Source code for jina.peapods.runtimes.base

import argparse

from ...logging import JinaLogger


[docs]class BaseRuntime: """A Jina Runtime is a procedure that blocks the main process once running (i.e. :meth:`run_forever`), therefore must be put into a separated thread/process. Any program/library/package/module that blocks the main process, can be formulated into a :class:`BaseRuntime` class and then be used in :class:`BasePea`. In the sequel, we call the main process/thread as ``M``, the process/thread blocked :class:`Runtime` as ``S``. In Jina, a :class:`BasePea` object is used to manage a :class:`Runtime` object's lifecycle. A :class:`BasePea` is a subclass of :class:`multiprocessing.Process` or :class:`threading.Thread`, it starts from ``M`` and once the ``S`` is spawned, it calls :class:`Runtime` methods in the following order: 0. :meth:`__init__` in ``M`` 1. :meth:`setup` in ``S`` 2. :meth:`run_forever` in ``S``. Note that this will block ``S``, step 3 won't be reached until it is unblocked by :meth:`cancel` 3. :meth:`teardown` in ``S``. Note that ``S`` is blocked by :meth:`run_forever`, this step won't be reached until step 2 is unblocked by :meth:`cancel` The :meth:`setup` and :meth:`teardown` pair together, which defines instructions that will be executed before and after. In subclasses, they are optional. The :meth:`run_forever` and :meth:`cancel` pair together, which introduces blocking to ``S`` and then unblocking from it. They are mandatory for all subclasses. Note that, there is no "exclusive" relation between :meth:`run_forever` and :meth:`teardown`, :meth:`teardown` is not about "cancelling", it is about "cleaning". Unlike other three methods that get invoked inside ``S``, the :meth:`cancel` is invoked in ``M`` to unblock ``S``. Therefore, :meth:`cancel` usually requires some special communication between ``M`` and ``S``, e.g. - Use :class:`threading.Event` or `multiprocessing.Event`, while :meth:`run_forever` polls for this event - Use ZMQ to send a message, while :meth:`run_forever` polls for this message - Use HTTP/REST to send a request, while :meth:`run_forever` listens to this request Note, another way to jump out from :meth:`run_forever` is raise exceptions from it. This will immediately move to :meth:`teardown`. .. note:: Rule of thumb on exception handling: if you are not sure if you should handle exception inside :meth:`run_forever`, :meth:`cancel`, :meth:`setup`, :meth:`teardown`, then DO NOT catch exception in them. Exception is MUCH better handled by :class:`BasePea`. .. seealso:: :class:`BasePea` for managing a :class:`Runtime` object's lifecycle. """
[docs] def run_forever(self): """Running the blocking procedure inside ``S``. Note, once this method is called, ``S`` is blocked. .. note:: If this method raises any exception, :meth:`teardown` will be called. .. seealso:: :meth:`cancel` for cancelling the forever loop. """ raise NotImplementedError
[docs] def cancel(self): """Cancelling :meth:`run_forever` from ``M``. :meth:`cancel` usually requires some special communication between ``M`` and ``S``, e.g. - Use :class:`threading.Event` or `multiprocessing.Event`, while :meth:`run_forever` polls for this event - Use ZMQ to send a message, while :meth:`run_forever` polls for this message - Use HTTP/REST to send a request, while :meth:`run_forever` listens to this request .. seealso:: :meth:`run_forever` for blocking the process/thread. """ raise NotImplementedError
[docs] def setup(self): """Method called to prepare the runtime inside ``S``. Optional in subclasses. The default implementation does nothing. .. note:: If this method raises any exception, then :meth:`run_forever` and :meth:`teardown` won't be called. .. note:: Unlike :meth:`__init__` called in ``M``, :meth:`setup` is called inside ``S``. """ pass
[docs] def teardown(self): """Method called immediately after :meth:`run_forever` is unblocked. You can tidy up things here. Optional in subclasses. The default implementation does nothing. .. note:: This method will only be called if the :meth:`setup` succeeds. """ self.logger.close()
def __init__(self, args: 'argparse.Namespace'): super().__init__() self.args = args if args.name: self.name = f'{args.name}/{self.__class__.__name__}' else: self.name = self.__class__.__name__ self.logger = JinaLogger(self.name, **vars(self.args))