Skip to content

Latest commit

 

History

History
103 lines (75 loc) · 2.81 KB

README.md

File metadata and controls

103 lines (75 loc) · 2.81 KB
Raw

Introduction

almson-regex is a simple library for writing readable regular expressions.

The goals of this library are:

  • Descriptive syntax
  • Good documentation
  • Support of all Java regex features

The following are not goals:

  • Type safety (keep it simple, use Strings!)
  • Extreme brevity (if you want super-compact, illegible regular expressions, write them the old way!)
  • Allow user to avoid learning regular expressions (fundamentally, you're still writing and reading regexes, but the descriptive names and good documentation makes it much easier!)

almson-regex is based on string operations, and is easy to use for all, some, or parts of your regular expressions. almson-regex is compiled with Java 8 and supports all Java 17 regex features, such as named capturing groups and glyph cluster matchers.

The documentation for the library doesn't replace knowledge of how regular expressions work. However, this library succeeds in making your regular expressions easy to read by those who do not have expert knowledge. For the best reference on Java regular expressions, see the java.util.regex.Pattern documentation.

Installation with Maven

<dependency>
    <groupId>net.almson</groupId>
    <artifactId>almson-regex</artifactId>
    <version>1.5.1</version>
</dependency>

Examples

Match leading or trailiing whitespace:

String regex = "^[ \t]+|[ \t]+$"

becomes:

String regex = 
    either(sequence(START_BOUNDARY, oneOrMore(HORIZONTAL_WHITESPACE)),
           sequence(oneOrMore(HORIZONTAL_WHITESPACE), END_BOUNDARY))

but since sequence simply does string concatenation, we can also write:

either(START_BOUNDARY + oneOrMore(HORIZONTAL_WHITESPACE),
       oneOrMore(HORIZONTAL_WHITESPACE) + END_BOUNDARY)

Match an IP address (simple version):

"\\b(\\d{1,3}\\.){3}\\d{1,3}\\b"

becomes:

WORD_BOUNDARY +
exactly(3, between(1, 3, DIGIT) + text(".")) +
between(1, 3, DIGIT) +
WORD_BOUNDARY

Match an email address (simple version):

"\\b(<user>[a-zA-Z0-9._%+-]+)@(?<domain>[A-Z0-9.-]+\.\\p{L}{2,})\\b"

becomes:

WORD_BOUNDARY + 
namedGroup("user", 
           oneOrMore(charclassUnion(LETTER, DIGIT, charclass('.', '_', '%', '+', '-')))) + 
text("@") + 
namedGroup("domain", 
           oneOrMore(charclassUnion(LETTER, DIGIT, charclass('.', '-'))) + 
text(".") + 
atLeast(2, LETTER)) + 
WORD_BOUNDARY

Select consecutive duplicates from a comma-delimited list

"(?<=,|^)([^,]*)(,\1)+(?=,|$)"

becomes:

precededBy (either (START_BOUNDARY, text (","))) + 
group (zeroOrMore(charclassComplement(charclass(',')))) + 
oneOrMore (text(",") + backreference(1)) + 
followedBy (either (text(","), END_BOUNDARY))