Everything You Need to Know About Regex (WIP)
Regex is amazing, super powerful, and is a topic that every programmer need to be familiar with. Regex is short for regular expressions, and it is a string that describes a pattern to be found in other pieces of text.
What is Regex Used For?
- Finding
- Replacing
- Validating
- Splitting
How Do We Use Regex?
Regex is ran using a regex engine. A regex engine has support for finding, replacing, validating, and splitting. Many programs you are already familiar for support regex in their search/find functionality including but not limited to the following:
- Notepad ++
- Visual Studio
Also, most programming languages support regex including C#, Python, and JavaScript! There are a few advanced regex topics that aren’t supported everywhere, but most of the basics are the same in every regex engine.
My favorite regex engine to use for writing and testing regex before implementing in my final code is RegExr. RegExr has a helpful reference list and you can visually see your matches as you make changes to you regex. You can even replace values to see what your final string will look like if you are using regex for replacements.
The Basics
Matching Characters
We will start by just matching a single word in a sentence to learn some of the basics.
Let’s use the sentence Where are we going to get food?
And let’s find the word food. This regex match is as easy. It is just as easy as doing a case-sensitive search in any program you are familiar with. To find the word food in the sentence, the regex is just food. Food will not work since regex is case sensitive.
There are some characters that cannot be used in regex without an escape character since they have special meanings in regex. The escape character in regex is \ . Here are the common escaped characters
| Description | Example |
|---|---|
| Reserved Characters +*?^$\.[]{}()|/ |
\+ |
| Octal Escape | \000 |
| Hexadecimal Escape | \xFF |
| Unicode Escape | \uFFFF |
| Extended Unicode Escape | \u{FFFF} |
| Control Character Escape | \cI |
| Tab | \t |
| Line Feed | \n |
| Vertical Tab | \v |
| Form Feed | \f |
| Carriage Return | \r |
| Null | \0 |
Matching Characters Using Quantifiers
In the word food, there are 2 os. Do we have to type out the word to match or is there a different way to do this matching? You guessed it. There is a different way! Now, matching the word food in a real scenario it would be best to just use the simplest most readable regex from above, food, but for examples and build on the basics here are other ways the match can be written, fo*d or fo+d. The * means match 0 or more of the previous character and + means match 1 or more of the previous character. So what happens if there was a typo in the text we were searching and food had 3 os? Well both of those would match the word foood as well which is not what we want. We can however use regex to say match on 2 occurrences exactly using fo{2}d.
Here are some other useful quantifiers that we can go into detail in later examples.
| Name | Ex |
|---|---|
| plus Matches 1 or more of the preceding token |
+ |
| star Matches 0 or more of the preceding token |
* |
| quantifier Matches the specified quantity of the previous token |
{1,3} |
| optional Matches 0 or 1 of the preceding token |
? |
| lazy Matches as few characters as possible. By default matches will match as many characters as possible. |
? |
| alternation Acts the the boolean OR. Matches one or the other. Can be used in groups. |
| |
Matching Characters Using Character Groups
Now let’s use the sentence Where are we going to get food to feed Max? , and the goal of matching 2 words, food and feed. These words are very similar. The only difference are the middle 2 letter, o and e.
To do this, let’s use a character group. Insead of matching just an o we also want to match an e. That can be written using square brackets, [] , with an o and e inside, [oe]. That regex matches an o or e. If we swap that into the regex we wrote before, we get f[oe]{2}d which matches both food and feed. We could also write this using alternation, f(oo|ee)d or f(o|e){2}d or many other ways.
Using Capturing Groups
Now that we know some basic matching techniques, let’s learn about capturing groups. Capturing groups come in handy when you want to replace data in a string or use a specific part of the match for another reason. Let’s start with a new example, say a list of names and email addresses. We need a tab delimited file with name in column 1 and phone number in column 2, but the phone number formatting is all over the place currently.
Example Data to Clean
Sally 2344447890
Joe 123-456 9999
Susan (789)999-9999
Samantha 888-888-8888
Max 234.123.1234
Matching Regex
([a-zA-Z]+)\s+\D*([0-9]{3})\D*([0-9]{3})\D*([0-9]{4})
List Data
$1\t$2-$3-$4\n
References
RegExr: Learn, Build, & Test RegEx