strange packet from server (wrong STDIN)...

Andre Garzia <[email protected]>
Newsgroups gmane.comp.web.fastcgi.devel
Message-ID <[email protected]>
Hi Folks,

I am new here. I am developing FastCGI libraries for the Runtime  
Revolution <http://www.runrev.com> language. This language is  
multiplatform and its script should run in all platforms without  
modification, so, I am not using fcgi dev kit and am building fastcgi  
support from the scratch. So far everything is fine. I am able to  
receive data from server and to reply to the server fine. My  
reference to create this really simple library was fcgi.py. I am  
testing with lighttp and WebStar (I am using Mac OS X 10.4.6)

The only issue so far is that when data is coming from a POST call, I  
am not receing the post data in a FCGI_STDIN (type: 5) packet but I  
am receiving an extra FCGI_PARAMS (type:4) packet with a very strange  
content. The contentLenght obtained by doing the bitwise operations  
is 0 but if I read more then there still content in the socket and my  
post data is there. I thought that this might be a stream record like  
FCGI_PARAMS and tried to force the data on my stream reading routines  
but it's no good. The bytes before the data do not resolve to key  
lenght and value lenght like they should if this was a stream record.

Can someone here please tell me how to implement FCGI_STDIN right? I  
must be doing something wrong, I can read all other packets and  
interpret them right. Also shouldn't the server send an empty  
FCGI_PARAMS and an empty FCGI_STDIN to flag my daemon that its ready  
for the answer? I am receiving only an empty FCGI_PARAMS, no STDIN  
ever comes... this happens both with lighttp and webstar.

If someone can point me to a good documentation source it would be  
great. The documents in fastcgi.com are not good for me for they all  
assume you're developing C/C++ web apps using the dev kit and not  
implementing your own DIY server.

Even a good example of library in Python or Ruby would be great.

Thanks in advance
Andre
___________________________________
fastcgi-developers mailing list
http://fastcgi.com/fastcgi-developers/
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.