Re: xpcom/gtkmozembed modules flags for tinymail
- From: Alexander Sack <asac jwsdot com>
- To: Philip Van Hoof <spam pvanhoof be>
- Cc: Alexander Sack <asac canonical com>, Loïc Minier <lool dooz org>, Steve Kowalik <steve kowalik canonical com>, tinymail-devel-list gnome org, Dirk-Jan Binnema <Dirk-Jan Binnema nokia com>
- Subject: Re: xpcom/gtkmozembed modules flags for tinymail
- Date: Tue, 4 Dec 2007 11:52:17 +0100
On Mon, Dec 03, 2007 at 07:55:52PM +0100, Philip Van Hoof wrote:
>
> On Mon, 2007-12-03 at 12:14 +0100, Loïc Minier wrote:
> > Hi Philip,
>
> Hey Loïc, I applied all your patches to both Tinymail's trunk and the
> devel/pvanhoof/bs branch.
>
> For future you can also post patches to our mailing list (if mailman
> blocks you because you're not subscribed, I'll add your From to the
> auto-accept list).
>
> http://mail.gnome.org/mailman/listinfo/tinymail-devel-list
>
>
> > Please find attached a patch for the AC_TNY_MOZ_CHECK m4 macro in
> > tinymail. It reworks some of the code of this macro, adds support for
> > libxul-embedding (which is how the modules will be named with the next
> > XUL Runner uploads to Ubuntu), and permits passing configure flags to
> > set the xpcom/nspr/nss/gtkmozembed pkg-config modules.
> >
> > Alexander Sack, who guided Steve Kowalik and myself through this
> > patching [1], also suggested that if you use a Mozilla engine >= 1.8,
> > then nss/nspr should be pulled over xpcom, and shouldn't be checked
> > directly. I didn't succeed at implementing this suggestion from him as
> > it seems libtinymail-camel uses nss/nspr directly.
>
> This is correct. Tinymail uses NSS for its SSL implementation in
>
> libtinymail-camel/camel-lite/camel/configure.ac
>
> You can opt for using OpenSSL in stead by passing --with-ssl=openssl to
> Tinymail's configure options. Doing that will make Tinymail not depend
> on NSS/NSPR, but instead on OpenSSL.
I think openssl would be a licensing issue because its incompatible
with GPL. Do you just use standard features that could easily make use
of gnutls?
>
> We have a very interesting bug in NSS/NSPR by the way. The coredump of
> our tester regretfully revealed few information. Other than these
> stacktraces.
Do you have a testcase or steps on how to reproduce?
- Alexander
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]