Parcourir la source

Resolved runeh's FIXME's

Ask Solem il y a 15 ans
Parent
commit
88dcdc69aa

+ 3 - 14
docs/getting-started/first-steps-with-only-celery.rst

@@ -45,18 +45,7 @@ default name.
 
 Let's create our ``celeryconfig.py``.
 
-FIXME: Is the invocation below something people are expected to do,
-appending cwd to sys.path? It seems like something that would usually
-be handled elsewhere?
-
-1. Start by making sure Python is able to import modules from the current
-   directory::
-
-        import os
-        import sys
-        sys.path.insert(0, os.getcwd())
-
-2. Configure how we communicate with the broker::
+1. Configure how we communicate with the broker::
 
         BROKER_HOST = "localhost"
         BROKER_PORT = 5672
@@ -64,12 +53,12 @@ be handled elsewhere?
         BROKER_PASSWORD = "mypassword"
         BROKER_VHOST = "myvhost"
 
-3. In this example we don't want to store the results of the tasks, so
+2. In this example we don't want to store the results of the tasks, so
    we'll use the simplest backend available; the AMQP backend::
 
         CELERY_BACKEND = "amqp"
 
-4. Finally, we list the modules to import, that is, all the modules
+3. Finally, we list the modules to import, that is, all the modules
    that contain tasks. This is so celery knows about what tasks it can
    be asked to perform. We only have a single task module,
    ``tasks.py``, which we added earlier::

+ 6 - 4
docs/userguide/executing.rst

@@ -75,6 +75,12 @@ task. There is built-in support for using ``pickle``, ``JSON`` and ``YAML``,
 and you can add your own custom serializers by registering them into the
 carrot serializer registry.
 
+The default serializer (pickle) supports Python objects, like ``datetime`` and
+any custom datatypes you define yourself. But since pickle has poor support
+outside of the Python language, you need to choose another serializer if you
+need to communicate with other languages. In that case, ``JSON`` is a very
+popular choice.
+
 The serialization method is sent with the message, so the worker knows how to
 deserialize any task. Of course, if you use a custom serializer, this must
 also be registered in the worker.
@@ -88,10 +94,6 @@ configuration directive.
 
     >>> add.apply_async(args=[10, 10], serializer="json")
 
-FIXME: in what cases does it make sense to use a different
-serializers? Same something about when you need to care about this.
-
-
 Connections and connection timeouts.
 ------------------------------------
 

+ 1 - 3
docs/userguide/remote-tasks.rst

@@ -19,9 +19,7 @@ or using POST::
 
     POST http://example.com/mytask
 
-FIXME: what is the content type? post as formencoded or as text/plain
-with json ? support both?
-
+**Note:** POST data has to be form encoded.
 Whether to use GET or POST is up to you and your requirements.
 
 The web page should then return a response in the following format