Regexp > Switch from Rubular to regex101

[adriens]

❔ Context

Actually rubular is the default website to provide regexp and permalink. regex101 seems to do the same job but a bit better.

Below example for Quarkus : https://regex101.com/r/NygtZr/1

[captn3m0]

Both are used in the codebase, but as comments. Regex101 is clearly better, but it doesn't support Ruby regexes, which is what our git automation uses.

Our other automation scripts are in python, so I use regex101 comments there: https://github.com/search?q=org%3Aendoflife-date+regex101&type=code

There is one crucial difference in the engines, which actually makes the Ruby one much less painful to use: Ruby lets you re-use named capturing groups.

As an example, if you see the mariadb regex here, it's a single regex that re-uses the capturing groups. Since only one is captured at any time, the regex works correctly, and only a single value is returned for each field. As can be seen here, on Rubular.

However, copying the same regex on regex101 will give an error, since Python marks it as an invalid regex, so it's not possible to document the mariadb regex on R101.

[adriens]

Amazing detailed feedback, thanks a lot !!!

[marcwrobel]

Closing, https://rubular.com/ is not used anymore on endoflife.date.

The ruby scripts in https://github.com/endoflife-date/release-data/ has all been replaced by Python scripts, so now all products are using https://regex101.com/.