kdeps-tools-wikipedia

version: 2.0.0
category: web
description: Look up information on Wikipedia. Use for general knowledge questions about people, places, companies, historical events, concepts, or any topic needing an encyclopedic answer. Input is a search query.
constraints: good for general knowledge; not for real-time/current information
outputFormat: the title and introduction of the two best-matching articles
seeAlso: web_search, web_scraper
parameters:
  query:
    type: string
    description: The topic or question to look up on Wikipedia
    required: true
python:
  timeout: 30s
  env:
    WIKI_URL: [KDEPS_WIKIPEDIA_URL, ""]  # optional: a MediaWiki api.php endpoint (default en.wikipedia.org)
  script: |
    import json, os, sys, urllib.error, urllib.parse, urllib.request

    query = json.loads(sys.argv[1])["query"]
    base = os.environ.get("WIKI_URL") or "https://en.wikipedia.org/w/api.php"
    # One request: the two best matches (generator=search) with the
    # plain-text introduction of each (prop=extracts; several need exintro).
    params = {
        "action": "query", "format": "json", "formatversion": "2",
        "generator": "search", "gsrsearch": query, "gsrlimit": "2",
        "prop": "extracts", "exintro": "1", "explaintext": "1", "exlimit": "2",
    }
    req = urllib.request.Request(base + "?" + urllib.parse.urlencode(params),
                                 headers={"User-Agent": "kdeps/agent"})
    try:
        with urllib.request.urlopen(req, timeout=25) as resp:
            data = json.load(resp)
    except urllib.error.HTTPError as e:
        sys.exit("wikipedia: responded with %d" % e.code)
    except (urllib.error.URLError, ValueError) as e:
        sys.exit("wikipedia: %s" % getattr(e, "reason", e))
    pages = sorted((data.get("query") or {}).get("pages") or [], key=lambda p: p.get("index", 0))
    if not pages:
        print("no wikipedia pages found")
    for p in pages:
        print("%s\n%s\n" % (p.get("title", ""), (p.get("extract") or "").strip()))