Scoring performance

Scoring performance

[Raymond Scholz]

Hi!

First of all I'd like to thank all Gnus hackers that noticeably speed
up Oort Gnus recently.

I've got a suggestion / question on scoring:

My all.SCORE contains scoring rules for certain headers and the body,
thus forcing Gnus to fetch the whole message to apply all rules.  But
as far as I can tell from the messages Gnus prints out, first the head
of a message is fetched, then the body (including all headers again, I
assume).  Fetching twice slows down the scoring process for my NNTP
groups.  Is this really necessary?

Cheers, Ray
-- 
/* Halley */ (Halley's comment.)

Re: Scoring performance

[Lars Magne Ingebrigtsen]

Raymond Scholz <ray-2002@zonix.de> writes:

> My all.SCORE contains scoring rules for certain headers and the body,
> thus forcing Gnus to fetch the whole message to apply all rules.  But
> as far as I can tell from the messages Gnus prints out, first the head
> of a message is fetched, then the body (including all headers again, I
> assume).  Fetching twice slows down the scoring process for my NNTP
> groups.  Is this really necessary?

Anything is a SMOP.  :-)

This behavior is mentioned in the "Scoring Tips" section.

-- 
(domestic pets only, the antidote for overdose, milk.)
   larsi@gnus.org * Lars Magne Ingebrigtsen

Re: Scoring performance

[Raymond Scholz]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:

> Anything is a SMOP.  :-)

SMOP? *google* Ahh!  "Often used ironically by the intended victim
when a suggestion for a program is made which seems easy to the
suggester, but is obviously (to the victim) a lot of work." (from the
jargon file).

> This behavior is mentioned in the "Scoring Tips" section.

Yes, but unifying "head" and "body" to "all" will render some of my
rules useless.  F.e. a rule, that will score down articles with lines
longer then 80 characters should be narrowed to the body instead of
the whole article.  

I've looked for the right place to insert an ugly hack, but
gnus-score.el is probably not the best start for an Elisp illiterate
like myself :)

Cheers, Ray
-- 
Sigmentation fault, core dumped.

Re: Scoring performance

[Lars Magne Ingebrigtsen]

Raymond Scholz <ray-2002@zonix.de> writes:

> SMOP? *google* Ahh!  "Often used ironically by the intended victim
> when a suggestion for a program is made which seems easy to the
> suggester, but is obviously (to the victim) a lot of work." (from the
> jargon file).

Yeah.  The last time I looked at this problem (which was at least a
couple of years ago), I came to the conclusion that it would require
refactoring the scoring code quite a bit.  But thinking it over now, I
don't really remember why.

I've had a fresh look over the code, and I think it can be done
without a total rewrite, but it would entail some rather ugly-looking
hacks...

-- 
(domestic pets only, the antidote for overdose, milk.)
   larsi@gnus.org * Lars Magne Ingebrigtsen