Method search

pattern/web/__init__.py:1294–1353

Returns a list of results from DuckDuckGo for the given query.

(self, query, type=SEARCH, start=None, count=None, sort=RELEVANCY, size=None, cached=True, **kwargs)

        SearchEngine.__init__(self, license or DUCKDUCKGO_LICENSE, throttle, language)

    def search(self, query, type=SEARCH, start=None, count=None, sort=RELEVANCY, size=None, cached=True, **kwargs):
        """" Returns a list of results from DuckDuckGo for the given query.
        """
        if type != SEARCH:
            raise SearchEngineTypeError
        # 1) Construct request URL.
        url = URL(DUCKDUCKGO, method=GET, query={
            "q": query,
            "o": "json"
        })
        # 2) Restrict language.
        if type == SEARCH and self.language is not None:
            url.query["kl"] = self.language
        # 3) Parse JSON response.
        kwargs.setdefault("unicode", True)
        kwargs.setdefault("throttle", self.throttle)
        data = url.download(cached=cached, **kwargs)
        data = json.loads(data)
        results = Results(DUCKDUCKGO, query, type)
        results.total = None
        for x in data.get("Results", []):
            if x.get("FirstURL"):
                r = Result(url=None)
                # Parse official website link.
                r.url    = self.format(x.get("FirstURL"))
                r.title  = self.format(data.get("Heading"))
                r.text   = self.format(data.get("Abstract"))
                r.author = self.format(data.get("AbstractSource"))
                r.type   = self.format(REFERENCE)
                results.append(r)
        for topic in data.get("RelatedTopics", []):
            for x in topic.get("Topics", [topic]):
                r = Result(url=None)
                r.url = x.get("FirstURL")
                # Parse title and type from URL (e.g., http://duckduckgo.com/d/Cats?kl=en).
                m = re.match(r"^http://duckduckgo.com/([a-z]/)?(.*?)(\?|$)", r.url)
                # Parse title: "Cats".
                s1 = m and m.group(2) or "" # Title: "Cats"
                s1 = u(decode_url(s1.encode("utf-8")))
                s1 = s1.strip().replace("_", " ")
                s1 = s1[:1].upper() + s1[1:]
                # Parse description; the part before the first "-" or "," was the link.
                s2 = x.get("Text", "").strip()
                s2 = re.sub(r" +", " ", s2)
                s2 = s2[:1].upper() + s2[1:] or ""
                s2 = s2.startswith(s1) \
                    and "<a href=\"%s\">%s</a>%s" % (r.url, s1, s2[len(s1):]) \
                     or re.sub(r"^(.*?)( - | or |, )(.*?)", "<a href=\"%s\">\\1</a>\\2\\3" % r.url, s2)
                # Parse type: "d/" => "definition".
                s3 = m and m.group(1) or ""
                s3 = {"c": CATEGORY, "d": DEFINITION}.get(s3.rstrip("/"), "")
                s3 = topic.get("Name", "").lower() or s3
                s3 = re.sub("^in ", "", s3)
                # Format result.
                r.url   = self.format(r.url)
                r.title = self.format(s1)
                r.text  = self.format(s2)
                r.type  = self.format(s3)