You can subscribe to this list here.
2002 |
Jan
(2) |
Feb
|
Mar
(3) |
Apr
(8) |
May
|
Jun
(1) |
Jul
|
Aug
|
Sep
|
Oct
(1) |
Nov
|
Dec
|
---|---|---|---|---|---|---|---|---|---|---|---|---|
2003 |
Jan
|
Feb
|
Mar
|
Apr
(2) |
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
2004 |
Jan
|
Feb
|
Mar
(3) |
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
2006 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
(2) |
Sep
(2) |
Oct
(1) |
Nov
(1) |
Dec
(2) |
2007 |
Jan
(1) |
Feb
(1) |
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
From: Boitumelo S. <heg...@oa...> - 2007-02-08 23:13:38
|
Hi, Vivagra 3. 35 Ciavlis 3. 75 Valvium 1. 25 Sovma 1. 15 Ambvien 2. 90 http://www.zonr*x.com Important: Remove "*" in the above link Forbidden Forest, and the rippling sails of the Durmstrang ship. An eagle owl flew through the coil of smoke rising from Hagrids chimney; it soared toward the castle, around the Owlery, and out of sight. Looking |
From: Amandine B. <ch...@pa...> - 2007-01-25 11:24:26
|
Good day, VlAG_GRA $1, 80 ClAL_LIS $3, 00 LEVl_ITRA $3, 35 http://www.rabietichb*com ( Important! Replace * with . ) -- Bagman swore loudly. Damn them! he said, looking quite distracted, and without another word, he Disapparated with a small pop! |
From: Solvej W. <se...@is...> - 2006-10-10 14:25:00
|
Hi, =20 VltAGRA for less http://www.vuigandesinkase.com =20 _____ =20 them. After which, knowing their record for veracity, they probably Im afraid that information is not mine to reveal. the loathsome details re Liokukae, let you see what information we |
From: Mila L. <ash...@am...> - 2006-09-15 18:01:42
|
Hi =20 QUIT OVE d RPA o YIN j G FOR YOU f R PH w ARM b AC o Y =20 S k AV x E up f to 50 m % wi j th http://www.zaxunherwunjsions.com |
From: Ponzio B. <ec...@ab...> - 2006-09-09 01:43:34
|
Hi =20 All yo i ur PH p AR v MAC j Y di y re o ctl g y from the ma e nu r fact r ur l er, Your ch m an z ce to ec b ono k miz a e up w to 50 k % wit p h http://www.xladiaswo.info |
From: Vilfred P. <jul...@c-...> - 2006-08-29 13:30:35
|
Hi, Go v od news for you. =20 P p HARMAC u Y d q irect g ly from the ma t nu z fact i urer, Eco n nom s ize up j to 6 y 0 % wi a th us http://gadefunkerwain.com , d=20 , z=20 , d=20 belt. And spoke. Svinjar takes care of his people. Svinjar is your friend. Svinjar brings you The Stainless Steel Rats and their magic music. Now let us |
From: Kyriakos S. <bou...@bu...> - 2006-08-28 09:38:32
|
Hi, Goo i d news for you. =20 PHA k RM l ACY di n rect r ly from the m l anuf e acture o r, Eco w nomiz c e up t v o 60 l % w k ith us http://puteryungan.com , u=20 , w=20 , d=20 Excellent. If I have only thirty days to go shouldnt you be telling me about what you want me to do? In good time, he said, sitting down across from me. Captain Varod |
From: Kamruddin Kareem-Q. <Kar...@mo...> - 2004-03-18 21:56:49
|
Shilad, Thanks for the response. The PyErr_Print() prints out the following: SystemError: bad argument to internal function I also realized that I was using Python 1.5 instead of Python 1.5.2. Could that be an issue? When I moved to Python 1.5.2, I got the following error Traceback (innermost last): File "examples/examples.py", line 10, in ? import xmlrpc File "./xmlrpc.py", line 37, in ? import _xmlrpc ImportError: dynamically linked modules not supported Any idea what is it refering to? Thanks Kareem -----Original Message----- From: Shilad Sen [mailto:sh...@so...] Sent: Thursday, March 18, 2004 12:44 AM To: Kamruddin Kareem-QKK002 Cc: py-...@li... Subject: Re: [py-xmlrpc-devel] problem running py-xmlrpc-0.8.8.3 Hi Kareem, It's possible that PyMethod_New isn't implemented the same way in older versions of Python. Can you add a call to "PyErr_Print();" after the PyMethod_New() to see what the error message is? Thanks, Shilad On Wed, Mar 17, 2004 at 05:44:24PM -0600, Kamruddin Kareem-QKK002 wrote: > Hi folks, > > I downloaded the source package for py-xmlrpc-0.8.8.3 and compiled it on > the SunOS platform with python 1.52. When I tried to run the examples.py > module, I got the following error > > "rpcFaultStr is NULL in xmlrpcInit" > > A quick debug effort points the source to the following code in > rpcFaultClass(void) function in rpcFault.c > > meth = PyMethod_New(func, NULL, klass); > if (meth == NULL) > return (NULL); > > Can anyone please suggest what could be wrong? I am just getting familiar > with the Python/C APIs. Any help will be greatly appreciated. > > Thanks > Kareem |
From: Shilad S. <sh...@so...> - 2004-03-18 06:44:21
|
Hi Kareem, It's possible that PyMethod_New isn't implemented the same way in older versions of Python. Can you add a call to "PyErr_Print();" after the PyMethod_New() to see what the error message is? Thanks, Shilad On Wed, Mar 17, 2004 at 05:44:24PM -0600, Kamruddin Kareem-QKK002 wrote: > Hi folks, > > I downloaded the source package for py-xmlrpc-0.8.8.3 and compiled it on > the SunOS platform with python 1.52. When I tried to run the examples.py > module, I got the following error > > "rpcFaultStr is NULL in xmlrpcInit" > > A quick debug effort points the source to the following code in > rpcFaultClass(void) function in rpcFault.c > > meth = PyMethod_New(func, NULL, klass); > if (meth == NULL) > return (NULL); > > Can anyone please suggest what could be wrong? I am just getting familiar > with the Python/C APIs. Any help will be greatly appreciated. > > Thanks > Kareem |
From: Kamruddin Kareem-Q. <Kar...@mo...> - 2004-03-17 23:44:29
|
Hi folks, I downloaded the source package for py-xmlrpc-0.8.8.3 and compiled it on the SunOS platform with python 1.52. When I tried to run the examples.py module, I got the following error "rpcFaultStr is NULL in xmlrpcInit" A quick debug effort points the source to the following code in rpcFaultClass(void) function in rpcFault.c meth = PyMethod_New(func, NULL, klass); if (meth == NULL) return (NULL); Can anyone please suggest what could be wrong? I am just getting familiar with the Python/C APIs. Any help will be greatly appreciated. Thanks Kareem |
From: Jesus C. A. <jc...@ar...> - 2003-04-24 19:37:54
|
> Release 0.8.8.3 is based on various changes to the cvs archives over > the last year. At last!! :-). -- Jesus Cea Avion _/_/ _/_/_/ _/_/_/ jc...@ar... http://www.argo.es/~jcea/ _/_/ _/_/ _/_/ _/_/ _/_/ _/_/ _/_/ _/_/_/_/_/ PGP Key Available at KeyServ _/_/ _/_/ _/_/ _/_/ _/_/ "Things are not so easy" _/_/ _/_/ _/_/ _/_/ _/_/ _/_/ "My name is Dump, Core Dump" _/_/_/ _/_/_/ _/_/ _/_/ "El amor es poner tu felicidad en la felicidad de otro" - Leibniz |
From: Shilad S. <sh...@so...> - 2003-04-21 19:17:22
|
Greetings, After over a year, I have finally put a new py-xmlrpc release together. Release 0.8.8.3 is based on various changes to the cvs archives over the last year. Among the interesting new features: - Ability to postpone server responses to a client request for an indefinite amount by rasing a xmlrpc.postpone exception. - A drop in compatiblity wrapper for xmlrpclib; "import pyxmlrpclib" instead of "import xmlrpclib". - Support for custom logging objects I've setup a win32 build environment (which was one of the difficulties in the past), so I expect releases to occur more frequently in the future. Please try it out and let me know what you think! Shilad sh...@so... Changelog: Public -------- - improved documentation in xmlrpc.py for xmlrpc.server class - created xmlrpc.postpone error allowing for delayed responses - created queueResponse to finish responding to a delayed response - created queueRequest to finish responding to a delayed request - create response example in examples.py - create response method in exampleServer in examples.py - added support for file logging object (thanks to Ross Brattain) - added buildCall, parseCall to help support pyxmlrpclib - added pyxmlrpclib.py for drop-in compatibility w/ xmlrpclib - added value attribute for DateTime object (6-tuple, not a string) - added data attribute for Base64 object Internal -------- - used buffConstant for appending constant length strings in rpcUtils.c (thanks to Christian Gafton). This speeds things up by 20% in many cases. - Fixed signed/unsigned problem and a 64 bit problem that could lead to weird bugs. Thanks to Marcin Cieslak for the patch. - Fixed windows server typo bug that lead to "No such file or directory" in bindAndListen. - Fixed off-by-one bug in parsing entities starting with '&#' |
From: Michael L. <m....@mi...> - 2002-10-26 15:15:54
|
-----Ursprüngliche Nachricht----- Von: Michael Linke Gesendet: Samstag, 26. Oktober 2002 17:03 An: crj...@us...; sh...@us... Betreff: :/ Diploma Thesis Problem: Strange Windows Problem with bind Hi List, a warm welcome from southwest of germany on a stormy saturday :) I have a nasty problem with some code for my diploma thesis (MySQL-RPC Sync betwenn tweo Databases). Program runs under linux fine, but on several out-of-the box installations it brings always the same error :/ I have to install it on a 98 Machine at my University (TU Ilmenau) on Wednesday, so do you have an idea why the machines produced this bind-error ? It should be pre-defined (as it is in linux) ... For any help i'am thankful ... :/ With best regards Michael P. Linke P.S.: Could this be any trouble, because of the "other" xml-rpc in the python22 installer included ? Theoretically it must not, but perhaps you know more ..... :-| Here is the code .... Codeline in validate.py: ----snip----- s.bindAndListen(PORT) ----snap----- Traceback (most recent call last): File "C:\Python22\Lib\site-packages\de\tuilmenau\replicator\examples\validate.py" , line 98, in ? main() File "C:\Python22\Lib\site-packages\de\tuilmenau\replicator\examples\validate.py" , line 30, in main s.bindAndListen(PORT) File "C:\PYTHON22\Lib\site-packages\xmlrpc.py", line 105, in bindAndListen def bindAndListen(self, port, queue=5): xmlrpc.error: (0, 'Error') ---------------------------------------------------------------------------- - Codeline in xmlrpc.py ----snip----- def bindAndListen(self, port, queue=5): self._o.bindAndListen(port, queue) ----snap----- _______________________________________________ m....@mi..., http://www.mindphazer.de MINDPHAZER.DE Försterstraße 24 66111 Saarbruecken Germany fon: +49 (0) 681 / 9 50 85 - 94 fax: +49 (0) 681 / 9 50 85 - 93 |
From: Stanley A. K. <sk...@cp...> - 2002-06-05 03:34:25
|
I tried to install the latest py-xmlrpc rpms on my Red Hat 7.2 system and all I got was an error message that "read manifest failed." It happened with both the binary and the src rpm. I also tried the next earlier binary and got the same message. Stan Klein |
From: Conrad S. <co...@he...> - 2002-04-24 18:30:58
|
Hi Shilad Here is a little patchlet to automate the choice of python version in the python/Makefile. If someone has a more complex setup (e.g. multiple python versions, or a user area install) that person would have to edit the makefile anyway, so this helps the common case a lot, without hurting more esoteric build environments. Cheers! Conrad -- *-----------------------------------------* | Conrad Steenberg | | Caltech, Mail Code 356-48 | | Pasadena, CA, 91125 | | e-mail: co...@he... | | Tel: (626) 395-8758 | *-----------------------------------------* |
From: Conrad S. <co...@he...> - 2002-04-23 23:28:55
|
Hi Shilad I started looking at ehs, and adding at one feature that I need: namely simple authentication. Attached is a patch that adds simple authentication headers to a request. It works here, but the transport may have to handle 401 responses (authentication required) more gracefully. The patch modifies the file examples/xmlrpc_client_example.c to use authentication, but this should probably become a separate example. Also, please look at the changes in http/ where the pasword and username gets base64 encoded, I'm not sure who owns returned strings for some of the calls, so it may leak memory in its current implementation. Comments welcome! Conrad On Thu, 2002-04-18 at 16:12, Shilad Sen wrote: > Conrad, > > I'm glad to here that you find the library useful. > > The main motivation for ehs was so that we could write services inhouse in C > that are fast. I took a look at several other options of C libraries... for > xml-rpc I looked at epinions and Eric Kidd's versions. > > Eric Kidd's xmlrpc library is pretty closely tied into libwww, which I > dislike (it's a bit big and clunky for my tastes). > > Epinions library is quite good, but I would need to code my own transport > libraries. There is no really good C server library that fits nicely into > complicated event handling programs (that I've found). > > So that's the motivation. > > As far as significant features... > > - I plan on including better support for HTTP including HTTPS, different > encodings, etc. > - I'll include an OPTION to use expat as the parser > - It will be available as a standalone C library for C programs, or other > languages can have C bindings > - It will be COMPLETELY backwards compatible with the xmlrpc.py module > > I'm even thinking about trying to go down the soap path, although I know > that will involve writing a schema validator. > > I actually think now that I may release another version or two under the > old codebase, though. > > Shilad > > > > > > Hi Shilad > > > > What are the motivation/significant features of ehs? > > > > Btw, I use py-xmlrpc every day and having gone through doing xmlrpc in > > C++ in the last two days\I appreciate it even more ;-) > > > > Cheers! > > > > Conrad > > > > On Fri, 2002-04-12 at 07:22, Shilad Sen wrote: > > > Hello, > > > > > > I finally got around to setting up a cvs archive for the py-xmlrpc library. > > > > > > There are two packages available. Py-xmlrpc, is the latest (unstable) > > > release under the old codebase. Ehs is the latest (really unstable) new > > > codebase. Please peruse the new codebase and let me know what you think. > > > > > > You can download the NEW stuff by running the following commands (for redhat > > > bash users). > > > > > > export CVS_RSH=ssh > > > cvs -d:pserver:ano...@cv...:/cvsroot/py-xmlrpc co ehs > > > > > > if you want to get the old codebase for any reason, the you would use: > > > > > > cvs -d:pserver:ano...@cv...:/cvsroot/py-xmlrpc co py-xmlrpc > > > > > > The new codebase is far along, but not well documented. I need suggestions on > > > it! > > > > > > Shilad > > _______________________________________________ > py-xmlrpc-devel mailing list > py-...@li... > https://lists.sourceforge.net/lists/listinfo/py-xmlrpc-devel -- *-----------------------------------------* | Conrad Steenberg | | Caltech, Mail Code 356-48 | | Pasadena, CA, 91125 | | e-mail: co...@he... | | Tel: (626) 395-8758 | *-----------------------------------------* |
From: Conrad S. <co...@he...> - 2002-04-19 15:11:11
|
On Thu, 2002-04-18 at 16:12, Shilad Sen wrote: > Conrad, > > I'm glad to here that you find the library useful. > > The main motivation for ehs was so that we could write services inhouse in C > that are fast. I took a look at several other options of C libraries... for > xml-rpc I looked at epinions and Eric Kidd's versions. > > Eric Kidd's xmlrpc library is pretty closely tied into libwww, which I > dislike (it's a bit big and clunky for my tastes). > > Epinions library is quite good, but I would need to code my own transport > libraries. There is no really good C server library that fits nicely into > complicated event handling programs (that I've found). Yes, xmlrpc-c is also unsupported AFAIK as Eric is busy with other things. Also, I wrote a simple client to my py-xmlrpc/apache-based server last night (oh the joys of jet-lag ;-) and found it seriously lacking performance - on my laptop connecting to a localhost server my python-based client does about 130 connections/sec, while the xmlrpc-c client does only 16/sec. > > So that's the motivation. > > As far as significant features... > > - I plan on including better support for HTTP including HTTPS, different > encodings, etc. > - I'll include an OPTION to use expat as the parser > - It will be available as a standalone C library for C programs, or other > languages can have C bindings > - It will be COMPLETELY backwards compatible with the xmlrpc.py module > > I'm even thinking about trying to go down the soap path, although I know > that will involve writing a schema validator. > > I actually think now that I may release another version or two under the > old codebase, though. Sounds good. Cheers Conrad > > Shilad > > > > > > Hi Shilad > > > > What are the motivation/significant features of ehs? > > > > Btw, I use py-xmlrpc every day and having gone through doing xmlrpc in > > C++ in the last two days\I appreciate it even more ;-) > > > > Cheers! > > > > Conrad > > > > On Fri, 2002-04-12 at 07:22, Shilad Sen wrote: > > > Hello, > > > > > > I finally got around to setting up a cvs archive for the py-xmlrpc library. > > > > > > There are two packages available. Py-xmlrpc, is the latest (unstable) > > > release under the old codebase. Ehs is the latest (really unstable) new > > > codebase. Please peruse the new codebase and let me know what you think. > > > > > > You can download the NEW stuff by running the following commands (for redhat > > > bash users). > > > > > > export CVS_RSH=ssh > > > cvs -d:pserver:ano...@cv...:/cvsroot/py-xmlrpc co ehs > > > > > > if you want to get the old codebase for any reason, the you would use: > > > > > > cvs -d:pserver:ano...@cv...:/cvsroot/py-xmlrpc co py-xmlrpc > > > > > > The new codebase is far along, but not well documented. I need suggestions on > > > it! > > > > > > Shilad > > _______________________________________________ > py-xmlrpc-devel mailing list > py-...@li... > https://lists.sourceforge.net/lists/listinfo/py-xmlrpc-devel -- |
From: Shilad S. <sh...@so...> - 2002-04-19 05:20:55
|
Hello all, Based on popular demand I added the ability for the server to "postpone" a response to a client if you are worrying about deadlock for a particular request. The idea is the following: You save away the client (src paramter), and raise a xmlrpc.postpone exception in the server handler. Then, whenever you are ready, you call either one of the following: server.queueResponse(client, result) server.queueFault(client, faultCode, faultString) There is an example in examples.py. All of this is checked in to the cvs archives. You can get at it by following the instructions at this page: http://sourceforge.net/cvs/?group_id=23992 This will be available in py-xmlrpc-0.8.9, whenever that is available. Any comments or suggestions? Shilad |
From: Shilad S. <sh...@so...> - 2002-04-18 23:14:26
|
Conrad, I'm glad to here that you find the library useful. The main motivation for ehs was so that we could write services inhouse in C that are fast. I took a look at several other options of C libraries... for xml-rpc I looked at epinions and Eric Kidd's versions. Eric Kidd's xmlrpc library is pretty closely tied into libwww, which I dislike (it's a bit big and clunky for my tastes). Epinions library is quite good, but I would need to code my own transport libraries. There is no really good C server library that fits nicely into complicated event handling programs (that I've found). So that's the motivation. As far as significant features... - I plan on including better support for HTTP including HTTPS, different encodings, etc. - I'll include an OPTION to use expat as the parser - It will be available as a standalone C library for C programs, or other languages can have C bindings - It will be COMPLETELY backwards compatible with the xmlrpc.py module I'm even thinking about trying to go down the soap path, although I know that will involve writing a schema validator. I actually think now that I may release another version or two under the old codebase, though. Shilad > > Hi Shilad > > What are the motivation/significant features of ehs? > > Btw, I use py-xmlrpc every day and having gone through doing xmlrpc in > C++ in the last two days\I appreciate it even more ;-) > > Cheers! > > Conrad > > On Fri, 2002-04-12 at 07:22, Shilad Sen wrote: > > Hello, > > > > I finally got around to setting up a cvs archive for the py-xmlrpc library. > > > > There are two packages available. Py-xmlrpc, is the latest (unstable) > > release under the old codebase. Ehs is the latest (really unstable) new > > codebase. Please peruse the new codebase and let me know what you think. > > > > You can download the NEW stuff by running the following commands (for redhat > > bash users). > > > > export CVS_RSH=ssh > > cvs -d:pserver:ano...@cv...:/cvsroot/py-xmlrpc co ehs > > > > if you want to get the old codebase for any reason, the you would use: > > > > cvs -d:pserver:ano...@cv...:/cvsroot/py-xmlrpc co py-xmlrpc > > > > The new codebase is far along, but not well documented. I need suggestions on > > it! > > > > Shilad |
From: Conrad S. <co...@he...> - 2002-04-16 07:30:12
|
Hi Shilad What are the motivation/significant features of ehs? Btw, I use py-xmlrpc every day and having gone through doing xmlrpc in C++ in the last two days\I appreciate it even more ;-) Cheers! Conrad On Fri, 2002-04-12 at 07:22, Shilad Sen wrote: > Hello, > > I finally got around to setting up a cvs archive for the py-xmlrpc library. > > There are two packages available. Py-xmlrpc, is the latest (unstable) > release under the old codebase. Ehs is the latest (really unstable) new > codebase. Please peruse the new codebase and let me know what you think. > > You can download the NEW stuff by running the following commands (for redhat > bash users). > > export CVS_RSH=ssh > cvs -d:pserver:ano...@cv...:/cvsroot/py-xmlrpc co ehs > > if you want to get the old codebase for any reason, the you would use: > > cvs -d:pserver:ano...@cv...:/cvsroot/py-xmlrpc co py-xmlrpc > > The new codebase is far along, but not well documented. I need suggestions on > it! > > Shilad > > _______________________________________________ > py-xmlrpc-devel mailing list > py-...@li... > https://lists.sourceforge.net/lists/listinfo/py-xmlrpc-devel -- |
From: Isaac C. <cl...@da...> - 2002-04-13 12:22:58
|
Hi folks, I just started working with py-xmlrpc a few days ago, and found it rather simple and intuitive to use. The most attracting stuff of it to me is the non-block IO -- I still remember the old times when I was using Java RMI, where parallel method invocations cant be achieved other than with threads, which proved quite a resource hog. Now with pr-xmlrpc around, parallelism is as simple and efficient as some select() calls. However, what I'm now worrying about is the authorization mechanism of pyxmlrpc, since in the examples.py coming with py-xmlrpc, user name and password is simply hard coded into the source. Does py-xmlrpc have some other, and hopefully better, mechanism of authorization? Or, can it make use of some external mechanism, say SSL, to do the dirty job? I know you guys are busy, just point me to some material or URL is greatly appreciated. Thanks :) -Clay |
From: Shilad S. <sh...@so...> - 2002-04-12 14:22:38
|
Hello, I finally got around to setting up a cvs archive for the py-xmlrpc library. There are two packages available. Py-xmlrpc, is the latest (unstable) release under the old codebase. Ehs is the latest (really unstable) new codebase. Please peruse the new codebase and let me know what you think. You can download the NEW stuff by running the following commands (for redhat bash users). export CVS_RSH=ssh cvs -d:pserver:ano...@cv...:/cvsroot/py-xmlrpc co ehs if you want to get the old codebase for any reason, the you would use: cvs -d:pserver:ano...@cv...:/cvsroot/py-xmlrpc co py-xmlrpc The new codebase is far along, but not well documented. I need suggestions on it! Shilad |
From: Shilad S. <sh...@so...> - 2002-03-21 18:37:29
|
There was an error in the linking order of some of the object files in release 0.8.8 that resulted in a segfault on module import. I have uploaded files for a bugfix release version (0.8.8.1). Let me know if you have any questions. Shilad |
From: Conrad S. <co...@he...> - 2002-03-19 01:36:45
|
On Mon, 2002-03-18 at 17:03, Shilad Sen wrote: > The latest version of py-xmlrpc has been released. There are many feature > improvements (mostly based on requests), and a few bugfixes. > > This is probably the last releases (outside of a bugfix release) under this > codebase. I have the new codebase ready to go, and will post it as 'unstable' > soon. > > Let me know if you have any questions / comments! Hi Shilad Thanks for the new release, it is appreciated :-) Attached is a patch that I use to define a new method, called parseRequestXML that takes input from a string and parses the XML-RPC call it contains. This method, along with buildResponse enables py-xmlrpc to be used in mod_python in Apache for building a robust and scalable application server. E.g. in the handler function that mod_python uses: def handler(req): pw = req.get_basic_auth_pw() user = req.connection.user data=req.read() decoded=xmlrpc.parseRequestXML(data) # "decoded" now contains the method name [0] and arguments [1] # Call the method, and receive the return values in e.g. retval response=xmlrpc.buildResponse(retval) req.write(response) The patch is against 0.8.7, but applies cleanly against 0.8.8 as well. Using a stock Apache/mod_python from Mandrake 8.1 on an PIII/933 the server can handle about 200 calls/s with each call serviced by a different process. This setup is very stable and provides crash protection for serving different clients. Anyway, I would like to contribute the patch to py-xmlrpc if this seems useful. Cheers! Conrad |
From: Shilad S. <sh...@so...> - 2002-03-19 01:03:21
|
The latest version of py-xmlrpc has been released. There are many feature improvements (mostly based on requests), and a few bugfixes. This is probably the last releases (outside of a bugfix release) under this codebase. I have the new codebase ready to go, and will post it as 'unstable' soon. Let me know if you have any questions / comments! Shilad |