Details
Diff Detail
- Repository
- rEFL core/efl
- Branch
- devs/cedric/forwarddoc
- Lint
Lint OK - Unit
No Unit Test Coverage - Build Status
Buildable 8671
Added some inline comments. The whole event forwarding thing seems to be previously-undocumented API, so it is important we put clear docs :)
src/lib/eo/efl_object.eo | ||
---|---|---|
252–264 | It's still not clear to me in which direction the event is forwarded. How about: Things that I am missing:
| |
268 | BEFORE and AFTER what? | |
270 | I cannot find a definition for Efl_Callback_Priority in any EO file. Once they are available, they should be referenced with @ (hyperlink) instead of $ (monospaced font), and using the proper EO syntax, which will probably be @Efl.Callback_Priority. |
I have also added some more text on the new Efl.Callback_Priority type too.
src/lib/eo/efl_object.eo | ||
---|---|---|
270 | I didn't even know that was possible! Nice improvement indeed. |
src/lib/eo/efl_object.eo | ||
---|---|---|
252 | This is still a bit obscure to me... how about: | |
254 | So confusing! So the event is unaffected, but it can be intercepted? How about: $new_obj still emits the original event. Whether this object emits the forwarded event before or after $new_obj can be controlled by the $priority parameter (see below). | |
266 | Wow, I didn't know @cref. What does it do? | |
267 | Do not change the name of the thing! We have been calling it "event forwarder", so keep calling it "event forwarder" (not "redirection handler"). | |
272 | I would call this original_obj, and the comment should be [[The object which emits the initial event]] Change my previous comments accordingly if you call this parameter something other than new_obj. |
src/lib/eo/efl_object.eo | ||
---|---|---|
254 | I am removing completely the reference to event interception as this is confusing and unrelated to forwarder (all events can be blocked from propagating there event further with efl_event_callback_stop). | |
266 | To my understanding @cref is a const pointer to an object that the caller keep control of after the call. I would for example not expect this call to keep a reference somewhere to desc once the call is done. Arguably, this is one of this things where onle @q66 and @felipealmeida will be able to give the exact definition. It also imply @in as const pointer can not be modified and so can't be an out. |