updated regex code

expert_report/.vscode/settings.json

@@ -0,0 +1,3 @@
+{
+    "python.formatting.provider": "black"
+}

expert_report/parse_expert_pdf.py

@@ -11,40 +11,32 @@ from parse_expert_pdf_utils import (
     address_extraction,
 )
 import tika
+import os
 from tika import parser
-import re
+
+tika.initVM()
 import warnings
-warnings.filterwarnings('ignore')
 
+warnings.filterwarnings("ignore")
+
+
+def main(path):
+    required_files = [file for file in os.listdir(path) if file.find(".pdf") != -1]
+    for idx, file in enumerate(required_files):
+        print(idx, file)
+        parsed_pdf = parser.from_file(file)
+        # print(parsed_pdf.keys())
 
-def main():
-    tika.initVM()
-    path = "/home/ftech/Code/xc/pdf_parser/pdfs/2018008353_Mail_Decision.pdf" #2018008353_Mail_Decision.pdf"
-    parsed_pdf = parser.from_file(path)
-    print(parsed_pdf.keys())
+        content = parsed_pdf["content"].strip().replace("\n", "")
+        # print(content)
+        date = date_extraction(content)
+        print(date)
+        address = address_extraction(content)
+        print(address)
 
-    content = parsed_pdf['content'].strip().replace('\n', '')
-    # print(content)
-    # case_number = case_number_extraction(content)
-    # print(case_number)
-    # expert_name = expert_name_extraction(content)
-    # print(expert_name)
-    # plaintiff = plaintiff_extraction(content)
-    # print(plaintiff)
-    # defendent = defendent_extraction(content)
-    # print(defendent)
-    # patent = patent_extraction(content)
-    # print(patent)
-    # on_behalf_of = on_behalf_of_extraction(content)
-    # print(on_behalf_of)
-    # acronym = acronym_extraction(content)
-    # print(acronym)
-    # pay = hourly_compensation(content)
-    # print(pay)
-    date = date_extraction(content)
-    print(date)
-    address = address_extraction(content)
-    print(address)
 
 if __name__ == "__main__":
-    main()
+    HOME_DIR = os.path.expanduser("~")
+    BASE_DIR = "/home/ftech/Code/pdf_parser/pdfs/"
+    path = os.path.join(HOME_DIR, BASE_DIR)
+    main(path)

expert_report/parse_expert_pdf_utils.py

@@ -1,19 +1,27 @@
-from collections import defaultdict
 import re
-from distutils.filelist import findall
 
 
 def date_extraction(content):
-    regex = r"((\d{2}\/\d{3}\,\d{3}\s)\d{1,2}\/\d{2}\/\d{4}|(Entered\:\s|:Filing\sDate\:\s|Date\:\s)[A-Z]\w+\s\d{1,2}\,\s\d{4})"
-    date = re.search(regex, content).groups()[0]
-    return date.strip()
+    """
+    extracts filing date from the documents.
+    """
+    pattern = r"((FILING\sDATE.*?)\d{1,2}\/\d{1,2}\/\d{2,4}|(Date\:\s)\w{1,9}\s\d{1,2}\,\s\d{4}|(Entered\:\s.*?)\w.*?\d{1,2}\,\s\d{4}|(Filed.*?)\d{1,2}\/\d{1,2}\/\d{1,2})"  # "((FILING\sDATE.*?)\d{1,2}\/\d{1,2}\/\d{2,4}|(Date\:\s)\w{1,9}\s\d{1,2}\,\s\d{4})"
+    try:
+        return re.search(pattern, content).groups()[0]
+    except:
+        return "None"
+
 
 def address_extraction(content):
-    regex = r"Address\:([\s\S].*)\s\d{6}\-\d{4}\swww" 
-    data = re.search(regex, content).groups()[0]
-    if data == None:
-        print("N")
-    return data.strip()
+    """
+    extracts address from the documents.
+    """
+    regex = r"Address\:\s([\s\S].*)www"
+    try:
+        print(re.search(regex, content).groups()[0])
+    except:
+        return "None"
+
 
 def case_number_extraction(content):
     # dict_case_numbers = defaultdict(int)