burzum / flux

Fluent Regular Expressions for PHP

FLUX (Fluent Regex) 0.4.0

by Selvin Ortiz

Description

Fluent Regular Expressions for PHP

---

@Changelog

---

0.4.0

• Adds Flux to the Sortiz\Tools namespace
• Implements PSR-2 Compliance (Tabs over Spaces)
• Updates version number on Flux and this readme file
• Updates the class instantiation with fully qualified class name on example.php
• Adds references to other repos that have ported flux
• Addresses concerns outlined in Issue #3

---

0.3.0

• Improves documentation with phone/date examples
• Adds the letters() method
• Renames the numbers() method to digits()
• Adds support for quantifiers for digits()
• Adds ignoreCase() and promotes it above inAnyCase()
• Improves the documented API

Thought hard about changing the name to FluentX any thoughts?

---

0.2.0

• Adds the either( $option1, $option2 [, $option3 ...] ) method to handle OR cases
• Updates the fluent example in this readme file
• Adds the license

---

0.1.0

Initial preview release

---

@Todo

• Add source code comments
• Add support for quantifiers
• Add language methods for more advanced use cases
• Add support for array/array replacements
• Add reference to repos that have ported Flux (*)
• Add license notes (*)
• Add contributing notes
• Add credits

@Examples

/**
 * Build a URL pattern then test w/ match() and do a replace()
 */
$url	= 'http://www.selvinortiz.com';
$flux	= new Sortiz\Tools\Flux();
$flux
	->startOfLine()
	->find('http')
	->maybe('s')
	->then('://')
	->maybe('www.')
	->anythingBut('.')
	->either('.co', '.com')
	->ignoreCase()
	->endOfLine();

// URL Pattern /^(http)(s)?(\:\/\/)(www\.)?([^\.]*)(.co|.com)$/i
echo $flux->match( $url ); // true
echo $flux->replace( 'https://$5$6', $url ); // https://selvinortiz.com

/**
 * Build a US Date pattern then test w/ match() and do a replace()
 */
$date	= 'Monday, Jul 22, 2013';
$flux	= new Sortiz\Tools\Flux();
$flux
	->startOfLine()
	->word()
	->then(', ')
	->letters(3)
	->then(' ')
	->digits(1, 2)
	->then(', ')
	->digits(4)
	->endOfLine();

// Date Pattern /^(\w+)(, )([a-zA-Z]{3})( )(\d{1,2})(, )(\d{4})$/
echo $flux->match( $date ) ? 'matched' : 'unmatched'; // matched
echo $flux->replace( '$3/$5/$7', $date ); // Jul/22/2013

/**
 * Build a US Phone Number pattern then test w/ match() and do a replace()
 */
$phone	= '[phone]';
$flux	= new Sortiz\Tools\Flux();
$flux
	->startOfLine()
	->find('(')
	->digits(3)
	->then(')')
	->maybe(' ')
	->digits(3)
	->anyOf(' -')
	->digits(4)
	->endOfLine();

// Phone Pattern /^(\()(\d{3})(\))( )?(\d{3})([ \-])(\d{4})$/
echo $flux->match( $phone ) ? 'matched' : 'unmatched'; // matched
echo $flux->replace( '$2.$5.$7', $phone ); // 612.424.0013

FLUX API

The flux API was designed to give you a fluent chainable object to build patterns with.

startOfLine()

Adds a beginning of line ^ modifier

endOfLine()

Adds an end of line $ modifier

find( $val ) & then( $val )

Allow you to augment the pattern with a required segment and it escapes regular expression characters

maybe( $val )

Allows you to augment the pattern with an optional segment

any( $val ) & anyOf( $val )

Allow you to create a set of characters to match

anything()

Adds a wild card (.*) segment to the pattern but it does not make dotAll() explicit

anythingBut( $val )

Will match anything but the characters in $val which is opposite of any() and anyOf

word()

Adds (\w+) to the pattern which will match a single word

letters( $min=null, $max=null )

Only matches characters in the alphabet and uses $min and $max to create a quantifier

digits( $mix=null, $max=null )

Only matches digits and uses $min and $max to create a quantifier like word()

range( $from, $to [, $from, $to ...])

Allows you to create a range character class like a-z0-9 by calling range('a', 'z', 0, 9)

orTry()

This is experimental and I don't have the implementation I feel comfortable with... yet!

ignoreCase() & inAnyCase()

Adds the i modifier to the pattern which will allow you to match in a case insensitive manner

dotAll()

Adds the s modifier to the pattern which will allow you to match a new line when using anything()

multiline()

Adds the m modifier to the pattern which will allow you to search across multiple lines

searchOneLine()

Removes the modifier added by multiline() if it was previously called

match( $subject )

Simply takes your $subject in, compares it against the pattern, and returns whether a it matched or not

replace( $replacement, $subject )

You can replace matched segments by using the $x format where x is the (int) position of the matched segment

---

Flux Elsewhere

There is interest in porting Flux to other languages/platforms like NodeJS, Groovy and Java they'll be listed here once available.

• NodeJS by James Brooks

Feedback

This is something that started as a weekend experiment but I would love to take it further so if you have any suggestions, please fire away!

The best way to get in touch with me is via twitter @selvinortiz we'll take if from there :)

MIT License

Flux is released under the MIT license which pretty much means you can do with it as you please and I won't get mad because I'm that nice; )