User:Colin M/creations

import json
import sys
import requests

SESSION = requests.Session()

BASE_URL = 'https://en.wikipedia.org/w/api.php'
BASE_PARAMS = dict(action='query',
        list='usercontribs',
        ucuser='Colin_M',
        ucnamespace=0,
        uclimit=500, # max
        format='json',
)

def get_contribs(**params):
    r = SESSION.get(url=BASE_URL, params={**BASE_PARAMS, **params})
    return r.json()['query']['usercontribs']

# Cases where content was placed at a title that used to hold a redirect. This form
# of article creation is not included by the ucshow=new filter.
expanded_redirects = get_contribs(uctag='mw-removed-redirect')
# These are pages that show up in the 'expanded_redirects' set, but which aren't
# page creations on my part - rather these were cases where someone converted
# a page to a redirect, and I undid that change.
TITLE_BLACKLIST = {
        'Muh Dikhai',
        'Allie X discography',
        'The Picture of Dorian Gray (2004 film)',
}
creations = get_contribs(ucshow='new')
contribs = sorted(creations + expanded_redirects, key=lambda c: c['timestamp'], reverse=True)

print(r"""{| class="wikitable sortable"
|-
! Title !! Date !! Size at creation""")

def row_for_contrib(c):
    return "|-\n| {} || {} || {:,}".format(
            '[[' + c['title'] + ']]',
            c['timestamp'],
            c['size'],
    )

for c in contribs:
    # For simplicity, let's just exclude all small-ish creations. It's easy to draw
    # a line that separates "real" articles from redirects and dab pages. (Whereas
    # separating the latter two categories is a lot messier.)
    if c['title'] in TITLE_BLACKLIST or c['size'] < 800:
        continue
    print(row_for_contrib(c))


print("|}")