sample-nltk

import nltk
import re
with open('/home/chetan/Documents/sample-certificate.txt','r') as file:
	text = file.read()
# print(text)
sentences = nltk.sent_tokenize(text)
tokenized_sentences = [nltk.word_tokenize(sentence) for sentence in sentences]
# print(tokenized_sentences)
sign_date = {}

for indecies, elements in enumerate(tokenized_sentences):
	for index, element in enumerate(elements):
		elements.append(element)
		if(element == "dated"):
			sign_date[indecies] = index 
			print(index)
		elif(element == ','):
			sign_date.append(index)
			sign_date[indecies] = index
		
print(elements)
regex = re.compile(r"BP(\d{8})")
result = regex.search(text)
print('Agreement Number: '+result.group())

tokenized_sentences has that list i sent you earlier.