Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
|
@@ -733,7 +733,8 @@ def strip_degenerate_html_tables(md: str) -> str:
|
|
| 733 |
return re.sub(r"\n{3,}", "\n\n", out)
|
| 734 |
|
| 735 |
|
| 736 |
-
|
|
|
|
| 737 |
|
| 738 |
|
| 739 |
def mask_non_monetary_long_numbers(md: str) -> str:
|
|
|
|
| 733 |
return re.sub(r"\n{3,}", "\n\n", out)
|
| 734 |
|
| 735 |
|
| 736 |
+
# Match long digit runs even when OCR glues trailing letters (e.g. 9257403599CCD).
|
| 737 |
+
_LONG_NUM_TOKEN = re.compile(r"(?<!\d)\d{7,}(?!\d)")
|
| 738 |
|
| 739 |
|
| 740 |
def mask_non_monetary_long_numbers(md: str) -> str:
|