CODEWITHSUNDEEP
phpregex
dabear3021
dabear3021

Reputation: 3

regex that handles a decimal point if it exists

I am using a php regular expression to try and solve an advanced text search query in mysql. The query needs to group smaller things to make them work properly. Currently I can handle things like "foo 7S" and group them together if I see the word "foo".

Currently, my regex looks as such

/(foo|bar|baz faz)\s+([a-zA-Z0-9]+)\b/i

However, if I am trying to search for something specific like "foo 7.10", then the ".10" is split out instead of being combined. My current regular expression is

/(foo|bar|baz faz)\s+([a-zA-Z0-9]+)(\.[a-zA-Z0-9]+)\b/i

This works for "foo 7.10" but doesn't match anything for "foo 7" or "foo 7S". Any help would be appreciated.

Upvotes: 0

Views: 89

Answers (1)

CertainPerformance
CertainPerformance

Reputation: 370729

You just need to make the group that captures the decimal part optional:

(foo|bar|baz faz)\s+([a-zA-Z0-9]+)(\.[a-zA-Z0-9]+)?\b
                                                  ^

https://regex101.com/r/QBFeuP/1

Note that since you're using the i flag, the regex is case-insensitive, so there's no need to use a-zA-z - those can be reduced to just a-z:

(foo|bar|baz faz)\s+([a-z0-9]+)(\.[a-z0-9]+)?\b

and if there is no chance of underscores being there, you can use \w instead:

(foo|bar|baz faz)\s+(\w+)(\.\w+)?\b

in which case the final \b is unnecessary, because both of the \w+s are greedy and will match as many characters as they can, which implies that the following position will necessarily be a word boundary.

You also might consider putting a word boundary at the very beginning, if it would be helpful:

\b(foo|bar|baz faz)\s+(\w+)(\.\w+)?

Upvotes: 2

Related Questions