fix: add input validation, retry limit, code formatting, and exclude i18n from secret detection

.pre-commit-config.yaml

@@ -35,7 +35,8 @@ repos:
             .*pnpm-lock\.yaml|
             .*alembic\.ini|
             .*alembic/versions/.*\.py|
-            .*\.mdx$
+            .*\.mdx$|
+            .*/messages/.*\.json$
           )$
 
   # Python Backend Hooks (surfsense_backend) - Using Ruff for linting and formatting

surfsense_backend/app/connectors/bookstack_connector.py

@@ -42,9 +42,7 @@ class BookStackConnector:
         self.token_secret = token_secret
         self._last_request_time = 0.0
 
-    def set_credentials(
+    def set_credentials(self, base_url: str, token_id: str, token_secret: str) -> None:
-        self, base_url: str, token_id: str, token_secret: str
-    ) -> None:
         """
         Set the BookStack credentials.
 
@@ -52,7 +50,17 @@ class BookStackConnector:
             base_url: BookStack instance base URL
             token_id: BookStack API Token ID
             token_secret: BookStack API Token Secret
+
+        Raises:
+            ValueError: If any required credential is missing or invalid
         """
+        if not base_url or not isinstance(base_url, str):
+            raise ValueError("base_url must be a non-empty string")
+        if not token_id or not isinstance(token_id, str):
+            raise ValueError("token_id must be a non-empty string")
+        if not token_secret or not isinstance(token_secret, str):
+            raise ValueError("token_secret must be a non-empty string")
+
         self.base_url = base_url.rstrip("/")
         self.token_id = token_id
         self.token_secret = token_secret
@@ -86,11 +94,15 @@ class BookStackConnector:
             time.sleep(self.REQUEST_INTERVAL - elapsed)
         self._last_request_time = time.time()
 
+    # Maximum retries for rate limit errors
+    MAX_RATE_LIMIT_RETRIES = 3
+
     def make_api_request(
         self,
         endpoint: str,
         params: dict[str, Any] | None = None,
         raw_response: bool = False,
+        _retry_count: int = 0,
     ) -> dict[str, Any] | str:
         """
         Make a request to the BookStack API.
@@ -128,9 +140,17 @@ class BookStackConnector:
 
         except requests.exceptions.HTTPError as e:
             if e.response.status_code == 429:
-                logger.warning("Rate limit exceeded, waiting 60 seconds...")
+                if _retry_count >= self.MAX_RATE_LIMIT_RETRIES:
+                    raise Exception(
+                        f"BookStack API rate limit exceeded after {self.MAX_RATE_LIMIT_RETRIES} retries"
+                    ) from e
+                logger.warning(
+                    f"Rate limit exceeded, waiting 60 seconds... (retry {_retry_count + 1}/{self.MAX_RATE_LIMIT_RETRIES})"
+                )
                 time.sleep(60)
-                return self.make_api_request(endpoint, params, raw_response)
+                return self.make_api_request(
+                    endpoint, params, raw_response, _retry_count + 1
+                )
             raise Exception(f"BookStack API request failed: {e!s}") from e
         except requests.exceptions.RequestException as e:
             raise Exception(f"BookStack API request failed: {e!s}") from e

surfsense_backend/app/tasks/connector_indexers/bookstack_indexer.py

@@ -87,7 +87,11 @@ async def index_bookstack_pages(
         bookstack_token_id = connector.config.get("BOOKSTACK_TOKEN_ID")
         bookstack_token_secret = connector.config.get("BOOKSTACK_TOKEN_SECRET")
 
-        if not bookstack_base_url or not bookstack_token_id or not bookstack_token_secret:
+        if (
+            not bookstack_base_url
+            or not bookstack_token_id
+            or not bookstack_token_secret
+        ):
             await task_logger.log_task_failure(
                 log_entry,
                 f"BookStack credentials not found in connector config for connector {connector_id}",
@@ -196,9 +200,7 @@ async def index_bookstack_pages(
                         page_id, use_markdown=True
                     )
                 except Exception as e:
-                    logger.warning(
+                    logger.warning(f"Failed to fetch content for page {page_name}: {e}")
-                        f"Failed to fetch content for page {page_name}: {e}"
-                    )
                     skipped_pages.append(f"{page_name} (content fetch error)")
                     documents_skipped += 1
                     continue
@@ -275,7 +277,9 @@ async def index_bookstack_pages(
                                 full_content, user_llm, summary_metadata
                             )
                         else:
-                            summary_content = f"BookStack Page: {page_name}\n\nBook ID: {book_id}\n\n"
+                            summary_content = (
+                                f"BookStack Page: {page_name}\n\nBook ID: {book_id}\n\n"
+                            )
                             if page_content:
                                 content_preview = page_content[:1000]
                                 if len(page_content) > 1000:
@@ -299,9 +303,7 @@ async def index_bookstack_pages(
                         existing_document.chunks = chunks
 
                         documents_indexed += 1
-                        logger.info(
+                        logger.info(f"Successfully updated BookStack page {page_name}")
-                            f"Successfully updated BookStack page {page_name}"
-                        )
                         continue
 
                 # Document doesn't exist - create new one
@@ -388,9 +390,7 @@ async def index_bookstack_pages(
             f"Final commit: Total {documents_indexed} BookStack pages processed"
         )
         await session.commit()
-        logger.info(
+        logger.info("Successfully committed all BookStack document changes to database")
-            "Successfully committed all BookStack document changes to database"
-        )
 
         # Log success
         await task_logger.log_task_success(