handle missing paragraphs in doc-extract

GooeyAI · Feb 13, 2024 · 6e22901 · 6e22901
1 parent 7f5adb9
commit 6e22901
Showing 1 changed file with 1 addition and 3 deletions.
diff --git a/daras_ai_v2/azure_doc_extract.py b/daras_ai_v2/azure_doc_extract.py
@@ -67,11 +67,9 @@ def azure_form_recognizer(url: str, model_id: str, params: dict = None):
 
 
 def extract_records(result: dict, page_num: int) -> list[dict]:
-    if not result:
-        return []
     table_polys = extract_tables(result, page_num)
     records = []
-    for para in result["paragraphs"]:
+    for para in result.get("paragraphs", []):
         try:
             if para["boundingRegions"][0]["pageNumber"] != page_num:
                 continue