Policy Control is a kind of simple script used to make flow control easier, especially when your transaction contains asynchronous procedure. This little script can help you construct long and complicated but mostly sequenced transaction policy.
There isn't a context manager for you so you have to implement it by yourself.
The context provided __getstate__ and __setstate__ method so it's very
handy to store it in persistent interface using pickle library.
A policy is such a file having syntax like this, assumed example.policy:
load!: policy_test
assert!: is_authenticated
await: modify_email -> address
apply: send_verf_email(address)
await: verf_email_recv
apply: save_current_user_info(address)
And there is a thing called context to record the state the policy transaction is running in. You have to store it in your own way to ensure the policy transaction to run normally.
When you want to run the policy file anew:
p = policy.load(file("example.policy"))
p.provide("modify_email", "john@example.com")
p.resume()
c = p.context
db.store(pickle.dumps(c))
A few days later finally you discovered a verification email in you mailbox:
p = policy.load(file("example.policy"))
# equivalent: p.load_context(db.retrieve())
p.context = db.retrieve()
p.provide("verf_email_recv")
p.resume()
And that's it. For more information about policy, read test samples in
/tests/policy_test.py.
I don't have many plans. Depending on my mood, maybe I will or will not:
- implement flow control command like
if/for(Under consideration) - implement some operators to complete simple operations (Never)
- make the policy script Turing complete and generic (Never)
- fix bugs (Sure)