At Tue, 15 Mar 2011 04:24:40 -0500,
Ted Zlatanov wrote:
> 
> On Tue, 15 Mar 2011 08:57:05 +0100 claudio.bley@gmail.com (Claudio Bley) wrote: 
> 
> CB> Ted Zlatanov wrote:
> >> Any update on the integration of our patches (I had some verify-flags
> >> support started, also the buffer-local hostname and initial C-level
> >> callback support)?  Or are you waiting for me?
> 
> CB> No, I'm not waiting for you, not at all. I haven't made any
> CB> progress. I'm just very busy these days and am not 25 anymore where I
> CB> could hack away all night...
> 
> I know the feeling :)
> 
> CB> I've send a mail to assign@gnu.org and are awaiting the papers now.
> 
> CB> I'll see if I can get something done this weekend.
> 
> OK.  I just wanted to make sure we're not waiting for each other.

I'm sorry that I'm responding so infrequently, but unfortunately I
really have little time working on this right now.

I made some minor changes and integrated your patch into my branch.

Basically, the changes are in correspondence to what Eli requested.

I backed out the changes to starttls.el, moved the pull/push functions
to w32.c.

Additionally, I reverted the change to gnutls-negotiate where I signal
an error now instead of returning an error value since I had realized
that some more code depended upon returning the given process. As the
GnuTLS functionality should be almost transparent to other libraries
making network connections I think this is the better solution. What
do you think about that?

Otherwise, your patch looks good so far. Except that I always got
hostname mismatches for www.google.no and www.google.com.

Btw, I could not reproduce the problem reported by Lars Magne
Ingebrigsten, ie. this

(progn
  (require 'gnutls)
  (url-retrieve "https://www.google.no" #'ignore)
  (url-retrieve "https://www.google.no" #'ignore))

works flawlessly and does not hang (tested on Windows and Linux).

- Claudio