Update patentwiz/preprocess_data.py

patentwiz/preprocess_data.py

@@ -86,7 +86,7 @@ def filter_rf_patents(patents, keywords=None, fields=None):
     """
     Filters patents based on keywords and specified fields.
     Parameters:
-        patents (list): List of patents.
+        patents (list): List of patent texts (as strings or structured data).
         keywords (list): Keywords to filter patents.
         fields (list): Fields to search for keywords (e.g., Title, Abstract, Claims).
     Returns:
@@ -99,14 +99,21 @@ def filter_rf_patents(patents, keywords=None, fields=None):
 
     filtered_patents = []
     for patent in patents:
-        for field in fields:
-            field_content = patent.get(field.lower(), "")
-            if any(keyword.lower() in field_content.lower() for keyword in keywords):
+        # If patent is a string, search for keywords in the entire text
+        if isinstance(patent, str):
+            if any(keyword.lower() in patent.lower() for keyword in keywords):
                 filtered_patents.append(patent)
-                break
+        # If patent is structured (e.g., dictionary), search within fields
+        elif isinstance(patent, dict):
+            for field in fields:
+                field_content = patent.get(field.lower(), "")
+                if any(keyword.lower() in field_content.lower() for keyword in keywords):
+                    filtered_patents.append(patent)
+                    break
     return filtered_patents
 
 
+
 def extract_patents(year, month, day, logging):
     """
     This function reads a patent file in XML format, splits it into individual patents, parses each