EECS 201 - Class 5: Regular Expressions

# Overview
1. Announcements
2. Review + Exercises
3. Q&A
4. Basic assignment
---

# Announcements
* Shell assignments due Feb 21
* Regex assignments due Mar 6
* Shell survey closing tonight!
---

## Misc special characters
* `.` matches _any_ single character
  * `...` matches three consecutive characters
--

* `|` for an OR between regexes
  * `hello|world` matches a string that is "hello" or "world"
--

* `\` for special expressions/escapes
  * `\b` matches the empty string at the edge of a "word"
  * There's more: check the GNU `grep` manual for the rest
--

* `(`, `)` enclose a whole expression as a _subexpression_
---

## Brackets
* `[`, `]` enclose a set to match for __one character__
  * `[abc]` matches 'a', 'b', or 'c'
* `-`: range
  * `[A-Za-z0-9]`: capital and lowercase numbers and digits
* `^`: not in set
  * `[^ab]`: everything not 'a' or 'b'
* Named classes
  * e.g. `[:alnum:]` (alphanumeric characters)
  * See the Grep documentation for more
--

* `[:alnum:]` is the actual name for the class, goes inside of `[`, `]`
  * `[[:alnum:]]` to actually use
  * to further illustrate: `[[:alpha:]0-5]`
---

## Quantifiers
* Specify how many of a preceding regex to match
* `?`: ≤1 time
* `*`: ≥0 times
* `+`: ≥1 times
* `{n}`: _n_ times
* `{n,}`: ≥_n_ times
* `{,m}`: ≤_m_ times
* `{n,m}`: _x_ times where _n _≤ _x_ ≤ _m_
---

## Exercise 1
* If you want to test these with `grep`, try using `grep -E`
  * Default `grep` uses BRE, which requires you to `\` escape a lot of things (more on this at the end)
* Write regexes that matches against:
  1. "hello" or "world"
  3. 3 of any character, "cat", then at least 5 of any character
---

## Exercise 2
Write regexes that match against:
1. 3 English vowels (a, e, i , o, u) in a row
2. 5 non-number characters in a row
3. "Odd" and a single digit odd number
4. "Even" and an even number
  * For simplicity's sake, leading 0s are allowed
---

## Anchors
* Perform _positional_ matching
* `^`: match empty string at the beginning of a line
* `$`: match empty string at the end of a line
---

## Exercise 3
Write regexes that match against:
1. File names that end in ".txt"
2. File names that start with "file" with an odd number after and ends in ".txt"
3. A phone number with an optional country code
  * In the +X (XXX) XXX-XXXX format
---

## Backreferences
* Match previous parenthesized `()` subexpression
* `\n`: match _n_ th parenthesized subexpression
---

## BRE vs ERE
* In BRE `?`, `+`, `{`, `|`, `(`, and `)` must be escaped with `\`
  * (and `}` if you're specifying an interval)
---

## Exercise 4
* `sed` is a command that performs text transformations
* The `s` command can perform search-and-replace
* `sed 's/hello/world/g'` replaces instances of "hello" with "world"
  * The `g` at the end allows for multiple replacements on a line
  * The first character specifies the delimiter: doesn't have to be `/`
  * `s/pattern/replacement/flags`
  * `-E` as an argument puts `sed` into ERE mode
* Write a `sed` command that replaces instances of four digit numbers with "(XXXX)"
  * To be a number, it can't have letters next to it
  * `\<` and `\>` are positional matches for word boundaries
---