by `gnus-fetch-field',
called by `gnus-fetch-original-field',
returns nil.

`gnus-fetch-original-field' calls `gnus-fetch-field'
inside `with-current-buffer gnus-original-article-buffer`
and the buffer referenced by `gnus-original-article-buffer'
_does_ contain the fields in the header.

> So this is working as designed.

Then how should bbdb extract the "Newsgroups" et al?

Thank you!

-- 
Sam Steingold (http://sds.podval.org/) on darwin Ns 10.3.2113
http://childpsy.net http://calmchildstories.com http://steingoldpsychology.com
https://iris.org.il https://memri.org https://mideasttruth.com
Man has 2 states: hungry/angry and sate/sleepy.  Catch him in transition.