Talk:AllPages

From Heroes 3 wiki
Revision as of 10:26, 10 September 2016 by Imahero (talk | contribs) (→‎Discussion)
Jump to navigation Jump to search

List generation

#!/usr/bin/env python3

import urllib.request
import json

base_url = 'http://heroes.thelazy.net/wiki/'
base_query = 'http://heroes.thelazy.net/wiki/api.php?action=query&list=allpages&aplimit=500&apfilterredir=nonredirects&format=json&apfrom='
continue_title = ''
titles = {}

while True:
    request = urllib.request.urlopen(base_query + urllib.parse.quote(continue_title))
    response = request.read()
    results = json.loads(response.decode())
    for page in results['query']['allpages']:
        title = page['title']
        titles[title] = base_url + title
    if 'query-continue' in results:
        continue_title = results['query-continue']['allpages']['apcontinue']
    else:
        break

title_queries = []
title_query = ''

for title, url in titles.items():
    if title != 'All Pages' and title != 'Main Page' and title != 'Search':
        title_query += title + '|'
        if len(title_query) > 50:
            title_queries.append(title_query.rstrip('|'))
            title_query = ''

prop_query = 'http://heroes.thelazy.net/wiki/api.php?action=query&prop=categories&format=json&titles='
titles_by_category = {}

for title_query in title_queries:
    request = urllib.request.urlopen(prop_query + urllib.parse.quote(title_query))
    response = request.read()
    results = json.loads(response.decode())
    for _, page in results['query']['pages'].items():
        if 'categories' not in page:
            print(page)
            continue
        for category in page['categories']:
            if category['title'] not in titles_by_category:
                titles_by_category[category['title']] = set()
            titles_by_category[category['title']].add(page['title'])

output_file = open('output.mediawiki', 'w')

for category, titles in sorted(titles_by_category.items()):
    output_file.write('[' + base_url + category.replace(' ', '_') + ' ' + category.split('Category:', 1).pop() + ']\n\n')
    for title in sorted(titles):
        output_file.write(':[' + base_url + title.replace(' ', '_') + ' ' + title + ']\n\n')

Discussion

Nice technical page, but I must ask is there any use for it? At least for me this seems quite useless, causing hinder rather than help. –Kapteeni Ruoska (talk) 06:11, 7 September 2016 (CEST)

I wanted to make sure there wasn't anything I was missing. When I'm browsing through the list it's easier to click these links than copy pasting the auto-generated titles from the api query.
--imahero 03:14, 8 September 2016 (CEST)
Sure, just wondering, as the wiki already has Special:AllPages, but perhaps there is a use for that. –Kapteeni Ruoska (talk) 07:27, 8 September 2016 (CEST)
Special:AllPages felt too clumsy for me to navigate. All I really wanted was a single list of all nonredirect links :)
--imahero 08:31, 8 September 2016 (CEST)

I may want to go back to using the external link syntax if breaking the orphaned pages is a big deal. For now I've just added the orphaned pages here and I'll just update the list every time I generate the AllPages list.
--imahero 15:19, 8 September 2016 (CEST)

I changed it back to use the external link syntax to make sure the orphaned pages list still works properly.
--imahero 22:48, 8 September 2016 (CEST)

It's now sorted by category.
--imahero 12:26, 10 September 2016 (CEST)