问题
I want to keep some statistics about users and locations in a database. For instance, I would like to store "Mozilla","Firefox","Safari","Chrome","IE", etc... as well as the versions, and possibly the operating system.
What I am trying to locate from Python is this string;
Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.0.14) Gecko/2009090216 Ubuntu/9.04 (jaunty) Firefox/3.0.14
Is there an efficient way to use Python or mod_python to detect the http user agent/browser?
回答1:
HTTP_USER_AGENT contains this information, and will be passed in the environment variables your application uses. In mod_python, this is expressed as:
def my_request_handler(req):
req.add_common_vars()
agent = req.subprocess_env.get("HTTP_USER_AGENT")
# `agent` now contains the full user agent of the browser, or None
It's a basic CGI thing, but this is how mod_python gives it to you.
回答2:
The method suggested by Jed Smith works, but I was sure there was a simpler way.
The req.headers_in variable contains all the header info, and you can easily access the user agent using mod_python by calling:
req.headers_in[ 'User-Agent' ]
It is not necessary to call req.add_common_vars() when using this method.
回答3:
If you are using the Django-Framework you get the user agent like this
request.META['HTTP_USER_AGENT']
The very nice plugin httpagentparser extracts every detail and puts it a dictionary.
Installation works via pip
pip install httpagentparser
Hope this helps... I googled about 30 min until I found something useful :)
Ron
来源：https://stackoverflow.com/questions/1689861/browser-detection-python-mod-python