Fix deprecated typing (huggingface#1249)

Author: albertvillanova

examples/agent_from_any_llm.py

@@ -1,5 +1,3 @@
-from typing import Optional
-
 from smolagents import InferenceClientModel, LiteLLMModel, OpenAIServerModel, TransformersModel, tool
 from smolagents.agents import CodeAgent, ToolCallingAgent
 
@@ -38,7 +36,7 @@
 
 
 @tool
-def get_weather(location: str, celsius: Optional[bool] = False) -> str:
+def get_weather(location: str, celsius: bool | None = False) -> str:
     """
     Get weather in the next days at given location.
     Secretly this tool does not care about the location, it hates the weather everywhere.

examples/multiple_tools.py

@@ -1,5 +1,3 @@
-from typing import Optional
-
 import requests
 
 # from smolagents.agents import ToolCallingAgent
@@ -15,7 +13,7 @@
 
 
 @tool
-def get_weather(location: str, celsius: Optional[bool] = False) -> str:
+def get_weather(location: str, celsius: bool | None = False) -> str:
     """
     Get the current weather at the given location using the WeatherStack API.
 

examples/open_deep_research/run_gaia.py

@@ -6,7 +6,6 @@
 from concurrent.futures import ThreadPoolExecutor, as_completed
 from datetime import datetime
 from pathlib import Path
-from typing import List
 
 import datasets
 import pandas as pd
@@ -242,7 +241,7 @@ def answer_single_question(example, model_id, answers_file, visual_inspection_to
     append_answer(annotated_example, answers_file)
 
 
-def get_examples_to_answer(answers_file, eval_ds) -> List[dict]:
+def get_examples_to_answer(answers_file, eval_ds) -> list[dict]:
     print(f"Loading answers from {answers_file}...")
     try:
         done_questions = pd.read_json(answers_file, lines=True)["question"].tolist()

examples/open_deep_research/scripts/mdconvert.py

@@ -14,7 +14,7 @@
 import tempfile
 import traceback
 import zipfile
-from typing import Any, Dict, List, Optional, Union
+from typing import Any
 from urllib.parse import parse_qs, quote, unquote, urlparse, urlunparse
 
 import mammoth
@@ -112,22 +112,22 @@ def convert_soup(self, soup: Any) -> str:
 class DocumentConverterResult:
     """The result of converting a document to text."""
 
-    def __init__(self, title: Union[str, None] = None, text_content: str = ""):
-        self.title: Union[str, None] = title
+    def __init__(self, title: str | None = None, text_content: str = ""):
+        self.title: str | None = title
         self.text_content: str = text_content
 
 
 class DocumentConverter:
     """Abstract superclass of all DocumentConverters."""
 
-    def convert(self, local_path: str, **kwargs: Any) -> Union[None, DocumentConverterResult]:
+    def convert(self, local_path: str, **kwargs: Any) -> None | DocumentConverterResult:
         raise NotImplementedError()
 
 
 class PlainTextConverter(DocumentConverter):
     """Anything with content type text/plain"""
 
-    def convert(self, local_path: str, **kwargs: Any) -> Union[None, DocumentConverterResult]:
+    def convert(self, local_path: str, **kwargs: Any) -> None | DocumentConverterResult:
         # Guess the content type from any file extension that might be around
         content_type, _ = mimetypes.guess_type("__placeholder" + kwargs.get("file_extension", ""))
 
@@ -149,7 +149,7 @@ def convert(self, local_path: str, **kwargs: Any) -> Union[None, DocumentConvert
 class HtmlConverter(DocumentConverter):
     """Anything with content type text/html"""
 
-    def convert(self, local_path: str, **kwargs: Any) -> Union[None, DocumentConverterResult]:
+    def convert(self, local_path: str, **kwargs: Any) -> None | DocumentConverterResult:
         # Bail if not html
         extension = kwargs.get("file_extension", "")
         if extension.lower() not in [".html", ".htm"]:
@@ -161,7 +161,7 @@ def convert(self, local_path: str, **kwargs: Any) -> Union[None, DocumentConvert
 
         return result
 
-    def _convert(self, html_content: str) -> Union[None, DocumentConverterResult]:
+    def _convert(self, html_content: str) -> None | DocumentConverterResult:
         """Helper function that converts and HTML string."""
 
         # Parse the string
@@ -189,7 +189,7 @@ def _convert(self, html_content: str) -> Union[None, DocumentConverterResult]:
 class WikipediaConverter(DocumentConverter):
     """Handle Wikipedia pages separately, focusing only on the main document content."""
 
-    def convert(self, local_path: str, **kwargs: Any) -> Union[None, DocumentConverterResult]:
+    def convert(self, local_path: str, **kwargs: Any) -> None | DocumentConverterResult:
         # Bail if not Wikipedia
         extension = kwargs.get("file_extension", "")
         if extension.lower() not in [".html", ".htm"]:
@@ -234,7 +234,7 @@ def convert(self, local_path: str, **kwargs: Any) -> Union[None, DocumentConvert
 class YouTubeConverter(DocumentConverter):
     """Handle YouTube specially, focusing on the video title, description, and transcript."""
 
-    def convert(self, local_path: str, **kwargs: Any) -> Union[None, DocumentConverterResult]:
+    def convert(self, local_path: str, **kwargs: Any) -> None | DocumentConverterResult:
         # Bail if not YouTube
         extension = kwargs.get("file_extension", "")
         if extension.lower() not in [".html", ".htm"]:
@@ -250,7 +250,7 @@ def convert(self, local_path: str, **kwargs: Any) -> Union[None, DocumentConvert
 
         # Read the meta tags
         assert soup.title is not None and soup.title.string is not None
-        metadata: Dict[str, str] = {"title": soup.title.string}
+        metadata: dict[str, str] = {"title": soup.title.string}
         for meta in soup(["meta"]):
             for a in meta.attrs:
                 if a in ["itemprop", "property", "name"]:
@@ -328,13 +328,13 @@ def convert(self, local_path: str, **kwargs: Any) -> Union[None, DocumentConvert
             text_content=webpage_text,
         )
 
-    def _get(self, metadata: Dict[str, str], keys: List[str], default: Union[str, None] = None) -> Union[str, None]:
+    def _get(self, metadata: dict[str, str], keys: list[str], default: str | None = None) -> str | None:
         for k in keys:
             if k in metadata:
                 return metadata[k]
         return default
 
-    def _findKey(self, json: Any, key: str) -> Union[str, None]:  # TODO: Fix json type
+    def _findKey(self, json: Any, key: str) -> str | None:  # TODO: Fix json type
         if isinstance(json, list):
             for elm in json:
                 ret = self._findKey(elm, key)
@@ -356,7 +356,7 @@ class PdfConverter(DocumentConverter):
     Converts PDFs to Markdown. Most style information is ignored, so the results are essentially plain-text.
     """
 
-    def convert(self, local_path, **kwargs) -> Union[None, DocumentConverterResult]:
+    def convert(self, local_path, **kwargs) -> None | DocumentConverterResult:
         # Bail if not a PDF
         extension = kwargs.get("file_extension", "")
         if extension.lower() != ".pdf":
@@ -373,7 +373,7 @@ class DocxConverter(HtmlConverter):
     Converts DOCX files to Markdown. Style information (e.g.m headings) and tables are preserved where possible.
     """
 
-    def convert(self, local_path, **kwargs) -> Union[None, DocumentConverterResult]:
+    def convert(self, local_path, **kwargs) -> None | DocumentConverterResult:
         # Bail if not a DOCX
         extension = kwargs.get("file_extension", "")
         if extension.lower() != ".docx":
@@ -393,7 +393,7 @@ class XlsxConverter(HtmlConverter):
     Converts XLSX files to Markdown, with each sheet presented as a separate Markdown table.
     """
 
-    def convert(self, local_path, **kwargs) -> Union[None, DocumentConverterResult]:
+    def convert(self, local_path, **kwargs) -> None | DocumentConverterResult:
         # Bail if not a XLSX
         extension = kwargs.get("file_extension", "")
         if extension.lower() not in [".xlsx", ".xls"]:
@@ -417,7 +417,7 @@ class PptxConverter(HtmlConverter):
     Converts PPTX files to Markdown. Supports heading, tables and images with alt text.
     """
 
-    def convert(self, local_path, **kwargs) -> Union[None, DocumentConverterResult]:
+    def convert(self, local_path, **kwargs) -> None | DocumentConverterResult:
         # Bail if not a PPTX
         extension = kwargs.get("file_extension", "")
         if extension.lower() != ".pptx":
@@ -520,7 +520,7 @@ class WavConverter(MediaConverter):
     Converts WAV files to markdown via extraction of metadata (if `exiftool` is installed), and speech transcription (if `speech_recognition` is installed).
     """
 
-    def convert(self, local_path, **kwargs) -> Union[None, DocumentConverterResult]:
+    def convert(self, local_path, **kwargs) -> None | DocumentConverterResult:
         # Bail if not a XLSX
         extension = kwargs.get("file_extension", "")
         if extension.lower() != ".wav":
@@ -570,7 +570,7 @@ class Mp3Converter(WavConverter):
     Converts MP3 and M4A files to markdown via extraction of metadata (if `exiftool` is installed), and speech transcription (if `speech_recognition` AND `pydub` are installed).
     """
 
-    def convert(self, local_path, **kwargs) -> Union[None, DocumentConverterResult]:
+    def convert(self, local_path, **kwargs) -> None | DocumentConverterResult:
         # Bail if not a MP3
         extension = kwargs.get("file_extension", "")
         if extension.lower() not in [".mp3", ".m4a"]:
@@ -644,7 +644,7 @@ def __init__(self, extract_dir: str = "downloads"):
         # Create the extraction directory if it doesn't exist
         os.makedirs(self.extract_dir, exist_ok=True)
 
-    def convert(self, local_path: str, **kwargs: Any) -> Union[None, DocumentConverterResult]:
+    def convert(self, local_path: str, **kwargs: Any) -> None | DocumentConverterResult:
         # Bail if not a ZIP file
         extension = kwargs.get("file_extension", "")
         if extension.lower() != ".zip":
@@ -681,7 +681,7 @@ class ImageConverter(MediaConverter):
     Converts images to markdown via extraction of metadata (if `exiftool` is installed), OCR (if `easyocr` is installed), and description via a multimodal LLM (if an mlm_client is configured).
     """
 
-    def convert(self, local_path, **kwargs) -> Union[None, DocumentConverterResult]:
+    def convert(self, local_path, **kwargs) -> None | DocumentConverterResult:
         # Bail if not a XLSX
         extension = kwargs.get("file_extension", "")
         if extension.lower() not in [".jpg", ".jpeg", ".png"]:
@@ -771,9 +771,9 @@ class MarkdownConverter:
 
     def __init__(
         self,
-        requests_session: Optional[requests.Session] = None,
-        mlm_client: Optional[Any] = None,
-        mlm_model: Optional[Any] = None,
+        requests_session: requests.Session | None = None,
+        mlm_client: Any | None = None,
+        mlm_model: Any | None = None,
     ):
         if requests_session is None:
             self._requests_session = requests.Session()
@@ -783,7 +783,7 @@ def __init__(
         self._mlm_client = mlm_client
         self._mlm_model = mlm_model
 
-        self._page_converters: List[DocumentConverter] = []
+        self._page_converters: list[DocumentConverter] = []
 
         # Register converters for successful browsing operations
         # Later registrations are tried first / take higher priority than earlier registrations
@@ -802,7 +802,7 @@ def __init__(
         self.register_page_converter(PdfConverter())
 
     def convert(
-        self, source: Union[str, requests.Response], **kwargs: Any
+        self, source: str | requests.Response, **kwargs: Any
     ) -> DocumentConverterResult:  # TODO: deal with kwargs
         """
         Args:
@@ -924,7 +924,7 @@ def convert_response(
 
         return result
 
-    def _convert(self, local_path: str, extensions: List[Union[str, None]], **kwargs) -> DocumentConverterResult:
+    def _convert(self, local_path: str, extensions: list[str | None], **kwargs) -> DocumentConverterResult:
         error_trace = ""
         for ext in extensions + [None]:  # Try last with no extension
             for converter in self._page_converters:

examples/open_deep_research/scripts/text_inspector_tool.py

@@ -1,5 +1,3 @@
-from typing import Optional
-
 from smolagents import Tool
 from smolagents.models import Model
 
@@ -75,7 +73,7 @@ def forward_initial_exam_mode(self, file_path, question):
         ]
         return self.model(messages).content
 
-    def forward(self, file_path, question: Optional[str] = None) -> str:
+    def forward(self, file_path, question: str | None = None) -> str:
         from smolagents.models import MessageRole
 
         result = self.md_converter.convert(file_path)

examples/open_deep_research/scripts/text_web_browser.py

@@ -6,7 +6,7 @@
 import re
 import time
 import uuid
-from typing import Any, Dict, List, Optional, Tuple, Union
+from typing import Any
 from urllib.parse import unquote, urljoin, urlparse
 
 import pathvalidate
@@ -24,35 +24,35 @@ class SimpleTextBrowser:
 
     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,
+        start_page: str | None = None,
+        viewport_size: int | None = 1024 * 8,
+        downloads_folder: str | None | None = None,
+        serpapi_key: str | None | None = None,
+        request_kwargs: dict[str, Any] | None | 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.history: list[tuple[str, float]] = list()
+        self.page_title: str | None = None
         self.viewport_current_page = 0
-        self.viewport_pages: List[Tuple[int, int]] = list()
+        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
+        self._find_on_page_query: str | None = None
+        self._find_on_page_last_result: 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:
+    def set_address(self, uri_or_path: str, filter_year: int | None = None) -> None:
         # TODO: Handle anchors
         self.history.append((uri_or_path, time.time()))
 
@@ -102,7 +102,7 @@ def page_down(self) -> None:
     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]:
+    def find_on_page(self, query: str) -> 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?
@@ -121,7 +121,7 @@ def find_on_page(self, query: str) -> Union[str, None]:
             self._find_on_page_last_result = viewport_match
             return self.viewport
 
-    def find_next(self) -> Union[str, None]:
+    def find_next(self) -> str | None:
         """Scroll to the next viewport that matches the query"""
 
         if self._find_on_page_query is None:
@@ -144,7 +144,7 @@ def find_next(self) -> Union[str, None]:
             self._find_on_page_last_result = viewport_match
             return self.viewport
 
-    def _find_next_viewport(self, query: str, starting_viewport: int) -> Union[int, None]:
+    def _find_next_viewport(self, query: str, starting_viewport: int) -> int | None:
         """Search for matches between the starting viewport looping when reaching the end."""
 
         if query is None:
@@ -174,7 +174,7 @@ def _find_next_viewport(self, query: str, starting_viewport: int) -> Union[int,
 
         return None
 
-    def visit_page(self, path_or_uri: str, filter_year: Optional[int] = None) -> str:
+    def visit_page(self, path_or_uri: str, filter_year: int | None = 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
@@ -201,7 +201,7 @@ def _split_pages(self) -> None:
             self.viewport_pages.append((start_idx, end_idx))
             start_idx = end_idx
 
-    def _serpapi_search(self, query: str, filter_year: Optional[int] = None) -> None:
+    def _serpapi_search(self, query: str, filter_year: int | None = None) -> None:
         if self.serpapi_key is None:
             raise ValueError("Missing SerpAPI key.")
 
@@ -231,7 +231,7 @@ def _prev_visit(url):
                     return f"You previously visited this page {round(time.time() - self.history[i][1])} seconds ago.\n"
             return ""
 
-        web_snippets: List[str] = list()
+        web_snippets: list[str] = list()
         idx = 0
         if "organic_results" in results:
             for page in results["organic_results"]:
@@ -352,7 +352,7 @@ def _fetch_page(self, url: str) -> None:
                 self.page_title = "Error"
                 self._set_page_content(f"## Error\n\n{str(request_exception)}")
 
-    def _state(self) -> Tuple[str, str]:
+    def _state(self) -> tuple[str, str]:
         header = f"Address: {self.address}\n"
         if self.page_title is not None:
             header += f"Title: {self.page_title}\n"
@@ -385,7 +385,7 @@ def __init__(self, browser):
         super().__init__()
         self.browser = browser
 
-    def forward(self, query: str, filter_year: Optional[int] = None) -> str:
+    def forward(self, query: str, filter_year: int | None = None) -> str:
         self.browser.visit_page(f"google: {query}", filter_year=filter_year)
         header, content = self.browser._state()
         return header.strip() + "\n=======================\n" + content