Regular Expressions

Using string methods and regular expressions to work with textual data

Data 100, Summer 2020 @ UC Berkeley

Josh Hug

LECTURE 8

Goals For Today

Goals For Today: Working With Text Data

• Canonicalizing text data.
• Extracting data from text.
• Using split.
• Using regular expressions.

​

String Canonicalization

Goal 1: Joining Tables with Mismatched Labels

join

???

A Joining Problem

join

???

To join our tables we’ll need to canonicalize the county names.

• Canonicalize: Convert data that has more than one possible presentation into a standard form.

Canonicalizing County Names

Canonicalization

Canonicalization:

• Replace each string with a unique representation.
• Feels very “hacky”, but messy problems often have messy solutions.

​

Can be done slightly better but not by much →

• Code is very brittle! Requires maintenance.

​

Tools used:�

Extracting From Text Using Split

Goal 2: Extracting Date Information

Suppose we want to extract times and dates from web server logs that look like the following:

​

​

​

​

There are existing libraries that do most of the work for us, but let’s try to do it from scratch.

• Will do together, just a little bit at a time.
• Let’s go check out lec08-working-with-text.ipynb.

169.237.46.168 - - [26/Jan/2014:10:47:58 -0800] "GET /stat141/Winter04/ HTTP/1.1" 200 2585 "http://anson.ucdavis.edu/courses/"

​

Extracting Date Information

​

​

​

​

169.237.46.168 - - [26/Jan/2014:10:47:58 -0800] "GET /stat141/Winter04/ HTTP/1.1" 200 2585 "http://anson.ucdavis.edu/courses/"

One possible solution:

Regular Expression Basics

Extracting Date Information

​

​

​

​

Earlier we saw that we can hack together code that uses split to extract info:

​

​

​

​

An alternate approach is to use a so-called “regular expression”:

• Implementation provided in the re library built into Python.
• We’ll spend some time today working up to expressions like shown below.

Regular Expressions

A formal language is a set of strings, typically described implicitly.

• Example: “The set of all strings of length < 10 that contain ‘horse’”

​

A regular language is a formal language that can be described by a regular expression (which we will define soon).

​

Example: [0-9]{3}-[0-9]{2}-[0-9]{4}

3 of any digit, then a dash, then 2 of any digit, then a dash, then 4 of any digit.

​

​

text = "My social security number is 123-45-6789.";

pattern = r"[0-9]{3}-[0-9]{2}-[0-9]{4}"

re.findall(pattern, text)

Regex101.com (or the online tutorial regexone.com)

There are a ton of nice resources out there to experiment with regular expressions (e.g. regex101.com, regexone.com, sublime text, python, etc).

​

I recommend trying out regex101.com, which provides a visually appealing and easy to use platform for experimenting with regular expressions.

• Example: https://regex101.com/r/1SREie/1

​

​

Regular Expression Syntax

The four basic operations for regular expressions.

• Can technically do anything with just these basic four (albeit tediously).

Regular Expression Syntax

AB*: A then zero or more copies of B: A, AB, ABB, ABBB

(AB)*: Zero or more copies of AB: ABABABAB, ABAB, AB,

Matches the empty string!

Puzzle: Use regex101.com to test! Or tinyurl.com/reg913z

Give a regular expression that matches moon, moooon, etc. Your expression should match any even number of os except zero (i.e. don’t match mn).

Puzzle Solution

Solution to puzzle on previous slide: moo(oo)*n

Regular Expression moo(oo)*n: https://tinyurl.com/reg913m

Give a regex that matches muun, muuuun, moon, moooon, etc. Your expression should match any even number of us or os except zero (i.e. don’t match mn).

Puzzle Solution

Solution to puzzle on previous slide: m(uu(uu)*|oo(oo)*)n

• Note: m(uu(uu)*)|(oo(oo)*)n is not correct! OR must be in parentheses!

Order of Operations in Regexes

m(uu(uu)*|oo(oo)*)n

• Matches starting with m and ending with n, with either of the following in the middle:
• uu(uu)*
• oo(oo)*

​

m(uu(uu)*)|(oo(oo)*)n

• Matches either of the following
• m followed by uu(uu)*
• oo(oo)* followed by n

​

In regexes | comes last.

Match examples:

muu

muuuu

oon

oooon

Match examples:

muun

muuuun

moon

moooon

More Advanced Regular Expressions Syntax

Expanded Regex Syntax

More Regular Expression Examples

Expanded Regex Puzzle: https://tinyurl.com/reg913w

Challenge: Give a regular expression for any lowercase string that has a repeated vowel (i.e. noon, peel, festoon, looop, etc).

Expanded Regex Puzzle Solution

Challenge: Give a regular expression for any lowercase string that has a repeated vowel (i.e. noon, peel, festoon, looop, etc): [a-z]*(aa|ee|ii|oo|uu)[a-z]*

Expanded Regex Syntax Puzzle: https://tinyurl.com/reg913v

Challenge: Give a regular expression for any string that contains both a lowercase letter and a number.

Click “run tests” to test your regex.

Expanded Regex Syntax Solution

Challenge: Give a regular expression for any string that contains both a lowercase letter and a number: (.*[0-9].*[a-z].*)|(.*[a-z].*[0-9].*)

Click “run tests” to test your regex.

Limitations of Regular Expressions

Writing regular expressions is like writing a program.

• Need to know the syntax well.
• Can be easier to write than to read.
• Can be difficult to debug.

​

Regular expressions sometimes jokingly referred to as a “write only language”.

​

Regular expressions are terrible at certain types of problems. Examples:

• For parsing a hierarchical structure, such as JSON, use a parser, not a regex!
• Complex features (e.g. valid email address).
• Counting (same number of instances of a and b). (impossible)
• Complex properties (palindromes, balanced parentheses). (impossible)�

"Some people, when confronted with a problem, think ‘I know, I'll use regular expressions.’ Now they have two problems." - Jamie Zawinski (Source)

Email Address Regular Expression (a probably bad idea)

The regular expression for email addresses (for the Perl programming language):

​

(?:(?:\r\n)?[ \t])*(?:(?:(?:[^()<>@,;:\\".\[\] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[\["()<>@,;:\\".\[\]]))|"(?:[^\"\r\\]|\\.|(?:(?:\r\n)?[ \t]))*"(?:(?: \r\n)?[ \t])*)(?:\.(?:(?:\r\n)?[ \t])*(?:[^()<>@,;:\\".\[\] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[\["()<>@,;:\\".\[\]]))|"(?:[^\"\r\\]|\\.|(?:(?:\r\n)?[ \t]))*"(?:(?:\r\n)?[ \t])*))*@(?:(?:\r\n)?[ \t])*(?:[^()<>@,;:\\".\[\] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[\["()<>@,;:\\".\[\]]))|\[([^\[\]\r\\]|\\.)*\ ](?:(?:\r\n)?[ \t])*)(?:\.(?:(?:\r\n)?[ \t])*(?:[^()<>@,;:\\".\[\] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[\["()<>@,;:\\".\[\]]))|\[([^\[\]\r\\]|\\.)*\](?: (?:\r\n)?[ \t])*))*|(?:[^()<>@,;:\\".\[\] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[\["()<>@,;:\\".\[\]]))|"(?:[^\"\r\\]|\\.|(?:(?:\r\n)?[ \t]))*"(?:(?:\r\n) ?[ \t])*)*\<(?:(?:\r\n)?[ \t])*(?:@(?:[^()<>@,;:\\".\[\] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[\["()<>@,;:\\".\[\]]))|\[([^\[\]\r\\]|\\.)*\](?:(?:\r\n)?[ \t])*)(?:\.(?:(?:\r\n)?[ \t])*(?:[^()<>@,;:\\".\[\] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[\["()<>@,;:\\".\[\]]))|\[([^\[\]\r\\]|\\.)*\](?:(?:\r\n)?[ \t] )*))*(?:,@(?:(?:\r\n)?[ \t])*(?:[^()<>@,;:\\".\[\] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[\["()<>@,;:\\".\[\]]))|\[([^\[\]\r\\]|\\.)*\](?:(?:\r\n)?[ \t])* )(?:\.(?:(?:\r\n)?[ \t])*(?:[^()<>@,;:\\".\[\] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[\["()<>@,;:\\".\[\]]))|\[([^\[\]\r\\]|\\.)*\](?:(?:\r\n)?[ \t])*))*) *:(?:(?:\r\n)?[ \t])*)?(?:[^()<>@,;:\\".\[\] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[\["()<>@,;:\\".\[\]]))|"(?:[^\"\r\\]|\\.|(?:(?:\r\n)?[ \t]))*"(?:(?:\r \n)?[ \t])*)(?:\.(?:(?:\r\n)?[ \t])*(?:[^()<>@,;:\\".\[\] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[\["()<>@,;:\\".\[\]]))|"(?:[^\"\r\\]|\\.|(?:(?:\r\n)?[ \t ]))*"(?:(?:\r\n)?[ \t])*))*@(?:(?:\r\n)?[ \t])*(?:[^()<>@,;:\\".\[\] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[\["()<>@,;:\\".\[\]]))|\[([^\[\]\r\\]|\\.)*\]( ?:(?:\r\n)?[ \t])*)(?:\.(?:(?:\r\n)?[ \t])*(?:[^()<>@,;:\\".\[\] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[\["()<>@,;:\\".\[\]]))|\[([^\[\]\r\\]|\\.)*\](?:(? :\r\n)?[ \t])*))*\>(?:(?:\r\n)?[ \t])*)|(?:[^()<>@,;:\\".\[\] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[\["()<>@,;:\\".\[\]]))|"(?:[^\"\r\\]|\\.|(?:(?:\r\n)? [ \t]))*"(?:(?:\r\n)?[ \t])*)*:(?:(?:\r\n)?[ \t])*(?:(?:(?:[^()<>@,;:\\".\[\] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[\["()<>@,;:\\".\[\]]))|"(?:[^\"\r\\]| \\.|(?:(?:\r\n)?[ \t]))*"(?:(?:\r\n)?[ \t])*)(?:\.(?:(?:\r\n)?[ \t])*(?:[^()<>@,;:\\".\[\] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[\["()<>@,;:\\".\[\]]))|" (?:[^\"\r\\]|\\.|(?:(?:\r\n)?[ \t]))*"(?:(?:\r\n)?[ \t])*))*@(?:(?:\r\n)?[ \t])*(?:[^()<>@,;:\\".\[\] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[\["()<>@,;:\\ ".\[\]]))|\[([^\[\]\r\\]|\\.)*\](?:(?:\r\n)?[ \t])*)(?:\.(?:(?:\r\n)?[ \t])*(?:[^()<>@,;:\\".\[\] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[\["()<>@,;:\\".\[ \]]))|\[([^\[\]\r\\]|\\.)*\](?:(?:\r\n)?[ \t])*))*|(?:[^()<>@,;:\\".\[\] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[\["()<>@,;:\\".\[\]]))|"(?:[^\"\r\\]|\\.|( ?:(?:\r\n)?[ \t]))*"(?:(?:\r\n)?[ \t])*)*\<(?:(?:\r\n)?[ \t])*(?:@(?:[^()<>@,;:\\".\[\] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[\["()<>@,;:\\".\[\]]))|\[([ ^\[\]\r\\]|\\.)*\](?:(?:\r\n)?[ \t])*)(?:\.(?:(?:\r\n)?[ \t])*(?:[^()<>@,;:\\".\[\] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[\["()<>@,;:\\".\[\]]))|\[([^\[\ ]\r\\]|\\.)*\](?:(?:\r\n)?[ \t])*))*(?:,@(?:(?:\r\n)?[ \t])*(?:[^()<>@,;:\\".\[\] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[\["()<>@,;:\\".\[\]]))|\[([^\[\]\ r\\]|\\.)*\](?:(?:\r\n)?[ \t])*)(?:\.(?:(?:\r\n)?[ \t])*(?:[^()<>@,;:\\".\[\] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[\["()<>@,;:\\".\[\]]))|\[([^\[\]\r\\] |\\.)*\](?:(?:\r\n)?[ \t])*))*)*:(?:(?:\r\n)?[ \t])*)?(?:[^()<>@,;:\\".\[\] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[\["()<>@,;:\\".\[\]]))|"(?:[^\"\r\\]|\\ .|(?:(?:\r\n)?[ \t]))*"(?:(?:\r\n)?[ \t])*)(?:\.(?:(?:\r\n)?[ \t])*(?:[^()<>@,;:\\".\[\] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[\["()<>@,;:\\".\[\]]))|"(? :[^\"\r\\]|\\.|(?:(?:\r\n)?[ \t]))*"(?:(?:\r\n)?[ \t])*))*@(?:(?:\r\n)?[ \t])*(?:[^()<>@,;:\\".\[\] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[\["()<>@,;:\\". \[\]]))|\[([^\[\]\r\\]|\\.)*\](?:(?:\r\n)?[ \t])*)(?:\.(?:(?:\r\n)?[ \t])*(?:[^()<>@,;:\\".\[\] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[\["()<>@,;:\\".\[\] ]))|\[([^\[\]\r\\]|\\.)*\](?:(?:\r\n)?[ \t])*))*\>(?:(?:\r\n)?[ \t])*)(?:,\s*(?:(?:[^()<>@,;:\\".\[\] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[\["()<>@,;:\\ ".\[\]]))|"(?:[^\"\r\\]|\\.|(?:(?:\r\n)?[ \t]))*"(?:(?:\r\n)?[ \t])*)(?:\.(?:(?:\r\n)?[ \t])*(?:[^()<>@,;:\\".\[\] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[ \["()<>@,;:\\".\[\]]))|"(?:[^\"\r\\]|\\.|(?:(?:\r\n)?[ \t]))*"(?:(?:\r\n)?[ \t])*))*@(?:(?:\r\n)?[ \t])*(?:[^()<>@,;:\\".\[\] \000-\031]+(?:(?:(?:\r\n)?[ \t ])+|\Z|(?=[\["()<>@,;:\\".\[\]]))|\[([^\[\]\r\\]|\\.)*\](?:(?:\r\n)?[ \t])*)(?:\.(?:(?:\r\n)?[ \t])*(?:[^()<>@,;:\\".\[\] \000-\031]+(?:(?:(?:\r\n)?[ \t])+| \Z|(?=[\["()<>@,;:\\".\[\]]))|\[([^\[\]\r\\]|\\.)*\](?:(?:\r\n)?[ \t])*))*|(?:[^()<>@,;:\\".\[\] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[\["()<>@,;:\\".\[\ ]]))|"(?:[^\"\r\\]|\\.|(?:(?:\r\n)?[ \t]))*"(?:(?:\r\n)?[ \t])*)*\<(?:(?:\r\n)?[ \t])*(?:@(?:[^()<>@,;:\\".\[\] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[\[" ()<>@,;:\\".\[\]]))|\[([^\[\]\r\\]|\\.)*\](?:(?:\r\n)?[ \t])*)(?:\.(?:(?:\r\n)?[ \t])*(?:[^()<>@,;:\\".\[\] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[\["()<> @,;:\\".\[\]]))|\[([^\[\]\r\\]|\\.)*\](?:(?:\r\n)?[ \t])*))*(?:,@(?:(?:\r\n)?[ \t])*(?:[^()<>@,;:\\".\[\] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[\["()<>@, ;:\\".\[\]]))|\[([^\[\]\r\\]|\\.)*\](?:(?:\r\n)?[ \t])*)(?:\.(?:(?:\r\n)?[ \t])*(?:[^()<>@,;:\\".\[\] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[\["()<>@,;:\\ ".\[\]]))|\[([^\[\]\r\\]|\\.)*\](?:(?:\r\n)?[ \t])*))*)*:(?:(?:\r\n)?[ \t])*)?(?:[^()<>@,;:\\".\[\] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[\["()<>@,;:\\". \[\]]))|"(?:[^\"\r\\]|\\.|(?:(?:\r\n)?[ \t]))*"(?:(?:\r\n)?[ \t])*)(?:\.(?:(?:\r\n)?[ \t])*(?:[^()<>@,;:\\".\[\] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z|(?=[\[ "()<>@,;:\\".\[\]]))|"(?:[^\"\r\\]|\\.|(?:(?:\r\n)?[ \t]))*"(?:(?:\r\n)?[ \t])*))*@(?:(?:\r\n)?[ \t])*(?:[^()<>@,;:\\".\[\] \000-\031]+(?:(?:(?:\r\n)?[ \t]) +|\Z|(?=[\["()<>@,;:\\".\[\]]))|\[([^\[\]\r\\]|\\.)*\](?:(?:\r\n)?[ \t])*)(?:\.(?:(?:\r\n)?[ \t])*(?:[^()<>@,;:\\".\[\] \000-\031]+(?:(?:(?:\r\n)?[ \t])+|\Z |(?=[\["()<>@,;:\\".\[\]]))|\[([^\[\]\r\\]|\\.)*\](?:(?:\r\n)?[ \t])*))*\>(?:(?:\r\n)?[ \t])*))*)?;\s*)

From: http://www.ex-parrot.com/~pdw/Mail-RFC822-Address.html

Even More Regular Expression Syntax

Suppose you want to match one of our special characters like . or [ or ]

• In these cases, you must “escape” the character using the backslash.
• You can think of the backslash as meaning “take this next character literally”.

Regular Expressions Puzzle: tinyurl.com/reg913a

Create a regular expression that matches the red portion below.

169.237.46.168 - - [26/Jan/2014:10:47:58 -0800] "GET /stat141/Winter04/ HTTP/1.1" 200 2585 "http://anson.ucdavis.edu/courses/"

Regular Expressions Puzzle Solution: tinyurl.com/reg913a

Create a regular expression that matches the red portion below: \[.*\]

169.237.46.168 - - [26/Jan/2014:10:47:58 -0800] "GET /stat141/Winter04/ HTTP/1.1" 200 2585 "http://anson.ucdavis.edu/courses/"

Quiz

Create a regular expression that matches anything inside of angle brackets <>, but none of the string outside of angle brackets.

• Example: <div><td valign="top">Moo</td></div>
• Moo should not match because it is not between < and >.
• Note: This is equivalent to the problem of matching HTML tags.

Regex101 link: https://tinyurl.com/reg913

Even More Regular Expression Features

A few additional common regex features are listed above.

• Won’t discuss these in class, but might come up in discussion or hw.
• There are even more out there!

The official guide is good! https://docs.python.org/3/howto/regex.html

​

5.*5 would match this!

Regular Expressions in Python

(and Regex Groups)

re.findall in Python

In Python, re.findall(pattern, text) will return a list of all matches.

​

​

text = "My social security number is 456-76-4295 bro, or actually maybe it’s 456-67-4295.";

pattern = r"[0-9]{3}-[0-9]{2}-[0-9]{4}"

m = re.findall(pattern, text)

print(m)

['456-76-4295', '456-67-4295']

re.sub in Python

In Python, re.sub(pattern, repl, text) will return text with all instances of pattern replaced by repl.

​

​

text = '<div><td valign="top">Moo</td></div>'

pattern = r"<[^>]+>"

cleaned = re.sub(pattern, '', text)

print(cleaned)

’Moo’

Raw Strings in Python

Note: When specifying a pattern, we strongly suggest using “raw strings”.

• A raw string is created using r“” or r‘’ instead of just “” or ‘’.
• The exact reason is a bit tedious.
• Rough idea: Regular expressions and Python strings both use \ as an escape character.
• Using non-raw strings leads to uglier regular expressions.

​

​

​

​

For more information see “The Backslash Plague” under https://docs.python.org/3/howto/regex.html.

​

Regular Expression Groups

Earlier we used parentheses to specify the order of operations.

​

Parenthesis have another meaning:

• Every set of parentheses specifies a so-called “group”.
• Regular expression matchers (e.g. re.findall, regex101.com) will return matches organized by groups. In Python, returned as tuples.

​

​

s = """Observations: 03:04:53 - Horse awakens.

03:05:14 - Horse goes back to sleep."""

pattern = "(\d\d):(\d\d):(\d\d) - (.*)"

matches = re.findall(pattern, s)

[('03', '04', '53', 'Horse awakens.'),

('03', '05', '14', 'Horse goes back to sleep.')]

​

Regex Puzzle

​

​

​

​

Fill in the regex below so that after code executes, day is “26”, month is “Jan”, and year is “2014”.

• See lec08-working-with-text.ipynb or https://tinyurl.com/reg913s.

169.237.46.168 - - [26/Jan/2014:10:47:58 -0800] "GET /stat141/Winter04/ HTTP/1.1" 200 2585 "http://anson.ucdavis.edu/courses/"

pattern = "YOUR REGEX HERE"

matches = re.findall(pattern, log[0])

day, month, year = matches[0]

​

log[0]:

Regex Puzzle (One Possible Solution)

​

​

​

​

Fill in the regex below so that after it executes, day is “26”, month is “Jan”, and year is “2014”.

169.237.46.168 - - [26/Jan/2014:10:47:58 -0800] "GET /stat141/Winter04/ HTTP/1.1" 200 2585 "http://anson.ucdavis.edu/courses/"

pattern = "\[(\d{2})/(\w{3})/(\d{4})"

matches = re.findall(pattern, log[0])

day, month, year = matches[0]

​

log[0]:

Extracting Date Information

​

​

​

​

With a little more work, we can do something similar and extract day, month, year, hour, minutes, seconds, and time zone all in one regular expression.

• Derivation is left as an exercise for you guys.

​

​

​

​

You will also see code that uses re.search instead of re.findall.

• Beyond the scope of our lecture today.

Case Studies on Police Data and Restaurant Data

See lec08-working-with-text.ipynb

Summary

Today we saw many different string manipulation tools.

• There are many many more!
• With just this basic set of tools, you can do most of what you’ll need.

​

​

​

Even More Regex Syntax (Bonus)

Optional (but Handy) Regex Concepts

These regex features aren’t going to be on an exam, but they are useful:

• Lookaround: match “good” if it’s not preceded by “not”: (?<!not )good
• Backreferences: match HTML tags of the same name: <(\w+)>.*</\1>
• Named groups: match a vowel as a named group: (?P<vowel>[aeiou])
• Free Space: Allow free space and comments in a pattern:�# Match a 20th or 21st century date in yyyy-mm-dd format

(19|20)\d\d # year (group 1)

[- /.] # separator

(0[1-9]|1[012]) # month (group 2)

[- /.] # separator

(0[1-9]|[12][0-9]|3[01]) # day (group 3)