Spaces:
Paused
Paused
Update app.py
Browse files
app.py
CHANGED
|
@@ -68,12 +68,12 @@ def get_azure_client() -> Optional[DocumentAnalysisClient]:
|
|
| 68 |
|
| 69 |
# --- Regex patterns for text-based PDF extraction ---
|
| 70 |
INVOICE_NO_RE = re.compile(
|
| 71 |
-
r"(?
|
| 72 |
re.IGNORECASE,
|
| 73 |
)
|
| 74 |
|
| 75 |
PREFIXED_INVOICE_RE = re.compile(
|
| 76 |
-
r"\b([A-Z]{2,4}[-/]\d{4,}(?
|
| 77 |
)
|
| 78 |
|
| 79 |
GST_LIKE_RE = re.compile(r"\b(GST[-\s]?\d+[A-Za-z0-9-]*)\b", re.IGNORECASE)
|
|
|
|
| 68 |
|
| 69 |
# --- Regex patterns for text-based PDF extraction ---
|
| 70 |
INVOICE_NO_RE = re.compile(
|
| 71 |
+
r"(?: Inv(?:Inoice)?\s*No\. ?|Invoice\s*No\. ?|Bill\s*No\.?|BILL\s*NO\.?|BILL\s*NO)\s*[:\-]?\s*([A-Za-z0-9\-/]+)",
|
| 72 |
re.IGNORECASE,
|
| 73 |
)
|
| 74 |
|
| 75 |
PREFIXED_INVOICE_RE = re.compile(
|
| 76 |
+
r"\b([A-Z]{2,4}[-/]\d{4,}(?:/\d+)?[A-Z]*)\b"
|
| 77 |
)
|
| 78 |
|
| 79 |
GST_LIKE_RE = re.compile(r"\b(GST[-\s]?\d+[A-Za-z0-9-]*)\b", re.IGNORECASE)
|