Hi, thanks again for the chi library. It simplifies things alot relative to the raw underlying openstack python commands I used a few years ago, but one issue I hit was chi.lease.create_lease does not document what happens when the requested resources are exhausted. Trying this function in an error condition, it returns None and prints a log message. While this is a decent C error reporting mechanism; and I do appreciate the improved error message relative the blazarclient message that is caught below, it is kind of an unexpected way to indicate an error in Python where exceptions are generally preferred (c.f. "Errors should never pass silently" - The Zen of Python). Perhaps instead of logging and returning None, you could perhaps instead use
raise RuntimeError(what_you_log_right_now) from blazar_exception
If you don't want to do this, please at least document the error handling for this routine.
Hi, thanks again for the chi library. It simplifies things alot relative to the raw underlying openstack python commands I used a few years ago, but one issue I hit was chi.lease.create_lease does not document what happens when the requested resources are exhausted. Trying this function in an error condition, it returns None and prints a log message. While this is a decent C error reporting mechanism; and I do appreciate the improved error message relative the blazarclient message that is caught below, it is kind of an unexpected way to indicate an error in Python where exceptions are generally preferred (c.f. "Errors should never pass silently" - The Zen of Python). Perhaps instead of logging and returning None, you could perhaps instead use
If you don't want to do this, please at least document the error handling for this routine.