Merge pull request #27 from ic-xu/main

fix: Fix sequence item 2: expected str instance, NoneType found exception when table output is set to markdown.

src/openparse/tables/parse.py

@@ -51,6 +51,11 @@ def _ingest_with_pymupdf(
         tabs = page.find_tables()
         for i, tab in enumerate(tabs.tables):
             headers = tab.header.names
+            for j, header in enumerate(headers):
+                if header is None:
+                    headers[j] = ""
+                else:
+                    headers[j] = header.strip()
             lines = tab.extract()
 
             if parsing_args.table_output_format == "str":

src/openparse/tables/pymupdf/parse.py

@@ -26,7 +26,7 @@ def output_to_markdown(headers: List[str], rows: List[List[str]]) -> str:
     markdown_output += "|---" * len(headers) + "|\n"
 
     for row in rows:
-        processed_row = [" " if cell in [None, ""] else cell for cell in row]
+        processed_row = [" " if cell in [None, ""] else cell.replace("\n", " ") for cell in row]
         markdown_output += "| " + " | ".join(processed_row) + " |\n"
 
     return markdown_output