Every mock VWS response hardcodes:
x-aws-region: us-east-2, us-west-2
There are 39 occurrences across src/, covering the success responses of each endpoint and every ValidatorError in src/mock_vws/_services_validators/exceptions.py.
Real Vuforia returned a different value on every response observed while probing the reco counts endpoint on 2026-08-08, across success responses, 400s and 401s:
x-aws-region: us-west-2, eu-west-1
This is not specific to the reco counts endpoint. It was the value on every response in that session, so the mock's value appears to be either stale or account-specific or region-specific.
Why it may not be worth changing
The header is informational and no caller should branch on it. Replacing one hardcoded pair with another only moves the guess, and the observation above is a single session against a single account, so us-west-2, eu-west-1 may be no more universally right than the current value. It is plausible that the value varies by the account's region or by which Vuforia deployment serves the request, in which case no fixed value is correct.
Suggested resolution
Pick one:
- Leave it, and note in
differences-to-vws.rst that the value is a fixed sample which real Vuforia does not always match. Cheapest, and honest. - Update it to the observed value, which is right for at least one account today and wrong in the same way as before for others.
- Make it configurable, which is almost certainly more machinery than an informational header deserves.
Option 1 is likely correct unless someone has evidence about what makes the value vary. The point of this issue is that the difference is recorded rather than that it is necessarily fixed.
Every mock VWS response hardcodes:
There are 39 occurrences across
src/, covering the success responses of each endpoint and everyValidatorErrorinsrc/mock_vws/_services_validators/exceptions.py.Real Vuforia returned a different value on every response observed while probing the reco counts endpoint on 2026-08-08, across success responses, 400s and 401s:
This is not specific to the reco counts endpoint. It was the value on every response in that session, so the mock's value appears to be either stale or account-specific or region-specific.
Why it may not be worth changing
The header is informational and no caller should branch on it. Replacing one hardcoded pair with another only moves the guess, and the observation above is a single session against a single account, so
us-west-2, eu-west-1may be no more universally right than the current value. It is plausible that the value varies by the account's region or by which Vuforia deployment serves the request, in which case no fixed value is correct.Suggested resolution
Pick one:
differences-to-vws.rstthat the value is a fixed sample which real Vuforia does not always match. Cheapest, and honest.Option 1 is likely correct unless someone has evidence about what makes the value vary. The point of this issue is that the difference is recorded rather than that it is necessarily fixed.