Track all feature flag calls #81
Conversation
If an analytics processor exists we previouly only tracked feature flag names that were present either from the environment or from the API. However, this gave us no information about calls to flags that either were wrong, or were handled by the default handler. With this change we now track all calls for more extensive coverage.
|
Hey @JonasAlfredsson, this is interesting. I can definitely understand the use case, however, I'm not sure how useful it would be right now since there is no functionality for surfacing this data, right? What was your intention here? |
|
My intention is to make tracking more accurate by also recording every time we call for a feature flag that is subsequently handled by the default handler instead of only when we get them via the API/Environment. If there is no way of seeing the calls to "invalid" feature names, then perhaps that is another feature request that should be placed on the server issue tracker (if there is such a one)?. |
|
Yep, for sure. We track all of our server feature requests / issues here: https://github.com/Flagsmith/flagsmith/issues |
|
Added a feature request there, where it could be discussed at a higher level if desired. |
If an analytics processor exists we previouly only tracked feature flag names that were present either from the environment or from the API. However, this gave us no information about calls to flags that either were wrong, or were handled by the default handler.
With this change we now track all calls for more extensive coverage.