http-parser

http-parser is a handy tool for parsing HTTP requests and responses in Python, and it's built in C under the MIT License. This software is based on the original http-parser created by Ryan Dahl.

How to Install http-parser

If you want to get started with http-parser, you can easily install it using pip. Just run:

pip install http-parser

If you prefer installing from source, here's what you need to do:

git clone git://github.com/benoitc/http-parser.git
cd http-parser && python setup.py install

Features of http-parser

This tool gives you access to a low-level parser called HttpParser, written in C, right in your Python programs. It also has a higher-level feature called HttpStream, which lets you work with a readable, sequential io.RawIOBase object.

User-Friendly Readers Available

The http-parser makes your job easier with three types of readers in its reader module:

• IterReader: For reading iterables.


• StringReader: For reading strings and StringIO objects.


• SocketReader: For reading sockets or similar objects (you'll need recv_into).

You can also use any io.RawIOBase object that suits your needs!

Create Your Own Callbacks!

You can pass your own callbacks to the C parser just like the original C http-parser. This means you can parse HTTP streams asynchronously! Here are some useful callbacks:

• on_message_begin()


• on_path(path)


• on_query_string(query_string)


• on_url(url)


• on_fragment(fragment)


• on_header_field(field, last_was_value)


• on_header_value(key, value)

#!/usr/bin/env python
import socket
from http_parser.http import HttpStream
from http_parser.reader import SocketReader

def main():
s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
try:
s.connect(('gunicorn.org', 80))
s.send(b"GET / HTTP/1.1\r\nHost: gunicorn.org\r\n\r\n")
r = SocketReader(s)
p = HttpStream(r)
printf(p.headers())
printf(p.body_file().read())
finally:
s.close()
if __name__ == "__main__":
main()