From b0b1a6777149b252cc606fbd4529dd11f33c70d7 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jakub=20Paw=C5=82owski?=
 <44931252+jkpawlowski96@users.noreply.github.com>
Date: Tue, 30 Apr 2024 20:29:22 +0200
Subject: [PATCH] community[patch]: Skip unexpected 404 HTTP Error in Arxiv
 download (#21042)

### Description:
When attempting to download PDF files from arXiv, an unexpected 404
error frequently occurs. This error halts the operation, regardless of
whether there are additional documents to process. As a solution, I
suggest implementing a mechanism to ignore and communicate this error
and continue processing the next document from the list.

Proposed Solution: To address the issue of unexpected 404 errors during
PDF downloads from arXiv, I propose implementing the following solution:

- Error Handling: Implement error handling mechanisms to catch and
handle 404 errors gracefully.
- Communication: Inform the user or logging system about the occurrence
of the 404 error.
- Continued Processing: After encountering a 404 error, continue
processing the remaining documents from the list without interruption.

This solution ensures that the application can handle unexpected errors
without terminating the entire operation. It promotes resilience and
robustness in the face of intermittent issues encountered during PDF
downloads from arXiv.

### Issue:
#20909
### Dependencies:
none

---------

Co-authored-by: Bagatur <22008038+baskaryan@users.noreply.github.com>
---
 .../langchain_community/utilities/arxiv.py    |  8 ++++
 .../document_loaders/test_arxiv.py            | 39 ++++++++++++++++++-
 2 files changed, 46 insertions(+), 1 deletion(-)

diff --git a/libs/community/langchain_community/utilities/arxiv.py b/libs/community/langchain_community/utilities/arxiv.py
index 383df6396e..2a0027fd5a 100644
--- a/libs/community/langchain_community/utilities/arxiv.py
+++ b/libs/community/langchain_community/utilities/arxiv.py
@@ -27,6 +27,7 @@ class ArxivAPIWrapper(BaseModel):
     Attributes:
         top_k_results: number of the top-scored document used for the arxiv tool
         ARXIV_MAX_QUERY_LENGTH: the cut limit on the query used for the arxiv tool.
+        continue_on_failure (bool): If True, continue loading other URLs on failure.
         load_max_docs: a limit to the number of loaded documents
         load_all_available_meta:
             if True: the `metadata` of the loaded Documents contains all available
@@ -54,6 +55,7 @@ class ArxivAPIWrapper(BaseModel):
     arxiv_exceptions: Any  # :meta private:
     top_k_results: int = 3
     ARXIV_MAX_QUERY_LENGTH: int = 300
+    continue_on_failure: bool = False
     load_max_docs: int = 100
     load_all_available_meta: bool = False
     doc_content_chars_max: Optional[int] = 4000
@@ -225,6 +227,12 @@ class ArxivAPIWrapper(BaseModel):
             except (FileNotFoundError, fitz.fitz.FileDataError) as f_ex:
                 logger.debug(f_ex)
                 continue
+            except Exception as e:
+                if self.continue_on_failure:
+                    logger.error(e)
+                    continue
+                else:
+                    raise e
             if self.load_all_available_meta:
                 extra_metadata = {
                     "entry_id": result.entry_id,
diff --git a/libs/community/tests/integration_tests/document_loaders/test_arxiv.py b/libs/community/tests/integration_tests/document_loaders/test_arxiv.py
index 765290642a..ceeb48d750 100644
--- a/libs/community/tests/integration_tests/document_loaders/test_arxiv.py
+++ b/libs/community/tests/integration_tests/document_loaders/test_arxiv.py
@@ -1,10 +1,17 @@
-from typing import List
+import shutil
+from http.client import HTTPMessage
+from pathlib import Path
+from typing import List, Union
+from unittest.mock import patch
+from urllib.error import HTTPError
 
 import pytest
 from langchain_core.documents import Document
 
 from langchain_community.document_loaders.arxiv import ArxivLoader
 
+EXAMPLE_HELLO_PDF_PATH = Path(__file__).parents[1] / "examples" / "hello.pdf"
+
 
 def assert_docs(docs: List[Document]) -> None:
     for doc in docs:
@@ -57,6 +64,36 @@ def test_load_returns_full_set_of_metadata() -> None:
         assert len(set(doc.metadata)) > 4
 
 
+def test_skip_http_error() -> None:
+    """Test skipping unexpected Http 404 error of a single doc"""
+    tmp_hello_pdf_path = Path(__file__).parent / "hello.pdf"
+
+    def first_download_fails() -> Union[HTTPError, str]:
+        if not hasattr(first_download_fails, "firstCall"):
+            first_download_fails.__setattr__("firstCall", False)
+            raise HTTPError(
+                url="", code=404, msg="Not Found", hdrs=HTTPMessage(), fp=None
+            )
+        else:
+            # Return temporary example pdf path
+            shutil.copy(EXAMPLE_HELLO_PDF_PATH, tmp_hello_pdf_path)
+            return str(tmp_hello_pdf_path.absolute())
+
+    with patch("arxiv.Result.download_pdf") as mock_download_pdf:
+        # Set up the mock to raise HTTP 404 error
+        mock_download_pdf.side_effect = first_download_fails
+        # Load documents
+        loader = ArxivLoader(
+            query="ChatGPT",
+            load_max_docs=2,
+            load_all_available_meta=True,
+            continue_on_failure=True,
+        )
+        docs = loader.load()
+        # Only 1 of 2 documents should be loaded
+        assert len(docs) == 1
+
+
 @pytest.mark.skip(reason="test could be flaky")
 def test_load_issue_9046() -> None:
     """Test for the fixed issue 9046"""