Welcome to the SRP Forum! Please refer to the SRP Forum FAQ post if you have any questions regarding how the forum works.
HTTP Framework Endpoint Toggle
Is there a way in the Framework to toggle off the visibility of a Endpoint (and method within)?
Basically, we are looking at better control of API 'modules' so we may not want all Endpoints or Methods to be available by default. This can obviously be done manually, by (rather than deploying the whole record) but that makes the installation a bit more tedious.
There will obviously be another consideration with regards of updatability once a particular site deviates from the standard but first things first....
Basically, we are looking at better control of API 'modules' so we may not want all Endpoints or Methods to be available by default. This can obviously be done manually, by (rather than deploying the whole record) but that makes the installation a bit more tedious.
There will obviously be another consideration with regards of updatability once a particular site deviates from the standard but first things first....
Comments
Even if we added a feature that toggled on/off the visibility, it would still be an extension to the resource configuration record. It seems like you were hoping to avoid that. What kind of a solution did you have in mind? Also, were you hoping this would be tied to security (i.e., authorization based) or is this still something that would be configuration based (i.e., specific to a given site)?
You are on point as always.
I havent thought too much about a design but I have been asked the question by the higher-ups in the past.
Basically an attempt to monetize some of the unique API endpoints. I like how I can deploy the full endpoint list via the configuration record but then that would indeed overwrite any visibility customisation. Food for thought...
And yes, ultimately I can see arguements for both access scenarios. At the start, I would say site based (i.e. they pay for a special integration like to a particular CRM) but I can see it might quickly go down the path of user level security.
I think I will park this until there is more pressure from above to come up with a solution.
Any high level statstical metrics on endpoint access in that roadmap?
I get we can keep logs but thats a LOT of ever expanding information when something as simple as daterange/endpoint/method/access count (or some variation therein) could be enough...
Once again, nothing concrete in plans yet. Just food for thought at this stage...