pr final comment, checking run

Author: christianabbet

src/entity/utils.py

@@ -7,7 +7,7 @@
 
 
 def pages_to_bytes(pdf_document: Document, page_start: int, page_end: int) -> BytesIO:
-    """Select pages from PDF.
+    """Extract a range of pages from a PDF document and return them as a BytesIO buffer.
 
     Args:
         pdf_document (Document): PDF to split.
@@ -22,6 +22,4 @@ def pages_to_bytes(pdf_document: Document, page_start: int, page_end: int) -> By
         for page_number in range(page_start, page_end + 1):
             # Insert the page into the new PDF
             select_pdf.insert_pdf(pdf_document, from_page=page_number - 1, to_page=page_number - 1)
-
-        # Extract bytes and close document
         return BytesIO(select_pdf.tobytes())

src/scripts/pixtral_extract_feature.py

@@ -129,7 +129,8 @@ def extract_feature(input_directory: Path, prompt: Path, prompt_version: str, gr
         matched_files = list(filter(lambda x: x.name == gt.filename, paths))
         if matched_files:
             gt = update_ground_truth(gt, document=matched_files[0], pixtral_interface=pixtral_interface)
-        # Compute new features
+
+        # Accumulate updated ground truth
         gt_list_new.append(gt)
 
     # Write updated items

src/utils/utility.py

@@ -91,7 +91,7 @@ def get_pdf_files(input_path: Path) -> list[Path]:
 
 
 def standardize_text(text: str) -> str:
-    """Standardize text by removing new lines, double spaces and lowercasing.
+    """Standardize text by removing newlines, collapsing whitespace, stripping accents, and lowercasing.
 
     Args:
         text (str): Text to standardize.