Update utilities/extract_text.py

utilities/extract_text.py

@@ -26,31 +26,36 @@
 import os
 import requests
 
-EXTRACT_TEXT_API = "https://point9-extract-text-and-table.hf.space/api/text"  # Replace with your space URL
+# Hardcoded API URL - DO NOT CHANGE
+EXTRACT_TEXT_API = "https://point9-extract-text-and-table.hf.space/api/text"
 
 def extract_text_remote(state):
-    filename = state["filename"]
-    path = state["temp_files"][filename]
+    """
+    Extract text from documents via API.
+    All file types are sent to the API (no local processing).
     
-    # Get the file extension
-    _, file_extension = os.path.splitext(filename)
-    
-    # If the filename in state doesn't have extension, check if it's in the temp_files path
-    if not file_extension:
-        # Try to get extension from the actual file path
-        _, file_extension = os.path.splitext(path)
-        if file_extension:
-            # Add extension to filename
-            filename = f"{filename}{file_extension}"
+    Args:
+        state: Dict with 'filename', 'temp_files', 'start_page', 'end_page'
     
+    Returns:
+        state: Dict with 'text' key containing extracted text
+    """
+    filename = state["filename"]
+    path = state["temp_files"][filename]
+
+    if not os.path.exists(path):
+        raise RuntimeError(f"File not found: {path}")
+
     with open(path, "rb") as f:
         files = {"file": (filename, f, "application/pdf")}
         data = {
-            "filename": filename,  # Now includes extension
+            "filename": filename,
             "start_page": state.get("start_page", 1),
             "end_page": state.get("end_page", 1)
         }
         headers = {"Authorization": f"Bearer {os.getenv('HUGGINGFACE_API_TOKEN')}"}
+        
+        # Call API and wait for response
         resp = requests.post(EXTRACT_TEXT_API, files=files, data=data, headers=headers)
 
     if resp.status_code != 200: