💄 improve formatting

Signed-off-by: peter szemraj <[email protected]>

pdf2text.py

@@ -556,7 +556,7 @@ def postprocess(text: str) -> str:
     return eval_and_replace(proc)
 
 
-def result2text(result) -> str:
+def result2text(result, as_text=False) -> str or list:
     """Convert OCR result to text"""
 
     full_doc = []
@@ -570,8 +570,7 @@ def result2text(result) -> str:
                     text += word.value + " "
         full_doc.append(text)
 
-    full_text = "\n".join(full_doc)
-    return full_text
+    return "\n".join(full_doc) if as_text else full_doc
 
 
 import warnings
@@ -603,8 +602,10 @@ def convert_PDF_to_Text(
     logging.info(f"running OCR on {len(doc)} pages")
     result = ocr_model(doc)
     raw_text = result2text(result)
-    proc_text = format_ocr_out(raw_text)
-    output_text = postprocess(proc_text)
+    proc_text = [format_ocr_out(r) for r in raw_text]
+    fin_text = [postprocess(t) for t in proc_text]
+
+    ocr_results = "\n\n".join(fin_text)
 
     fn_rt = time.perf_counter() - st
 
@@ -614,8 +615,8 @@ def convert_PDF_to_Text(
         "num_pages": len(doc),
         "runtime": round(fn_rt, 2),
         "date": str(date.today()),
-        "converted_text": output_text,
-        "length": len(output_text),
+        "converted_text": ocr_results,
+        "length": len(ocr_results),
     }
 
     return results_dict