Regular Expression

Four tips for learning regular expressions bit.ly/Xn7UH

What we call 'regular expressions' today includes more general patterns than the original definition from the 1950's.

The escape character \ makes special characters literal. For example, \. matches a period.

Four tips for learning regular expressions johndcook.com/blog/2008/01/1…

.* will match as much as it can, but then backtrack if the next atom can also be matched. Backtracking on some engines is slow.

(Fine print: whether . matches a newline character depends on the implementation and configuration options.)

A dot matches any single character. For example, /r.de/ matches 'ride', 'rode' etc.

+ means one or more occurrences. For example, /ab+c/ matches 'abc' or 'abbc' or 'abbbbbbc'.

Finding all English words with four or more consecutive vowels: johndcook.com/blog/2019/10/1…

? means a character or expression is optional, i.e. zero or one occurrences. Example: /ab?c/ matches 'ac' and 'abc' but not 'abbc'.

* means zero or more occurrences. Ex: /ab*c/ matches 'ac', 'abc', 'abbc', etc.

Languages have minor variations in their regex support. @RegexTip mostly sticks to the common features of Perl, C#, Python, JavaScript, ...

Create a diagram of a regular expression regexper.com

'Everybody stand back. I know regular expressions.' xkcd.com/208/

Unicode Regular Expressions Standard: unicode.org/reports/tr18/

Why Gnu grep is fast lists.freebsd.org/pipermail/free…

Regular expressions in Haskell wiki.haskell.org/Regular_expres…

Comparing regular expressions in Perl, Python, and Emacs johndcook.com/blog/regex-per…

Properties such as \p{Alphabetic} can be easier to read than [A-Za-z]. Works better with internationalization as well.

Perl Unicode properties: perldoc.perl.org/perluniprops.h…

Adding a + after a quantifier tells it to not give anything back when the regex engine backtracks.