def _create_duckduckgo_parser()

in src/smolagents/default_tools.py [0:0]

• 39 lines of code
• 19 McCabe index (conditional complexity)

    def _create_duckduckgo_parser(self):
        from html.parser import HTMLParser

        class SimpleResultParser(HTMLParser):
            def __init__(self):
                super().__init__()
                self.results = []
                self.current = {}
                self.capture_title = False
                self.capture_description = False
                self.capture_link = False

            def handle_starttag(self, tag, attrs):
                attrs = dict(attrs)
                if tag == "a" and attrs.get("class") == "result-link":
                    self.capture_title = True
                elif tag == "td" and attrs.get("class") == "result-snippet":
                    self.capture_description = True
                elif tag == "span" and attrs.get("class") == "link-text":
                    self.capture_link = True

            def handle_endtag(self, tag):
                if tag == "a" and self.capture_title:
                    self.capture_title = False
                elif tag == "td" and self.capture_description:
                    self.capture_description = False
                elif tag == "span" and self.capture_link:
                    self.capture_link = False
                elif tag == "tr":
                    # Store current result if all parts are present
                    if {"title", "description", "link"} <= self.current.keys():
                        self.current["description"] = " ".join(self.current["description"])
                        self.results.append(self.current)
                        self.current = {}

            def handle_data(self, data):
                if self.capture_title:
                    self.current["title"] = data.strip()
                elif self.capture_description:
                    self.current.setdefault("description", [])
                    self.current["description"].append(data.strip())
                elif self.capture_link:
                    self.current["link"] = "https://" + data.strip()

        return SimpleResultParser()