User:TweetsFactsAndQueries/Queries/“controversies” articles with the most sitelinks

Originally posted on Twitter and on Mastodon.

# “controversies” articles with the most sitelinks (language versions)
# note: technically, this counts all sitelinks, not only Wikipedia ones (e.g. also Wikimedia Commons categories) –
# the top results are correct, though, so I don’t think it’s worth complicating the query further to fix this
SELECT ?controversies ?controversiesLabel ?sitelinks WHERE {
  # find all “controversies” items using CirrusSearch (SPARQL isn’t the right tool for this job)
  SERVICE wikibase:mwapi {
    bd:serviceParam wikibase:endpoint "www.wikidata.org";
                    wikibase:api "Search";
                    # note: the “inlabel” keyword also searches aliases
                    # exclude scholarly articles (Q13442814), there’s plenty of them and we don’t care about them here
                    # also exclude categories (Q4167836, P4224); we’ll filter them out again below, but excluding some here speeds things up
                    mwapi:srsearch "inlabel:controversies@en -haswbstatement:P31=Q13442814 -haswbstatement:P31=Q4167836 -haswbstatement:P4224";
                    # get as many search results at once as possible
                    mwapi:srlimit "max".
    ?controversies wikibase:apiOutputItem mwapi:title.
  }
  # get the number of sitelinks and the English label
  ?controversies wikibase:sitelinks ?sitelinks;
                 rdfs:label ?controversiesLabel.
  FILTER(LANG(?controversiesLabel) = "en")
  # filter out categories (-haswbstatement:P31=Q4167836 doesn’t exclude categories that are instance of a subclass of Q4167836)
  FILTER(!STRSTARTS(?controversiesLabel, "Category:"@en))
  # also filter out templates (only one match at the moment, not worth adding this to the search string)
  FILTER(!STRSTARTS(?controversiesLabel, "Template:"@en))
  # filter out the item about the “controversy” concept
  FILTER(?controversies != wd:Q1255828)
  # require at least one sitelink (article)
  FILTER(?sitelinks > 0)
}
ORDER BY DESC(?sitelinks)

Try it!