lp://staging/~radix/landscape-client/reactor-refactor-3

Branch merges

Propose for merging

No branches dependent on this one.

Related source package recipes

No recipes using this branch. (?)

Create packaging recipe

Related snap packages

Related bugs

Related blueprints

683. By Christopher Armstrong on 2013-05-16

snapshot. this is WAY too much. :-(

682. By Christopher Armstrong on 2013-05-16

merge from trunk

681. By Christopher Armstrong on 2013-05-15

lint

680. By Christopher Armstrong on 2013-05-15

revert an unrelated part of the remove-reactor refactoring (still needs to happen, but not in this branch)

679. By Christopher Armstrong on 2013-05-15

holy crap... all the tests pass.

678. By Christopher Armstrong on 2013-05-14

- get rid of a pointless/untested KeyboardInterrupt handling clause which
  added a dependency of the event handling code on the reactor
- add an EventSystem class which only provides event logic, not reactor
  interfaces

677. By Christopher Armstrong on 2013-05-14

lint

676. By Christopher Armstrong on 2013-05-14

rename TwistedReactor to LandscapeReactor.

675. By Christopher Armstrong on 2013-05-14

Merge remove-component-publisher-subclasses [f=] [r=tribaal,free.ekanayaka] [a=Christopher Armstrong]
Simplify publishing of AMP objects by doing two things:

- get rid of the necessity of subclassing ComponentPublisher
- allow implementations of brokered objects to use a @remote decorator to declare which methods should be exposed over AMP.

674. By Free Ekanayaka on 2013-05-14

Merge amp-test-readability [f=1170669] [r=tribaal,radix] [a=Free Ekanayaka]
This is the last branch I have around cleaning up the AMP-based machinery in the client. It essentially just change the tests:

- Drop the Words class, which had a bunch of ad-hoc methods used to exercise the behavior of the MethodCall protocol. Essentially each method was associated with a test, but the context was not clear by reading the test itself (you'd have to go and read the implementation in the Words class).

- Modify most methods to define the remotely-exposed method being tested, so the context is clear.

- Partially re-wrote RemoteObjectTest, which had a bunch of tests which were only duplicating tests from MethodCallTest. Now this test class has only tests that are really specific to RemoteObject.