rfc2045-encode-string: should signal an error on non-ASCII strings

rfc2045-encode-string: should signal an error on non-ASCII strings

[Ivan Shmakov]

	Shouldn't `rfc2045-encode-string' signal an error in case either
	PARAM or VALUE contain non-ASCII characters?

	BTW, [1] specifies CVS as the mean of obtaining the development
	sources for Gnus.  However, looking at [2] makes me think that
	it's GNU Arch that is used to hold the sources nowadays.  Can
	the appropriate GNU Arch archive be mentioned on [1] as well?

[1] http://gnus.org/distribution.html
[2] http://quimby.gnus.org/cgi-bin/cvsweb.cgi/gnus/lisp/rfc2045.el?rev=7.8&content-type=text/x-cvsweb-markup

Re: rfc2045-encode-string: should signal an error on non-ASCII strings

[Miles Bader]

Ivan Shmakov <ivan@theory.asu.ru> writes:
> 	BTW, [1] specifies CVS as the mean of obtaining the development
> 	sources for Gnus.  However, looking at [2] makes me think that
> 	it's GNU Arch that is used to hold the sources nowadays.

The "canonical" sources are still in CVS, but there are arch (and git)
repositories that are synced with the CVS repository.

-Miles

-- 
Occam's razor split hairs so well, I bought the whole argument!

Re: rfc2045-encode-string: should signal an error on non-ASCII strings

[Ivan Shmakov]

>>>>> Miles Bader <miles@gnu.org> writes:

 >> BTW, [1] specifies CVS as the mean of obtaining the development
 >> sources for Gnus.  However, looking at [2] makes me think that it's
 >> GNU Arch that is used to hold the sources nowadays.

 > The "canonical" sources are still in CVS, but there are arch (and
 > git) repositories that are synced with the CVS repository.

	Are these repositories public?  And if yes, how could they be
	reached?

Arch repository for Gnus (was: rfc2045-encode-string: should signal an error on non-ASCII strings)

[Reiner Steib]

On Fri, Jan 11 2008, Ivan Shmakov wrote:

>>>>>> Miles Bader <miles@gnu.org> writes:
>  > The "canonical" sources are still in CVS, but there are arch (and
>  > git) repositories that are synced with the CVS repository.
>
> 	Are these repositories public?  And if yes, how could they be
> 	reached?

Miles will surely tell you the location (he posted it on the list, but
I don't know it off hand).

As for <http://gnus.org/distribution.html>, you should mail your
suggestion to Lars, as only he can change this page (AFAIK).  The text
should maybe include minimal instructions as for cvs.

Bye, Reiner.
-- 
       ,,,
      (o o)
---ooO-(_)-Ooo---  |  PGP key available  |  http://rsteib.home.pages.de/

Re: rfc2045-encode-string: should signal an error on non-ASCII strings

[Katsumi Yamaoka]

>>>>> Ivan Shmakov wrote:
> 	Shouldn't `rfc2045-encode-string' signal an error in case either
> 	PARAM or VALUE contain non-ASCII characters?

Gnus doesn't use `rfc2045-encode-string' for non-ASCII characters.
I didn't find no Lisp module that uses it for non-ASCII characters
in the Emacs trunk either.

Re: rfc2045-encode-string: should signal an error on non-ASCII strings

[Ivan Shmakov]

>>>>> Katsumi Yamaoka <yamaoka@jpl.org> writes:
>>>>> Ivan Shmakov wrote:

 >> Shouldn't `rfc2045-encode-string' signal an error in case either
 >> PARAM or VALUE contain non-ASCII characters?

 > Gnus doesn't use `rfc2045-encode-string' for non-ASCII characters.  I
 > didn't find no Lisp module that uses it for non-ASCII characters in
 > the Emacs trunk either.

	Yes.  Indeed, it was my old Emacs 21.4.1 to use it that way and
	produce the MIME output damaged quite badly as a consequence.

	However, shouldn't the error be signalled, so that some useful
	diagnostics be shown instead of producing garbage on output (on
	this certain garbage on input)?  I guess, it's quite likely this
	library may get used by some other, not necessarily correct,
	code.

	(A wishlist priority item.)