-
Notifications
You must be signed in to change notification settings - Fork 21
No error when multiple step implementations exist #55
Comments
Gauge should also have this behaviour consistent across languages. |
Are you sure? I didn't study the whole architecture of gauge and runner communication. But as far as I understood: Gauge simply loads the runner, and calls the steps from the specs file, and has no information about the actual stepNames. But maybe I am wrong. |
Gauge does most of the orchestration, but the core isn't aware of implementation (which is language specific). What I meant here is - this problem exists in other language runners as well, we need to address it there as well. The implementation for this would be at language runner level.(In this case, the change would be in gauge-csharp). Then there is also the aspect of IDE plugins being aware of these rules, but we should address command line first IMO. |
This could be treated similar to Unimplemented Steps. The flow would then look like:
This will ensure that Thoughts? |
Update gauge-proto to include duplicate step implementation error. If any language runner sends duplicate step implementation error, Gauge core will recognize it and report it accordingly. Ref getgauge/gauge-java#43 Ref getgauge/gauge-csharp#55 Ref getgauge/gauge-ruby#20
Fix will be available in nightly >= 2016-06-30 |
When I have two step implementations with the same stepText:
... the runner gives no error about ambiguous step definiton. In my case it simply executed DuplicateStep2.
In our case we had 2 step implementations with the same stepText in different assemblies.
Expected behavior:
The text was updated successfully, but these errors were encountered: