view gpp/oembed/core.py @ 334:6805d15cda13

Adding a script I had to write on the fly to filter out posts from the posts csv file that had no parent topics. MyISAM let me get away with that, but InnoDB won't.
author Brian Neal <bgneal@gmail.com>
date Sat, 26 Feb 2011 01:28:22 +0000
parents 8fd4984d5c3b
children 6d6fdc58487c
line wrap: on
line source
"""
This module contains core functionality for the oembed application.
"""
from __future__ import with_statement
import urllib
import urllib2
import gzip
try:
    from cStringIO import StringIO
except ImportError:
    from StringIO import StringIO

import django.utils.simplejson as json


USER_AGENT = 'gremmies python oembed'


def get_oembed(api_endpoint, url, format='json', **opts):
    """
    Perform the GET request to retrieve the embedded media data from the given
    API endpoint for the given URL. Return the result as a Python dictionary.

    format specifies the response format, and should be 'json' or 'xml'.
    opts are any additional GET options that should be present in the GET
    request.

    """
    opts['url'] = url
    opts['format'] = format
    api_url = "%s?%s" % (api_endpoint, urllib.urlencode(opts))

    headers = {
        'User-Agent': USER_AGENT, 
        'Accept-Encoding': 'gzip',
    }
    request = urllib2.Request(api_url, headers=headers)

    opener = urllib2.build_opener()
    f = opener.open(request)
    headers = f.info()
    result = f.read()
    f.close()

    if headers.get('content-encoding') == 'gzip':
        with gzip.GzipFile(fileobj=StringIO(result)) as f:
            result = f.read()

    return json.loads(result)

if __name__ == "__main__":
    try:
        print get_oembed("http://www.youtube.com/oembed",
                #"http://www.youtube.com/watch?v=7_IMzJldOf4")
                "http://www.youtube.com/watch?v=99999999999")
    except urllib2.HTTPError, e:
        print e