All Apps and Add-ons

Invalid regex when using `|`

boris
Path Finder

I need to capture both "D" OR "=" using (D|=).

Editing the regex of the field extractor this works:

=(?P<FIELDNAME>[a-zA-Z0-9_.]+)[_.]geom

But why does this not work?

(D|=)(?P<FIELDNAME>[a-zA-Z0-9_.]+)[_.]geom

Its returns:

Invalid regex: no named extraction at position 0 (i.e., "(D|=)(?P<F..."). Expected "(?P<variable>pattern)"
1 Solution

boris
Path Finder

Not sure why but the same regex using "(x|y)" that did not work in the Field Extractor works OK when directly done in Search:

index=main sourcetype="mapfluence" | head 10000 | rex "(%3D|=)(?P<FIELDNAME>[^=^(%3D)]+)[_.]geom" | top 50 FIELDNAME

View solution in original post

boris
Path Finder

Not sure why but the same regex using "(x|y)" that did not work in the Field Extractor works OK when directly done in Search:

index=main sourcetype="mapfluence" | head 10000 | rex "(%3D|=)(?P<FIELDNAME>[^=^(%3D)]+)[_.]geom" | top 50 FIELDNAME

dwaddle
SplunkTrust
SplunkTrust

I think it's a bug in the field extractor. It is making an assumption that every regex capturing group has a field name. The rex search command makes no such assumption.

dwaddle
SplunkTrust
SplunkTrust

Sounds like a quirk in the field extractor, assuming that every capturing group has to have a named variable to go with it. A suitable workaround might be:

[|=](?P<FIELDNAME>[a-zA-Z0-9_.]+)[_.]geom
Get Updates on the Splunk Community!

AI for AppInspect

We’re excited to announce two new updates to AppInspect designed to save you time and make the app approval ...

App Platform's 2025 Year in Review: A Year of Innovation, Growth, and Community

As we step into 2026, it’s the perfect moment to reflect on what an extraordinary year 2025 was for the Splunk ...

Operationalizing Entity Risk Score with Enterprise Security 8.3+

Overview Enterprise Security 8.3 introduces a powerful new feature called “Entity Risk Scoring” (ERS) for ...