Text File URLS

#!/usr/bin/env python
"""Download video meta-info for a given video urls from input file.

Input files is in Firefox'es bookmarks export file
"""
import csv
import re
import sys
import urlparse
from BeautifulSoup import BeautifulSoup
from gdata.youtube.service import YouTubeService

# parse bookmarks.html
#with open(sys.argv[1]) as bookmark_file:
#    soup = BeautifulSoup(bookmark_file.read())

# extract youtube video urls
#video_url_regex = re.compile('http://www.youtube.com/watch')
#urls = [link['href'] for link in soup('a', href=video_url_regex)]

# parse text file
urls = open(sys.argv[1]).readlines()

### video order doesn't change ###

# extract video ids from the urls
ids = []
for video_url in urls:
    url = urlparse.urlparse(video_url)
    video_id = urlparse.parse_qs(url.query).get('v')
    if not video_id: continue # no video_id in the url
    ids.append(video_id[0])

### video order doesn't change ###

# remove duplicates but changes the order of the list
#ids = list(set(ids))

# print total number of video_ids
counter = len(ids)
print counter

# get some statistics for the videos
yt_service = YouTubeService()
yt_service.developer_key = 'AI39si4yOmI0GEhSTXH0nkiVDf6tQjCkqoys5BBYLKEr-PQxWJ0IlwnUJAcdxpocGLBBCapdYeMLIsB7KVC_OA8gYK0VKV726g'
#NOTE: you don't need to authenticate for readonly requests
yt_service.ssl = True #NOTE: it works for readonly requests
#yt_service.debug = True # show requests

writer = csv.writer(open(sys.argv[2], 'wb'))
writer.writerow(("video_id", "title", "view_count", "favorites", "comments", "average", "num_raters", "author", "published", "tags"))
for video_id in ids:
    try:
        entry = yt_service.GetYouTubeVideoEntry(video_id=video_id)
        dir(entry.rating)
        ['FindExtensions', 'ToString', '_AddMembersToElementTree', '_BecomeChildElement', '_ConvertElementAttributeToMember', '_ConvertElementTreeToMember', '_HarvestElementTree', '_ToElementTree', '__class__', '__delattr__', '__dict__', '__doc__', '__format__', '__getattribute__', '__hash__', '__init__', '__module__', '__new__', '__reduce__', '__reduce_ex__', '__repr__', '__setattr__', '__sizeof__', '__str__', '__subclasshook__', '__weakref__', '_attributes', '_children', '_namespace', '_tag', 'average', 'extension_attributes', 'extension_elements', 'max', 'min', 'num_raters', 'text']
        comments = yt_service.GetYouTubeVideoCommentFeed(video_id=video_id)
    except Exception, e:
        print "############################"
        print "## Videos Remaining:", counter, "##"
        print "############################"
        counter -= 1
        print "##"
        print >>sys.stderr, "## Failed to retrieve entry"
        error = sys.stderr, "Failed to retrieve entry video_id=%s: %s" %(video_id, e)
        print "############################"
        print ""
        print ""
        writer.writerow((video_id, e))
    else:
        print "############################"
        print "## Videos Remaining:", counter, "##"
        print "############################"
        counter -= 1
        title = entry.media.title.text
        print "## Title:", title
        view_count = entry.statistics.view_count
        print "## View count:", view_count
        favorites = entry.statistics.favorite_count
        print "## Favorite Count:", favorites
        comments = comments.total_results.text
        print "## Comment Count:", comments
        if entry.rating is None: # skip it
            average = 0
        else:
            average = entry.rating.average
        print "## Average Rating:", average
        if entry.rating is None: # skip it
            num_raters = 0
        else:
            num_raters = entry.rating.num_raters
        print "## Number of Raters:", num_raters
        author = entry.author[0].name.text
        print "## Autor:", author
        published = entry.published.text
        print "## Published on:", published
        tags = entry.media.keywords.text
        print "## Tags:", tags
        print "############################"
        print ""
        print ""
        writer.writerow((video_id, title, view_count, favorites, comments, average, num_raters, author, published, tags))