What Next?

You might be wondering what you can do with the finished SRL expression, since Grep and most other tools only digest conventional regular expressions.

If you just need a regular expression on the fly, you can always enter the SRL expression into the test at the SRL website (as described earlier in this article), and then copy the resulting regular expression to Grep or another tool.

Also, some languages have already begun to implement SRL support. You will find special SRL libraries for JavaScript, PHP, Python, and C++ at GitHub under the MIT license [3]. Java and C# libraries are still pending.

The functions and classes of these SRL libraries accept an SRL expression, evaluate it, and convert it into a regular expression. In the case of PHP, the user just has to create an SRL object and check the text using the method isMatching():

$srl = new SRL('one of "eE" literally "rror"');$srl?isMatching('Error'); // is True

In addition to featured keywords like literally, SRL also offers the keywords shown in Tables 1 to 6. You'll find a detailed reference and many other examples on the official SRL homepage [1].

Conclusions

SRL is built on the philosophy that, if regular expressions are easier to read, errors will stand out more quickly. It is worth noting, however, that SRL expressions are also complex and difficult to understand if you aren't accustomed to the syntax. SRL does not currently feature comments, which would help to add clarity, and recursion is also missing. Classic Unix text tools such as Grep do not yet provide SRL support; however, you can convert the expression at the SRL website or use an SRL library with some programming languages.

Despite the problems, SRL is still worth a look. If you only use regular expressions occasionally, or if you are using them for the first time, you will get to your destination much faster with SRL. Even regex old-timers might find they can make their expressions more legible with SRL.

In the future, developer Karim Geiger wants to add support for additional programming languages and also standardize the SRL language to define the syntax and commands more clearly. In the long run, he imagines a kind of compiler that translates regular expressions into SRL. He has stated that a Bash version is not planned but is conceivable.

SRL commands are based on English. Geiger has resisted the suggestion to translate the natural language commands to other written languages. He fears that language proliferation would introduce a complexity that could lead to incompatible versions.