[JDEV] MSN Transport
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hey everybody... I've gotten a working version of MSN-T with MSNP8... None of the new features of MSNP8 though.. The only noticeable difference from the old version is it takes longer to login (needs to setup two SSL connections). This transport will work after October 15th. If anybody would like a copy of the code, email or Jabber me (same address) Btw, please CC, as I'm not subscribed to jdev.. Thanks.. -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.2 (GNU/Linux) iD8DBQE/XY+QcyzaDGfRzi4RAvapAKDJGp1H5U71+H7O3EMyYj0NSRYwbACfYnzM yJIGLHlrmipEucPw+j8Vf1s= =kvsu -END PGP SIGNATURE- ___ jdev mailing list [EMAIL PROTECTED] http://mailman.jabber.org/listinfo/jdev
[JDEV] MSN-T pre-release
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Version 1.2.8pre3 is up for download on http://msn-transport.jabberstudio.org This gives MSNP8 support.. The latest additions to CVS include support for non-hotmail accounts, and fixes to presence and the some subscription oddities.. If you're running the old MSN-T and are able to test, please do so... Any bug reports to me. Thanks.. -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.2 (GNU/Linux) iD8DBQE/YksQcyzaDGfRzi4RAvCnAJ0akrqDJbVONI1XnfMHBcRaFyPBPgCg7MxN dMuZKLA6IXgB9hz2CjsTFgk= =RPs8 -END PGP SIGNATURE- ___ jdev mailing list [EMAIL PROTECTED] http://mailman.jabber.org/listinfo/jdev
[JDEV] MSN-T 1.2.8pre4
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hi... MSN-T 1.2.8pre4 has been released on http://msn-transport.jabberstudio.org Please test this.. If you are still having problems building or running it contact me.. If you have problems, run the transport with -D for heaps of debugging info.. Then mail it all to me.. Seriously... ;) You might want to search it for passwords, etc though, as they'll be in there in several places... Thanks for testing! -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.2 (GNU/Linux) iD4DBQE/bNLbcyzaDGfRzi4RApkJAJY/X3Z6JJKp07M7OdO1J8wvvE1QAKC+xXab QPL1gU8/CMA+8zKA0b3P0A== =60fh -END PGP SIGNATURE- ___ jdev mailing list [EMAIL PROTECTED] http://mailman.jabber.org/listinfo/jdev
[JDEV] Re: MSN-T 1.2.8pre4
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 D'oh! When I adjusted configure.in to let you specify the path to curl (--with-curl=/path/to/curl) I forgot to add -lcurl to the libs... =| That's what I get for testing with LD_PRELOAD already set.. Ok.. Hopefully the next build will run.. =) For now you can either apply the attached patch, or run the transport with export LD_PRELOAD=/usr/local/lib/libcurl.so.2 (or whatever your library is called) James -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.2 (GNU/Linux) iD8DBQE/biLIcyzaDGfRzi4RAh8RAJwOpknmBmsXgDrly4igRN/F2+x0IwCglOPr Ra3XH7J1nskj5897dfNXZf8= =3v4a -END PGP SIGNATURE- --- msn-transport-cvs/configure.in 2003-09-22 08:04:47.0 +1000 +++ msn-transport-new/configure.in 2003-09-22 08:05:32.0 +1000 @@ -1,7 +1,7 @@ AC_INIT(src/init.c) AM_CONFIG_HEADER(include/config.h) -AM_INIT_AUTOMAKE(msn-transport,1.2.8pre4) +AM_INIT_AUTOMAKE(msn-transport,1.2.8pre5) AC_PROG_CC AM_SANITY_CHECK @@ -50,6 +50,7 @@ AC_MSG_CHECKING(for curl.h) if test -f "$CURL_PATH/curl.h"; then AC_MSG_RESULT(Found); +LIBS="$LIBS -lcurl" else AC_MSG_RESULT(Error); AC_MSG_ERROR([
Re: [JDEV] MSN-T 1.2.8pre5
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Anybody still getting this error, please send me the output from running with
debug turned on (-D at the command line).
Also, please apply this patch (it will be pre7 in CVS), it turns off the
status_nick feature for now, I was having some weird crashes with it. I'll
wait for a stable version to be released before I put it back in.
Thanks..
James
Morgan McMillian wrote:
> Hi, I'm attempting to setup the lastes version of msn-t. It compiled
> just fine however I get the following error whenever I try to login.
>
> 20030923T02:45:20: [alert] (-internal): Error processing packet! USR 8 TWN S
lc=1033,id=507,tw=40,fs=1,ru=http%3A%2F%2Fmessenger%2Emsn%2Ecom,ct=1064285120,kpp=1,kv=5,ver=2.1.0173.1,tpf=827471f38ab31e4050cfef71af65e9b9
>
> I'm not quite sure what to even make of this error. Any ideas?
> Suggestions? I've attached my configuration and the debug log...
>
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.2.2 (GNU/Linux)
iD8DBQE/dMQ7cyzaDGfRzi4RAgdNAKCyc0/FQ92gKnrabjCl43liOAd7zgCg40Sg
g58abg2Qjx3AFcDQUuUGm9o=
=FmWO
-END PGP SIGNATURE-
diff -ur msn-transport-cvs/configure.in msn-transport-new/configure.in
--- msn-transport-cvs/configure.in 2003-09-26 18:29:40.0 +1000
+++ msn-transport-new/configure.in 2003-09-26 22:44:43.0 +1000
@@ -18,7 +18,6 @@
AC_MSG_CHECKING(for jabberd.h)
if test -f "$JABBERD_PATH/jabberd.h"; then
AC_MSG_RESULT(Found);
-LIBS="$LIBS -lcurl"
else
AC_MSG_RESULT(Error);
AC_MSG_ERROR([
@@ -51,6 +50,7 @@
AC_MSG_CHECKING(for curl.h)
if test -f "$CURL_PATH/curl.h"; then
AC_MSG_RESULT(Found);
+LIBS="$LIBS -lcurl"
else
AC_MSG_RESULT(Error);
AC_MSG_ERROR([
diff -ur msn-transport-cvs/include/session.h msn-transport-new/include/session.h
--- msn-transport-cvs/include/session.h 2003-09-26 18:29:40.0 +1000
+++ msn-transport-new/include/session.h 2003-09-26 21:59:59.0 +1000
@@ -59,7 +59,7 @@
xht rooms;
xht chats;
xht invites;
-char *user, *nick, *status_msg;
+char *user, *nick/*, *status_msg*/;
char *pass;
int exit_flag, attemps, connected, ref;
unsigned long int currentcontact, numcontacts;
diff -ur msn-transport-cvs/src/chat.c msn-transport-new/src/chat.c
--- msn-transport-cvs/src/chat.c 2003-09-19 16:35:22.0 +1000
+++ msn-transport-new/src/chat.c 2003-09-26 22:29:27.0 +1000
@@ -542,6 +542,7 @@
if (jp->to->user != NULL)
{
char *to = mt_jid2mid(jp->p,jp->to);
+lowercase(to);
if (to == NULL || strcmp(to,s->user) == 0)
{
jutil_error(jp->x,TERROR_BAD);
Binary files msn-transport-cvs/src/chat.o and msn-transport-new/src/chat.o differ
diff -ur msn-transport-cvs/src/conf_room.c msn-transport-new/src/conf_room.c
--- msn-transport-cvs/src/conf_room.c 2003-09-19 16:35:22.0 +1000
+++ msn-transport-new/src/conf_room.c 2003-09-26 22:33:58.0 +1000
@@ -49,6 +49,7 @@
mti ti = s->ti;
sbroom r;
+lowercase(jp->to->user);
r = (sbroom) xhash_get(s->rooms,jp->to->user);
if (r != NULL && r->state == sb_READY)
Binary files msn-transport-cvs/src/conf_room.o and msn-transport-new/src/conf_room.o differ
Binary files msn-transport-cvs/src/msntrans.so and msn-transport-new/src/msntrans.so differ
Binary files msn-transport-cvs/src/ns.o and msn-transport-new/src/ns.o differ
diff -ur msn-transport-cvs/src/presence.c msn-transport-new/src/presence.c
--- msn-transport-cvs/src/presence.c 2003-09-26 18:29:40.0 +1000
+++ msn-transport-new/src/presence.c 2003-09-26 22:03:19.0 +1000
@@ -41,7 +41,7 @@
lowercase(jp->from->user);
lowercase(jp->from->server);
-mt_update_nick(s, jp);
+//mt_update_nick(s, jp);
switch (jpacket_subtype(jp))
{
@@ -55,11 +55,11 @@
s->p_db = ppdb_insert(s->p_db,jp->from,jp->x);
state = mt_show2state(xmlnode_get_tag_data(jp->x,"show"));
-if (s->connected /* && state != s->state */)
+if (s->connected && state != s->state) // NICK_STATUS
{
mt_stream_register(s->st,&mt_presence_chg,(void *) s);
mt_cmd_chg(s->st,mt_state2char(state));
-mt_cmd_rea(s->st,s->user,s->status_msg);
+//mt_cmd_rea(s->st,s->user,s->status_msg);
}
s->state = state;
Binary files msn-transport-cvs/src/presence.o and msn-transport-new/src/presence.o differ
Binary files msn-transport-cvs/src/receive.o and msn-transport-new/src/receive.o differ
diff -ur msn-transport-cvs/src/register.c msn-transport-new/src/register.c
--- msn-transport-cvs/src/register.c 2003-09-26 18:29:42.0 +1000
+++ msn-transport-new/src/register.c 2003-09-26 22:02:06.0 +1000
@@ -184,9 +184,10 @@
mt_free(s->nick);
s->nick = mt_strdup(tmp);
if (rcon == 0) {
-log_debug(ZONE,"Sending nickname & status for session %
[JDEV] MSN-T 1.2.8pre7
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 MSN-T 1.2.8pre7 is now available for download from http://msn-transport.jabberstudio.org This should fix all the problems people are currently having with the transport. Those of you who were getting segfaults or Server Connect Errors on login, the problem is that your OpenSSL certificates are not setup properly. You can either set them up correctly (sorry, can't help you there), or set the option in msn.xml (thanks to David Carter). Also many build problems have been fixed (hopefully), so cvsup or, grab the tarball.. :) Full ChangeLog from 1.2.0 to 1.2.8pre7: * Updated the protocol to MSNP8 * Improved the build system somewhat, it should now work on a clean jabberd1.4.2 install without needing to move files around * Fixed bug with case sensitive session names (now logging on with Jabber ID [EMAIL PROTECTED] and [EMAIL PROTECTED] will work) * Fixed bug with case sensitive group names (sending to [EMAIL PROTECTED] and [EMAIL PROTECTED] will work) Please download it, crash it and break it in as many ways as you can.. =) If you have any problems, mail or jabber me (same address). Thanks for testing everybody.. -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.2 (GNU/Linux) iD8DBQE/d4PPcyzaDGfRzi4RAhJYAKDPKcrnoS06T4ktI2xqUO9Fq1xZ8ACgq/f1 wToGNeFc/Xi+pp5mChL5Jsk= =NaKV -END PGP SIGNATURE- ___ jdev mailing list [EMAIL PROTECTED] http://mailman.jabber.org/listinfo/jdev
[JDEV] MSN-T 1.2.8pre8
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Sorry guys.. Replace all instances of pre7 in the last email with pre8 And it's not there just yet. But it should be soon.. :$ - --- James -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.2 (GNU/Linux) iD8DBQE/d5k7cyzaDGfRzi4RApnJAJ9tQT81N7AIw/yBXITZwS6jVF9IRQCfSOua 1520LVDOt2RSgGZfvAsVFws= =3SNs -END PGP SIGNATURE- ___ jdev mailing list [EMAIL PROTECTED] http://mailman.jabber.org/listinfo/jdev
[JDEV] Email outage
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hi, if anybody sent any emails to me and I haven't replied yet, please resend them. My email server silently dropped two days worth of mail. :| Also, MSN-T 1.2.8pre10 is up for download. This fixes a bunch of things (all listed on http://msn-transport.jabberstudio.org) I'm hoping that this will be the last pre-release, so if anybody is still having problems. Please tell me. If not, I'll release around October 10th as a final. - --- James -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.2 (GNU/Linux) iD8DBQE/e9uKcyzaDGfRzi4RArnsAJ44kkvNVI063Y3XjoCWToCoNYl9HQCfYvA/ VI7YoH3y3KRX4UNb7S3mOP4= =8sQi -END PGP SIGNATURE- ___ jdev mailing list [EMAIL PROTECTED] http://mailman.jabber.org/listinfo/jdev
[JDEV] MSN-t 1.2.8rc1
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Ok, I'd classify this version as ready for production machines. I haven't had a single problem with it, and I don't know of any current issues. It's already running on several servers including tipic.com Almost no changes from pre10 to rc1, just a few cosmetic things in the code that have no effect on how it runs. So if you're using pre10, you don't need to bother bringing down your server, these two are exactly the same. So, if you're not running it yet, now is the time to upgrade. I've only got a few more days to work out any possible problems. Have a look at the new Trouble-Shooting section of the README file if you have problems.. - --- James -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.2 (GNU/Linux) iD8DBQE/iRatcyzaDGfRzi4RApGJAKDjVhWcRiVUjuw34pZF6RV3Qz7NEACfYLUU up/34rAtrNT5+EdCd1og1Og= =JN6Q -END PGP SIGNATURE- ___ jdev mailing list [EMAIL PROTECTED] http://mailman.jabber.org/listinfo/jdev
[JDEV] Re: MSN-t 1.2.8rc1
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 > Hi, > > I have just applied the latest version MSN-t 1.2.8rc1 > with CURL 7.10.4. Everything works fine except when I > delete the msn agent from winjab. I got the following > error when using Winjab. The Jabber is running on > Redhat 7.3. Will be the cause since I set > ? No, only effects the login.. Have you tried using any other clients, it doesn't look like Winjab is sending the right XML. Try Psi (http://psi.affinix.com), that works fine for me, removes it from my roster, and deregisters me with the transport. - --- James -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.2 (GNU/Linux) iD8DBQE/jTrBcyzaDGfRzi4RAqvjAJ4jloD2xRSQfwVIbdvu2upfrPrKkgCfVwMY 6e7OALvXMQ6ET7+QjlJMGS4= =2GUY -END PGP SIGNATURE- ___ jdev mailing list [EMAIL PROTECTED] http://mailman.jabber.org/listinfo/jdev
[JDEV] Re: Jabber Component Runtime (JCR) Pre-Release
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Wow! This sounds great.. :) And to think I was going to try and do all that myself in a few days time... =) Thanks guys, I'll work on porting MSN-t to this sometime soon, for the next release after 1.2.8 - --- James -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.2 (GNU/Linux) iD8DBQE/kc+WcyzaDGfRzi4RAow9AKC8cOUUq4SEse+Si7o+6vjyTSksnwCcDJgl yr8jXLPzXtziatvLrVjCccc= =3b8S -END PGP SIGNATURE- ___ jdev mailing list [EMAIL PROTECTED] http://mailman.jabber.org/listinfo/jdev
[JDEV] MSN-T not sending presences to real MSN users
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Oops... Accidentally sent this to jadmin first.. :) > I'm using the MSN gateway. When I : > - remove an MSN contact > - add it again > My MSN contact doesn't see me as online anymore, but I see him as > online. > I discovered that deleting an MSN contact removes the jid='[EMAIL PROTECTED]' subscription='both'/> from my jid.xml file in the > spool dir of my MSN gateway. > But when I add it again, the tag isn't re-added to the file. > > If I add it manually, to the file, I receive a message "We unsubscribed > from [EMAIL PROTECTED], and the line is removed from the file > again. > > Can someone explain this behaviour ? Whenever your jabber client sends remove authorisation packet to an MSN user, MSN-t adds that user to your block list. If your client supports XML streaming, this is what the packet looks like for me My client by default doesn't send a remove auth packet to users when they're removed from my list, so if I delete somebody, they are still subscribed to my presence. Have a look and see if yours does. The reason MSN-t does this is because there is that after authorising an MSN user, the only way to stop them from seeing your presence is to block them. Blocking them also means that you will not receive messages or other auth requests from them. The quick fix is to resend auth to those people, your client should have an option for this. In Psi it's right-click -> Authorisation -> Resend Auth Hope all that helps. - --- James -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.2 (GNU/Linux) iD8DBQE/rBGecyzaDGfRzi4RAqqIAJ0cs4/t8H22QpJxm+BVVrODNMJS0wCgpc72 4RKoI+K31wttRpK2GAWCpcY= =l59n -END PGP SIGNATURE- ___ jdev mailing list [EMAIL PROTECTED] http://mailman.jabber.org/listinfo/jdev
[JDEV] MSN-t update
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 It's been a while since there was a real change, so I thought I'd do some coding. As somebody pointed out, in the passport auth section there were a few possible buffer overrun exploits, plus the code was a bit of an ugly mess. I've cleaned it up, and rewritten it according to the latest guide at hypothetic.org (which is a simpler scheme, so it's cleaner in itself). I was also able to make it cache a few things, and reuse some Curl handles, so hopefully things will speed up for users logging in. I've only tested this on my server, but it seems to work fine for me, please reply here with success or failures so I can keep working on it. You can get the code by CVS, or as a tarball from http://msn-tranport.jabberstudio.org Thanks, and good luck. =) - --- James -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.2 (GNU/Linux) iD8DBQE/sMs3cyzaDGfRzi4RAsK+AKCuWx/q+sYyt1KRdnmitW0E9isDvgCg12Db vpJrO0NDqQMTwZWEKGSJqsQ= =v2n4 -END PGP SIGNATURE- ___ jdev mailing list [EMAIL PROTECTED] http://mailman.jabber.org/listinfo/jdev
[JDEV] msn-transport-1.2.1 on jabberd1.4.2 not working!!
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Try http://msn-transport.jabberstudio.org Have a look at the README file for instructions for setting it up, I don't know of any binary packages you can download, so you'll have to compile it yourself. Good luck! If you have any troubles, check the troubleshooting section of the README, if you still can't fix it, mail and ask. - --- James -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.2 (GNU/Linux) iD8DBQE/vJoncyzaDGfRzi4RAoFqAKCAHsRLXYxrdBTfElPg9+6Jv4WdEwCfUh0H uWLPWUmASFjLO+zzr870txc= =vL+X -END PGP SIGNATURE- ___ jdev mailing list [EMAIL PROTECTED] http://mailman.jabber.org/listinfo/jdev
[JDEV] MSN-t crashes
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 I've been tracking down the source of the crashes in MSN-t for anywise.org and it turned out that it was actually segfaulting inside the Curl library itself. This problem only occurs with Curl 7.10.8, it has been fixed in Curl CVS, and wasn't present in 7.10.7 Anybody using 7.10.8 who has the Curl source handy can apply this patch to curl-src/lib/transfer.c to fix the problem (an free() on an uninitialised variable) http://cvs.php.net/diff.php/curl/lib/transfer.c?r1=1.182&r2=1.183&ty=h Otherwise, try to find a binary of latest CVS, or downgrade to 7.10.7 If this doesn't fix the crashes for some of you, *please* send me a backtrace so I can debug it. Run ulimit -c 1 Then start MSN-t, when it crashes, run gdb jabberd -c corefile >> bt And send me the backtrace. Thanks Good luck people! - --- James -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.2 (GNU/Linux) iD8DBQE/vafycyzaDGfRzi4RAtz0AJ4hSIqSIYSJ4EFTgUSUpSh88KI/QwCgyvoB knriXCMrfKFYrIFWMh+o4Yw= =dc7E -END PGP SIGNATURE- ___ jdev mailing list [EMAIL PROTECTED] http://mailman.jabber.org/listinfo/jdev
[JDEV] MSN-t update: proxy & curl
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 A few small changes. It's now a bit easier to tell if you're having a crash inside libcurl (a message is printed just before and after the call to curl) Also, support for HTTPS proxies is now in there properly, this means that you can specify in the config file a proxy for curl to use in it's HTTP SSL connections. Note that you still need to be able to make a direct connection to the messenger servers, this is *only* for the passport authentication. cvs update, or grab the tarball from http://msn-transport.jabberstudio.org if you need these fixes And just one more plea/reminder, if you're still having crashes, please send me a backtrace. There are better instructions in the new readme on the site. Thanks - --- James -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.2 (GNU/Linux) iD8DBQE/vbchcyzaDGfRzi4RAkseAKCOrB0kWM4pNW9YJx5RlMEzaFCCegCg7n+X xT0JwBznn7Jjh7F0xmY30Oo= =9Orq -END PGP SIGNATURE- ___ jdev mailing list [EMAIL PROTECTED] http://mailman.jabber.org/listinfo/jdev
[jdev] Re: MSN: an open protocol?
> It states "an open protocol in MSN's case", but is MSN like completely open? > I thought it was a closed protocol, or isn't it completely closed? Or did > the author meant "an open protocol in Jabbers case". The very first version of the MSN protocol was submitted as a draft protocol to the IETF (I think). There was a public draft of MSNP2. Microsoft never updated the draft with any of their modifications to the protocol as it went from MSNP2 to MSNP10. It eventually expired, and can now only be found in at a few websites. http://www.hypothetic.org/docs/msn is one. On October 15th 2003 Microsoft discontinued support on it's servers for any protocols less than MSNP8. So the draft is now useless anyway, and MSN is effectively a closed protocol. Critical information to use the servers had to be reversed engineered from the official client (which I'm pretty sure is illegal in the US under DMCA), and Microsoft says that you are not allowed to connect to their servers without a license. So far I don't know of any attempts to block third party software from accessing their network, so they're at least more "open" than AOL/Yahoo in that respect. I maintain the MSN transport, and it supports MSNP8, I've never had to modify it to fix something to Microsoft broke. MSN-t has no license to access the MSN network at the moment, but I've had no trouble connecting using information from the website listed above (which documents MSNP8). Hope that clears it up for you.. --- James ___ jdev mailing list [EMAIL PROTECTED] https://jabberstudio.org/mailman/listinfo/jdev
[jdev] New MSN-t release with experimental file transfer support
Hi. I've just put a new release of MSN-t online with very experimental file transfer support. Any files that a MSN user sends will be accepted and downloaded to the server that MSN-t runs on. There are a few problems with the way this works now, most notably the fact that all files are accepted, and that all the data is read into ram before being saved to disk. I'll update this soon to actually send the file to the Jabber user that it's intended for, and then hopefully eventually we'll have two-way file transfer. Anybody interested in contributing some code would be most welcome, testers aren't needed at the moment, but will be later. Also, does anybody know of any existing C code that either uses the Jabberd1.4 API, or would be compatibly with it, that does file transfers? Have fun with it everybody.. http://msn-transport.jabberstudio.org --- James ___ jdev mailing list [EMAIL PROTECTED] https://jabberstudio.org/mailman/listinfo/jdev
[jdev] New MSN-t in the works
Hi. Just thought I'd put the word out that I'm working on a new version of MSN-t. I've started rewriting the transport from scratch in Python using Twisted (unrelated to the dead twisted-pymsnt on Jabberstudio). It's still undergoing heavy testing, and lots doesn't work yet (notably registration of new users. When it is released it will be a drop-in replacement for the current MSN-t. So you can remove Jabberd1.4 completely (no need for XDB services, unless you have another transport that needs them). The new version accesses the msn.host.com spool directory directly, and you'll be able to swap between the Python & C version (in case something breaks.). The main reason I started this was because I want to get rid of Jabberd1.4.3, JCR doesn't have the mtq functions needed for MSN-t to work, and I didn't want to put yet more work into maintaining that codebase (which had some bugs that I hadn't been able to track down), as well as working directly with the MSN protocol. The twisted framework already has classes for working with MSN, as well as Oscar (AOL/ICQ), and others. So maybe other people could pick up this code and adapt it to that. Check http://msn-transport.jabberstudio.org every now and then for a new release. I'll probably put some code up there to download in the next few weeks. If you want a copy now, just email me, but it's not ready to be used yet. --- James ___ jdev mailing list [EMAIL PROTECTED] https://jabberstudio.org/mailman/listinfo/jdev
Re: [jdev] New MSN-t in the works
> Contact me directly ... I had the same problem when porting Yahoo to > JCR. I replaced the 'mtq' functions with glib-2.0 equivalents to get the > same functionality, if you are interested. Yeah, it might be useful to have a JCR version as well as the Python one. Could you let me know how you did it? I don't have any experience with glib. > On another note ... I'm looking into using Gaim plugins as the > "transport layer" and building a generic transport framework. The Gaim > plugin API has most of the functions we would need for a transport, and > this may be a way to streamline the development of transports for Jabber. > > Of course, I'm very interested in MSN-t with Twisted. Perhaps that would > be a way to make a set of the legacy transports ... AIM, MSN, and Yahoo. > Anything that will make the transport more usable and esaier to maintain. I think twisted would be a lot easier to work with than the GAIM plugins. It's a framework for creating networked applications that may use multiple protocols. http://www.twistedmatrix.com for more info. --- James ___ jdev mailing list [EMAIL PROTECTED] https://jabberstudio.org/mailman/listinfo/jdev
[jdev] Re: New MSN-t in the works
On Fri, 13 Aug 2004 06:20 am, Peter Saint-Andre <[EMAIL PROTECTED]> > James Bunton <[EMAIL PROTECTED]> wrote: > > > Contact me directly ... I had the same problem when porting Yahoo to > > > JCR. I replaced the 'mtq' functions with glib-2.0 equivalents to get > > > the same functionality, if you are interested. > > > > Yeah, it might be useful to have a JCR version as well as the Python one. > > Could you let me know how you did it? I don't have any experience with > > glib. > > > > > On another note ... I'm looking into using Gaim plugins as the > > > "transport layer" and building a generic transport framework. The Gaim > > > plugin API has most of the functions we would need for a transport, and > > > this may be a way to streamline the development of transports for > > > Jabber. > > > > > > Of course, I'm very interested in MSN-t with Twisted. Perhaps that > > > would be a way to make a set of the legacy transports ... AIM, MSN, and > > > Yahoo. Anything that will make the transport more usable and esaier to > > > maintain. > > > > I think twisted would be a lot easier to work with than the GAIM plugins. > > It's > > a framework for creating networked applications that may use multiple > > protocols. http://www.twistedmatrix.com for more info. > > Would it be possible to somehow build the GAIM plugins into Twisted? I > know that Twisted has support for lots of protocols, but the closed IM > protocols are not on that list AFAIK. > > /psa It would probably be possible to build the GAIM plugins into twisted, but it would be messy, and that's what I'm trying to avoid =P Twisted has support for the Oscar, MSN, IRC and Jabber IM protocols. It also has support for lots of others, like pop3, smtp, etc (maybe for later on =)) Right now I have the new transport working. It shouldn't be too hard to adapt for AOL/ICQ, all the legacy protocol code is in one directory, so that can be swapped out. I've still got a few critical things. Disco/Browse/Agents (think it's still worth supporting the last two?), as well as in band registration and groupchat. Oh yeah, I wanted to ask. Has anybody got any good ideas for a better way to do Jabber<->MSN groupchat? The MSN protocol makes no separation between groupchats and ordinary chats, and extra people can be invited into a room at any time. So the only way I can think of handling it is.. * MSN contact invites a new person into the room. * MSN-t notices, converts the room to a conference and sends Jabber user an invite to the virtual conference room * Jabber joins the conference and receives any buffered events with jabber:x:delay * If the Jabber user doesn't join the conference within (maybe 3 minutes?) then they get a message from the transport with a transcript of the conference room so far, and they leave the conference on the MSN side. Any objections/suggestions? For starting an MSN groupchat on the jabber side, just join any random room on conference.msn.host.com and invite your msn contacts to it. --- James ___ jdev mailing list [EMAIL PROTECTED] https://jabberstudio.org/mailman/listinfo/jdev
[jdev] Re: New MSN-t in the works
On Sat, 14 Aug 2004 03:00 am, Trejkaz Xaoza <[EMAIL PROTECTED]> wrote: > That already sounds infinitely better than what we currently have. :-) > > Does the MSN protocol really give the user no way to reject an invite? I > always thought it was odd that an MSN user was able to get me into a > conference even if I didn't want to be there. I always figured it would be > better if it sent an invite, and only joined me to the real conference when > I joined to the transported one. :-/ Yup. In the MSN protocol all conversation map to a switchboard session. It corresponds directly to a room in Jabber MUC, or IRC. When you double-click on a contact in the MSN client, it joins a new room, then invites that person. Anybody in the room can at any time invite other people to the room (annoying I think, that you can be moved from a private to public place without permission). I want to make sure that a only ever gets sent to that one contact. --- James ___ jdev mailing list [EMAIL PROTECTED] https://jabberstudio.org/mailman/listinfo/jdev
[jdev] Re: New MSN-t in the works
On Sat, 14 Aug 2004 03:00 am, "Tijl Houtbeckers" <[EMAIL PROTECTED]> wrote: > On Fri, 13 Aug 2004 10:07:51 +1000, James Bunton <[EMAIL PROTECTED]> > > wrote: > > Oh yeah, I wanted to ask. Has anybody got any good ideas for a better > > way to > > do Jabber<->MSN groupchat? The MSN protocol makes no separation between > > groupchats and ordinary chats, and extra people can be invited into a > > room at > > any time. So the only way I can think of handling it is.. > > You could drop groupchat and use SHIM (JEP-0131) to indicate there are > multiple recipients to a message. That would require clients to implement > parts of SHIM though, and to be smart about it when they write a reply to > a message or start their own mini-conferences. (they should add their own > SHIM headers, particulary when they do the inviting themselves). That actually sounds quite good. But it's still in experimental, so there isn't going to be any client support for quite a while. Also how would that work for when the user has, for example Contact A Contact B Contact C Contact D User has a one-to-one chat with contact A User has a conference with all four User has another conference running with Contact A and B I think it would probably require to much MSN specific code in the clients to support that. Thanks for the suggestion though. --- James ___ jdev mailing list [EMAIL PROTECTED] https://jabberstudio.org/mailman/listinfo/jdev
[jdev] New MSN-t available for testing
Hi everybody. Please check out the new MSN transport available. It has been completely re-written using Python and Twisted. See here for details. http://msn-transport.jabberstudio.org There are lots of niceties fixed. About the only things not implemented that were in the old version are typing notification (coming soon), and support for MSN notification messages (probably coming soon). If you're in a position to test please do try it out. It's fully forwards and backwards compatible with the old transport (so you can swap between them at will, the spool files are compatible). Maqi, did you say we could get a wiki setup for documentation of this transport? That would be great thanks.. =) --- James ___ jdev mailing list [EMAIL PROTECTED] https://jabberstudio.org/mailman/listinfo/jdev
[jdev] PyMSNt 0.3 release + docs!
http://msn-transport.jabberstudio.org has a new release of PyMSNt with mostly bugfixes. There's also a wiki that's been setup to help everybody get the transport configured and installed. Please read it if you have a chance and let me know if you have any comments or suggestions. Thanks. There's also some documentation on how to use MSN transport from a user's perspective. That probably needs to be expanded on in many areas too. Feel free to link to this from wherever if you want to. --- James ___ jdev mailing list [EMAIL PROTECTED] https://jabberstudio.org/mailman/listinfo/jdev
[jdev] Re: [jadmin] PyMSNt 0.3 release + docs!
On Wed, 25 Aug 2004 03:00 am, Bruce Wang <[EMAIL PROTECTED]> wrote: > I got a bug related to UTF-8 encoding/or i18n charsets > after I send two Chinese characters using Psi > peer is one using MSN Messenger > > File "/usr/lib/python2.3/site-packages/twisted/xish/utility.py", line > 50, in callback > > File "/usr/lib/python2.3/site-packages/twisted/xish/utility.py", > line 34, in __call__ > > File "main.py", line 99, in onMessage > self.sessions[froj.userhost()].onMessage(el) > File "/home/bruce/sandbox/PyMSNt/src/jabw.py", line 97, in onMessage > body = str(child) > exceptions.UnicodeEncodeError: 'ascii' codec can't encode characters > in position 0-1: ordinal not in range(128) > > -- > simple is good Yes. At the moment the transport doesn't support unicode at all. I'm looking into how to get it working properly with UTF-16. If anybody wants to help out, have a look in jabw.py for the parts that need to be changed. --- James ___ jdev mailing list [EMAIL PROTECTED] https://jabberstudio.org/mailman/listinfo/jdev
[jdev] PyMSNt 0.5 released
Lots of fixes.. Here's the release summary. Testing would be greatly appreciated =) * Unicode support! (please test thoroughly with as many clients as you can) * Found the cause of the XML parsing bug. Anybody having exceptions thrown with particular Jabber clients should check out the server admin page and apply the domish.py patch there * Ghost sessions should happen less * If a contact changes their nick you'll see it * Other misc bugfixes Enjoy! :) --- James ___ jdev mailing list [EMAIL PROTECTED] https://jabberstudio.org/mailman/listinfo/jdev
Re: [jdev] PyMSNt 0.5 released
On Fri, 3 Sep 2004 08:23 am, Peter Millard <[EMAIL PROTECTED]> wrote: > Message: 4 > Date: Thu, 2 Sep 2004 15:01:41 -0600 > From: Peter Millard <[EMAIL PROTECTED]> > Subject: Re: [jdev] PyMSNt 0.5 released > To: Jabber software development list <[EMAIL PROTECTED]> > Message-ID: <[EMAIL PROTECTED]> > Content-Type: text/plain; charset=US-ASCII > > On Thu, 2 Sep 2004 07:49:22 +1000, James Bunton <[EMAIL PROTECTED]> wrote: > > * If a contact changes their nick you'll see it > > This COULD be the cause of the serious performance problems we're > seeing on jabber.org. I'm wondering what the motivation is behind this > change was? Jabber has ALWAYS valued MY setting of your nickname, over > yours. The problem that I think we're seeing is that some MSN clients > change folks nicknames as they change their presence (like from "pgm" > to "pgm-away"). This results in a change to everyone's roster who has > me in there. This is BAD since it can easily over-whelm XDB data > stores. (It's no so bad until you have 200k+ users :). > > I'm hoping this change can be rolled OUT of this release? And possibly > have server admins revert back to an older version until this can be > fixed. No, that's definitely not the problem. All that happens when an MSN user changes their nick (or all that's supposed to happen) is a MSN nickname I agree with you entirely, I really dislike the official MSN client, with one of the main reasons being because of the nickname thing. But it's sometimes important to be able to see what people have as their nicknames, so it's mapped to a status message. Which is pretty much what most MSNers use it for anyway. > Not sure if any other BIG public servers are getting hit by this, but > I've seen our load our main server skyrocket today. I _believe_ this > is the cause. hmm.. I'll look at it. I did notice a bug like this ages ago, but I thought I fixed it. --- James ___ jdev mailing list [EMAIL PROTECTED] https://jabberstudio.org/mailman/listinfo/jdev
[jdev] ANNOUNCE: PyMSNt 0.6
2004-09-04 - PyMSNt 0.6 released * Unicode now works with the official MSN client * Service ID translation works (jabber:iq:gateway) * Moved all strings seen by users into config.py for ease of translation (this means you'll have to start from the distribution config file if you're upgrading) * Made mail notifications an admin option * No more exceptions if a message is sent before the transport has logged you in * Lots of other little fixes. Please test it as hard as you can. Do all sorts of nasty stuff (send login/logout XML with messages, etc, over & over) to see what breaks =P Go to http://msn-transport.jabberstudio.org to download it. Happy testing! -- James ___ jdev mailing list [EMAIL PROTECTED] https://jabberstudio.org/mailman/listinfo/jdev
[jdev] ANNOUNCE: PyMSNt 0.6.1
Sorry.. I had it working, then added that line to fix something else, and forgot to test unicode again afterwards.. It should be fine now. --- James ___ jdev mailing list [EMAIL PROTECTED] https://jabberstudio.org/mailman/listinfo/jdev
[jdev] Proposal for a solution to transport rosters
This a proposal for a quick and easy solution to the current issues with transport rosters. The current situation is: * A user with an account on a legacy service will have a legacy contact list that will need to be synchronised with their Jabber contact list by the gateway * The current way that gateways do this is illegal according to XMPP. It also no longer works in Jabberd2s3 (and it shouldn't, it's a security flaw when it does) * There are no existing protocols for shared roster groups, and we need a way for this to work quickly, so that users can see their legacy contacts without hassle. * A user should not have to authorise all their legacy system contacts on Jabber. They have already authorised them on the legacy service. My proposal is an extension to the presence subscription packets (allowed by XMPP) which will work in such a way that existing clients will still function (the user will just have to authorise all their contacts again), but modified clients would work securely without bothering the user. An example flow with a modified client follows: A user has been using MSN Messenger, and has acquired a large contact list on this service. The user has heard about Jabber and wants to try it out. They still want to be able to talk to their MSN friends, so they will use the MSN transport on their server (host.com) * User registers with msn.host.com * The transport obtains the user's MSN contacts from MSN servers and begins the import process * The transport sends a series of packets looking like this: * The user's client notices the import tag, and checks to see if the user's contact list contains msn.host.com. It does, so the client then prompts the user in order to double-check that this entity is allowed to send roster imports to the user's contact list. * The user gives the affirmative. From now on all presence type=subscribe packets originating from the msn.host.com domain will be automatically authorised by the client. * The effect for the user is that by registering with the MSN gateway, and answering yes to one prompt, they now have their entire MSN contact list available. If the user had been using an existing client, they would need to answer yes to every subscription request, but they will still receive their contact list at the end of it. That's the advantage of using this method, any client will support it, and all can be easily modified. Rules for the client: * If the client receives a presence subscription packet with an import tag, but the originating domain is not on the user's contact list the client MUST ignore the import tag, and treat the presence packet as normal. (this prevents arbitrary Jabber users from auto-authorising themselves) * A client MUST check with the user at least once before auto-importing any contacts. The client SHOULD remember the user's answer for the duration of the session and MAY choose to remember the answer forever. (If the latter, then the transport will be able to transparently keep the user's contact list in sync, if for example it is modified using another legacy client) * A client MUST NOT auto-authorise any contacts that do not have an import tag Rules for the transport: * The transport SHOULD send a presence packet with an import tag if the user has already authorised that contact on the legacy service. * The transport MUST NOT send a presence packet with an import tag in any other case (eg, when a legacy user requests subscription for the first time) Comment/Questions? --- James ___ jdev mailing list [EMAIL PROTECTED] https://jabberstudio.org/mailman/listinfo/jdev
[jdev] ANNOUNCE: PyMSNt 0.7
2004-09-05 - PyMSNt 0.7 release Wow, lots of changes =) * Twisted is in the download tarball for this release (I'll keep it there until all the patches I've made make it into a Twisted release) * The transport recognises xml:lang. Meaning if you send your initial presence with an xml:lang="nl" attribute, the transport will send any messages back in Dutch! Translations would be much appreciated. * Support for the Roster-Import(1) proposal. Check out Remko's Psi patches to see this in action. It's great! Iris patch(2) and Psi patch(3) * The above change also means that PyMSNt will now work with Jabberd2s3 * Fixed many bugs (no longer crashes on incoming MSN alerts, and more) (1) http://msn-transport.jabberstudio.org/roster-import.txt (2) http://www.cs.kuleuven.ac.be/~remko/psi/patches/autoimport_iris.diff (3) http://www.cs.kuleuven.ac.be/~remko/psi/patches/autoimport_psi.diff I've been running this since release since yesterday. It works fine for me. Enjoy! =) --- James ___ jdev mailing list [EMAIL PROTECTED] https://jabberstudio.org/mailman/listinfo/jdev
[jdev] Re: Proposal for a solution to transport rosters
Mike wrote: > Hi James & all others. > > What is 'wrong' with using JEP-0093 (jabber:x:roster) after a disco or caps > to the client? I am using '93 within my yahoo-transport. > > The only change in the logic would be: > > 1 User registers with msn.host.com > 2 Transport does disco or caps comparison to see if '93 is supported, client > advertises the jabber:x:disco feature. > 3 If '93 is supported then provide a roster exchange list to which the > client can decide what to do with (user configurable), or go through old > method of sending a subscription for each > (or just sending a '93 anyway which is what I do). > > As you are already checking for the registered hostname the logic is easier > and a lot of clients already support '93 so you loose nothing in real terms. > > Please correct me if I have made a major error in logic. > > Thanks > > Mike The main reason is that not all clients support jabber:x:roster (including all current releases of Psi). I was not aware that those that did advertised their support with service discovery (it doesn't seem to be in the JEP). My reasoning for not using it is that if the client doesn't support it, then they do not see their roster at all. To fix this, we say send the presence type="subscribe" messages anyway. But this means that the clients that do support x:roster receive these anyway, and the user still gets lots of prompts. My idea would allow for XMPP compatibility (which needed to be fixed by changing any presence subscribed into presence subscribe), as well as compatibility with all clients that support subscriptions, and a better user experience for newer clients. Older clients would still show the roster though (which is very important). I just think it's cleaner to integrate the importing into the presence subscription packets, since they pretty much have to be sent at some point anyway. Eventually of course we'll have proper shared groups functionality, then we can drop all the subscription stuff completely, as there will be no need to even have the MSN contacts stored on the Jabber user's roster, they can just be sent at login. =) That's my hopes anyway :) --- James ___ jdev mailing list [EMAIL PROTECTED] https://jabberstudio.org/mailman/listinfo/jdev
[jdev] ANNOUNCE: PyMSNt 0.7.3 released
PyMSNt 0.7.3 has been released. This version doesn't bring any major new features (look for 0.8 which will) New stuff: * Easier for new transport developers to get started (thanks Daniel!) * Configurable spool directory and PID file locations * Much improved debug logging (exceptions now go to console and debug log file) * Updated subscription stuff to partially comply with roster-subsync (for the first few use cases) * More translations * Periodically ping the MSN server (will help for servers on ADSL links like me) * Runs tests against Twisted on startup to make sure you're using the patched version * Other misc fixes You'll need the new patched Twisted to run this. So grab that too. http://msn-transport.jabberstudio.org Please mail any bug reports to me. Have fun everyone! --- James ___ jdev mailing list [EMAIL PROTECTED] https://jabberstudio.org/mailman/listinfo/jdev
[jdev] Re: Gaim -> transport
Trejkaz Xaoza wrote: > This protocol-specific effort is exactly the issue. When a protocol > changes, Trillian and GAIM are the _first_ clients to adapt. By linking > straight to their protocol plugins, we get the fixes at the same time. We > don't have to wait for someone to port the fixes to Python. > > I agree that something like Twisted would be good if it were just as fast > moving as GAIM, though... only perhaps in a more efficient runtime than > Python. :-/ > > TX There's nothing stopping somebody from grabbing the latest version of PyMSNt, and adapting it to use libgaim for the legacy protocol. You'd just have to make a python wrapper for libgaim instead of a Perl one. btw, what's wrong with Python? It runs as fast as perl for me. --- James ___ jdev mailing list [EMAIL PROTECTED] https://jabberstudio.org/mailman/listinfo/jdev
[jdev] PyMSNt 0.8 released
Hi everybody. After quite a long delay (been busy at school and work) I've finally gotten around to making another release of PyMSNt. New in this release: * Message Events (JEP0022) support, aka Typing Notification * Beginnings of HTTPS proxy support (not functional yet) * Checks PID file on startup to make sure another instance of PyMSNt isn't running * Multiple resources support should work (No really! It should) * Fixed some memory leaks and other bugs It runs stable enough for me, and there are a lot of fixes over the previous version. Please upgrade to test it out, and let me know about any problems you have. Thanks. --- James ___ jdev mailing list [EMAIL PROTECTED] http://mail.jabber.org/mailman/listinfo/jdev
[jdev] PyMSNt 0.9 preview
Hi everyone. A few things: For the jadmin & jdev people, from now on everything related to PyMSNt will be only posted to [EMAIL PROTECTED] (along with Daniel's PyAIM/ICQ stuff) I suggest you subscribe if you're interested :) Also, there's a preview of PyMSNt 0.9 available for download from my website http://delx.cjb.net/PyMSNt-0.9-dev.zip Done so far for 0.9: * HTTPS proxy support * Switch to using xish for all XML parsing * Make an option of registration * Review subscription code (WRT eJabberd) * Don't reconnect if disconnect reason is connected from another location * jabber:x:delay in groupchat messages * Send error messages as errors * Better config format (XML) * Fix groupchat Remaining to do... * Keep track of MSN contact list versions and get the SYN command to use them * Finish MSN Notification messages (NOT message, send as headline events) The last two may or may not get done in time for release. Basically, there's a few cleanups, lots of bugfixes (memory leak still there), as well as a new config file format (XML no less :P) and a couple of small features. I'm running the transport now, it seems stable to me, but of course will need testing, especially on the features listed above :) I welcome any bug reports and feedback. Also, if anybody has a copy of MSN Messenger 7.0beta that they can test for compatibility, that would be great. I have now Windows machine, so I can't do it. Any questions, comments, or bug reports, please don't hesitate to contact me. Thanks for testing :) --- James ___ jdev mailing list jdev@jabber.org http://mail.jabber.org/mailman/listinfo/jdev
[jdev] Transport Registration
Hi everybody. I just cooked up a neat little PHP script that will register you with a Jabber gateway. If the Jabber gateway supports roster-subsync (That's Py*t), then your contacts will even be added to your list for you :) Here's what it does.. 1) Log into Jabber server 2) Send disco#items to the server's JID 3) Send disco#info to the JID of everything returned from the above request 4) Save to $gatewayjid When we find something that has category="gateway" and type="msn" 5) Send jabber:iq:register to $gatewayjid 6) Accept auth from $gatewayjid 7) Autoauthorise anything in the $gatewayjid domain that has roster-subsync flag 8) Respond with "I am not here" to any messages sent from MSN users in this time 9) Wait for 15 seconds 10) Disconnect from Jabber server If there is a delay of more than 15 seconds between any of these steps, bail out & warn the user Please feel free to try it out http://msn-transport.jabberstudio.org/tarballs/webreg.tar.gz to download and try for yourself. I suggest you install php4-mhash (otherwise passwords are sent in cleartext). Apart from that it should just work in any PHP enabled web server. Oh yeah, anybody with design skills, please have a look at index.php & style.css to make it prettier :) Edit config.php to allow registration with ICQ, AIM, etc. I'll make a better way of doing this soon. Thanks Enjoy! --- James ___ jdev mailing list jdev@jabber.org http://mail.jabber.org/mailman/listinfo/jdev
[jdev] PyMSNt 0.9.3 released!
Hi everybody. I've decided to release PyMSNt 0.9.3 http://msn-transport.jabberstudio.org to download, and read on for details. It's been running very nicely on SAPO (http://messenger.sapo.pt) servers for a week or so now, so I figure it should be fine for everybody else too :) Anyway, changes in this release: * Lots and lots and lots and lots of stability fixes. No really, heaps. * Support for Twisted 2.0 if you want to try that out. * An option to only log things to debug.log if there's an error. * You can send packets to the transport in wrappers from other server components and they will work. * You can add the tag to a message if you don't wish to have it bounced upon an error (useful for scripts) * Lots and lots more bugfixes. * A few new languages. If yours isn't here, please have a look at lang.py and send me a translation :) The documentation has been updated to reflect anything new. Please have a look if you have troubles. Upgrades shouldn't require anything more than a "cvs update" though. Also, you might want to check out the web registration form while you're at it. http://msn-transport.jabberstudio.org/webreg for a live demo. If you have any trouble getting this working, or any bugs to report, please contact me at [EMAIL PROTECTED] (email or Jabber). Now, back to work on avatars :) --- James ___ jdev mailing list jdev@jabber.org http://mail.jabber.org/mailman/listinfo/jdev
[jdev] Ad-Hoc Commands permissions
Hi all. I'm implementing a few ad-hoc commands in PyMSNt. What should I send as a response if the requesting client doesn't have permission to execute the command? At the moment it goes like this http://jabber.org/protocol/commands"; node="connectusers" /> to="[EMAIL PROTECTED]/Powerbook" > Psi (with ad-hoc patches) doesn't display any error though, and JEP-0050 doesn't actually mention an error case for lack of permissions. --- James ___ jdev mailing list jdev@jabber.org http://mail.jabber.org/mailman/listinfo/jdev
[jdev] PyMSNt 0.9.4 released, PyMSNt 0.10-rc1 available, epoll-0.5 released!
Hi all! PyMSNt 0.9.4 has been released for those of you not wanting to try out the avatars branch yet. It just contains bugfixes to 0.9.3. PyMSNt 0.10 release candidate 1 is available. This is the latest branch, with avatars, ad-hoc commands, entity capabilities and all sorts of other goodies available. You can now view MSN avatars with JEP-0008. See the developer documentation for more details on things that are available. If there's no complains about rc1 there'll be 0.10 in a few days time. epoll-0.5 is also available. This allows Linux servers to perform better. Check out http://msn-transport.jabberstudio.org Enjoy :) --- James ___ jdev mailing list jdev@jabber.org http://mail.jabber.org/mailman/listinfo/jdev
[jdev] PyMSNt 0.10 released! Now with avatars!
Hi everybody. PyMSNt 0.10 has been released and declared stable. Its handling several thousand concurrent users on SAPO (http://messenger.sapo.pt) servers right now. This release brings all sorts of niceties: * Limited ad-hoc commands * More stability * Fixes to contact list handling (this may mean you have to redelete a few contacts as a once-off) * Clustering support (requires an extra component. Join py-transports if you're interested) * MSN7 personal message -> Jabber status, and MSN nick -> Jabber vCard nick * tag in presence for willing clients. * And of course, vCard and (limited) JEP-8 avatars! Grab it from: http://msn-transport.jabberstudio.org Have a look at the developer docs to see how to make sure your client works well with the transport. At the moment the user guide isn't fully up to date. As always, please mail me, (or preferably [EMAIL PROTECTED]) with any bug reports or requests for assistance, or just to let me know its all working :) Enjoy! --- James
[jdev] PyMSNt 0.10.2 released
Hi all. PyMSNt 0.10.2 has been released. This release just has a few critical fixes, and is recommended for all users of 0.10.x http://msn-transport.jabberstudio.org Please remember to read through the documentation on that page. As always, feedback on the program and docs are welcome. Also of note is that the transport now uses SVN. The repository is available for anonymous public access at svn://delx.cjb.net/pymsnt In svn://delx.cjb.net/pymsnt/branches/0_11_x is some preliminary file transfer code. It works for receiving files. I'll be polishing this up so that it works with more clients, and so that you can send as well. If you want to track the progress of file transfer, its probably a good idea to join py-transports. Also, you're best off not checking out trunk :) Enjoy! --- James
[jdev] PyMSNt 0.11 release
Hi all. PyMSNt 0.11 is being released. Major new features: * File transfer support * Removed annoying twistd stuff * Log verbosity controls * Should run better than 0.10.x on Windows * Automatically chooses reactor * Files don't get erased if the filesystem is full Lots of other little things have changed. Be sure to read through the documentation at http://msn-transport.jabberstudio.org This is especially important for client developers, because a new namespace is being used for roster-subsync. The old code had a nasty bug in it, and to avoid clients accidentally blocking all their contacts, you should only pay attention to roster-subsyncs from the new xmlns. Check the developer docs for details. I'll be releasing a new version of webreg to match this release within a few weeks. Including that patch to allow it to work with Google Talk (DNS SRV support) I'll be away and out of touch for a week. I'll get onto any bug reports when I get home. This is the same code from svn trunk that I haven't received any bug reports on for a few weeks though, so it should work just fine :) Please direct any bug reports to the [EMAIL PROTECTED] mailing list. I'm more likely to miss something in jadmin or jdev when I get back. Enjoy! --- James
[jdev] PyMSNt 0.11.1 released!
Hi all. Here are the changes since version 0.11 * * Avatars, and other things, work better with Google Talk * * Adding/removing contacts is more reliable. * * SVN revision is checked on startup and is sent with jabber:iq:version results * * Transport doesn't 'randomly' change status on disconnection from MSN * * Reconnect automatically * * Groupchat timeout is configurable * * Compatibility with Twisted 2.2 * * MSN file send failure message should go to both parties. With correct wording. * * More reliable avatar and file transfers. * * config option for service discovery name (MSN GW1, GW2, etc) * * Use vCard full name (FN) instead of NICKNAME if empty * * Fixed chunked messages * * Fixed error messages missing XMPP_STANZAS * * Fixed exception with MSNConnection.connectionFailed I recommend all users of any version of PyMSNt upgrade to this one. It is the most stable release so far. If you wish to disable file transfer, all you have to do is comment the option in config.xml If you are running any version since 0.10.1 you can run this command to upgrade svn switch svn://delx.cjb.net/pymsnt/tags/pymsnt-0.11.1 Please visit http://delx.cjb.net/pymsnt for more information, or to download a tarball. Make sure you read the documentation on that page, or included with the tarball, for information about setting the transport up. Have fun! :) --- James
[jdev] PyMSNt 0.11.2 released!
PyMSNt 0.11.2 is primarily a bugfix release. All users of PyMSNt are advised to upgrade. 2006-10-18 - PyMSNt 0.11.2 released! * Fixed problem caused by small MSN protocol change * Use md5 hashes for spool directory * Moved avatar to root of spool directory * Supported Twisted versions: 1.3, 2.0.1, 2.1, 2.2, 2.4 * Fixed licensing problems for the Debian people * kqueue apparently doesn't work too well, so it is no longer the default reactor on BSD --- James
Re: [jdev] help us fix jabberd2 and get $500
On Fri, May 23, 2008 at 07:10:10AM -0600, Peter Saint-Andre wrote: > so I'll add back the per-message footers (which BTW prevent a > message from being properly signed with digital signatures, oh well). I've seen some mailing lists where signatures still work. On fink-devel the text appended by sourceforge is just marked as not signed. Mutt says "Warning: Part of this message has not been signed." and marks the signed content with "This data has been signed" and "End of signed data" This works for multipart/signed MIME messages by simply adding an extra text/plain section to the end. --- James pgpg2wQO80axo.pgp Description: PGP signature
