69 — Patterns: finding matches with regex

"Find all phone numbers" needs a pattern, not a fixed word. The standard re library matches shapes: three digits, a hyphen, four digits — and anything that looks like an email address.

#Steps

1. \d matches any digit and {3} means exactly three, so \d{3}-\d{4} is three digits, a hyphen, four digits. re.findall returns every match:

   import re
   text = "Call 010-1234-5678 now"
   phones = re.findall(r"\d{3}-\d{4}", text)
   show phones
   

   copyrun it →

   ['010-1234']
   

   copy
2. An email is one-or-more allowed characters, @, then one-or-more more. + means one or more; brackets list the allowed characters:

   import re
   text = "Write to [email protected] or [email protected]"
   emails = re.findall(r"[A-Za-z0-9._%+-]+@[A-Za-z0-9.-]+", text)
   show emails
   

   copyrun it →

   ['[email protected]', '[email protected]']
   

   copy
3. Create a small file of contacts:

   Mina 010-1234-5678 [email protected]
   Jun 010-9876-5432 [email protected]
   Office 02-3456-7890 [email protected]
   

   copy
4. The full program reads the file with open(...).read(). The try / except from guide 68 reports a missing file:

   # contacts.nme — find phone numbers and emails in a text file.
   # Run: nme r contacts
   # re.findall searches a whole file with one pattern.
   
   import re
   
   ask file_name, "Text file to scan (for example contacts.txt): "
   
   try:
       text = open(file_name).read()
   except FileNotFoundError:
       show f"{file_name} is not in this folder."
   else:
       phones = re.findall(r"\d{3}-\d{4}", text)
       emails = re.findall(r"[A-Za-z0-9._%+-]+@[A-Za-z0-9.-]+", text)
   
       show f"Found {len(phones)} phone numbers:"
       for phone in phones:
           show f"  {phone}"
   
       show f"Found {len(emails)} email addresses:"
       for email in emails:
           show f"  {email}"
   
       both = 0
       for line in text.splitlines():
           if re.search(r"\d{3}-\d{4}", line) and re.search(r"[A-Za-z0-9._%+-]+@[A-Za-z0-9.-]+", line):
               both = both + 1
       show f"{both} of {len(text.splitlines())} lines have both a phone and an email."
   

   copyrun it →
5. Run it:

   printf 'contacts.txt\n' | nme r contacts
   

   copy

   Text file to scan (for example contacts.txt): Found 3 phone numbers:
     010-1234
     010-9876
     456-7890
   Found 3 email addresses:
     [email protected]
     [email protected]
     [email protected]
   3 of 3 lines have both a phone and an email.
   

   copy
6. re.findall collects every match in the text. re.search(pattern, line) answers one yes/no — it returns a match object or None — which is how the program counts lines that have both.

#Try it yourself

Add a row with a second phone style and widen the pattern, or search for a name like Mina. Report lines that have a phone but no email.

#What you learned

• import re loads the standard regular-expression library.
• re.findall(pattern, text) returns every match in the text as a list.
• \d is any digit, {3} means exactly three, and + means one or more.
• re.search(pattern, line) tests one line and returns None when there is no match.