Regex pattern to check JIRA ticket numbers before specific delimiter in commit messages

[JackHero77]

I need help with a regex pattern that can validate JIRA ticket IDs in git commit messages. Right now I’m working with this pattern: [A-Z][A-Z_0-9]+-[0-9]

Here’s an example of what my commit messages look like:

PROJ-123 TASK-456 BUG-789:: some description here with MISC-999 inside

The problem is I only want to match the JIRA IDs that appear before the :: separator. So in the example above, I want to capture PROJ-123, TASK-456, and BUG-789 but ignore MISC-999 since it comes after the :: delimiter.

How can I modify my regex to only validate ticket numbers in the first part of the commit message before the double colon?

[Laura219]

use ([A-Z][A-Z_0-9]+-[0-9]+)(?=.*::) to grab all JIRA IDs before ::. you might wanna split the string first tho - it’ll save u headaches if the format gets weird later.

[WhisperingWind]

To match only the part before the :: in commit messages, try ^[A-Z][A-Z_0-9]+-[0-9]+(?=::). This pattern anchors to the start and captures JIRA IDs before the delimiter. Alternatively, you can split the commit message at :: first, then apply your original regex on the first segment. This approach often results in cleaner code and better accommodates various commit formats.

[SkippingLeaf]

You could try a non-greedy match with your pattern: ([A-Z][A-Z_0-9]+-[0-9]+)(?=.*?::) might work. But honestly, I’d just extract everything before the :: first with string operations, then run your original regex on that. Regex-only solutions get messy when commit formats vary between teams or people forget the delimiter. Plus it’s way easier to debug string splitting than complex lookaheads when things break.