From c32112799bebcf177cfaa847d21a3c106b448e2a Mon Sep 17 00:00:00 2001
From: Zuzanna Rachuba <zuziarachuba@gmail.com>
Date: Thu, 31 Oct 2024 12:46:23 +0100
Subject: [PATCH] Regular expressions for dates and postal codes

---
 projekt.py | 43 ++++++++++++++++++++++++++++++++++++++-----
 1 file changed, 38 insertions(+), 5 deletions(-)

diff --git a/projekt.py b/projekt.py
index 0a36fc9..2d9ce13 100644
--- a/projekt.py
+++ b/projekt.py
@@ -1,7 +1,10 @@
 import fitz
-import pdfplumber
+import re 
+
 my_path = r"C:\Users\DELL\Downloads\A-24VU-00511.PDF.pdf"
 
+postal_code_pattern = r'\b(?!0)([1-9]\d?-\d{3}|[1-9]\d{4}|[1-9]\d? \d{3})\b'
+
 with fitz.open(my_path) as doc:
     order_numbers = []
     order_dates = []
@@ -12,10 +15,40 @@ with fitz.open(my_path) as doc:
 
         if "Your ref. no. PO" in text:
             parts = text.split("Your ref. no. PO") 
-            for part in parts[1:]: #idziemy od 2 elementu, żeby nie brać tekstu
+            for part in parts[1:]:
                 order_number = part.split()[0].strip()
                 order_numbers.append(order_number)
-        
-                   
-print("Reference numbers: ", order_numbers)
 
+        date_pattern = r'\b\d{2}/\d{2}/\d{2,4}\b'
+        matches = re.findall(date_pattern, text)
+        order_dates.extend(matches)
+
+
+        lines = text.splitlines()
+        for i in range(1, len(lines)):  
+            line = lines[i]
+
+            if re.search(postal_code_pattern, line):
+            
+                line_2 = lines[i-2] if i>0 else ''
+                line_above = lines[i - 1] if i > 0 else ''
+                line_below = lines[i + 1] if i + 1 < len(lines) else ''
+                
+         
+                if not re.search(r'[^0-9a-zA-Z \-]', line):
+                    buyers.append({
+                        'line_2': line_2,
+                        'line_above': line_above,
+                        'postal_code_line': line,
+                        'line_below': line_below
+                    })
+
+print("Reference numbers:", order_numbers)
+print("Document dates:", order_dates)
+print("Buyers: ")
+for context in buyers:
+    print("Company: ", context['line_2'])
+    print("St. :", context['line_above'])
+    print("City: ", context['postal_code_line'])
+    print("Country:", context['line_below'])
+    print()