Compose, decompose and analyze regular expressions.

This is an experimental project aimed at converting regular expressions into an object model and back again. An object model gives you the ability to inspect a regular expression as a series of structured objects, allowing you to easily inspect and manipulate them in order to compose and generate new regular expressions. Another advantage is objects can help to prevent invalid regular expressions through strict interfaces and argument checking.

A future aim of this project is to systematically execute each sub-expression against an input string. This would essentially allow you to "debug" a regex to see where it is failing.

Regular expressions are composed of a wide variety of special sequences and characters. While a .NET regex expert might know the names for each of these terms, I do not expect the average developer to be familiar with them. For that reason, I will quickly list each class found in RegexUp along with some examples:

• Literal - a, b, c
• Alternation - a|b|c
• CaptureGroup - (abc), (?<name>abc), (?'name'a)
• CharacterGroup - [abc], [^a-zA-Z0-9]
• Quantifiers - [0-9]+, .*, (abc)?, a{0,5}
• Anchor - ^, $, \b, \G
• Escape - \t, \n, \u0037
• CharacterClass - \w, \s, \p{IsBasicLatin}, \P{Lu}
• Backreference - (abc).*\1, (?<name>abc).*\k<name>
• NonCaptureGroup - (?:abc)
• OptionsGroup - (?im-nx:abc)
• BalancedGroup - ((?'Open'\()[^)]*)+((?'Close-Open')[^()]*)+
• PositiveLookaheadAssertion - a(?=bc)
• NegativeLookaheadAssertion - abc(?!d)
• PositiveLookbehindAssertion - (?<=a)bc
• NegativeLookbehindAssertion - (?<!a)bc
• NonbacktrackingAssertion - a(?>bc)?bc
• InlineOptions - A(?i)bc
• InlineComment - ab(?#The next character to match will be 'c'.)c
• XModeComment - [0-9a-fA-F] # Hexidecimal digit (Only enabled if IgnorePatternWhitespace, a.k.a, x mode, is set)

Regular expressions are a very compact notation. Building the same regular expressions using classes and method calls is significantly more verbose. Please see this example unit test that constructs a regular expression for checking for valid URLs.

Manipulating objects to generate an otherwise static regular expression requires runtime overhead that's unnecessary. As such, one use case for RegexUp is to generate regular expressions in a test project and then simply copy the generated regular expression string to your production code. Another use case is if you need to build your regex at runtime based on user input. Basically, you might consider using RegexUp if you find yourself building up complex regular expressions via string concatenation.

One of the most powerful features of this library is its ability to decompose regular expressions. It includes a small, yet powerful parser that can build an object model from an existing regular expression. Consider this quick example:

// Overly simplistic email validator
var emailRegex = RegularExpression.From(new Regex(@".+@.+\..+"));
// Overly simplistic phone validator
var phoneRegex = RegularExpression.From(new Regex(@"[0-9]{3}\-[0-9]{0,3}[0-9]{4}"));
// Now build a regex combining them
var combined = RegularExpression.Of(
  Alternation.Of(
    emailRegex.AsExpression(),
    phoneRegex.AsExpression()
  )
);

The beauty here is you can write small expressions you can actually comprehend and then stitch them together. This avoids the explosively large code blocks you'd otherwise end up with, like in the Quick Example.

Rather than try to manipulate an existing object model, a cleaner approach is to use the visitor pattern and build up a new model as you go. You can inherit from the ExpressionVisitor class and call the void Visit(IVisitableExpression expression) method to kick off the process. Internally, the same visitor pattern is used to generate the regex from the object model.

As this project is currently only experimental, I have not made it consumable via NuGet. In order to use this code, you must copy the source code into your project. If you have a use for this project and would like to experiment, please let me know and I will publish it to NuGet upon request.

This is free and unencumbered software released into the public domain.

Anyone is free to copy, modify, publish, use, compile, sell, or distribute this software, either in source code form or as a compiled binary, for any purpose, commercial or non-commercial, and by any means.

In jurisdictions that recognize copyright laws, the author or authors of this software dedicate any and all copyright interest in the software to the public domain. We make this dedication for the benefit of the public at large and to the detriment of our heirs and successors. We intend this dedication to be an overt act of relinquishment in perpetuity of all present and future rights to this software under copyright law.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

For more information, please refer to http://unlicense.org