Mailman 3 April 2004 - Twisted-web

vhost.NameVirtualHost implementation for Nevow
by David Reid April 10, 2004

April 10, 2004

Last night at about 4 in the morning I was saddened to discover that vhost.NameVirtualHost breaks horribly when using Nevow. So in a sleep deprived stupor I attempted to reimplement it and failed, deleted the results. Woke up this morning around 2:30pm and made it work. This JUST implements NameVirtualHost which is the only part of vhost I use. Anyway, here it is *drum roll please* *clicks the attach button, selects the file and hits ok* -David

2 1

really beginner example
by Ellers April 9, 2004

April 9, 2004

Hello all, I've been getting into Twisted apps recently and saw in the docs that use of Woven is not recommended and that Nevow should be used instead. Thats cool. I've been trying to get into it, and have read lots of docs and examples, but its seriously hard going. I'm not trying to dump on anyone, but the examples don't start out with really simple building blocks then build on that into something big and usable as-is. Something I can say "I need to do a multistage form - ah theres an example", and "I need to store user sessions in a persistent way, so they're there after a database restart - ah an example..." etc. That is, start with really really basic apps, adding extra details and features that someone will commonly want. For example, it might start with a really simple html template example, using minimal Nevow stuff - just a renderer or at most a page, and everything else from Twisted. Maybe this isn't the "done" thing, but I think it would be really helpful to someone coming from now understanding Twisted classes to see how to use a minimal set of Nevow classes to do something cool with Nevow. I would've thought Nevow would work well in this way, but tell me if I'm wrong and I should really be using the whole Nevow framework, not just the xhtml templating part. Right now I want to use my own Site object, not a subclass of the Nevow site object, and use the Nevow classes to generate html based on templates from files. I'm sure this is possible, but I just can't figure it out. So, here is my question... I have this in a file "hello_template.html": <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd"> <html lang="en" xml:lang="en" xmlns="http://www.w3.org/1999/xhtml"> <head> <title nevow:data="title">This is title</title> </head> <body> This page is called </body> </html> and I have the following python code. The function demo1() does what I expect, but what I really want is for demo2() to print to stdout the xhtml after it has been substituted with the title. << from nevow import rend class HelloPage(rend.Page): docFactory = rend.htmlfile('hello_template.html') def data_title(self, context, data): return "Hello Nevow World" def data_empty(self, context, data): return [] def someMethodToWriteToAFileLikeObject( self, writer ): writer.write( "not yet implemented\n" ) def demo1( ): """This will print the html just as it is in the template file - cool""" import sys d = rend.htmlfile('hello_template.html', '../shared/Assets/template') d.load() print "\nDocument:\n" d.getDoc( ).writexml(sys.stdout) print "\n" def demo2( ): """This does not work yet, but what I want it to do, is put the substituted title into the template then return the composite result. """ import sys p = HelloPage( ) p.someMethodToWriteToAFileLikeObject( sys.stdout ) if __name__ == "__main__": demo1( ) demo2( ) >> I know that in normal usage there isn't much point doing sys.stdout.write with the xhtml content, but there has to be some room between that extreme and using a framework where I can't even tell how the content is getting sent back to the http stream! Thanks to all for any suggestions, and for your patience with this beginner... Ellers

5 10

[PATCH] Posting forms to deferred configurables
by Alex Levy April 9, 2004

April 9, 2004

Attached is a patch that allows posting a form to a deferred configurable. However, it doesn't cover other places where locateConfigurable is used, because I didn't want to bite off more than I can chew. My use case is strange: sometimes a page that represents a directory receives a form post that was intended for its index page. Example: 1. user browses "/somedir/", posts form 2. form post goes to "/somedir/freeform_post..." 3. form post really should go to "/somedir/index/freeform_post..." So the directory page first needs to call self.locateChild, which returns a deferred, which means the configurable has to be deferred as well. So I'd like to know what the rest of you think about this. Maybe there's a better way to handle all of this. Maybe not? Discuss. -- Alex Levy WWW: http://mesozoic.geecs.org "Never let your sense of morals prevent you from doing what is right." -- Salvor Hardin, Isaac Asimov's _Foundation_

1 0

Re: [PATCH] nevow.static.File directory listing
by Lee Harr April 6, 2004

April 6, 2004

>Hmm, that's strange. Here /images is redirected to /images/ and I get a >directory list as expected. Could you paste the entire stack trace >please. > Ok, I got the latest code from svn, and it works fine for me also. _________________________________________________________________ Add photos to your e-mail with MSN 8. Get 2 months FREE*. http://join.msn.com/?page=features/featuredemail

1 0

Re: [PATCH] nevow.static.File directory listing
by Lee Harr April 6, 2004

April 6, 2004

>Everything at divmod recently moved to subversion. Use the following: > > svn co svn://divmod.org/svn/Nevow/trunk Nevow > Ah. Ok. I am getting the cvs link from the bottom of the page here: http://www.divmod.org/Home/Projects/Nevow/ >Hmm, that's strange. Here /images is redirected to /images/ and I get a >directory list as expected. Could you paste the entire stack trace >please. > 2004/04/05 22:10 EDT [HTTPChannel,2,127.0.0.1] Traceback (most recent call last): File "/usr/local/lib/python2.3/site-packages/twisted/protocols/http.py", line 554, in requestReceived self.process() File "/usr/local/lib/python2.3/site-packages/nevow/appserver.py", line 130, in process return self.site.getResourceFor( File "/usr/local/lib/python2.3/site-packages/twisted/internet/defer.py", line 189, in addCallback callbackKeywords=kw) File "/usr/local/lib/python2.3/site-packages/twisted/internet/defer.py", line 180, in addCallbacks self._runCallbacks() --- <exception caught here> --- File "/usr/local/lib/python2.3/site-packages/twisted/internet/defer.py", line 313, in _runCallbacks self.result = callback(self.result, *args, **kw) File "/usr/local/lib/python2.3/site-packages/nevow/appserver.py", line 133, in <lambda> lambda resource: inevow.IResource(resource).renderHTTP(self) File "/usr/local/lib/python2.3/site-packages/nevow/static.py", line 290, in renderHTTP f = self.openForReading() File "/usr/local/lib/python2.3/site-packages/nevow/static.py", line 253, in openForReading return self.fp.open() File "/usr/local/lib/python2.3/site-packages/twisted/python/filepath.py", line 138, in open return open(self.path, mode+'b') exceptions.IOError: [Errno 21] Is a directory 2004/04/05 22:10 EDT [HTTPChannel,2,127.0.0.1] html is not a string: None 2004/04/05 22:10 EDT [HTTPChannel,2,127.0.0.1] 127.0.0.1 - - [06/Apr/2004:02:10:53 +0000] "GET /images HTTP/1.1" 500 8478 "-" "Mozilla/5.0 (compatible; Konqueror/3.2; FreeBSD) (KHTML, like Gecko)" _________________________________________________________________ Add photos to your e-mail with MSN 8. Get 2 months FREE*. http://join.msn.com/?page=features/featuredemail

1 0

Re: [PATCH] nevow.static.File directory listing
by Lee Harr April 5, 2004

April 5, 2004

>Aha! All becomes clear. > >>If there is a better way, I am certainly willing to use it. > >Yep, it's much simpler too :) > >Although locateChild() is the "official" API for URL traversal, >rend.Page adds some more convenient ways of specifying children. The one >you probably want is so define a child_images attribute on your root >page's class: > > from twisted.application import internet, service > from twisted.web import static > > from nevow import appserver > from nevow import rend > from nevow import static > > class RootPage(rend.Page): > > # Mount the 'images' directory at the '/images' URl > child_images = static.File('images') > > docFactory = rend.htmlstr( > ''' > <html> > <body> > <img src="/images/TwistedLogoFull.png" /> > Blah, blah, blah > </body> > </html> > ''' > ) > > application = service.Application('static') > webServer = internet.TCPServer( > 8000, > appserver.NevowSite(RootPage())) > webServer.setServiceParent(application) > Ok. Cool. That looks nice. With that example, and unpatched nevow, I still get an error when trying to browse to http://localhost/images (note, no trailing slash) The error is the same as in my last post ... File "/usr/local/lib/python2.3/site-packages/twisted/python/filepath.py", line 138, in open return open(self.path, mode+'b') exceptions.IOError: [Errno 21] Is a directory Also... trying to get the Nevow module from anonymous cvs is giving me an error: $cvs -d:pserver:anon@divmod.org:/cvs co Nevow cvs server: cannot find module `Nevow' - ignored cvs [checkout aborted]: cannot expand modules Thank you for your time. _________________________________________________________________ STOP MORE SPAM with the new MSN 8 and get 2 months FREE* http://join.msn.com/?page=features/junkmail

3 2

Re: [PATCH] nevow.static.File directory listing
by Lee Harr April 5, 2004

April 5, 2004

>Can you explain *why* you have the following code (copied from your >original post): > > > from twisted.application import service > from twisted.application import internet > > > class ADir(rend.Page): > def locateChild(self, request, segments): > path = '/'.join(segments) > return static.File(path), () > > > application = service.Application('ADirlist') > webservice = internet.TCPServer( > 8080, > appserver.NevowSite(ADir()) > ) > webservice.setServiceParent(application) > > >What is the use case? Why not let static.File handle everything? > Well... it is certainly possible that I am going about this all wrong. I was trying to figure out how to serve images from the nevow server to go along with my web pages. This was the first way I figured to do that. If there is a better way, I am certainly willing to use it. Is this code flawed in some way? When I try to run it without that patch, I get an error like this trying to load the root url: 2004/04/04 20:52 EDT [HTTPChannel,0,127.0.0.1] Traceback (most recent call last): File "/usr/local/lib/python2.3/site-packages/twisted/protocols/http.py", line 554, in requestReceived self.process() File "/usr/local/lib/python2.3/site-packages/nevow/appserver.py", line 130, in process return self.site.getResourceFor( File "/usr/local/lib/python2.3/site-packages/twisted/internet/defer.py", line 189, in addCallback callbackKeywords=kw) File "/usr/local/lib/python2.3/site-packages/twisted/internet/defer.py", line 180, in addCallbacks self._runCallbacks() --- <exception caught here> --- File "/usr/local/lib/python2.3/site-packages/twisted/internet/defer.py", line 313, in _runCallbacks self.result = callback(self.result, *args, **kw) File "/usr/local/lib/python2.3/site-packages/nevow/appserver.py", line 133, in <lambda> lambda resource: inevow.IResource(resource).renderHTTP(self) File "/usr/local/lib/python2.3/site-packages/nevow/static.py", line 290, in renderHTTP f = self.openForReading() File "/usr/local/lib/python2.3/site-packages/nevow/static.py", line 253, in openForReading return self.fp.open() File "/usr/local/lib/python2.3/site-packages/twisted/python/filepath.py", line 138, in open return open(self.path, mode+'b') exceptions.IOError: [Errno 21] Is a directory 2004/04/04 20:52 EDT [HTTPChannel,0,127.0.0.1] html is not a string: None _________________________________________________________________ Tired of spam? Get advanced junk mail protection with MSN 8. http://join.msn.com/?page=features/junkmail

3 3

Performance of twisted web with HTTP/1.1 vs. HTTP/1.0
by Jason Sibre April 5, 2004

April 5, 2004

Hi folks, I'm using Quixote, and have recently started using Twisted instead of/in addition to Medusa to add SSL support to a web app. Ever since doing so, I've had the feeling that Twisted was slower than Medusa, but whenever I ran profiles of the app using ab (apache bench), the numbers indicated that twisted was comparable to medusa (a little slower, but only very slightly, and not enough that I should notice it 'seat of the pants', and not surprising considering it seems to be a more robust implementation). Today I noticed the problem again, (on a page that had lots of little icon type images) and decided that I *wasn't* going insane, it was definitely slower. Some more ab tests, and performance looked comparable to Medusa. Mind you, even though I'm adding Twisted because I want the SSL support, my tests are not using that SSL. In fact, at one stage, I even removed the code from my start script that enabled SSL to make sure that wasn't the problem. ab uses HTTP/1.0 to test an app, and of course most browsers use HTTP/1.1. I set up another, more sophisticated, test suite (ACT), where I could control the protocol, and ran the tests through it, and I confirmed that the protocol level was definitely the culprit. I ran the test using HTTP/1.0, and again using HTTP/1.1 (nothing else changed). Using HTTP/1.0, twisted served a small image about 131 times per second, but using HTTP/1.1 it dropped to about 5 requests per second. Yes, "Five." Has anyone else seen this type of behavior? Any ideas if it's something I may be doing wrong, or is this a known issue with Twisted? I'm running Twisted 1.2.0 (latest stable release) on Linux with Python 2.3.3 and have looked in the archives of twisted-web and twisted-users, and haven't seen anything about this. I'm not a member of this list, so if you have a reply, please cc me directly. Thanks, Jason Sibre

3 2

IRenderer interface for Nevow
by Matt Goodall April 5, 2004

April 5, 2004

Hi, While testing out an idea, I noticed that I was implementing IRendererFactory just to turn a class instance into a renderer. For instance: class ISummaryView(compy.Interface): pass class FooSummaryView(compy.Adapter): __implements__ = ISummaryView, IRendererFactory def __call__(self, context, data): return whatever That didn't make sense to me and I also had no reason to add IRendererFactory's renderer() method. I then noticed that quite a few classes (freeform in particular) that implement IRendererFactory were actually just renderers and didn't have or need a renderer() method either. So ... I created a IRenderer marker interface and changed most uses of IRendererFactory to IRenderer. See the attached diff (against svn trunk). It seems to work (i.e. no more tests fail than usual ;-)) but I didn't want to commit it without some feedback first. You'll notice that a couple of test_flatstan tests changed slightly to make them pass. I don't think the renderer function signature was correct in the context but please shout if you can think of a legitimate reason why it should be supported. James Knight (aka foom) suggested that IRenderer may not be needed at all and that adapting anything callable would probably be enough. That is probably correct but a) it's too early in the morning to think about it and b) I'd rather make the change it two steps. OK, let me know what you think to all this. Cheers, Matt -- Matt Goodall, Pollenation Internet Ltd w: http://www.pollenation.net e: matt(a)pollenation.net Any views expressed are my own and do not necessarily reflect the views of my employer.

1 0

Re: [PATCH] nevow.static.File directory listing
by Lee Harr April 4, 2004

April 4, 2004

>> >> I was getting an infinitely recursive redirect >> > >> > >> > Actually, this patch (to nevow.static.File) will also >> > take care of the redirect: >> > >> > >> > >> > 274c274,281 >> > < return self.redirect(request) >> > --- >> > >> >> if not request.uri.endswith('/'): >> >> self.redirect(request) >> >>You want to return the self.redirect(request). Otherwise the >>DirectoryLister will still get rendered to the body of the HTTP result, >>to no effect. > >It should probably also reuse the directoryListing() method, making the >change: > > if self.fp.isdir(): >- return self.redirect(request) >+ if not request.uri.endswith('/'): >+ return self.redirect(request) >+ return self.directoryListing() > This one works for me. Without the check, it goes in to a loop and never returns anything. _________________________________________________________________ Add photos to your messages with MSN 8. Get 2 months FREE*. http://join.msn.com/?page=features/featuredemail

2 1