jacob-mathison/doczyai-pipelines
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Activity

Merged in optimization/tin_npi (pull request #790)

Browse Source 
Optimization/tin npi

* updated signature page and added for tin  npi

* updated signature page

* removed print statements

* removed print statement

* updated test for extract signature page

* updated data type


Approved-by: Katon Minhas
This commit is contained in:
VenkataKrishna Reddy Avula
2025-11-26 17:21:44 +00:00
committed by Katon Minhas
parent 4334800fa3
commit 2482adb465
3 changed files with 44 additions and 32 deletions
Download Patch File Download Diff File Expand all files Collapse all files
fieldExtraction/src/utils/string_utils.py
+18 -14
View File
@@ -379,6 +379,12 @@ def extract_signature_page(text_dict: dict, filename: str) -> dict:
re.IGNORECASE,
)
keywords = [
"signature and information",
"signature authorization",
"in witness whereof",
]
# First pass: Try to match the full textract pattern
for page_num, page_text in text_dict.items():
if textract_pattern.search(page_text):
@@ -386,22 +392,20 @@ def extract_signature_page(text_dict: dict, filename: str) -> dict:
# If no pages found with textract pattern, try with keyword
if not signature_pages:
page_items = list(text_dict.items())
for i, (page_num, page_text) in enumerate(page_items):
# Iterate through each page in the text dictionary and check for keywords
for page_num, page_text in text_dict.items():
lower_text = page_text.lower()
if (
"signature page follow" in lower_text
or "signature authorization" in lower_text
):
# curr_page = int(float(page_num))
if page_num not in signature_pages:
signature_pages[page_num] = page_text # current page
if i + 1 < len(page_items):
next_page_num, next_page_text = page_items[i + 1]
# next_page = int(float(next_page_num))
if next_page_num not in signature_pages:
signature_pages[next_page_num] = next_page_text # next page
if any(keyword in lower_text for keyword in keywords):
signature_pages[page_num] = page_text
# If no signature pages found, try checking for the word "signature page follows" and return the next page only if it exists
if not signature_pages:
for page_num, page_text in text_dict.items():
if "signature page follows" in page_text.lower():
next_page_num = str(int(page_num) + 1)
if next_page_num in text_dict:
signature_pages[next_page_num] = text_dict[next_page_num]
return signature_pages