David T-G wrote:
>
> Hokay. As below, I just wanted to make sure of what we were discussing.
> I've changed the subject line to help :-)
Good, thanks!
> You realize what you're asking, right? With PGP/MIME, mutt simply has to
> look at the headers to see what it must do, and that's already done, and
> it's pretty resistant to problems. With inline signing, mutt now has to
> look at the body for these magic strings
[...]
Yes, I'm aware of those potential problems.
> Now, we clearly see that it's not impossible; that's what esc-P does for
> us. It could cause problems, though, if it were the default behavior,
> and it would mean reading every mail body for parsing as well. [Is that
> more expensive in an mbox file? I dunno; I'm not a programmer.]
It's an additional checking, so it *is* more expensive. *How* expensive
it is, I don't know. (See below.)
> Someday :-)
Yeah, this seems to be the most essential word in out thread. :-)
> % mail). It would be so much easier for everyone if check-traditional-pgp
> % would become a variable ("set always-check-traditional-pgp=yes"), so
> % that every mail will be checked automatically on demand (instead of
> % handling 50% of all mail accesses via macros and doing some terrible
> % workarounds for the other 50%).
> Hmmm...
With this I mean, you can
- press <enter> in the index
- press <down>, <up>, <pgup>, <pgdown> in the pager
- press <space> at the end of a message
- delete (in various ways) a mail
- jump to a mail
- ...
to view a mail. Do you really want do set a macro for each of these (and
surely more!) keybindings? Not really, at least not me. So a message-hook
seems to be the only senseful place to implement a "check-traditional on
demand". With macros it's only a nice workaround. IMHO.
> Understood. I think that all is well but I'm still working on my patch
> cocktail for .27 and won't have a use for force_traditional anyway (since
> I don't use 8-bit chars; I'm a boring American).
BTW: Thanks a lot for your work! :-)
I wasn't really aware of the fact that all of the traditional PGP stuff
is highly in employment. I started using mutt together with making my
first PGP experience about 4 months ago. Somehow I rather thought that
no one of the developer is interested in implementing this "strongly
deprecated" old-style -- beside of the users wish. Now I see that this
was rank nonsense. I beg for pardon, for my postings sound assumedly
rude.
> % Concerning the single Esc-P (or another bounded key): It's (clearly)
> % intricately.
> You've lost me here. Is it just me, or did you not complete your
> thought?
It's ok for a single mail or two. But I receive a lot of them. And
pressing Esc-P (or so) everytime to be able to read it (most of them are
encryptet!), *is* circumstantial.
> % Concerning folder-hook: This may a long time on large folders, so it's
> % not really good for me.
[...]
> That's interesting... It took longer by the clock to *not* check a
> thousand messages for traditional pgp :-)
Maybe there were no or not many traditional style mail in it?
I tested it in a folder with about 2100 mails. Entering the folder (with
no folder-hooks) in about 2 seconds, but checking traditional with
macro index <F8> \
"<tag-pattern>.<enter><tag-prefix><check-traditional-pgp><tag-prefix><tag-message>" \
"check-traditional-pgp for complete folder"
(sorry for the long line) takes about 9 seconds. And there are only
about 650 old-style signed and/or encrypted mails in it.
But normaly there is no need to check the complete folder at once. A
checking on demand (i.e. only if a mail is accessed) would suffice. But
maybe exactly here is the problem... in detecting the demand.
> I can't answer your latter point, since I haven't seen any examples of
> failure, but I agree that the current limitations mean lots of very
> similar macros. It's all we have for now, though.
It's not only the fact that the macros all are similar. The other thing
is to cover all possibilities. If someone implements a new way in
entering a mail it doesn't work here again. Again a pro for "on demand".
> % The most perfect way in my eyes is a $always-check-traditional-pgp
> % variable.
> Fair enough; to each his own. I don't think I'd mind having such a
> variable, either, since I could leave it off.
You really want everyone interested in handling with the old PGP style
write lots of macros and finally still forgetting some? IMHO all those
people want old-style work everytime, not just for some special
keybindings (and not for some other).
But if you say it's hard to implement the "on demand" (I really don't
have any clue about the mutt sources and PGP at all), macros are way
better than nothing. (Up to now I just thought it was your [or the
author's] intention only to support macros.)
Good night (for the Europeans),
happy evening (for you Americans),
and happy slaying (for Buffy!) ;-)
-volker
--
http://die-Moells.de/ * http://Stama90.de/ * http://ScriptDale.de/
"The C Programming Language -- A language which combines the
flexibility of assembly language with the power of assembly language."
