Regex Cheat Sheet

The character class is the most basic regex concept. It converts a small group of characters to match a larger number of characters.

For example:

1. [xyz]  matches x or y or z
2. [^pqr] matches any character _except_ p, q, or r (negation)
3. [a-zA-Z] matches a through z or A through Z, inclusive (range)

To operate on all items in a group, use a group expression. You can use a group expression to apply an operator to a group or to find a specified text before or after each member of the group, for example. The grouping operator is the parentheses, while the "|" is used to divide the elements.

For example:

• My (red|pink|blue) dress
  • Here "My red dress", "My blue kite", and "My pink dress" match the expression. "My yellow kite" or "My dress" do not match.

Let’s look into some of these operators.

More examples:

1. x(yz) parentheses create a capturing group with value yz
2. p(?:qr)* using ?: you disable the capturing group
3. a(?<ok>bc) using ?<ok> we put a name to the group

Quantifiers specify how many characters or expressions should be matched.

The quantifiers ( * + {}) are also known as greedy operators. They expand the match as far as they can via the input text.

For example:

1. <[^<>]+>  matches any of the characters except < or > included one or more times inside < and >
2. \w+?\d\d\w+  matches abcdef42ghijklmnfhaeij

More examples for quantifiers

Based on the current position in the string, determines whether the match will succeed or fail.

For example:

1. ^\d{3} matched 444 in 444-888-999-..
2. -\d{3}\Z matches -220 in 110-220
3. \babc\b performs a "whole words only" search
4. \Bend\w*\b matches "ends", "ender" in "end sends endure lender"
5. ^Hello me$ matches the string Hello me

Now, that you have got some brief idea about the characters, quantifiers and groups. Let’s look into some combined examples.

1. ^(\d*)[.,](\d+)$ matches numbers like 12,3 or 12.3
2. ^[a-zA-Z0-9 ]*$ matches any alphanumeric with spaces.
3. ^[\s]*(.*?)[\s]*$ matches the text by avoiding the extra spaces
4. ^([a-zA-Z0-9._%-]+@[a-zA-Z0-9.-]+\.[a-zA-Z]{2,6})*$ could be used for matching email
5. (https?)://(www)?.?(\\w+).(\\w+)/?(\\w+)? can be used matching URLs

When the regex engine processes the lookaround expression, it first creates a substring from the present place to the beginning (lookbehind) or end (lookahead) of the original string, and then runs Regex. With the use of the lookaround pattern, IsMatch on that picked substring. A positive or negative assertion might be used to assess the outcome's success.

Examples:

1. (?=\d{10})\d{4} matches 2348 in 2348856787
2. (?<=\d)rat matches mat in 2mat
3. (?!theatre)the\w+ matches theme
4. \w{3}(?<!mon)ster matches munster

A character class is a combination of a small number of characters and a large number of characters. Only within bracket expressions can we use POSIX character classes. To generate regular expressions, the POSIX standard supports the following character classes.

Examples:

1. [8[:alpha:]]+ sample match could beWellDone88
2. [[:alpha:]\d]+ sample match could beкошка99
3. [[:alnum:]]{10} sample match could beABC1275851
4. [[:alnum:]]{10} sample match could be кошка67810
5. [[:punct:]]+ sample match could be ?!.,:;
6. [[:punct:]]+ sample match could be ‽,:〽⁆

The following are examples of Unicode property escapes:

1. \p{prop=value}: All characters with the prop property have the value value.
2. \P{prop=value}: All characters without a property prop with the value value are matched. Match \p{bin_prop}: all characters with the bin prop binary property set to True.
3. \P{bin_prop}: All characters with the binary attribute bin prop set to False will be matched.

A flag is a parameter that can be added to a regex to change how it searches. A flag modifies a regular expression's default searching behaviour. It does a regex search in a unique method. A single lowercase alphabetic character is used to represent a flag. There are six flags in the JavaScript regex, each providing a different purpose.

When employing the forward slashes / to build an expression, flags come after the second slash. This can be expressed in general notation as follows: \pattern\flag

For example, if the flag i was added to the regex /a/, the result would be /a/i.

To provide a regex with many flags, we write them one by one (without any spaces or other delimiters).

If we gave the flags i and g to the regex /a/, for example, we'd write /a/ig (or equivalently /a/gi, as the order doesn't matter).

Note: The sequence in which flags appear is irrelevant; flags simply change the behaviour of searching, thus placing one before the other makes no difference.

The following engines enable recursion: PCRE (C, PHP, R...) Perl Ruby 2+ Python via the alternate regex package JGSoft (not available in a programming language)

The most common application of recursion is to match balanced constructions.

a(?R)?z, a(?0)?z, and a\g<0>?z are all regexes that match one or more letters followed by the same number of letters z.

Examples:

• (\((?R)?\)) match parentheses like ((()))
• (\((?R)*\)) match parentheses like (()()())
• \w{3}\d{4}(?R)? matches patterns like ccc8888ggg9999

Because two "equivalent" regexes might have substantial changes in processing performance, you should understand how your regex engine works.

1. It is feasible to construct regexes that match in exponential time, but you must essentially TRY to do so.
2. Regexes that run in quadratic time are more commonly created by accident.
3. Problems of many kinds
   • Recompilation (from forgetting to compile regexes used multiple times)
   • The Middle Dot-star (which causes backtracking)
     • The first approach, use a character class that is negated.
     • Use reluctant quantifiers as a second option.

• When you require parentheses but not capture, use non-capturing groups.
• Do a quick check before attempting a match, if the regex is very complex, e.g.
  • Does an email address contain '@'?
• Present the most likely option(s) first, e.g.
  • light green|dark green|brown|yellow|green|pink leaf
• Minimize the amount of looping
  • \d\d\d\d\d\d is faster than \d{6}
  • aaaaaa+ is faster than a{6,}
• Avoid obvious backtracking, e.g.
  • Mr|Ms|Mrs should be M(?:rs?|s)
  • Good night|Good morning should be Good (?:night|morning)

Conclusion

In this blog, you came across some of the interesting regex command concepts that would be very helpful in the matching of different kinds of strings. You also learned about some of the pitfalls that could happen when using regex and that could affect the performance of your application too. Further to get over these pitfalls, the blog also discusses some of the common tips that could help you in overcoming these obstacles.

Useful Resources

• Technical Interview Questions
• Coding Interview Questions
• Interview Resources
• DSA- Programming
• Mock Interview