Mailman 3 November 2001 - Twisted

[Twisted-Python] More on XMLRPC
by Magnus Heino 21 Nov '01

21 Nov '01

Hi. I just want to make sure that my conclusions are right: If I create an application using widgets, resource.Resource.getChild is used to traverse the URL to find the callable object. This traverse will always take place if resource.Resource.isLeaf is set to 0. The server.Request.process will then assume that we want to return html, and call render() on the object. After processing all childs, the page will be returned to the browser. The xmlrpc-implementation in xmlrpc.XMLRPC sets isLeaf to 1, which means that that the url wont be traversed, the only thing you can do is to use the _getFunction and traverse to the function manually, and return it. The useage of _getFunction is provided by overriding render. This means that you can't use anything in guard.ResourceGuard to protect your resources. Override isLeaf in the xmlrpc-class, and the url traversal wil take place, but server.Request isn't aware of anything but html, which means that it returns html to xmlrpc, which is not what we want. Not good at all... or am I wrong? Seems as if widgets and guards are only usable if you always, always only want to return html? /Magnus

2 1

[Twisted-Python] ONJava.com: Atomic File Transactions, Part 1 [Nov. 07, 2001]
by Itamar Shtull-Trauring 19 Nov '01

19 Nov '01

This would be very useful to have in Twisted (e.g. we could have a transactional dirdbm): <http://www.onjava.com/pub/a/onjava/2001/11/07/atomic.html>

2 1

[Twisted-Python] Docs/Examples
by Magnus Heino 19 Nov '01

19 Nov '01

Hi. Are there any documentation/examples/tutorials/applications that use twisted available? I can't find anything useful, except for browsing the source-code... is that the way to go, or have I missed something? /Magnus

3 4

[Twisted-Python] Docs/Examples
by Magnus Heino 19 Nov '01

19 Nov '01

Hi. Are there any documentation/examples/tutorials/applications that use twisted available? I can't find anything useful, except for browsing the source-code... is that the way to go, or have I missed something? /Magnus

1 0

[Twisted-Python] Possible patch to twisted.python.hook
by Itamar Shtull-Trauring 18 Nov '01

18 Nov '01

The following patch makes post hook get the result of the hooked function. THis lets you do cool stuff like pre/post-conditions (like synchronization, they can be optional), but it is most likely rather slower than the current hook, which means synchronized classes will be slower. I have a working implementation of pre/post conditions based on this, if anyone's interested. *** hook.py Sun Nov 18 20:02:59 2001 --- hook.py-new Sun Nov 11 11:31:17 2001 *************** *** 48,55 **** --- 48,58 ---- none of the hooks raise an exception). Hooks will be executed in the order in which they are added. + Pre-functions should have the signature (self, args, kwargs). + Post-functions should have the signature (success, result, self, args, kwargs). """ + # System Imports import string *************** *** 99,114 **** hooked_func = """ import %(module)s def %(name)s(*args, **kw): klazz = %(module)s.%(klass)s for preMethod in klazz.%(preName)s: apply(preMethod, args, kw) try: ! return apply(klazz.%(originalName)s, args, kw) ! finally: ! for postMethod in klazz.%(postName)s: ! apply(postMethod, args, kw) """ _PRE = '__hook_pre_%s_%s_%s__' --- 102,128 ---- hooked_func = """ import %(module)s + import sys def %(name)s(*args, **kw): klazz = %(module)s.%(klass)s for preMethod in klazz.%(preName)s: apply(preMethod, args, kw) + try: ! result = apply(klazz.%(originalName)s, args, kw) ! success = 1 ! except: ! success = 0 ! result = sys.last_value ! ! for postMethod in klazz.%(postName)s: ! apply(postMethod, (success, result) + args, kw) ! ! if success: ! return result ! else: ! raise result """ _PRE = '__hook_pre_%s_%s_%s__'

1 0

[Twisted-Python] Win32 select in internet.main
by Itamar Shtull-Trauring 18 Nov '01

18 Nov '01

HappyFool pointed out on IRC that the fix for empty read and write lists in select.select on windows is only done for iterate(), not run(). Perhaps the following patch should be used? Index: twisted/internet/main.py =================================================================== RCS file: /cvs/Twisted/twisted/internet/main.py,v retrieving revision 1.42 diff -c -r1.42 main.py *** twisted/internet/main.py 2001/10/30 23:55:39 1.42 --- twisted/internet/main.py 2001/11/18 11:54:58 *************** *** 440,454 **** del writes[writer] if platform.getType() == 'win32': ! """redefine iterate on WinXX to handle wierd error case when passing empty lists to select """ ! def iterate(): ! for delayed in delayeds: ! delayed.runUntilCurrent() ! if (not reads) and (not writes): ! pass else: ! doSelect(0) class _Win32Waker(styles.Ephemeral): """I am a workaround for the lack of pipes on win32. --- 440,455 ---- del writes[writer] if platform.getType() == 'win32': ! """redefine select on WinXX to handle wierd error case when passing empty lists to select """ ! def _select(r, w, e, timeout=None, _realselect=select.select): ! if (not r) and (not w): ! return [], [], [] else: ! return _realselect(r, w, e, timeout) ! ! select.select = _select ! class _Win32Waker(styles.Ephemeral): """I am a workaround for the lack of pipes on win32.

1 0

[Twisted-Python] Process Producer, Request Consumer
by Kevin Turner 18 Nov '01

18 Nov '01

So somebody asked on the channel today, "how do I get the output of a system command displayed on my .epy web page?" Basically, the  type of thing, if I remember my SSI. Now, given the pieces that are in Twisted, it seems like this should be at most two lines long: 1) create Process 2) designate the web Request as a consumer of the Process. except that I don't know how to do #2 quite so succinctly. Can anyone speak on this subject? - Kevin -- The moon is waxing crescent, 5.7% illuminated, 2.3 days old.

3 4

[Twisted-Python] Re: [Twisted-commits] CVS: Twisted/twisted/web server.py,1.22,1.23
by Itamar Shtull-Trauring 15 Nov '01

15 Nov '01

Itamar Shtull-Trauring wrote: > + > + # if this is a "HEAD" request, we shouldn't return any data > + if self.method == "HEAD": > + self.write = lambda data: None > + I just thought of this - is this gonna break spreadable web servers?

2 1

[Twisted-Python] We are Unable to Reach your Web Page
by adam.brett＠mail.internetseer.com 14 Nov '01

14 Nov '01

There appears to be a problem in reaching the following Web page from Philadelphia, Pennsylvania: http://twistedmatrix.com/pipermail/twisted-python/2001-August/000160.html Error: Time Out This error is most commonly due to an Internet connection problem and does not necessarily indicate that your server is down. As recommended by the robot Guidelines, this email is to explain our research activity and to alert you about the connectivity error we encountered. InternetSeer, the world's largest web site monitoring service, does not store or publish the content of your pages, but rather uses the information to update our ongoing Web Connectivity Study. To learn more about our study results or to request InternetSeer to continue monitoring your Web site in this way, visit: http://scclick.internetseer.com/sitecheck/clickthrough.jsp?I5s57g5f5g5f5j5h… If you find this information about the availability of your Web site useful, we would appreciate your feedback. Should you prefer not to receive these error notices in the future please let us know by replying to this email and placing "remove" in the subject line. You will be removed from receiving further email error notices. Adam Brett Analyst Manager cs-adam.brett(a)mail.internetseer.com InternetSeer.com "Your Remote Web Site Monitor" http://www.internetseer.com PS. We will send you a courtesy email when we are able to reach your Web page again. ##twisted-python(a)twistedmatrix.com##

3 2

[Twisted-Python] non-blocking and non-blocking.
by Kevin Turner 14 Nov '01

14 Nov '01

So, if you run something like the following through manhole: def spam(n=300): for i in xrange(n): print "iiIiiIiI", i spam() the whole mess comes back when the command has finished completion, and receiving that much data seems to take the protocol-decoder a while to decode. Long enough that it can cause a hiccup, or outright stall, in the GUI. So it might be nice to do that processing in incremental bit so it shares time better. In twisted.python terms, it'd be in a Looping construct, or in GTK terms, you'd add it on an idle handler. Gurgf. My brain, it is not doing the thinking at this time, so this post just is to say "don't forget about this; put it on the TODO list or something." -- The moon is waning crescent, 3.3% illuminated, 27.8 days old.

2 1