Re: allout encryption and non-ascii characters

["Ken Manheimer" <ken.manheimer@gmail.com>]

i still am getting successful encryption/decryption turnaround of your
text with a current cvs checkout.  in the process i realized one
elusive pitfall which could lead to a false negative, and want to warn
you about it.

you need to have some of the non-ascii text outside of an encrypted
topic.  otherwise, emacs auto-encrypts during the save, hiding the
encoded characters from emacs non-ascii detection.  it also
auto-decrypts the current topic after the save, if it was plain text
before, hiding the fact that the encryption happened from the user.
you would notice that the non-ascii text isn't preserved across the
save, but could miss that *emacs* could miss the non-ascii
characters/alternate encoding because of the auto-encryption...

thanks again...

ken


On 11/7/06, Ken Manheimer <ken.manheimer@gmail.com> wrote:
> i've replied to david, and i'm hoping we can track down the
> discrepancy between our systems - i'm unable to reproduce his problem
> with his text and my posted version of allout.  in the meanwhile, the
> request for testers still goes - encryption of non-ascii text with the
> (attached) allout version with the fix (after saving and revisiting
> the text file, to ensure that the buffer is set to the right
> encoding).
>
> ken
>
> On 11/6/06, David Smith <davidsmith@acm.org> wrote:
>
> > Thanks very much for working on the coding-system related
> > bug. Unfortunately, with the attached test file, it still
> > fails. I've attached my test file; it's encoded with
> > mule-utf-8, I'm using GNU Emacs 22.0.50.1 (i486-pc-linux-gnu,
> > GTK+ Version 2.8.20) of 2006-10-23 on pacem, modified by Debian.
> >
> > Any pgg hackers with a clue on this?
> >
> > David
> >
> > "Ken Manheimer" <ken.manheimer@gmail.com> writes:
> >
> > > though i haven't heard that anyone else has tested it, i would hate
> > > for this allout patch to miss the release, and am confident enough
> > > about its merits to ask that it be applied.
> > >
> > > the crucial thing the patch does is enable topic encryption of
> > > non-ascii encodings.  the encoding fix is only a few lines, but the
> > > patch also eliminates a tradeoff in the last fix i submitted,
> > > clarifies a variable name and some docstrings in the process, and
> > > rectifies some assorted boundary-condition behaviors, as well.
> > > --
> > > ken
> > > ken.manheimer@gmail.com
> > > http://myriadicity.net
> > >
> > > 2006-11-05  Ken Manheimer  <ken.manheimer@gmail.com>
> > >
> > >       * allout.el (allout-doublecheck-at-and-shallower): Clarify
> > >       docstring.
> > >       (allout-inhibit-aberrance-doublecheck): Rename from
> > >       allout-during-yank-processing.
> > >       (allout-do-doublecheck): Track allout-inhibit-aberrance-doublecheck
> > >       name change.
> > >       (allout-ascend): Provide for unusual case where some topic after
> > >       the first in file is at lower depth than the first.
> > >       (allout-shift-in): Ensure the offspring of the new containing
> > >       topic are exposed.
> > >       (allout-encrypt-string): Preserve the coding-system of the text,
> > >       according to that of the containing buffer.
> > >
> > >
> > >
> > > On 11/4/06, Ken Manheimer <ken.manheimer@gmail.com> wrote:
> > >> i have had some success with getting allout topic encryption to
> > >> encrypt text so that characters in an alternate coding set are
> > >> preserved.  i am looking for people to test it - i am so unfamiliar
> > >> with coding sets in general that i don't really know how to be sure it
> > >> works generally!  (i am excited, though, that i was able to round-trip
> > >> some text with an elaborate <'> apostrophe that isn't preserved in the
> > >> ascii character set, but is preserved in iso-2022-7bit.)
> > >>
> > >> i'm attaching a full copy of the revised allout.el for testing - make
> > >> sure you're not getting byte code from an old version when giving it a
> > >> go.  and when you do test it, be sure that the file you're working
> > >> with has a coding set which preserves the characters on rereading -
> > >> the encryption depends on the buffer being in the right coding set.
> > >>
> > >> let me know whether or not it works for you, and if possible, the
> > >> coding set of the trial ('Esc-x buffer-file-coding-set' will tell
> > >> you).  if i get good confirmation that this works, i'll submit a
> > >> proper patch.
> > >>
> > >> thanks!
> > >>
> > >> On 11/1/06, Ken Manheimer <ken.manheimer@gmail.com> wrote:
> > >> > allout's use of pgg for encryption doesn't provide for non-ascii text,
> > >> > and encoding is a realm where i seem to have less than zero
> > >> > cluefulness.  can anyone help me solve the problem posed below?
> > >> >
> > >> > ken
> > >> >
> > >> > On 11/1/06, an david smith wrote:
> > >> >
> > >> > > Hi Ken,
> > >> > >
> > >> > > I've been an allout user for a very long time. It's wonderful
> > >> > > software. Thank you.
> > >> > >
> > >> > > Today I thought I'd try out the encryption support as I finally
> > >> > > have a need for it but it doesn't properly handle non-ascii
> > >> > > characters. pgg-output-buffer is created inside of pgg-gpg with
> > >> > > mode of raw-text or binary and that is never converted back
> > >> > > into the charset of the original cleartext. I do a lot of work
> > >> > > in Japanese and so this is critical.
> > >> > >
> > >> > > I look at how gnus uses pgg and its charset handling but even
> > >> > > in edebug I couldn't quite see how it was doing it correctly
> > >> > > compared to how allout's method.
> > >> > >
> > >> > > If you have any insight I would really appreciate it. I will
> > >> > > try to debug this in my own time but as you are the
> > >> > > maintainer/author of the software involved, I hope you can at
> > >> > > least nudge in me the right direction towards a fix.
> > >
> > >
> >
> > --
> >   David D. Smith
> >
> >
> > Test Text
> > //_. Encrypt This
> > 筆は剣より強い。
> >
> >
> >
> >
>
>
> --
> ken
> ken.manheimer@gmail.com
> http://myriadicity.net
>
>
>


-- 
ken
ken.manheimer@gmail.com
http://myriadicity.net