Lecture 28

Regular expressions

MCS 260 Fall 2020
David Dumas

Reminders

Loose end: Recursion pros and cons

Often can solve a problem with recursion or with loops (an iterative solution). Why use recursion?

Regular expressions

Today we'll learn about the module re in Python, which supports a text searching language known as regular expressions or regexes.

Some of its key functions include:

• Searching for text matching a pattern
• Replacing text matching a pattern

Minimal example

Regexes are a mini programming language for specifying patterns of text. Dialects of regex are supported in many programming languages. We'll cover the Python dialect.

Simplest usage: Find and replace a substring.

import re
s = "Avocado is usually considered a vegetable."
print(re.sub("vegetable","fruit",s))
    

The first argument of re.sub is a pattern.

Unless it contains characters with special meaning in a regex pattern, the pattern just matches substrings equal to the pattern.

• "vegetable" matches the string "vegetable"
• "foo" matches the string "foo"

raw strings

Recall that backslash \ in a string starts an escape sequence in Python, and \\ represents a single backslash character in the string. If your string contains a lot of backslashes, you may want to disable escape sequences.

You can do so by putting the letter r immediately before the quotation mark(s). This is known as a raw string. In a raw string, a single \ represents the \ character.

special characters in patterns

• . — matches any character except newline
• \s — matches any whitespace character
• \d — matches a decimal digit
• + — previous item must repeat 1 or more times
• * — previous item must repeat 0 or more times
• ? — previous item must repeat 0 or 1 times
• {n} — previous item must appear n times

Example problem

Replace any price in whole dollars (written like $2 or $1999) with the string -PRICE-.

Note: $ is a special character. To match a dollar sign, use \$.

Matching and searching

What if you don't want to replace a regex, just find it?

• re.match(pattern,string) — does string begin with a match to pattern? Return a match object or None.
• re.search(pattern,string) — does string contain a match to the pattern? Return a match object or None.
• re.findall(pattern,string) — return a list of all non-overlapping matches as strings.

Match objects

If a match is found, then the match object has a method .group() that returns the full text of the match.

.start() and .end() return the indices where the match begins and ends in the string.

Parentheses

A part of a pattern in parentheses is a group. A group is treated as a unit for operators like +,*,?.

e.g. pattern (ha)+ matches ha or haha or hahaha but does not match Haha or h or hah.

Matched groups are available from the match object using .group(1), .group(2), etc..

Example problem

Find all of the phone numbers in a string that are written in the format 319-555-1012, and split each one into area code (e.g. 319), exchange (e.g. 555), and line number (e.g. 1012).