Regular expressions have been one of my favorite programming tools since I first discovered them. They are wonderfully robust and things can usually be done with them in many ways. For example, here are multiple ways to match an IPv4 address:
- ^\d\d?\d?\.\d\d?\d?\.\d\d?\d?\.\d\d?\d?$
- ^\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3}$
- ^(\d{1,3}\.){3}\d{1,3}$
- ^([0-9]{1,3}\.){3}[0-9]{1,3}$
One of my major annoyances though has always been lists. I have always done them like ^(REGEX,)*REGEX$.
For example, I would do a list of IP addresses like this: ^(\d{1,3}\.){3}\d{1,3},)*\d{1,3}\.){3}\d{1,3}$.
I recently realized however that a list can much more elegantly be done as follows: ^(REGEX(,|$))+(?<!,)$. I would describe this as working by:
- ^: Start of the statement (test string)
- (REGEX(,|$))+: A list of items separated by either a comma or EOS (end of statement). If we keep this regular expression as not-multi-line (the default), then the EOS can only happen at the end of the statement.
- (?<!,): This is a look-behind assertion saying that the last character before the EOS cannot be a comma. If we didn’t have this, the list could look like this, with a comma at the end: “ITEM,ITEM,ITEM,”.
- $: The end of the statement
So the new version of the IP address list would look like this ^((\d{1,3}\.){3}\d{1,3}(,|$))+(?<!,)$ instead of this ^((\d{1,3}\.){3}\d{1,3},)*(\d{1,3}\.){3}\d{1,3}$.
Also, since an IP address is just a list of numbers separated by periods, it could also look like this: ^(\d{1,3}(\.|$)){4}(?<!\.)$.
