Method _extract_from_html

IntelOSINT.py:15346–15577

(html_text, fallback_url)

            return merged

        def _extract_from_html(html_text, fallback_url):
            scripts = re.findall(r"<script[^>]*>(.*?)</script>", html_text, flags=re.DOTALL | re.IGNORECASE)
            json_scripts = re.findall(r"<script[^>]*type=[\"']application/json[\"'][^>]*>(.*?)</script>", html_text, flags=re.DOTALL | re.IGNORECASE)
            ldjson_scripts = re.findall(r"<script[^>]*type=[\"']application/ld\\+json[\"'][^>]*>(.*?)</script>", html_text, flags=re.DOTALL | re.IGNORECASE)
            next_json = re.findall(r"<script[^>]*id=[\"']__NEXT_DATA__[\"'][^>]*>(.*?)</script>", html_text, flags=re.DOTALL | re.IGNORECASE)
            parsed = []
            script_images = []

            def _extract_json_obj(blob, start_idx=0):
                i = start_idx
                while i < len(blob) and blob[i].isspace():
                    i += 1
                if i >= len(blob) or blob[i] != "{":
                    return ""
                depth = 0
                in_string = False
                esc = False
                quote = ""
                j = i
                while j < len(blob):
                    ch = blob[j]
                    if in_string:
                        if esc:
                            esc = False
                        elif ch == "\\":
                            esc = True
                        elif ch == quote:
                            in_string = False
                    else:
                        if ch in ("'", '"'):
                            in_string = True
                            quote = ch
                        elif ch == "{":
                            depth += 1
                        elif ch == "}":
                            depth -= 1
                            if depth == 0:
                                return blob[i:j + 1]
                    j += 1
                return ""

            prefixes = (
                "window.__NUXT__",
                "window.__INITIAL_STATE__",
                "__NUXT__",
                "window.__NEXT_DATA__",
                "__NEXT_DATA__",
                "window.__STATE__"
            )

            def _append_json_candidates(text):
                for pfx in prefixes:
                    pat = pfx + r"\s*=\s*"
                    for m in re.finditer(pat, text, flags=re.IGNORECASE):
                        start = m.end()
                        js = text[start:start + 60000]
                        obj_txt = _extract_json_obj(js, 0)
                        if obj_txt: