more detailed search

app_agents/web_agent.py

@@ -18,12 +18,31 @@ from smolagents import  InferenceClientModel, CodeAgent, DuckDuckGoSearchTool, G
 load_dotenv()
 model = InferenceClientModel(model_id="Qwen/Qwen2.5-Coder-32B-Instruct", provider="together")
 
+from app_tools.web_search_tools import(
+    SearchInformationTool,
+    NavigationalSearchTool,
+    VisitTool,
+    PageUpTool,
+    PageDownTool,
+    FinderTool,
+    FindNextTool,
+    ArchiveSearchTool,
+)
+
 web_agent = CodeAgent(
     model=model,
     tools=[
         # DuckDuckGoSearchTool(),
         GoogleSearchTool("serper"),
-        VisitWebpageTool()
+        VisitWebpageTool(),
+        SearchInformationTool(),
+        NavigationalSearchTool(),
+        VisitTool(),
+        PageUpTool(),
+        PageDownTool(),
+        FinderTool(),
+        FindNextTool(),
+        ArchiveSearchTool(),
     ],
     name="web_agent",
     description="Runs web searches for you. Give it your query as an argument.",

app_tools/browser.py

@@ -0,0 +1,354 @@
+# Shamelessly stolen from Microsoft Autogen team: thanks to them for this great resource!
+# https://github.com/microsoft/autogen/blob/gaia_multiagent_v01_march_1st/autogen/browser_utils.py
+import json
+import os
+import requests
+import re
+import io
+import uuid
+import mimetypes
+import time
+import pathlib
+import pathvalidate
+from urllib.parse import urljoin, urlparse, unquote, parse_qs
+from urllib.request import url2pathname
+from typing import Any, Dict, List, Optional, Union, Tuple
+from .mdconvert import MarkdownConverter, UnsupportedFormatException, FileConversionException
+from serpapi import GoogleSearch
+from .cookies import COOKIES
+
+
+class SimpleTextBrowser:
+    """(In preview) An extremely simple text-based web browser comparable to Lynx. Suitable for Agentic use."""
+
+    def __init__(
+        self,
+        start_page: Optional[str] = None,
+        viewport_size: Optional[int] = 1024 * 8,
+        downloads_folder: Optional[Union[str, None]] = None,
+        serpapi_key: Optional[Union[str, None]] = None,
+        request_kwargs: Optional[Union[Dict[str, Any], None]] = None,
+    ):
+        self.start_page: str = start_page if start_page else "about:blank"
+        self.viewport_size = viewport_size  # Applies only to the standard uri types
+        self.downloads_folder = downloads_folder
+        self.history: List[Tuple[str, float]] = list()
+        self.page_title: Optional[str] = None
+        self.viewport_current_page = 0
+        self.viewport_pages: List[Tuple[int, int]] = list()
+        self.set_address(self.start_page)
+        self.serpapi_key = serpapi_key
+        self.request_kwargs = request_kwargs
+        self.request_kwargs["cookies"] = COOKIES
+        self._mdconvert = MarkdownConverter()
+        self._page_content: str = ""
+
+        self._find_on_page_query: Union[str, None] = None
+        self._find_on_page_last_result: Union[int, None] = None  # Location of the last result
+
+    @property
+    def address(self) -> str:
+        """Return the address of the current page."""
+        return self.history[-1][0]
+
+    def set_address(self, uri_or_path: str, filter_year: Optional[int] = None) -> None:
+        # TODO: Handle anchors
+        self.history.append((uri_or_path, time.time()))
+
+        # Handle special URIs
+        if uri_or_path == "about:blank":
+            self._set_page_content("")
+        elif uri_or_path.startswith("google:"):
+            self._serpapi_search(uri_or_path[len("google:"):].strip(), filter_year=filter_year)
+        else:
+            if (
+                not uri_or_path.startswith("http:")
+                and not uri_or_path.startswith("https:")
+                and not uri_or_path.startswith("file:")
+            ):
+                if len(self.history) > 1:
+                    prior_address = self.history[-2][0]
+                    uri_or_path = urljoin(prior_address, uri_or_path)
+                    # Update the address with the fully-qualified path
+                    self.history[-1] = (uri_or_path, self.history[-1][1])
+            self._fetch_page(uri_or_path)
+
+        self.viewport_current_page = 0
+        self.find_on_page_query = None
+        self.find_on_page_viewport = None
+
+    @property
+    def viewport(self) -> str:
+        """Return the content of the current viewport."""
+        bounds = self.viewport_pages[self.viewport_current_page]
+        return self.page_content[bounds[0] : bounds[1]]
+
+    @property
+    def page_content(self) -> str:
+        """Return the full contents of the current page."""
+        return self._page_content
+
+    def _set_page_content(self, content: str) -> None:
+        """Sets the text content of the current page."""
+        self._page_content = content
+        self._split_pages()
+        if self.viewport_current_page >= len(self.viewport_pages):
+            self.viewport_current_page = len(self.viewport_pages) - 1
+
+    def page_down(self) -> None:
+        self.viewport_current_page = min(self.viewport_current_page + 1, len(self.viewport_pages) - 1)
+
+    def page_up(self) -> None:
+        self.viewport_current_page = max(self.viewport_current_page - 1, 0)
+
+    def find_on_page(self, query: str) -> Union[str, None]:
+        """Searches for the query from the current viewport forward, looping back to the start if necessary."""
+
+        # Did we get here via a previous find_on_page search with the same query?
+        # If so, map to find_next
+        if query == self._find_on_page_query and self.viewport_current_page == self._find_on_page_last_result:
+            return self.find_next()
+
+        # Ok it's a new search start from the current viewport
+        self._find_on_page_query = query
+        viewport_match = self._find_next_viewport(query, self.viewport_current_page)
+        if viewport_match is None:
+            self._find_on_page_last_result = None
+            return None
+        else:
+            self.viewport_current_page = viewport_match
+            self._find_on_page_last_result = viewport_match
+            return self.viewport
+
+    def find_next(self) -> None:
+        """Scroll to the next viewport that matches the query"""
+
+        if self._find_on_page_query is None:
+            return None
+
+        starting_viewport = self._find_on_page_last_result
+        if starting_viewport is None:
+            starting_viewport = 0
+        else:
+            starting_viewport += 1
+            if starting_viewport >= len(self.viewport_pages):
+                starting_viewport = 0
+
+        viewport_match = self._find_next_viewport(self._find_on_page_query, starting_viewport)
+        if viewport_match is None:
+            self._find_on_page_last_result = None
+            return None
+        else:
+            self.viewport_current_page = viewport_match
+            self._find_on_page_last_result = viewport_match
+            return self.viewport
+
+    def _find_next_viewport(self, query: str, starting_viewport: int) -> Union[int, None]:
+        """Search for matches between the starting viewport looping when reaching the end."""
+
+        if query is None:
+            return None
+
+        # Normalize the query, and convert to a regular expression
+        nquery = re.sub(r"\*", "__STAR__", query)
+        nquery = " " + (" ".join(re.split(r"\W+", nquery))).strip() + " "
+        nquery = nquery.replace(" __STAR__ ", "__STAR__ ")  # Merge isolated stars with prior word
+        nquery = nquery.replace("__STAR__", ".*").lower()
+
+        if nquery.strip() == "":
+            return None
+
+        idxs = list()
+        idxs.extend(range(starting_viewport, len(self.viewport_pages)))
+        idxs.extend(range(0, starting_viewport))
+
+        for i in idxs:
+            bounds = self.viewport_pages[i]
+            content = self.page_content[bounds[0] : bounds[1]]
+
+            # TODO: Remove markdown links and images
+            ncontent = " " + (" ".join(re.split(r"\W+", content))).strip().lower() + " "
+            if re.search(nquery, ncontent):
+                return i
+
+        return None
+
+    def visit_page(self, path_or_uri: str, filter_year: Optional[int] = None) -> str:
+        """Update the address, visit the page, and return the content of the viewport."""
+        self.set_address(path_or_uri, filter_year=filter_year)
+        return self.viewport
+
+    def _split_pages(self) -> None:
+        # Do not split search results
+        if self.address.startswith("google:"):
+            self.viewport_pages = [(0, len(self._page_content))]
+            return
+
+        # Handle empty pages
+        if len(self._page_content) == 0:
+            self.viewport_pages = [(0, 0)]
+            return
+
+        # Break the viewport into pages
+        self.viewport_pages = []
+        start_idx = 0
+        while start_idx < len(self._page_content):
+            end_idx = min(start_idx + self.viewport_size, len(self._page_content))  # type: ignore[operator]
+            # Adjust to end on a space
+            while end_idx < len(self._page_content) and self._page_content[end_idx - 1] not in [" ", "\t", "\r", "\n"]:
+                end_idx += 1
+            self.viewport_pages.append((start_idx, end_idx))
+            start_idx = end_idx
+
+
+    def _serpapi_search(self, query: str, filter_year: Optional[int] = None) -> None:
+        if self.serpapi_key is None:
+            raise ValueError("Missing SerpAPI key.")
+        
+        params = {
+            "engine": "google",
+            "q": query,
+            "api_key": self.serpapi_key,
+        }
+        if filter_year is not None:
+            params["tbs"] = f"cdr:1,cd_min:01/01/{filter_year},cd_max:12/31/{filter_year}"
+
+        search = GoogleSearch(params)
+        results = search.get_dict()
+        self.page_title = f"{query} - Search"
+        if "organic_results" not in results.keys():
+            raise Exception(f"'organic_results' key not found in results: {results}. Use a less restrictive query.")
+        if len(results['organic_results']) == 0:
+            year_filter_message = f" with filter year={filter_year}" if filter_year is not None else ""
+            self._set_page_content(f"No results found for '{query}'{year_filter_message}. Try with a more general query, or remove the year filter.")
+            return
+
+        def _prev_visit(url):
+            for i in range(len(self.history) - 1, -1, -1):
+                if self.history[i][0] == url:
+                    return f"You previously visited this page {round(time.time() - self.history[i][1])} seconds ago.\n"
+            return ""
+
+        web_snippets: List[str] = list()
+        idx = 0
+        if "organic_results" in results:
+            for page in results["organic_results"]:
+                idx += 1
+                date_published = ""
+                if "date" in page:
+                    date_published = "\nDate published: " + page["date"]
+
+                source = ""
+                if "source" in page:
+                    source = "\nSource: " + page["source"]
+
+                snippet = ""
+                if "snippet" in page:
+                    snippet = "\n" + page["snippet"]
+
+                redacted_version = f"{idx}. [{page['title']}]({page['link']}){date_published}{source}\n{_prev_visit(page['link'])}{snippet}"
+
+                redacted_version = redacted_version.replace("Your browser can't play this video.", "")
+                web_snippets.append(redacted_version)
+
+
+        content = (
+            f"A Google search for '{query}' found {len(web_snippets)} results:\n\n## Web Results\n"
+            + "\n\n".join(web_snippets)
+        )
+
+        self._set_page_content(content)
+
+
+    def _fetch_page(self, url: str) -> None:
+        download_path = ""
+        try:
+            if url.startswith("file://"):
+                download_path = os.path.normcase(os.path.normpath(unquote(url[7:])))
+                res = self._mdconvert.convert_local(download_path)
+                self.page_title = res.title
+                self._set_page_content(res.text_content)
+            else:
+                # Prepare the request parameters
+                request_kwargs = self.request_kwargs.copy() if self.request_kwargs is not None else {}
+                request_kwargs["stream"] = True
+
+                # Send a HTTP request to the URL
+                response = requests.get(url, **request_kwargs)
+                response.raise_for_status()
+
+                # If the HTTP request was successful
+                content_type = response.headers.get("content-type", "")
+
+                # Text or HTML
+                if "text/" in content_type.lower():
+                    res = self._mdconvert.convert_response(response)
+                    self.page_title = res.title
+                    self._set_page_content(res.text_content)
+                # A download
+                else:
+                    # Try producing a safe filename
+                    fname = None
+                    download_path = None
+                    try:
+                        fname = pathvalidate.sanitize_filename(os.path.basename(urlparse(url).path)).strip()
+                        download_path = os.path.abspath(os.path.join(self.downloads_folder, fname))
+
+                        suffix = 0
+                        while os.path.exists(download_path) and suffix < 1000:
+                            suffix += 1
+                            base, ext = os.path.splitext(fname)
+                            new_fname = f"{base}__{suffix}{ext}"
+                            download_path = os.path.abspath(os.path.join(self.downloads_folder, new_fname))
+
+                    except NameError:
+                        pass
+
+                    # No suitable name, so make one
+                    if fname is None:
+                        extension = mimetypes.guess_extension(content_type)
+                        if extension is None:
+                            extension = ".download"
+                        fname = str(uuid.uuid4()) + extension
+                        download_path = os.path.abspath(os.path.join(self.downloads_folder, fname))
+
+                    # Open a file for writing
+                    with open(download_path, "wb") as fh:
+                        for chunk in response.iter_content(chunk_size=512):
+                            fh.write(chunk)
+
+                    # Render it
+                    local_uri = pathlib.Path(download_path).as_uri()
+                    self.set_address(local_uri)
+
+
+        except UnsupportedFormatException as e:
+            print(e)
+            self.page_title = ("Download complete.",)
+            self._set_page_content(f"# Download complete\n\nSaved file to '{download_path}'")
+        except FileConversionException as e:
+            print(e)
+            self.page_title = ("Download complete.",)
+            self._set_page_content(f"# Download complete\n\nSaved file to '{download_path}'")
+        except FileNotFoundError:
+            self.page_title = "Error 404"
+            self._set_page_content(f"## Error 404\n\nFile not found: {download_path}")
+        except requests.exceptions.RequestException as request_exception:
+            try:
+                self.page_title = f"Error {response.status_code}"
+
+                # If the error was rendered in HTML we might as well render it
+                content_type = response.headers.get("content-type", "")
+                if content_type is not None and "text/html" in content_type.lower():
+                    res = self._mdconvert.convert(response)
+                    self.page_title = f"Error {response.status_code}"
+                    self._set_page_content(f"## Error {response.status_code}\n\n{res.text_content}")
+                else:
+                    text = ""
+                    for chunk in response.iter_content(chunk_size=512, decode_unicode=True):
+                        text += chunk
+                    self.page_title = f"Error {response.status_code}"
+                    self._set_page_content(f"## Error {response.status_code}\n\n{text}")
+            except NameError:
+                self.page_title = f"Error"
+                self._set_page_content(f"## Error\n\n{str(request_exception)}")

app_tools/cookies.py

@@ -0,0 +1,713 @@
+COOKIES_LIST = [
+    {
+        "domain": ".youtube.com",
+        "expirationDate": 1718884961,
+        "hostOnly": False,
+        "httpOnly": False,
+        "name": "ST-xuwub9",
+        "path": "/",
+        "sameSite": None,
+        "secure": False,
+        "session": False,
+        "storeId": None,
+        "value": "session_logininfo=AFmmF2swRAIgf4gadACOuWOcipI1anW-dakEjtidNLkufnOC8uml7EECIDh2YisqWELDBJPTGUysCucJ3I0wjXxYjVHro1LHrdW0%3AQUQ3MjNmd2Jiajl3OWZYRnpFNnZlWWV5ZGJWZ0hpcmp4LVVPU280bk4zOS03Z0ozZG9fOFhWZ0dXaVo3NG1wTEg1b3hGaG10TFBlaFBnTlJfbER5bEp0aFhoNS1OLVhYNFRZT2F6ajgzOFpDbGhlUjZpMWRETlFFRjFfTTRiM0RnNTROSkdmMTFMVjFic1VuZ2trbGp4aktDa0JJUC1BWDh3"
+    },
+    {
+        "domain": ".youtube.com",
+        "expirationDate": 1753004444.745411,
+        "hostOnly": False,
+        "httpOnly": True,
+        "name": "__Secure-YEC",
+        "path": "/",
+        "sameSite": "lax",
+        "secure": True,
+        "session": False,
+        "storeId": None,
+        "value": "CgtRVnI5LW1zRHlQVSjbtNCzBjIhCgJGUhIbEhcSFRMLFBUWFwwYGRobHB0eHw4PIBAREiAk"
+    },
+    {
+        "domain": ".youtube.com",
+        "expirationDate": 1753434620.050824,
+        "hostOnly": False,
+        "httpOnly": True,
+        "name": "__Secure-3PSID",
+        "path": "/",
+        "sameSite": "no_restriction",
+        "secure": True,
+        "session": False,
+        "storeId": None,
+        "value": "g.a000kwibeLUu8Ea9Y-vLun7u3kU5VNJVuMAZl_jdfJaNm50JyDBB4ezJ_bdWu46a7YwObVn44wACgYKAakSARQSFQHGX2MicJcTzecTKH6bHzqU6TMbTxoVAUF8yKqQYK-MoI6Ql3vI2oYTB3E-0076"
+    },
+    {
+        "domain": ".youtube.com",
+        "expirationDate": 1750420959.974642,
+        "hostOnly": False,
+        "httpOnly": False,
+        "name": "SIDCC",
+        "path": "/",
+        "sameSite": None,
+        "secure": False,
+        "session": False,
+        "storeId": None,
+        "value": "AKEyXzWQZauHKOo8t87zoEcjaVNIYUX54ohoWXT-tX4aAhEuZzIIptxZAcNkHuG2oDXYL6t-lw"
+    },
+    {
+        "domain": ".youtube.com",
+        "expirationDate": 1753434620.050652,
+        "hostOnly": False,
+        "httpOnly": False,
+        "name": "SID",
+        "path": "/",
+        "sameSite": None,
+        "secure": False,
+        "session": False,
+        "storeId": None,
+        "value": "g.a000kwibeLUu8Ea9Y-vLun7u3kU5VNJVuMAZl_jdfJaNm50JyDBB6VHrZcC3gBAsFPbCQ0gF5AACgYKAYkSARQSFQHGX2Mi9kt0gHg5CxCYSkLQGHWaeBoVAUF8yKre_V6r3jZVak6JV4o2Q0FL0076"
+    },
+    {
+        "domain": ".youtube.com",
+        "expirationDate": 1750420958.397534,
+        "hostOnly": False,
+        "httpOnly": True,
+        "name": "__Secure-1PSIDTS",
+        "path": "/",
+        "sameSite": None,
+        "secure": True,
+        "session": False,
+        "storeId": None,
+        "value": "sidts-CjIB3EgAEkYL2L-GfrEzW5Dfy62S9oefGNLgst78S_986htCnGcfkxECch_9oz-qytSsZBAA"
+    },
+    {
+        "domain": ".youtube.com",
+        "expirationDate": 1753433494.44729,
+        "hostOnly": False,
+        "httpOnly": False,
+        "name": "_ga_M0180HEFCY",
+        "path": "/",
+        "sameSite": None,
+        "secure": False,
+        "session": False,
+        "storeId": None,
+        "value": "GS1.1.1718871908.1.0.1718873494.0.0.0"
+    },
+    {
+        "domain": ".youtube.com",
+        "expirationDate": 1753434620.050933,
+        "hostOnly": False,
+        "httpOnly": False,
+        "name": "SAPISID",
+        "path": "/",
+        "sameSite": None,
+        "secure": True,
+        "session": False,
+        "storeId": None,
+        "value": "mfeuiC-HraNJ-A03/ASXvCPNJSw7yTFgd6"
+    },
+    {
+        "domain": ".youtube.com",
+        "expirationDate": 1750420959.974764,
+        "hostOnly": False,
+        "httpOnly": True,
+        "name": "__Secure-1PSIDCC",
+        "path": "/",
+        "sameSite": None,
+        "secure": True,
+        "session": False,
+        "storeId": None,
+        "value": "AKEyXzWHDSoXGCZpZhPxRrnC7B1s8zGIUjeMVyvgtQfsm1fs92lXPtFEI_td9LBUyqVUe0xK"
+    },
+    {
+        "domain": ".youtube.com",
+        "expirationDate": 1753434620.050881,
+        "hostOnly": False,
+        "httpOnly": True,
+        "name": "SSID",
+        "path": "/",
+        "sameSite": None,
+        "secure": True,
+        "session": False,
+        "storeId": None,
+        "value": "AmlwXHnQvOQ10LVd-"
+    },
+    {
+        "domain": ".youtube.com",
+        "expirationDate": 1753434620.050959,
+        "hostOnly": False,
+        "httpOnly": False,
+        "name": "__Secure-1PAPISID",
+        "path": "/",
+        "sameSite": None,
+        "secure": True,
+        "session": False,
+        "storeId": None,
+        "value": "mfeuiC-HraNJ-A03/ASXvCPNJSw7yTFgd6"
+    },
+    {
+        "domain": ".youtube.com",
+        "expirationDate": 1753434620.050795,
+        "hostOnly": False,
+        "httpOnly": True,
+        "name": "__Secure-1PSID",
+        "path": "/",
+        "sameSite": None,
+        "secure": True,
+        "session": False,
+        "storeId": None,
+        "value": "g.a000kwibeLUu8Ea9Y-vLun7u3kU5VNJVuMAZl_jdfJaNm50JyDBBrlk7lRpKQGywAHEon7WGQAACgYKAQsSARQSFQHGX2MirAmnSRdZl6GPG6KLd4hOihoVAUF8yKoV17Tcj1a_OenIOkf2wBjO0076"
+    },
+    {
+        "domain": ".youtube.com",
+        "expirationDate": 1753434620.050993,
+        "hostOnly": False,
+        "httpOnly": False,
+        "name": "__Secure-3PAPISID",
+        "path": "/",
+        "sameSite": "no_restriction",
+        "secure": True,
+        "session": False,
+        "storeId": None,
+        "value": "mfeuiC-HraNJ-A03/ASXvCPNJSw7yTFgd6"
+    },
+    {
+        "domain": ".youtube.com",
+        "expirationDate": 1750420959.974815,
+        "hostOnly": False,
+        "httpOnly": True,
+        "name": "__Secure-3PSIDCC",
+        "path": "/",
+        "sameSite": "no_restriction",
+        "secure": True,
+        "session": False,
+        "storeId": None,
+        "value": "AKEyXzXM5UjKUEXwSHVmRAIo6hGHA4G63adj3EE1VdNriD0f38jZQbsUKiD4LQbA3BValmTFDg"
+    },
+    {
+        "domain": ".youtube.com",
+        "expirationDate": 1750420958.397647,
+        "hostOnly": False,
+        "httpOnly": True,
+        "name": "__Secure-3PSIDTS",
+        "path": "/",
+        "sameSite": "no_restriction",
+        "secure": True,
+        "session": False,
+        "storeId": None,
+        "value": "sidts-CjIB3EgAEkYL2L-GfrEzW5Dfy62S9oefGNLgst78S_986htCnGcfkxECch_9oz-qytSsZBAA"
+    },
+    {
+        "domain": ".youtube.com",
+        "expirationDate": 1753434620.050908,
+        "hostOnly": False,
+        "httpOnly": False,
+        "name": "APISID",
+        "path": "/",
+        "sameSite": None,
+        "secure": False,
+        "session": False,
+        "storeId": None,
+        "value": "IlQWLPjdNqziwCrV/ANG7Z4x5FF-IBxbZk"
+    },
+    {
+        "domain": ".youtube.com",
+        "expirationDate": 1753434620.050855,
+        "hostOnly": False,
+        "httpOnly": True,
+        "name": "HSID",
+        "path": "/",
+        "sameSite": None,
+        "secure": False,
+        "session": False,
+        "storeId": None,
+        "value": "AasA7hmRuTFv7vjoq"
+    },
+    {
+        "domain": ".youtube.com",
+        "expirationDate": 1753435873.577793,
+        "hostOnly": False,
+        "httpOnly": True,
+        "name": "LOGIN_INFO",
+        "path": "/",
+        "sameSite": "no_restriction",
+        "secure": True,
+        "session": False,
+        "storeId": None,
+        "value": "AFmmF2swRAIgf4gadACOuWOcipI1anW-dakEjtidNLkufnOC8uml7EECIDh2YisqWELDBJPTGUysCucJ3I0wjXxYjVHro1LHrdW0:QUQ3MjNmd2Jiajl3OWZYRnpFNnZlWWV5ZGJWZ0hpcmp4LVVPU280bk4zOS03Z0ozZG9fOFhWZ0dXaVo3NG1wTEg1b3hGaG10TFBlaFBnTlJfbER5bEp0aFhoNS1OLVhYNFRZT2F6ajgzOFpDbGhlUjZpMWRETlFFRjFfTTRiM0RnNTROSkdmMTFMVjFic1VuZ2trbGp4aktDa0JJUC1BWDh3"
+    },
+    {
+        "domain": ".youtube.com",
+        "expirationDate": 1753444956.555608,
+        "hostOnly": False,
+        "httpOnly": False,
+        "name": "PREF",
+        "path": "/",
+        "sameSite": None,
+        "secure": True,
+        "session": False,
+        "storeId": None,
+        "value": "f4=4000000&f6=40000000&tz=Europe.Paris&f5=30000&f7=100"
+    }
+]
+
+COOKIES_LIST += [
+    {
+        "domain": ".www.researchgate.net",
+        "hostOnly": False,
+        "httpOnly": True,
+        "name": "isInstIp",
+        "path": "/",
+        "sameSite": None,
+        "secure": True,
+        "session": True,
+        "storeId": None,
+        "value": "False"
+    },
+    {
+        "domain": ".researchgate.net",
+        "expirationDate": 1734423981,
+        "hostOnly": False,
+        "httpOnly": False,
+        "name": "__eoi",
+        "path": "/",
+        "sameSite": None,
+        "secure": False,
+        "session": False,
+        "storeId": None,
+        "value": "ID=c26f752377373146:T=1718871981:RT=1718884914:S=AA-AfjZw-T_OOX2kW2LLaFzXImgc"
+    },
+    {
+        "domain": ".www.researchgate.net",
+        "expirationDate": 1753444909.646103,
+        "hostOnly": False,
+        "httpOnly": True,
+        "name": "ptc",
+        "path": "/",
+        "sameSite": None,
+        "secure": True,
+        "session": False,
+        "storeId": None,
+        "value": "RG1.8947708639250500550.1718872043"
+    },
+    {
+        "domain": ".researchgate.net",
+        "expirationDate": 1750507578,
+        "hostOnly": False,
+        "httpOnly": False,
+        "name": "euconsent-v2-didomi",
+        "path": "/",
+        "sameSite": "lax",
+        "secure": True,
+        "session": False,
+        "storeId": None,
+        "value": "CQAgmoAQAgmoAAHABBENA5EsAP_gAEPgAAYgJ2pB5G5UTWlBIG53YMskIAUFhFBoQEAgAACAAwIBSBIAIIwEAGAAIAgAICACAAIAIBIAIABAGAAAAAAAYIAAIAAIAAAQIAAKIAAAAAAAAgBQAAgIAgggEAAAgEBEABAAgAAAEIIAQNgACgAAACCAAAAAAAABAAAAAAAAQAAAAAAAYCQAAAJIAAAAACAIABAIAAAAAAAAAAAAAAAABBAAIJ2wPIAFAAXABQAFQALgAcAA8ACAAEgALwAZAA0ACIAEcAJgAUgAqgBcADEAGgAPQAfgBEACOAE4AMMAZYA0QBsgDkAHOAO4AfsBBwEIAItARwBHQC6gHUAO2Ae0A_4CHQEXgJ2AUOAo8BT4CpQFqALYAXmAwQBkgDLAGXANjAhCBG8CbAE3gJ1gTtAA.f_wACHwAAAAA"
+    },
+    {
+        "domain": ".researchgate.net",
+        "expirationDate": 1718885236,
+        "hostOnly": False,
+        "httpOnly": False,
+        "name": "_gat",
+        "path": "/",
+        "sameSite": None,
+        "secure": False,
+        "session": False,
+        "storeId": None,
+        "value": "1"
+    },
+    {
+        "domain": "www.researchgate.net",
+        "expirationDate": 1721477183,
+        "hostOnly": True,
+        "httpOnly": False,
+        "name": "_pbjs_userid_consent_data",
+        "path": "/",
+        "sameSite": "lax",
+        "secure": False,
+        "session": False,
+        "storeId": None,
+        "value": "3524755945110770"
+    },
+    {
+        "domain": ".researchgate.net",
+        "expirationDate": 1752567981,
+        "hostOnly": False,
+        "httpOnly": False,
+        "name": "__gads",
+        "path": "/",
+        "sameSite": None,
+        "secure": False,
+        "session": False,
+        "storeId": None,
+        "value": "ID=eca2adb88969c830:T=1718871981:RT=1718884914:S=ALNI_MY2qZchynrhWX6hWMlaI87Pcj9riQ"
+    },
+    {
+        "domain": ".researchgate.net",
+        "expirationDate": 1718886709.646173,
+        "hostOnly": False,
+        "httpOnly": True,
+        "name": "__cf_bm",
+        "path": "/",
+        "sameSite": "no_restriction",
+        "secure": True,
+        "session": False,
+        "storeId": None,
+        "value": "IkQ_J4ciBzKQduRvjqsfSmQu8UygDWbHeROO5JVccfo-1718884909-1.0.1.1-qvNGEdbfI0HfhFP6kwe7R7mkTqODNhFuKhs72lLly6K2BOPMG3kbahpQFGvPK0U8FUfkznkq65gngd1sWj7sDA"
+    },
+    {
+        "domain": ".researchgate.net",
+        "expirationDate": 1752567981,
+        "hostOnly": False,
+        "httpOnly": False,
+        "name": "__gpi",
+        "path": "/",
+        "sameSite": None,
+        "secure": False,
+        "session": False,
+        "storeId": None,
+        "value": "UID=00000e4e9aa2e6f2:T=1718871981:RT=1718884914:S=ALNI_MYFNrgzkKn7K6Bd2y8hC6GJCvDiSg"
+    },
+    {
+        "domain": ".researchgate.net",
+        "hostOnly": False,
+        "httpOnly": True,
+        "name": "_cfuvid",
+        "path": "/",
+        "sameSite": "no_restriction",
+        "secure": True,
+        "session": True,
+        "storeId": None,
+        "value": "_GPmGZkBymiH3UiqTqzakEpi98br3nfFUWC2_u_wqkc-1718884909785-0.0.1.1-604800000"
+    },
+    {
+        "domain": ".researchgate.net",
+        "expirationDate": 1753445177.271667,
+        "hostOnly": False,
+        "httpOnly": False,
+        "name": "_ga",
+        "path": "/",
+        "sameSite": None,
+        "secure": False,
+        "session": False,
+        "storeId": None,
+        "value": "GA1.1.1525244793.1718885177"
+    },
+    {
+        "domain": ".researchgate.net",
+        "expirationDate": 1753445177.271482,
+        "hostOnly": False,
+        "httpOnly": False,
+        "name": "_ga_4P31SJ70EJ",
+        "path": "/",
+        "sameSite": None,
+        "secure": False,
+        "session": False,
+        "storeId": None,
+        "value": "GS1.1.1718885177.1.0.1718885177.0.0.0"
+    },
+    {
+        "domain": ".researchgate.net",
+        "expirationDate": 1718971576,
+        "hostOnly": False,
+        "httpOnly": False,
+        "name": "_gid",
+        "path": "/",
+        "sameSite": None,
+        "secure": False,
+        "session": False,
+        "storeId": None,
+        "value": "GA1.2.854907463.1718885177"
+    },
+    {
+        "domain": ".www.researchgate.net",
+        "expirationDate": 1750407982.506505,
+        "hostOnly": False,
+        "httpOnly": True,
+        "name": "did",
+        "path": "/",
+        "sameSite": None,
+        "secure": True,
+        "session": False,
+        "storeId": None,
+        "value": "1dWLO3C6am8l667Q4VUlBo0O1LI49Qi2Vw21SJEXHavBDYT56DI9007W5rYGVFVH"
+    },
+    {
+        "domain": ".researchgate.net",
+        "expirationDate": 1750507578,
+        "hostOnly": False,
+        "httpOnly": False,
+        "name": "didomi_token",
+        "path": "/",
+        "sameSite": "lax",
+        "secure": True,
+        "session": False,
+        "storeId": None,
+        "value": "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"
+    },
+    {
+        "domain": ".www.researchgate.net",
+        "hostOnly": False,
+        "httpOnly": True,
+        "name": "hasPdpNext",
+        "path": "/",
+        "sameSite": None,
+        "secure": True,
+        "session": True,
+        "storeId": None,
+        "value": "False"
+    },
+    {
+        "domain": ".researchgate.net",
+        "expirationDate": 1750421183,
+        "hostOnly": False,
+        "httpOnly": False,
+        "name": "ph_phc_ma1XTQyee96N1GML6qUTgLQRiDifnRcE9STiHTZ0CfZ_posthog",
+        "path": "/",
+        "sameSite": "lax",
+        "secure": True,
+        "session": False,
+        "storeId": None,
+        "value": "%7B%22distinct_id%22%3A%220190358a-56a1-7313-83b0-d13dddeac787%22%2C%22%24sesid%22%3A%5B1718885183223%2C%220190358a-56a1-7313-83b0-d13b2b87778d%22%2C1718885176993%5D%2C%22%24session_is_sampled%22%3Atrue%7D"
+    },
+    {
+        "domain": ".www.researchgate.net",
+        "hostOnly": False,
+        "httpOnly": True,
+        "name": "sid",
+        "path": "/",
+        "sameSite": None,
+        "secure": True,
+        "session": True,
+        "storeId": None,
+        "value": "qmH5Lc4f0CUJ3zeaxORcV0S8I8V1MuCFZtcIQqPYtv1XPejrbSLAQRbT50PL40TqeKQ1XsQDWt9gtYVzuL80bRmPjw6jn3cQ0ikNqW40maHcQ3JL2Vfa8ZZf0j7p35eJ"
+    }
+]
+
+COOKIES_LIST += [
+    {
+        "domain": "github.com",
+        "hostOnly": True,
+        "httpOnly": True,
+        "name": "_gh_sess",
+        "path": "/",
+        "sameSite": "lax",
+        "secure": True,
+        "session": True,
+        "storeId": None,
+        "value": "P%2Fmof1avuqwHaUQUIJR%2FZYn7jqbT7lgGuTGjp1BGAFIG5UpNDusEE3b8dRjz0eATE5xPdPjLYFqMs%2FI9AOalKX4YuYfSEEnxCMawU01099b4o9Xzzcv%2BmecrmO0Q8q%2Bdq1h8SIv6nvPP7HzlFesl8ysafb9b%2F0q6dTArKdSOurasza8UgLSYD08ofA50Pcm0IG7CTzF8ZCizrGgGTMi%2F%2B7L3E17jav5PM1Sf2vQKg15Gbg1QIOppJJHzlufgQoZigqFv%2BWznaws0Tt7Y2lSFCw%3D%3D--CJRhqMXJnwOaJgk4--DhUErlL4GdROikEjKD4O9g%3D%3D"
+    },
+    {
+        "domain": ".github.com",
+        "expirationDate": 1750408875.763785,
+        "hostOnly": False,
+        "httpOnly": False,
+        "name": "_octo",
+        "path": "/",
+        "sameSite": "lax",
+        "secure": True,
+        "session": False,
+        "storeId": None,
+        "value": "GH1.1.728652011.1718872875"
+    },
+    {
+        "domain": ".github.com",
+        "expirationDate": 1750408875.763926,
+        "hostOnly": False,
+        "httpOnly": True,
+        "name": "logged_in",
+        "path": "/",
+        "sameSite": "lax",
+        "secure": True,
+        "session": False,
+        "storeId": None,
+        "value": "no"
+    },
+    {
+        "domain": ".github.com",
+        "hostOnly": False,
+        "httpOnly": False,
+        "name": "preferred_color_mode",
+        "path": "/",
+        "sameSite": "lax",
+        "secure": True,
+        "session": True,
+        "storeId": None,
+        "value": "dark"
+    },
+    {
+        "domain": ".github.com",
+        "hostOnly": False,
+        "httpOnly": False,
+        "name": "tz",
+        "path": "/",
+        "sameSite": "lax",
+        "secure": True,
+        "session": True,
+        "storeId": None,
+        "value": "Europe%2FParis"
+    }
+]
+
+COOKIES_LIST += [
+    {
+        "domain": ".web.archive.org",
+        "expirationDate": 1718886430,
+        "hostOnly": False,
+        "httpOnly": False,
+        "name": "_gat",
+        "path": "/web/20201123221659/http://orcid.org/",
+        "sameSite": None,
+        "secure": False,
+        "session": False,
+        "storeId": None,
+        "value": "1"
+    },
+    {
+        "domain": ".web.archive.org",
+        "expirationDate": 1718972770,
+        "hostOnly": False,
+        "httpOnly": False,
+        "name": "_gid",
+        "path": "/web/20201123221659/http://orcid.org/",
+        "sameSite": None,
+        "secure": False,
+        "session": False,
+        "storeId": None,
+        "value": "GA1.2.402246368.1606169825"
+    },
+    {
+        "domain": ".web.archive.org",
+        "expirationDate": 1753446370.315621,
+        "hostOnly": False,
+        "httpOnly": False,
+        "name": "_ga",
+        "path": "/web/20201123221659/http://orcid.org/",
+        "sameSite": None,
+        "secure": False,
+        "session": False,
+        "storeId": None,
+        "value": "GA1.2.1301409987.1606169825"
+    },
+    {
+        "domain": ".web.archive.org",
+        "expirationDate": 1750422367,
+        "hostOnly": False,
+        "httpOnly": False,
+        "name": "_hjid",
+        "path": "/web/20201123221659/http://orcid.org/",
+        "sameSite": "lax",
+        "secure": False,
+        "session": False,
+        "storeId": None,
+        "value": "07f80263-a631-4bf4-8ffd-8fc8912085e2"
+    },
+    {
+        "domain": ".web.archive.org",
+        "expirationDate": 1718888167,
+        "hostOnly": False,
+        "httpOnly": False,
+        "name": "_hjFirstSeen",
+        "path": "/web/20201123221659/http://orcid.org/",
+        "sameSite": "lax",
+        "secure": False,
+        "session": False,
+        "storeId": None,
+        "value": "1"
+    }
+]
+COOKIES_LIST += [
+    {
+        "domain": "orcid.org",
+        "hostOnly": True,
+        "httpOnly": False,
+        "name": "AWSELBCORS",
+        "path": "/",
+        "sameSite": "no_restriction",
+        "secure": True,
+        "session": True,
+        "storeId": None,
+        "value": "CBD1D7FF1216388FA48838CBCA4774FD22800B8FB548A40EF92BB0994D5B77A8410307CDEAA69C52236663F2BF89B252C17BC0FCDF790FD59771BDDF6EA8CA4CFD29D8733F"
+    },
+    {
+        "domain": ".orcid.org",
+        "expirationDate": 1753452454.637671,
+        "hostOnly": False,
+        "httpOnly": False,
+        "name": "_ga_9R61FWK9H5",
+        "path": "/",
+        "sameSite": None,
+        "secure": False,
+        "session": False,
+        "storeId": None,
+        "value": "GS1.1.1718892454.1.0.1718892454.0.0.0"
+    },
+    {
+        "domain": ".orcid.org",
+        "expirationDate": 1753452454.63421,
+        "hostOnly": False,
+        "httpOnly": False,
+        "name": "_ga",
+        "path": "/",
+        "sameSite": None,
+        "secure": False,
+        "session": False,
+        "storeId": None,
+        "value": "GA1.1.2021310691.1718892455"
+    },
+    {
+        "domain": "orcid.org",
+        "hostOnly": True,
+        "httpOnly": False,
+        "name": "AWSELB",
+        "path": "/",
+        "sameSite": None,
+        "secure": False,
+        "session": True,
+        "storeId": None,
+        "value": "CBD1D7FF1216388FA48838CBCA4774FD22800B8FB548A40EF92BB0994D5B77A8410307CDEAA69C52236663F2BF89B252C17BC0FCDF790FD59771BDDF6EA8CA4CFD29D8733F"
+    },
+    {
+        "domain": ".orcid.org",
+        "expirationDate": 1750428454,
+        "hostOnly": False,
+        "httpOnly": False,
+        "name": "OptanonAlertBoxClosed",
+        "path": "/",
+        "sameSite": "lax",
+        "secure": False,
+        "session": False,
+        "storeId": None,
+        "value": "2024-06-20T14:07:34.583Z"
+    },
+    {
+        "domain": ".orcid.org",
+        "expirationDate": 1750428454,
+        "hostOnly": False,
+        "httpOnly": False,
+        "name": "OptanonConsent",
+        "path": "/",
+        "sameSite": "lax",
+        "secure": False,
+        "session": False,
+        "storeId": None,
+        "value": "isGpcEnabled=0&datestamp=Thu+Jun+20+2024+16%3A07%3A34+GMT%2B0200+(heure+d%E2%80%99%C3%A9t%C3%A9+d%E2%80%99Europe+centrale)&version=202310.2.0&browserGpcFlag=0&isIABGlobal=False&hosts=&landingPath=NotLandingPage&groups=C0001%3A1%2CC0003%3A1%2CC0002%3A1%2CC0004%3A1"
+    },
+    {
+        "domain": "orcid.org",
+        "hostOnly": True,
+        "httpOnly": False,
+        "name": "XSRF-TOKEN",
+        "path": "/",
+        "sameSite": None,
+        "secure": True,
+        "session": True,
+        "storeId": None,
+        "value": "6957be7a-bcb4-4d59-a522-ea9b6b210ed9"
+    }
+]
+from requests.cookies import RequestsCookieJar
+
+# Create a RequestsCookieJar instance
+COOKIES = RequestsCookieJar()
+
+# Add cookies to the jar
+for cookie in COOKIES_LIST:
+    COOKIES.set(cookie['name'], cookie['value'], domain=cookie['domain'], path=cookie['path'])

app_tools/mdconvert.py

@@ -0,0 +1,659 @@
+# ruff: noqa: E722
+import json
+import os
+import requests
+import re
+import markdownify
+import mimetypes
+import html
+import puremagic
+import tempfile
+import copy
+import mammoth
+import pptx
+import pandas as pd
+import traceback
+
+from urllib.parse import urlparse, parse_qs
+from bs4 import BeautifulSoup
+from typing import Any, Dict, List, Optional, Union
+import pdfminer
+import pdfminer.high_level
+from youtube_transcript_api import YouTubeTranscriptApi
+
+
+
+class DocumentConverterResult:
+    """The result of converting a document to text."""
+
+    def __init__(self, title: Union[str, None] = None, text_content: str = ""):
+        self.title = title
+        self.text_content = text_content
+
+
+class DocumentConverter:
+    def convert(self, local_path, **kwargs) -> Union[None, DocumentConverterResult]:
+        raise NotImplementedError()
+
+
+class PlainTextConverter(DocumentConverter):
+    """Anything with content type text/plain"""
+
+    def convert(self, local_path, **kwargs) -> Union[None, DocumentConverterResult]:
+        extension = kwargs.get("file_extension", "")
+        if extension == "":
+            return None
+
+        content_type, encoding = mimetypes.guess_type("__placeholder" + extension)
+
+        text_content = ""
+        with open(local_path, "rt") as fh:
+            text_content = fh.read()
+
+        return DocumentConverterResult(
+            title=None,
+            text_content=text_content,
+        )
+
+
+class HtmlConverter(DocumentConverter):
+    """Anything with content type text/html"""
+
+    def convert(self, local_path, **kwargs) -> Union[None, DocumentConverterResult]:
+        # Bail if not html
+        extension = kwargs.get("file_extension", "")
+        if extension.lower() not in [".html", ".htm"]:
+            return None
+
+        result = None
+        with open(local_path, "rt") as fh:
+            result = self._convert(fh.read())
+
+        return result
+
+    def _convert(self, html_content) -> Union[None, DocumentConverterResult]:
+        """Helper function that converts and HTML string."""
+
+        # Parse the string
+        soup = BeautifulSoup(html_content, "html.parser")
+
+        # Remove javascript and style blocks
+        for script in soup(["script", "style"]):
+            script.extract()
+
+        # Print only the main content
+        body_elm = soup.find("body")
+        webpage_text = ""
+        if body_elm:
+            webpage_text = markdownify.MarkdownConverter().convert_soup(body_elm)
+        else:
+            webpage_text = markdownify.MarkdownConverter().convert_soup(soup)
+
+        return DocumentConverterResult(
+            title=None if soup.title is None else soup.title.string,
+            text_content=webpage_text,
+        )
+
+
+class WikipediaConverter(DocumentConverter):
+    """Handle Wikipedia pages separately, focusing only on the main document content."""
+
+    def convert(self, local_path, **kwargs) -> Union[None, DocumentConverterResult]:
+        # Bail if not Wikipedia
+        extension = kwargs.get("file_extension", "")
+        if extension.lower() not in [".html", ".htm"]:
+            return None
+        url = kwargs.get("url", "")
+        if not re.search(r"^https?:\/\/[a-zA-Z]{2,3}\.wikipedia.org\/", url):
+            return None
+
+        # Parse the file
+        soup = None
+        with open(local_path, "rt") as fh:
+            soup = BeautifulSoup(fh.read(), "html.parser")
+
+        # Remove javascript and style blocks
+        for script in soup(["script", "style"]):
+            script.extract()
+
+        # Print only the main content
+        body_elm = soup.find("div", {"id": "mw-content-text"})
+        title_elm = soup.find("span", {"class": "mw-page-title-main"})
+
+        webpage_text = ""
+        if body_elm:
+            # What's the title
+            main_title = soup.title.string
+            if title_elm and len(title_elm) > 0:
+                main_title = title_elm.string
+
+            # Convert the page
+            webpage_text = "# " + main_title + "\n\n" + markdownify.MarkdownConverter().convert_soup(body_elm)
+        else:
+            webpage_text = markdownify.MarkdownConverter().convert_soup(soup)
+
+        return DocumentConverterResult(
+            title=soup.title.string,
+            text_content=webpage_text,
+        )
+
+
+class YouTubeConverter(DocumentConverter):
+    """Handle YouTube specially, focusing on the video title, description, and transcript."""
+
+    def convert(self, local_path, **kwargs) -> Union[None, DocumentConverterResult]:
+        # Bail if not YouTube
+        extension = kwargs.get("file_extension", "")
+        if extension.lower() not in [".html", ".htm"]:
+            return None
+        url = kwargs.get("url", "")
+        if not url.startswith("https://www.youtube.com/watch?"):
+            return None
+
+        # Parse the file
+        soup = None
+        with open(local_path, "rt") as fh:
+            soup = BeautifulSoup(fh.read(), "html.parser")
+
+        # Read the meta tags
+        metadata = {"title": soup.title.string}
+        for meta in soup(["meta"]):
+            for a in meta.attrs:
+                if a in ["itemprop", "property", "name"]:
+                    metadata[meta[a]] = meta.get("content", "")
+                    break
+
+        # We can also try to read the full description. This is more prone to breaking, since it reaches into the page implementation
+        try:
+            for script in soup(["script"]):
+                content = script.text
+                if "ytInitialData" in content:
+                    lines = re.split(r"\r?\n", content)
+                    obj_start = lines[0].find("{")
+                    obj_end = lines[0].rfind("}")
+                    if obj_start >= 0 and obj_end >= 0:
+                        data = json.loads(lines[0][obj_start : obj_end + 1])
+                        attrdesc = self._findKey(data, "attributedDescriptionBodyText")
+                        if attrdesc:
+                            metadata["description"] = attrdesc["content"]
+                    break
+        except:
+            pass
+
+        # Start preparing the page
+        webpage_text = "# YouTube\n"
+
+        title = self._get(metadata, ["title", "og:title", "name"])
+        if title:
+            webpage_text += f"\n## {title}\n"
+
+        stats = ""
+        views = self._get(metadata, ["interactionCount"])
+        if views:
+            stats += f"- **Views:** {views}\n"
+
+        keywords = self._get(metadata, ["keywords"])
+        if keywords:
+            stats += f"- **Keywords:** {keywords}\n"
+
+        runtime = self._get(metadata, ["duration"])
+        if runtime:
+            stats += f"- **Runtime:** {runtime}\n"
+
+        if len(stats) > 0:
+            webpage_text += f"\n### Video Metadata\n{stats}\n"
+
+        description = self._get(metadata, ["description", "og:description"])
+        if description:
+            webpage_text += f"\n### Description\n{description}\n"
+
+        transcript_text = ""
+        parsed_url = urlparse(url)
+        params = parse_qs(parsed_url.query)
+
+        video_id = params["v"][0]
+        # Must be a single transcript.
+        print("VIDDDD ID:", video_id)
+        transcript = YouTubeTranscriptApi.get_transcript(video_id)
+        transcript_text = " ".join([part["text"] for part in transcript])
+        # Alternative formatting:
+        # formatter = TextFormatter()
+        # formatter.format_transcript(transcript)
+        if transcript_text:
+            webpage_text += f"\n### Transcript\n{transcript_text}\n"
+
+        return DocumentConverterResult(
+            title=title if title else soup.title.string,
+            text_content=webpage_text,
+        )
+
+    def _get(self, json, keys, default=None):
+        for k in keys:
+            if k in json:
+                return json[k]
+        return default
+
+    def _findKey(self, json, key):
+        if isinstance(json, list):
+            for elm in json:
+                ret = self._findKey(elm, key)
+                if ret is not None:
+                    return ret
+        elif isinstance(json, dict):
+            for k in json:
+                if k == key:
+                    return json[k]
+                else:
+                    ret = self._findKey(json[k], key)
+                    if ret is not None:
+                        return ret
+        return None
+
+
+class PdfConverter(DocumentConverter):
+    def convert(self, local_path, **kwargs) -> Union[None, DocumentConverterResult]:
+        # Bail if not a PDF
+        extension = kwargs.get("file_extension", "")
+        if extension.lower() != ".pdf":
+            return None
+
+        return DocumentConverterResult(
+            title=None,
+            text_content=pdfminer.high_level.extract_text(local_path),
+        )
+
+from huggingface_hub import InferenceClient
+class AudioConverter(DocumentConverter):
+    def __init__(self):
+        super().__init__()
+        self.client = InferenceClient("distil-whisper/distil-large-v3")
+
+    def convert(self, local_path, **kwargs) -> Union[None, DocumentConverterResult]:
+        # Bail if not an audio file
+        extension = kwargs.get("file_extension", "")
+        if extension.lower() not in [".wav", ".mp3", ".flac", ".m4a"]:
+            return None
+        try:
+            result = self.client.automatic_speech_recognition(audio=local_path).text
+        except Exception as e:
+            print("Exception in decoding audio:", e)
+            from openai import OpenAI
+            oai_client = OpenAI()
+            from pathlib import Path
+            result = oai_client.audio.transcriptions.create(
+                model="whisper-1", 
+                file=Path(local_path)
+            ).text
+
+        return DocumentConverterResult(
+            title=None,
+            text_content=result,
+        )
+
+
+class DocxConverter(HtmlConverter):
+    def convert(self, local_path, **kwargs) -> Union[None, DocumentConverterResult]:
+        # Bail if not a DOCX
+        extension = kwargs.get("file_extension", "")
+        if extension.lower() != ".docx":
+            return None
+
+        result = None
+        with open(local_path, "rb") as docx_file:
+            result = mammoth.convert_to_html(docx_file)
+            html_content = result.value
+            result = self._convert(html_content)
+
+        return result
+
+
+class XlsxConverter(HtmlConverter):
+    def convert(self, local_path, **kwargs) -> Union[None, DocumentConverterResult]:
+        # Bail if not a XLSX
+        extension = kwargs.get("file_extension", "")
+
+        if extension.lower() not in [".xlsx", ".xls"]:
+            return None
+
+        sheets = pd.read_excel(local_path, sheet_name=None)
+        md_content = ""
+        for s in sheets:
+            md_content += f"## {s}\n"
+            html_content = sheets[s].to_html(index=False)
+            md_content += self._convert(html_content).text_content.strip() + "\n\n"
+
+        return DocumentConverterResult(
+            title=None,
+            text_content=md_content.strip(),
+        )
+
+
+import xml.etree.ElementTree as ET
+class XmlConverter(DocumentConverter):
+    def convert(self, local_path, **kwargs) -> None | DocumentConverterResult:
+        # Parse the XML string
+        extension = kwargs.get("file_extension", "")
+
+        if extension.lower() not in [".xml"]:
+            return None
+        
+        xml_string = ""
+        with open(local_path, "rt") as fh:
+            xml_string = fh.read()
+        
+        def extract_table_from_html_like(xml_root):
+            table = xml_root.find('.//table')
+            if table is None:
+                raise ValueError("No table found in the XML")
+
+            headers = [th.text for th in table.find('thead').findall('th')]
+            rows = [[td.text for td in tr.findall('td')] for tr in table.find('tbody').findall('tr')]
+            
+            # Create markdown table
+            markdown = '| ' + ' | '.join(headers) + ' |\n'
+            markdown += '| ' + ' | '.join(['---'] * len(headers)) + ' |\n'
+            for row in rows:
+                markdown += '| ' + ' | '.join(row) + ' |\n'
+
+        def extract_table_from_wordml(xml_root, namespaces):
+            # Parse the XML content
+            root = xml_root
+            namespace = {'w': 'http://schemas.microsoft.com/office/word/2003/wordml'}
+            
+            # Extract text content
+            body = root.find('w:body', namespace)
+            paragraphs = body.findall('.//w:p', namespace)
+            text_content = []
+            for para in paragraphs:
+                texts = para.findall('.//w:t', namespace)
+                for text in texts:
+                    text_content.append(text.text)
+            
+            return '\n'.join(text_content)
+
+        # Parse the XML string
+        root = ET.fromstring(xml_string)
+        namespaces = {'w': 'http://schemas.microsoft.com/office/word/2003/wordml'}
+        
+        if root.tag.endswith('wordDocument'):
+            markdown = extract_table_from_wordml(root, namespaces)
+        else:
+            markdown = extract_table_from_html_like(root)
+
+        return DocumentConverterResult(
+            title=None,
+            text_content=markdown.strip(),
+        )
+
+class PptxConverter(HtmlConverter):
+    def convert(self, local_path, **kwargs) -> Union[None, DocumentConverterResult]:
+        # Bail if not a PPTX
+        extension = kwargs.get("file_extension", "")
+        if extension.lower() != ".pptx":
+            return None
+
+        md_content = ""
+
+        presentation = pptx.Presentation(local_path)
+        slide_num = 0
+        for slide in presentation.slides:
+            slide_num += 1
+
+            md_content += f"\n\n<!-- Slide number: {slide_num} -->\n"
+
+            title = slide.shapes.title
+            for shape in slide.shapes:
+                # Pictures
+                if self._is_picture(shape):
+                    # https://github.com/scanny/python-pptx/pull/512#issuecomment-1713100069
+                    alt_text = ""
+                    try:
+                        alt_text = shape._element._nvXxPr.cNvPr.attrib.get("descr", "")
+                    except:
+                        pass
+
+                    # A placeholder name
+                    filename = re.sub(r"\W", "", shape.name) + ".jpg"
+                    # try:
+                    #    filename = shape.image.filename
+                    # except:
+                    #    pass
+
+                    md_content += "\n![" + (alt_text if alt_text else shape.name) + "](" + filename + ")\n"
+
+                # Tables
+                if self._is_table(shape):
+                    html_table = "<html><body><table>"
+                    first_row = True
+                    for row in shape.table.rows:
+                        html_table += "<tr>"
+                        for cell in row.cells:
+                            if first_row:
+                                html_table += "<th>" + html.escape(cell.text) + "</th>"
+                            else:
+                                html_table += "<td>" + html.escape(cell.text) + "</td>"
+                        html_table += "</tr>"
+                        first_row = False
+                    html_table += "</table></body></html>"
+                    md_content += "\n" + self._convert(html_table).text_content.strip() + "\n"
+
+                # Text areas
+                elif shape.has_text_frame:
+                    if shape == title:
+                        md_content += "# " + shape.text.lstrip() + " "
+                    else:
+                        md_content += shape.text + " "
+
+            md_content = md_content.strip()
+
+            if slide.has_notes_slide:
+                md_content += "\n\n### Notes:\n"
+                notes_frame = slide.notes_slide.notes_text_frame
+                if notes_frame is not None:
+                    md_content += notes_frame.text
+                md_content = md_content.strip()
+
+        return DocumentConverterResult(
+            title=None,
+            text_content=md_content.strip(),
+        )
+
+    def _is_picture(self, shape):
+        if shape.shape_type == pptx.enum.shapes.MSO_SHAPE_TYPE.PICTURE:
+            return True
+        if shape.shape_type == pptx.enum.shapes.MSO_SHAPE_TYPE.PLACEHOLDER:
+            if hasattr(shape, "image"):
+                return True
+        return False
+
+    def _is_table(self, shape):
+        if shape.shape_type == pptx.enum.shapes.MSO_SHAPE_TYPE.TABLE:
+            return True
+        return False
+
+class FileConversionException(Exception):
+    pass
+
+class UnsupportedFormatException(Exception):
+    pass
+
+class MarkdownConverter:
+    """(In preview) An extremely simple text-based document reader, suitable for LLM use.
+    This reader will convert common file-types or webpages to Markdown."""
+
+    def __init__(
+        self,
+        requests_session: Optional[requests.Session] = None,
+    ):
+        if requests_session is None:
+            self._requests_session = requests.Session()
+        else:
+            self._requests_session = requests_session
+
+
+        self._page_converters: List[DocumentConverter] = []
+
+        # Register converters for successful browsing operations
+        # Later registrations are tried first / take higher priority than earlier registrations
+        # To this end, the most specific converters should appear below the most generic converters
+        self.register_page_converter(WikipediaConverter())
+        self.register_page_converter(XmlConverter())
+        self.register_page_converter(YouTubeConverter())
+        self.register_page_converter(DocxConverter())
+        self.register_page_converter(XlsxConverter())
+        self.register_page_converter(PptxConverter())
+        # self.register_page_converter(ImageConverter())
+        self.register_page_converter(PdfConverter())
+        self.register_page_converter(AudioConverter())
+        self.register_page_converter(HtmlConverter())
+        self.register_page_converter(PlainTextConverter())
+
+    def convert(self, source, **kwargs):
+        """
+        Args:
+            - source: can be a string representing a path or url, or a requests.response object
+            - extension: specifies the file extension to use when interpreting the file. If None, infer from source (path, uri, content-type, etc.)
+        """
+
+        # Local path or url
+        if isinstance(source, str):
+            if source.startswith("http://") or source.startswith("https://") or source.startswith("file://"):
+                return self.convert_url(source, **kwargs)
+            else:
+                return self.convert_local(source, **kwargs)
+        # Request response
+        elif isinstance(source, requests.Response):
+            return self.convert_response(source, **kwargs)
+
+    def convert_local(self, path, **kwargs):
+        # Prepare a list of extensions to try (in order of priority)
+        ext = kwargs.get("file_extension")
+        extensions = [ext] if ext is not None else []
+
+        # Get extension alternatives from the path and puremagic
+        base, ext = os.path.splitext(path)
+        self._append_ext(extensions, ext)
+        self._append_ext(extensions, self._guess_ext_magic(path))
+
+        # Convert
+        return self._convert(path, extensions, **kwargs)
+
+    def convert_url(self, url, **kwargs):
+        # Send a HTTP request to the URL
+        user_agent = "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/119.0.0.0 Safari/537.36 Edg/119.0.0.0"
+        response = self._requests_session.get(url, stream=True, headers={"User-Agent": user_agent})
+        response.raise_for_status()
+        return self.convert_response(response, **kwargs)
+
+    def convert_response(self, response, **kwargs):
+        # Prepare a list of extensions to try (in order of priority)
+        ext = kwargs.get("file_extension")
+        extensions = [ext] if ext is not None else []
+
+        # Guess from the mimetype
+        content_type = response.headers.get("content-type", "").split(";")[0]
+        self._append_ext(extensions, mimetypes.guess_extension(content_type))
+
+        # Read the content disposition if there is one
+        content_disposition = response.headers.get("content-disposition", "")
+        m = re.search(r"filename=([^;]+)", content_disposition)
+        if m:
+            base, ext = os.path.splitext(m.group(1).strip("\"'"))
+            self._append_ext(extensions, ext)
+
+        # Read from the extension from the path
+        base, ext = os.path.splitext(urlparse(response.url).path)
+        self._append_ext(extensions, ext)
+
+        # Save the file locally to a temporary file. It will be deleted before this method exits
+        handle, temp_path = tempfile.mkstemp()
+        fh = os.fdopen(handle, "wb")
+        result = None
+        try:
+            # Download the file
+            for chunk in response.iter_content(chunk_size=512):
+                fh.write(chunk)
+            fh.close()
+
+            # Use puremagic to check for more extension options
+            self._append_ext(extensions, self._guess_ext_magic(temp_path))
+
+            # Convert
+            result = self._convert(temp_path, extensions, url=response.url)
+        except Exception as e:
+            print(f"Error in converting: {e}")
+
+        # Clean up
+        finally:
+            try:
+                fh.close()
+            except:
+                pass
+            os.unlink(temp_path)
+
+        return result
+
+    def _convert(self, local_path, extensions, **kwargs):
+        error_trace = ""
+        for ext in extensions:
+            for converter in self._page_converters:
+                _kwargs = copy.deepcopy(kwargs)
+                _kwargs.update({"file_extension": ext})
+                # If we hit an error log it and keep trying
+                try:
+                    res = converter.convert(local_path, **_kwargs)
+                    if res is not None:
+                        # Normalize the content
+                        res.text_content = "\n".join([line.rstrip() for line in re.split(r"\r?\n", res.text_content)])
+                        res.text_content = re.sub(r"\n{3,}", "\n\n", res.text_content)
+
+                        # Todo
+                        return res
+                except Exception as e:
+                    error_trace = ("\n\n" + traceback.format_exc()).strip()
+
+
+        # If we got this far without success, report any exceptions
+        if len(error_trace) > 0:
+            raise FileConversionException(
+                f"Could not convert '{local_path}' to Markdown. File type was recognized as {extensions}. While converting the file, the following error was encountered:\n\n{error_trace}"
+            )
+
+        # Nothing can handle it!
+        # raise UnsupportedFormatException(
+        #     f"Could not convert '{local_path}' to Markdown. The formats {extensions} are not supported."
+        # )
+        res = PlainTextConverter().convert(local_path, **kwargs)
+        return res
+
+    def _append_ext(self, extensions, ext):
+        """Append a unique non-None, non-empty extension to a list of extensions."""
+        if ext is None:
+            return
+        ext = ext.strip()
+        if ext == "":
+            return
+        # if ext not in extensions:
+        if True:
+            extensions.append(ext)
+
+    def _guess_ext_magic(self, path):
+        """Use puremagic (a Python implementation of libmagic) to guess a file's extension based on the first few bytes."""
+        # Use puremagic to guess
+        try:
+            guesses = puremagic.magic_file(path)
+            if len(guesses) > 0:
+                ext = guesses[0].extension.strip()
+                if len(ext) > 0:
+                    return ext
+        except FileNotFoundError:
+            pass
+        except IsADirectoryError:
+            pass
+        except PermissionError:
+            pass
+        return None
+
+    def register_page_converter(self, converter: DocumentConverter) -> None:
+        """Register a page text converter."""
+        self._page_converters.append(converter)

app_tools/web_search_tools.py

@@ -0,0 +1,209 @@
+# Shamelessly stolen from Microsoft Autogen team and HuggingFace: thanks to them for this great resource!
+# https://github.com/microsoft/autogen/blob/gaia_multiagent_v01_march_1st/autogen/browser_utils.py
+# https://github.com/aymeric-roucher/GAIA/blob/a66aefc857d484a051a5eb66b49575dfaadff266/scripts/tools/web_surfer.py
+
+import os
+import re
+from typing import Tuple, Optional
+from smolagents import Tool
+import time
+from dotenv import load_dotenv
+import requests
+from pypdf import PdfReader
+from markdownify import markdownify as md
+import mimetypes
+from .browser import SimpleTextBrowser
+
+load_dotenv(override=True)
+
+user_agent = "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/119.0.0.0 Safari/537.36 Edg/119.0.0.0"
+
+browser_config = {
+    "viewport_size": 1024 * 5,
+    "downloads_folder": "coding",
+    "request_kwargs": {
+        "headers": {"User-Agent": user_agent},
+        "timeout": 300,
+    },
+}
+
+browser_config["serpapi_key"] = os.environ["SERPER_API_KEY"]
+
+browser = SimpleTextBrowser(**browser_config)
+
+# Helper functions
+def _browser_state() -> Tuple[str, str]:
+    header = f"Address: {browser.address}\n"
+    if browser.page_title is not None:
+        header += f"Title: {browser.page_title}\n"
+
+    current_page = browser.viewport_current_page
+    total_pages = len(browser.viewport_pages)
+
+    address = browser.address
+    for i in range(len(browser.history)-2,-1,-1): # Start from the second last
+        if browser.history[i][0] == address:
+            header += f"You previously visited this page {round(time.time() - browser.history[i][1])} seconds ago.\n"
+            break
+
+    header += f"Viewport position: Showing page {current_page+1} of {total_pages}.\n"
+    return (header, browser.viewport)
+
+
+class SearchInformationTool(Tool):
+    name="informational_web_search"
+    description="Perform an INFORMATIONAL web search query then return the search results."
+    inputs = {
+        "query": {
+            "type": "string",
+            "description": "The informational web search query to perform."
+        }
+    }
+    inputs["filter_year"]= {
+        "type": "string",
+        "description": "[Optional parameter]: filter the search results to only include pages from a specific year. For example, '2020' will only include pages from 2020. Make sure to use this parameter if you're trying to search for articles from a specific date!",
+        "nullable": True
+    }
+    output_type = "string"
+
+    def forward(self, query: str, filter_year: Optional[int] = None) -> str:
+        browser.visit_page(f"google: {query}", filter_year=filter_year)
+        header, content = _browser_state()
+        return header.strip() + "\n=======================\n" + content
+
+
+class NavigationalSearchTool(Tool):
+    name="navigational_web_search"
+    description="Perform a NAVIGATIONAL web search query then immediately navigate to the top result. Useful, for example, to navigate to a particular Wikipedia article or other known destination. Equivalent to Google's \"I'm Feeling Lucky\" button."
+    inputs = {"query": {"type": "string", "description": "The navigational web search query to perform."}}
+    output_type = "string"
+
+    def forward(self, query: str) -> str:
+        browser.visit_page(f"google: {query}")
+
+        # Extract the first line
+        m = re.search(r"\[.*?\]\((http.*?)\)", browser.page_content)
+        if m:
+            browser.visit_page(m.group(1))
+
+        # Return where we ended up
+        header, content = _browser_state()
+        return header.strip() + "\n=======================\n" + content
+
+
+class VisitTool(Tool):
+    name="visit_page"
+    description="Visit a webpage at a given URL and return its text."
+    inputs = {"url": {"type": "string", "description": "The relative or absolute url of the webapge to visit."}}
+    output_type = "string"
+
+    def forward(self, url: str) -> str:
+        browser.visit_page(url)
+        header, content = _browser_state()
+        return header.strip() + "\n=======================\n" + content
+
+
+class DownloadTool(Tool):
+    name="download_file"
+    description="""
+Download a file at a given URL. The file should be of this format: [".xlsx", ".pptx", ".wav", ".mp3", ".png", ".docx"]
+After using this tool, for further inspection of this page you should return the download path to your manager via final_answer, and they will be able to inspect it.
+DO NOT use this tool for .pdf or .txt or .htm files: for these types of files use visit_page with the file url instead."""
+    inputs = {"url": {"type": "string", "description": "The relative or absolute url of the file to be downloaded."}}
+    output_type = "string"
+
+    def forward(self, url: str) -> str:
+        if "arxiv" in url:
+            url = url.replace("abs", "pdf")
+        response = requests.get(url)
+        content_type = response.headers.get("content-type", "")
+        extension = mimetypes.guess_extension(content_type)
+        if extension and isinstance(extension, str):
+            new_path = f"./downloads/file{extension}"
+        else:
+            new_path = "./downloads/file.object"
+
+        with open(new_path, "wb") as f:
+            f.write(response.content)
+
+        if "pdf" in extension or "txt" in extension or "htm" in extension:
+            raise Exception("Do not use this tool for pdf or txt or html files: use visit_page instead.")
+
+        return f"File was downloaded and saved under path {new_path}."
+    
+
+class PageUpTool(Tool):
+    name="page_up"
+    description="Scroll the viewport UP one page-length in the current webpage and return the new viewport content."
+    output_type = "string"
+    inputs = {}
+
+    def forward(self) -> str:
+        browser.page_up()
+        header, content = _browser_state()
+        return header.strip() + "\n=======================\n" + content
+
+class ArchiveSearchTool(Tool):
+    name="find_archived_url"
+    description="Given a url, searches the Wayback Machine and returns the archived version of the url that's closest in time to the desired date."
+    inputs={
+        "url": {"type": "string", "description": "The url you need the archive for."},
+        "date": {"type": "string", "description": "The date that you want to find the archive for. Give this date in the format 'YYYYMMDD', for instance '27 June 2008' is written as '20080627'."}
+    }
+    output_type = "string"
+
+    def forward(self, url, date) -> str:
+        archive_url = f"https://archive.org/wayback/available?url={url}&timestamp={date}"
+        response = requests.get(archive_url).json()
+        try:
+            closest = response["archived_snapshots"]["closest"]
+        except:
+            raise Exception(f"Your url was not archived on Wayback Machine, try a different url.")
+        target_url = closest["url"]
+        browser.visit_page(target_url)
+        header, content = _browser_state()
+        return f"Web archive for url {url}, snapshot taken at date {closest['timestamp'][:8]}:\n" + header.strip() + "\n=======================\n" + content
+
+
+class PageDownTool(Tool):
+    name="page_down"
+    description="Scroll the viewport DOWN one page-length in the current webpage and return the new viewport content."
+    output_type = "string"
+    inputs = {}
+
+    def forward(self, ) -> str:
+        browser.page_down()
+        header, content = _browser_state()
+        return header.strip() + "\n=======================\n" + content
+
+
+class FinderTool(Tool):
+    name="find_on_page_ctrl_f"
+    description="Scroll the viewport to the first occurrence of the search string. This is equivalent to Ctrl+F."
+    inputs = {"search_string": {"type": "string", "description": "The string to search for on the page. This search string supports wildcards like '*'" }}
+    output_type = "string"
+
+    def forward(self, search_string: str) -> str:
+        find_result = browser.find_on_page(search_string)
+        header, content = _browser_state()
+
+        if find_result is None:
+            return header.strip() + f"\n=======================\nThe search string '{search_string}' was not found on this page."
+        else:
+            return header.strip() + "\n=======================\n" + content
+
+
+class FindNextTool(Tool):
+    name="find_next"
+    description="Scroll the viewport to next occurrence of the search string. This is equivalent to finding the next match in a Ctrl+F search."
+    inputs = {}
+    output_type = "string"
+
+    def forward(self, ) -> str:
+        find_result = browser.find_next()
+        header, content = _browser_state()
+
+        if find_result is None:
+            return header.strip() + "\n=======================\nThe search string was not found on this page."
+        else:
+            return header.strip() + "\n=======================\n" + content

requirements.txt

@@ -12,4 +12,12 @@ huggingface_hub
 markdownify
 transformers
 gradio_tools
-langchain
+langchain
+pypdf
+pathvalidate
+puremagic
+mammoth
+python-pptx
+pdfminer.six
+youtube-transcript-api
+google-search-results