Re: gnus-treat-[hide-citation[-maybe]]

[Dave Abrahams <dave@boostpro.com>]

on Sat Sep 10 2011, Lars Magne Ingebrigtsen <larsi-AT-gnus.org> wrote:

> Dave Abrahams <dave@boostpro.com> writes:
>
>> Furthermore, I'd like that capability for all of the `treat-' things.
>> You give me the "integer" option which says "apply if the thing is less
>> than n characters," which seems like it must've been designed for an era
>> when CPU cycles were much more expensive than they are now, or for
>> people who get truly enormous messages.
>
> That era is approaching again, as Emacs moves to cell phones...  :-)

Meh; my phone is more powerful than that era's desktops

>> But for any `treat-' action whose business is hiding information
>> that's going to get in my way, I usually only want it when the thing
>> in question gets really big.  For example, I want to see peoples'
>> signatures, but if their company forces them to write an obnoxiously
>> long disclaimer in the signature area, well, I don't want to see that.
>
> Yes, having that treatment function only trigger on long signatures
> would be nice.  So a way to specify a, er, sort of algorithm-specific
> input to each treatment function would be nice.

One easy change would be to make negative numbers set a minimum length
for applicability.

> Anyway, I've made the doc string of `gnus-treat-hide-citation-maybe'
> more sensible.

"Hide cited text according to certain conditions?"

That's not much help.  Those "certain conditions"
(head,first,last,integer,predicate) seem to apply to all treatments.  If
so, saying "according to certain conditions" is more confusing than
helpful: it implies that the available conditions are different for this
treatment

Here's how I currently see these two treatments:

gnus-treat-hide-citation-maybe: completely hide cited text
gnus-treat-hide-citation:       hide some portion of cited text

is this an inaccurate understanding?

-- 
Dave Abrahams
BoostPro Computing
http://www.boostpro.com