curl-library
Re: help with pycurl/win32 and HTTP PUT
Date: Wed, 9 Feb 2005 08:47:13 -0600
Kjetil-
Thanks for the reply. I actually was looking through some of the
pycurl tests and found something set up in a similar fashion (tho
using writers) and set up my code pretty much how you did the example
below. The thing that is strange is that I keep getting:
"<HTML><HEAD><TITLE>Bad request</TITLE></HEAD>
<BODY><H1>Bad request</H1>
Your browser sent a query this server could not understand.
</BODY></HTML>"
It appears to call the read_callback 7 times with a size of 16384.
That total byte count is larger than the file itself (which is about
99kbytes). Once the 7th time is reached I get a pause for several
seconds before it throws me back that error message.
I have tried sending the same file to the same server with the curl
command line tool on Windows and it works just fine. I'm not sure
what I'm doing wrong here because inserting some print statements
shows that the full file is being read in.
Thanks,
Jason
On Wed, 9 Feb 2005 13:10:22 +0100 (CET), Kjetil Jacobsen
<kjetilja_at_cs.uit.no> wrote:
> hello,
>
> there is some information wrt callbacks and pycurl at
>
> http://pycurl.sf.net/doc/callbacks.html
>
> in particular, READDATA cannot be used when READFUNCTION is specified.
> fortunately, this limitation can be overcome by using a class to store the
> file reference (instead of using READDATA) and then set a class method as
> the callback function like this:
>
> class filereader:
>
> def __init__(self, f):
> self.f = f
>
> def read_callback(self, size):
> return self.f.read(size)
>
> import pycurl
> c = pycurl.Curl()
> f = filereader(open('somefile'))
> c.setopt(c.URL, 'http://somehere/')
> c.setopt(c.READFUNCTION, f.read_callback)
>
> ...
>
> regards,
>
> - kjetil
>
> _______________________________________________
> http://cool.haxx.se/mailman/listinfo/curl-and-python
>
Received on 2005-02-09