This is most probably FAQ, but I swear I couldn't find it - please have
mercy on a nevow/twisted newbie :)
If I make some changes in the python code for a nevow application, is
there a way to refresh and test those changes without killing/restarting
my twisted server?
Static.File doesn't seem to support any of the cache features that
modern browsers use.
There are various HTTP headers that govern this, including Etag,
Cache-Control, Expires, etc. The Twisted server returns a
comparatively minimal set of headers (hit twistedmatrix.com and look
at the headers).
Also, as far as I can tell, a "good citizen" HTTPD changes the
response from a "200" to another response code that indicates "nothing
new here", under certain circumstances. OR, it lets the browser figure
that out... but it has to give the browser enough headers for it to do
Is there a subclass of Static.File somewhere in Twisted that supports
these features, or am I perhaps looking in the wrong place for this
Has anyone done anything like Archetypes from the zope world with
Nevow? I.e take some schema description for a type and automatically
from that make bare-bones views and editors for the object?
We've had a lot of problems going from glibc 2.0 to glibc 2.1.
People claim binary compatibility. Except for functions they
don't like. -- Peter Van Eynde, comp.lang.lisp
Short story: bizarre problem with a corrupted popdown list in Firefox
and Mozilla on unix systems.
I'm trying to render some unicode strings in a popdown, basing my code
on the formpost2.py example in nevow 0.4.1:
label="Choose a composer",
composer = annotate.autocallable(composer)
The full code is here:
This seems to work great. Well, at least the unicode strings seem to
roundtrip from the server to the browser, and there aren't any
That is, it works great in Safari, Konqueror, IE6, and win32 Firefox,
but with Firefox (1.0.2) on OS X and Linux and Mozilla (1.7.6) on
Linux the popdown gets corrupted and clicking on the popdown button is
not effective, you have to click to the left of the button to popdown
the menu, and the menu is misaligned. The scrollbar in the dropdown
list even disappears on Linux.
Unless my code is doing something nasty, I'm guessing this is a
Firefox bug. So the next question is, how do I report a bug like
this? Will they take a nevow script as a test case?
We are using XSLT with twisted/nevow and want to pass xml through
fillslots or stan generated API without getting escaped.
Any suggestions or pointers to how to do?
> erik swan
> splunk technologies
> 548 Fourth St. San Francisco
>> spelunk vb: explore natural caves
>> splunk vb : explore IT infrastructure
I'm just starting with Nevow, and wanted to know if this is possible,
and if so get a few leads.
1) to authenticate users (from a form with a "username" and a
"password" field) based on data stored in a database.
2) after authentication is performed, the user is sent to a URL based
on some of the data from the database. For example, if the database
query revealed him to be named "John Smith", he shall be refered to
http://appserver.com/JohnDoe. Subsequently, he should remain at this
URL, and all links should refer to addresses "below" his personal
address. For example, http://appserver.com/JohnDoe/configuration,
Thanks in advance,
I've just updated to nevow 0.4.1 and I enjoy the improvements, especially on the
LivePage API. But i've found something that doesn't work for me in the new way
to render livepage "glue" : the render_liveglue method uses the url.here()
function, and as I'm using an Apache proxy to access the site it gives me
"http://localhost:8080[...]" instead of the URL I'm accessing.
For now I've set manually the liveglue.js in my header, and I think it's a clean
way to do it, but I was wondering if there was a another way to get "good" URL
(which means the one the browser see).
Anyway thanks for that great work !
This message was sent using IMP, the Internet Messaging Program.
I must be a total moron:
I have this:
return dbpool.runQuery("SELECT english FROM lang WHERE spanish LIKE 'hola'")
def words(ctx, data):
docFactory = loaders.stan(tags.html[
tags.div(data="words")[ words ]])
And I get this at the terminal, which is correct:
2005/03/10 23:30 GMT [HTTPChannel,0,184.108.40.206] 220.127.116.11 - -
[10/Mar/2005:23:30:09 +0000] "GET / HTTP/1.1" 200 126 "-" "Mozilla/4.0
(compatible; MSIE 6.0; Windows NT 5.0)"
2005/03/10 23:30 GMT [-] hello
But I get a big red None on the web.