Merged in bugfix/molina_nv_prov_info_issues (pull request #728)

Bugfix/molina nv prov info issues to main

* updated tin and npi pattern

* updated relevant pages search


Approved-by: Katon Minhas

fieldExtraction/constants/regex_patterns.py

@@ -4,10 +4,13 @@ BACKTICK_PATTERN = r"`([^`]+)`"
 TRIPLE_BACKTICK_PATTERN = r"```([^`]+)```"
 
 # TIN and NPI Regex Patterns
-TIN_PATTERN = (
-    r"\b\d{9}\b|\b\d{2}[-.]\d{7}\b|\b\d{3}[-.]\d{2}[-.]\d{4}\b|\b\d{2}-\d{4}-\d{3}\b"
-)
-NPI_PATTERN = r"\b\d{10}\b"
+# TIN_PATTERN = (
+#     r"\b\d{9}\b|\b\d{2}[-.]\d{7}\b|\b\d{3}[-.]\d{2}[-.]\d{4}\b|\b\d{2}-\d{4}-\d{3}\b"
+# )
+# NPI_PATTERN = r"\b\d{10}\b"
+TIN_PATTERN = r"\b(?:\d[\s\-.]*){9}\b"
+NPI_PATTERN = r"\b(?:\d\s*){10}\b"
+
 
 # OCR-Correctable Patterns
 # Handles common OCR misinterpretations where letters are mistaken for digits.

fieldExtraction/src/investment/tin_npi_funcs.py

@@ -682,10 +682,12 @@ def run_provider_info_fields(
             page_num
             for page_num, page_text in text_dict.items()
             if any(
-                tin.replace("-", "") in page_text.replace("-", "") for tin in all_tins
+                re.sub(r"[\s-]", "", tin) in re.sub(r"[\s-]", "", page_text)
+                for tin in all_tins
             )
             or any(
-                npi.replace("-", "") in page_text.replace("-", "") for npi in all_npis
+                re.sub(r"[\s-]", "", npi) in re.sub(r"[\s-]", "", page_text)
+                for npi in all_npis
             )
         ]