User defined digest types in Gnus

User defined digest types in Gnus

[Per Abrahamsen]

There is some weird format digests in `rec.arts.sf.tv.babylon-5.info'
that I would like to split.  I don't think Gnus should support these
by default, as they are probably only of interest to Babylon 5 fans.

For situations like this, it would be nice if there was a well
documented way to add new digest types to Gnus.

`nndoc-type-alist' almost provides this, with two exceptions:

1) It should be documented (although it seems pretty straight-forward).

2) Each digest type description should contain a function that returns
non-nil if the current buffer is a digest of that type.  This would
avoid the hard-coding of the digest types in `nndoc-guess-digest-type'. 

Re: User defined digest types in Gnus

[Lars Magne Ingebrigtsen]

abraham@dina.kvl.dk (Per Abrahamsen) writes:

> For situations like this, it would be nice if there was a well
> documented way to add new digest types to Gnus.
> 
> `nndoc-type-alist' almost provides this, with two exceptions:
> 
> 1) It should be documented (although it seems pretty straight-forward).
> 
> 2) Each digest type description should contain a function that returns
> non-nil if the current buffer is a digest of that type.  This would
> avoid the hard-coding of the digest types in `nndoc-guess-digest-type'. 

Yup.  I've now added this to the Red Gnus todo list.

-- 
  "Yes.  The journey through the human heart 
     would have to wait until some other time."

Re: User defined digest types in Gnus

[Edward J. Sabol]

Excerpts from mail: (04-Mar-96) User defined digest types in Gnus by Per Abrahamsen
> There is some weird format digests in `rec.arts.sf.tv.babylon-5.info'
> that I would like to split. I don't think Gnus should support these by
> default, as they are probably only of interest to Babylon 5 fans.

Yeah, I've come across non-standard digest formats fairly often myself. In
the past, I've usually ended up modifying the source code...

> For situations like this, it would be nice if there was a well
> documented way to add new digest types to Gnus.
> 
> `nndoc-type-alist' almost provides this, with two exceptions:
> 
> 1) It should be documented (although it seems pretty straight-forward).
> 
> 2) Each digest type description should contain a function that returns
> non-nil if the current buffer is a digest of that type. This would avoid
> the hard-coding of the digest types in `nndoc-guess-digest-type'.

Wow, this is a great idea and long overdue. Maybe we can classify this as a
bug and sneak it into September Gnus rather than waiting for Red?

Later,
Ed