Timeout Implementation without schema values and entry point values

Commencez par les exigences du contrat et les lignes 511-522 référencées de src/rpdk/core/project.py, puis examinez les classes générées et les modèles du framework pour trouver les données relatives aux timeouts et aux temps de requête. L'issue est terminée lorsque les valeurs requises et le comportement du timeout sont clairement spécifiés et implémentés de manière cohérente pour les points d'entrée des handlers.

Looking at the contract as written, how do I guarantee the following lines?(see quoted below). From what I can tell, no plugin(this one included) contains the original start time of the resource call. Is it my handler's problem to track original call time in the callback context? Shouldnt this be included on the generated classes or within the framework Models? I can then just interrogate and respond with error, or even better let framework respond with error... Right now as written in contract - we dont know when to return since we do not have the start time and the resource schema of timeoutinMinutes is not exposed to us via any generated classes, correct? I see it is used to set maximum time on assuming the role only. That doesnt allow us to honor contract for less than 60 minutes for a single request: see here

As I see it we need at least 2 values to make the implementation somewhat generic:

• Schema TimeoutInMinutes time needs to be exposed to internal AWS python plugin. Also you need to expose original request time as well. Both of these together will allow you to expose a real variable handlerEndTime that is then read/used by customer handlers to exit reinvoke logic with a timeout error. This is the lates time that a handler can run based on its own schema.
• Handler current entry time - this gets reset for every handler native call - we can use it to make sure private handlers return in correct time frame of 60 seconds or 30 seconds based on the different Action types.

Contract Lines

A create handler MUST return FAILED progress event if it can't reach the desired-state within the timeout specified in the resource schema.

An update handler MUST return FAILED progress event if it can't reach the desired-state within the timeout specified in the resource schema.

A delete handler MUST return FAILED progress event if it can't reach the desired-state within the timeout specified in the resource schema.

Originally posted by @carpnick in https://github.com/aws-cloudformation/cloudformation-cli-python-plugin/discussions/243