Regex Tester

A tiny language for describing shapes of text

A regular expression is a small, dense language whose only purpose is to describe patterns in text. Once the vocabulary clicks, an enormous amount of everyday work collapses into a single line: pulling every URL out of a document, validating that a field looks like a date, splitting a log line into its component fields, or finding every function definition in a file. The density is both its power and its reputation — a regex that would take thirty lines of hand-written parsing is one line of symbols, and that line is genuinely hard to read six months later. The two habits that make regexes maintainable are to comment them, and to test them against real data rather than the three examples you happened to think of.

Know when not to use one

Regular expressions are the wrong tool for parsing nested structures, and this is not a matter of taste. HTML, XML and JSON can nest arbitrarily deeply, and regular expressions — in their formal sense — provably cannot count nesting depth, which is why the famous Stack Overflow answer about parsing HTML with regex descends into cosmic horror. The pattern that works for your ten test cases will fail the moment a real document nests one level deeper or puts an angle bracket inside an attribute. Use a real parser for structured formats. Keep regexes for what they are genuinely superb at: flat, line-oriented, pattern-shaped text — log files, CSV fields, validation, search and replace.