Would you mind comment on the Gajim bug report as well? On Mon, 2005-12-05 at 12:05 +1100, James Bunton wrote: > As far as I can tell from the vCard avatars JEP, PyMSNt's behaviour is > correct. > It says to generate a hash of the image data. It doesn't mention > hashing the base64ed image data anywhere that I can see. > > http://www.jabber.org/jeps/jep-0153.html > If you can find anything to contradict that please let me know. > > --- > > James > > > > On 05/12/2005, at 11:25 AM, Oscar Hellstr?m wrote: > > > Hi. > > > > I have found a problem when using Gajim (SVN checkout) and PyMSN-t. > > Gajim caches the avatars and displays them in the roster (kinda like > > gaim?) which works nice when you start gajim. However, gajim then > > stores > > the avatar hash as a sha of the base64 encoded image data while PyMSN-t > > does send a sha of the image data. Therefore the sha does not match and > > the avatars does not get loaded the way they should when Gajim is > > restarted. > > > > There is a bug report for Gajim (thought about filing it against > > PyMSN-t) on http://trac.gajim.org/ticket/1137 about this, but the > > developer(s) proposes that PyMSN-t is patched to calculate the hash > > from > > the base64 encoded data. > > > > What ya all think about this? > > -- > > Oscar Hellstr?m, [EMAIL PROTECTED] > > web: personal.oscarh.net > > jid: [EMAIL PROTECTED] > > icq: 52604556 > > _______________________________________________ > > py-transports mailing list > > py-transports@blathersource.org > > http://www.modevia.com/cgi-bin/mailman/listinfo/py-transports > > _______________________________________________ > py-transports mailing list > py-transports@blathersource.org > http://www.modevia.com/cgi-bin/mailman/listinfo/py-transports -- Oscar Hellstr?m, [EMAIL PROTECTED] web: personal.oscarh.net jid: [EMAIL PROTECTED] icq: 52604556 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part Url : http://modevia.com/pipermail/py-transports/attachments/20051205/83c86cdf/attachment.pgp From [EMAIL PROTECTED] Mon Dec 5 08:10:10 2005 From: [EMAIL PROTECTED] (Norman Rasmussen) Date: Mon Dec 5 08:10:17 2005 Subject: [py-transports] PyMSN-t avtar hash In-Reply-To: <[EMAIL PROTECTED]> References: <[EMAIL PROTECTED]> <[EMAIL PROTECTED]> <[EMAIL PROTECTED]> Message-ID: <[EMAIL PROTECTED]>
I added my 2c on the trac ticket, I think gajim needs to be fixed here, and not PyMSNt. There are existing client implementations like Psi-Pedrito which cached the avatar data fine. On 12/5/05, Oscar Hellstr?m <[EMAIL PROTECTED]> wrote: > Would you mind comment on the Gajim bug report as well? > > On Mon, 2005-12-05 at 12:05 +1100, James Bunton wrote: > > As far as I can tell from the vCard avatars JEP, PyMSNt's behaviour is > > correct. > > It says to generate a hash of the image data. It doesn't mention > > hashing the base64ed image data anywhere that I can see. > > > > http://www.jabber.org/jeps/jep-0153.html > > If you can find anything to contradict that please let me know. > > > > --- > > > > James > > > > > > > > On 05/12/2005, at 11:25 AM, Oscar Hellstr?m wrote: > > > > > Hi. > > > > > > I have found a problem when using Gajim (SVN checkout) and PyMSN-t. > > > Gajim caches the avatars and displays them in the roster (kinda like > > > gaim?) which works nice when you start gajim. However, gajim then > > > stores > > > the avatar hash as a sha of the base64 encoded image data while PyMSN-t > > > does send a sha of the image data. Therefore the sha does not match and > > > the avatars does not get loaded the way they should when Gajim is > > > restarted. > > > > > > There is a bug report for Gajim (thought about filing it against > > > PyMSN-t) on http://trac.gajim.org/ticket/1137 about this, but the > > > developer(s) proposes that PyMSN-t is patched to calculate the hash > > > from > > > the base64 encoded data. > > > > > > What ya all think about this? > > > -- > > > Oscar Hellstr?m, [EMAIL PROTECTED] > > > web: personal.oscarh.net > > > jid: [EMAIL PROTECTED] > > > icq: 52604556 > > > _______________________________________________ > > > py-transports mailing list > > > py-transports@blathersource.org > > > http://www.modevia.com/cgi-bin/mailman/listinfo/py-transports > > > > _______________________________________________ > > py-transports mailing list > > py-transports@blathersource.org > > http://www.modevia.com/cgi-bin/mailman/listinfo/py-transports > -- > Oscar Hellstr?m, [EMAIL PROTECTED] > web: personal.oscarh.net > jid: [EMAIL PROTECTED] > icq: 52604556 > > > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.4.1 (GNU/Linux) > > iD8DBQBDk5PfknyQDD4PIdkRAngsAKCiLoEUS/qtTt6+XpQQK005qnV37wCeLP18 > ekjnavLttO7Fj6p18O7cDoU= > =PC/g > -----END PGP SIGNATURE----- > > > _______________________________________________ > py-transports mailing list > py-transports@blathersource.org > http://www.modevia.com/cgi-bin/mailman/listinfo/py-transports > > > -- - Norman Rasmussen - Email: [EMAIL PROTECTED] - Home page: http://norman.rasmussen.co.za/ From [EMAIL PROTECTED] Mon Dec 5 06:11:08 2005 From: [EMAIL PROTECTED] (Jeff Licquia) Date: Mon Dec 5 13:31:26 2005 Subject: [py-transports] PyMSN-t avtar hash In-Reply-To: <[EMAIL PROTECTED]> References: <[EMAIL PROTECTED]> <[EMAIL PROTECTED]> Message-ID: <[EMAIL PROTECTED]> On Mon, 2005-12-05 at 12:05 +1100, James Bunton wrote: > As far as I can tell from the vCard avatars JEP, PyMSNt's behaviour is > correct. > It says to generate a hash of the image data. It doesn't mention > hashing the base64ed image data anywhere that I can see. > > http://www.jabber.org/jeps/jep-0153.html > If you can find anything to contradict that please let me know. The JEP seems to indicate that all image data must be base64-encoded. Nowhere does it indicate that image data must be decoded before processing, implying that the data must stay in base64 unless otherwise indicated. Thus, I would interpret silence on this point as requiring base64 encoding for the checksum. On the other hand, my reading of the source for JOPL (the Exodus library) seems to indicate that JOPL interprets the spec the same way PyMSNt does. See: http://www.jabberstudio.org/cgi-bin/viewcvs.cgi/exodus/trunk/jopl/Avatar.pas?rev=2272&view=auto Iris (Psi, Kopete) doesn't seem to use the hash at all. I was unable to find other examples of avatar handling in client source. The JEP should probably be clearer on this point.
