Add ability to pass extra params for Arxiv api

crewai_tools/tools/arxiv_paper_tool/Examples.md

@@ -41,6 +41,10 @@ tool = ArxivPaperTool(
     download_pdfs=True,
     save_dir=save_dir,
     use_title_as_filename=True
+    extra_params = {
+        "sortBy": "relevance",
+        "sortOrder": "descending"
+    }
 )
 tool.result_as_answer = True #Required,otherwise
 

crewai_tools/tools/arxiv_paper_tool/README.md

@@ -29,6 +29,7 @@ This tool:
 | `download_pdfs`         | `bool` | ❌        | Whether to download the corresponding PDFs. Defaults to `False`.                  |
 | `save_dir`              | `str`  | ❌        | Directory to save PDFs (created if it doesn’t exist). Defaults to `./arxiv_pdfs`. |
 | `use_title_as_filename` | `bool` | ❌        | Use the paper title as the filename (sanitized). Defaults to `False`.             |
+| `extra_params`          | `dict[str, str]` | ❌        | Extend or override the query parameters used for a search, see options at [Arxiv's API documentation](https://info.arxiv.org/help/api/user-manual.html#311-query-interface) |
 
 ---
 
@@ -100,9 +101,24 @@ result = tool._run(
 print(result)
 ```
 
+### Example 5: Order results by the most recently submitted
+
+```python
+tool = ArxivPaperTool(
+    extra_params={
+        "sortBy": "submittedDate",
+        "sortOrder": "descending"
+    }
+)
+result = tool._run(
+    search_query="explainable ai",
+)
+print(result)
+```
+
 ---
 
-### Example 5: All Options Combined
+### Example 6: All Options Combined
 
 ```python
 tool = ArxivPaperTool(

crewai_tools/tools/arxiv_paper_tool/arxiv_paper_tool.py

@@ -29,11 +29,18 @@ class ArxivPaperTool(BaseTool):
     package_dependencies: List[str] = ["pydantic"]
     env_vars: List[EnvVar] = []
 
-    def __init__(self, download_pdfs=False, save_dir="./arxiv_pdfs", use_title_as_filename=False):
+    def __init__(
+        self,
+        download_pdfs=False,
+        save_dir="./arxiv_pdfs",
+        use_title_as_filename=False,
+        extra_params=None,
+    ):
         super().__init__()
         self.download_pdfs = download_pdfs
         self.save_dir = save_dir
         self.use_title_as_filename = use_title_as_filename
+        self.extra_params = extra_params
 
     def _run(self, search_query: str, max_results: int = 5) -> str:
         try:
@@ -68,7 +75,16 @@ def _run(self, search_query: str, max_results: int = 5) -> str:
 
 
     def fetch_arxiv_data(self, search_query: str, max_results: int) -> List[dict]:
-        api_url = f"{self.BASE_API_URL}?search_query={urllib.parse.quote(search_query)}&start=0&max_results={max_results}"
+        params = {
+            'search_query': search_query,
+            'start': 0,
+            'max_results': max_results,
+        }
+        if self.extra_params:
+            params = {**params, **self.extra_params}
+
+        query = urllib.parse.urlencode(params)
+        api_url = f"{self.BASE_API_URL}?{query}"
         logger.info(f"Fetching data from Arxiv API: {api_url}")
 
         try:

crewai_tools/tools/arxiv_paper_tool/arxiv_paper_tool_test.py

@@ -111,3 +111,35 @@ def test_run_with_max_results(mock_fetch, tool):
 
     result = tool._run(search_query="test", max_results=100)
     assert result.count("Title:") == 100
+
+
+@patch("urllib.request.urlopen")
+def test_fetch_arxiv_data_with_extra_params(mock_urlopen):
+    mock_response = MagicMock()
+    mock_response.status = 200
+    mock_response.read.return_value = mock_arxiv_response().encode("utf-8")
+    mock_urlopen.return_value.__enter__.return_value = mock_response
+
+    tool = ArxivPaperTool(
+        extra_params = {
+            "sortBy": "lastUpdatedDate",
+            "sortOrder": "descending",
+            "start": 10,
+            }
+    )
+    tool.fetch_arxiv_data("transformer", 1)
+
+    expected_url = "".join(
+        [
+            "http://export.arxiv.org/api/query",
+            "?search_query=transformer",
+            "&start=10",
+            "&max_results=1",
+            "&sortBy=lastUpdatedDate",
+            "&sortOrder=descending",
+        ]
+    )
+    mock_urlopen.assert_called_once_with(
+        expected_url,
+        timeout=10,
+    )