i missed the example until your recent message. as far as i can tell, the problem you're describing is one of configuration management - it's the versions of the stuff you're using. we need to guard against that - but claiming that it "just doesn't work" is unnecessarily incindiary! specifics below...
On 1/9/06, Dave Love <[EMAIL PROTECTED]> wrote: > Ken Manheimer <[EMAIL PROTECTED]> writes: > > > it would be helpful to me to find out how to repeat that bug, so i > > could repair it. > > Find allout.el, go to the first heading (e.g. with the menu), use > Headings -> Toggle Topic Encryption from the menu, type `foo' as > passphrase and `bar' as the hint to get: > > ;;;_~* Provide > -----BEGIN PGP MESSAGE----- > Version: GnuPG v1.4.1 (GNU/Linux) > > jA0EAwMC4k9RaxuRboJgyRabxt2CW0in+vaRHnZ6jysmkDfz1KQ7 > =TImu > -----END PGP MESSAGE----- > > Use the toggle menu item again, and entering `foo' as the passphrase, > it says `Passphrase differs from established' and fails to decrypt > when I try again. I did that test in Emacs 21 on Debian stable(-ish), > `gpg (GnuPG) 1.4.1'. It was the same when I originally tried with the > development Emacs on Debian testing. i did this in emacs 22.0.50, with no problems. i tried to do it in emacs 21.4.1 with a special version of allout, and was refused due to the version of pgg. i tried it in that same 21.4.1 but added the load path for the new version of pgg, and it worked fine. i certainly can't control every aspect of the environment. perhaps i can add something that refuses to do the encryption operations with versions of pgg prior to some proven-ok one. i wish you would qualify your objections a bit more carefully, though - i thought this new version of allout was targeted solely to the new version of emacs! > > it defaults to using symmetric-mode encoding. describe-key (`C-h k > > C-c x') will tell you that a single universal argument (ie, C-u) will > > use key-pair encryption, and a doubled universal argument will do the > > symmetric-mode encryption but disregard the cache. > > Oh. I hadn't examined the keymap, but C-c x is reserved for users > according to the Lisp manual. `allout-passphrase-verifier-handling' > appeared to be documented as controlling the behaviour. several people have mentioned this. i'm trying to figure out what best to do - but do note that the standard outline-mode uses \C-c bindings, also. (i will probably make the key prefix a customizable setting and default it to something not \C-c, but need to find out whether the outline-mode precedent justifies leaving it the way it is.) > >> with non-ASCII, potentially also causing data loss, though I don't > >> know what's actually done. > > > > this is something i need to understand better. it's a responsibility > > of the pgg routines, but my changes may have left that off. > > No. What, say, pgg-encode tries to do for an interactive call (which > is tested wrongly) can lead to data loss anyway since it doesn't check > that it's valid. Check that with handa, since people accept what he > says. could you spell out the details a bit here? i'm losing track of your prepositions - "since it doesn't check that it's valid" - what the heck (pardon my french) does the second "it" refer to? > > allout doesn't handle whitespace-delimited outlines. (i also have an > > unreleased minor-mode block-wise outlining package - outdent.el - > > which i use for python programming.) i don't know about the outline > > variables that major modes set. can you tell me about them? > > See python.el for the example. If it's not done in the installed > version, try <URL:http://www.loveshack.ukfsn.org/emacs/python.el>. I > don't recall if I contributed that for the old python-mode.el, but I > couldn't get even the bugs in that fixed for use with Emacs. > > > i agree. there are some provisions of which you might not be aware - > > the allout-mode function has substantial documentation, > > I don't understand why the help text for allout-mode looks different > from what I remember before, but I don't have time or inclination to > understand it all, I'm afraid. This was just an aside. i don't mean to waste your time. you're raising serious objections, though, and i'm trying to address them. i expect it will not be hard to have allout refuse to do encryption with pgg versions prior to some reasonable baseline. i would love to get the attention of People Who Know to pgg, to iron out whatever coding-system flaws there are. ken manheimer [EMAIL PROTECTED] _______________________________________________ emacs-pretest-bug mailing list [email protected] http://lists.gnu.org/mailman/listinfo/emacs-pretest-bug
