How to Use Clock

Date problem

coopy is based on re-execute actions performed in the past. When you call datetime.now() inside an ‘business’ method, when your actions are executed in restore process, datetime.now() will be executed again. This behaviour will produce unexpected results.

Why use Clock?

Clock uses coopy timestamp. When you execute a ‘business’ method, coopy takes the current timestamp and persist inside action object. Clock object has his timestamp updated with action timestamp so in a restore process, Clock will have the original timestamp, and not the timestamp from the re-execution process.

Wrong code:

def create_page(self, wikipage):
    page = None
    wikipage.last_modify = datetime.now()
    ....

Right code:

from coopy import clock
def create_page(self, wikipage):
    page = None
    wikipage.last_modify = self._clock.now()
    ....

Clock API

Clock-aware code validation

coopy has a validation mechanism that will not accept obvious code errors such as calling datetime.now() inside a system method.

API

Take note that a _clock attribute is injected on your system instance and the API is always called via self._clock.

For Clock instances

clock.now()

Return datetime.now()

Return type:datetime

clock.utcnow()

Return datetime.now()

Return type:datetime

clock.today()

Return date.today()

Return type:date