Submissions sorted by number of interested attendees

Below is a sorted (by number of interested attendees) list of all pages in Category:Submissions as of 23:18, 4 August 2013 (UTC)

Note: The list is not necessarily accurate, as the script used to generate it is a quick hack (see below).

rank / (number of interested attendees) / link to submission

How to update this page

Any Mac or Linux user should be able to update this page by saving the below script as sortsubs.py and running it by typing python sortsubs.py in a command line. The script will save the sorted list to a file called sorted_subs.txt. Then click here and replace the list with the one from the file. Don't forget to replace the timestamp with ~~~~~!

Note: Windows users should first install Python and add the Python installation folder to the Path system variable (instructions | video), in order to be able to run the command above from the DOS command line.

Source code: Note: If the script doesn't work without logging in, uncomment the login-related lines by removing the two '''.

#!/usr/bin/env python
import urllib
import json
import re
from getpass import getpass
import sys

wiki_url = 'wikimania2013.wikimedia.org'
api_url = 'http://' + wiki_url + '/w/api.php'

if __name__ == '__main__':
    # login
    '''
    print 'Login to ' + wiki_url
    username = raw_input('Username: ')
    password = getpass()
    url = api_url + '?action=login&lgname=' + username + '&lgpassword=' + password
    urllib.urlopen(url)
    '''

    print 'Processing the submissions...'

    # fetch all entries in the "Submissions" category
    url = api_url + '?action=query&format=json&list=categorymembers' + \
                    '&cmlimit=500&cmnamespace=0&cmtitle=Category:Submissions'
    res = json.loads(urllib.urlopen(url).read())
    all_submissions = res['query']['categorymembers']
    # fetch withdrawn submissions
    url = api_url + '?action=query&format=json&list=categorymembers' + \
                    '&cmlimit=500&cmnamespace=0&cmtitle=Category:Withdrawn_submissions'
    res = json.loads(urllib.urlopen(url).read())
    withdrawn = res['query']['categorymembers']
    # fetch submissions not accepted
    url = api_url + '?action=query&format=json&list=categorymembers' + \
                    '&cmlimit=500&cmnamespace=0&cmtitle=Category:Submissions_not_accepted'
    res = json.loads(urllib.urlopen(url).read())
    not_accepted = res['query']['categorymembers']

    submissions = []
    for i,p in enumerate(all_submissions):
        # Print progress
        sys.stdout.write('\r')
        sys.stdout.flush()
        sys.stdout.write( str(round(float(i+1)/len(all_submissions)*100, 1)) + '% --> ' + \
                          str(i+1) + '/' + str(len(all_submissions)) )

        # Filter out unwanted pages        
        submission = p['title'].encode('utf8')
        if not re.match('^Submissions/.{3,}', submission):
            continue

        # Don't include withdrawn or not accepted submissions
        # TODO: maybe use the same approach than the "accepted" check below?
        # Then again, a Python loop with a simple comparison might be faster than an API query.
        removed = False
        for j,q in enumerate(withdrawn):
            if p['pageid'] == q['pageid']:
                removed = True
                break
        if not removed:
            for j,q in enumerate(not_accepted):
                if p['pageid'] == q['pageid']:
                    removed = True
                    break
        if removed:
            continue

        # Get the interested attendees count from the submission
        url = api_url + '?action=query&format=json&prop=revisions&titles=%s&rvprop=content'\
              % urllib.quote(submission)
        res = json.loads(urllib.urlopen(url).read())
        if '-1' in res['query']['pages']:
            print(res)
            continue
        num = len( re.findall('#[^:*#]*?\[\[([a-z]*:)?User:.+?\]\]', repr(res)) )

        # Check if this is an accepted submission
        accepted = bool(re.search('\[\[[Cc]ategory: *[Aa]ccepted submissions\]\]|\{\{[Aa]ccepted submission', repr(res)))

        submissions.append((num, p['title'], accepted))

    submissions.sort()
    submissions.reverse()

    f = open('sorted_subs.txt', 'w')

    for c, n, a in submissions:
        acc = ""
        if a:
            acc = "{{ok}} "
        f.write( '# (%d) %s[[%s]]\n' % (c, acc, re.sub('(Submissions/)(.+)','\\1\\2|\\2',n.encode('utf8'))) )
    f.close()

    print '\nDone.'