Some Eclipse Foundation services are deprecated, or will be soon. Please ensure you've read this important communication.
Bug 334197 - [client] Clean up story for loading/initializing the service registry
Summary: [client] Clean up story for loading/initializing the service registry
Status: RESOLVED FIXED
Alias: None
Product: Orion
Classification: ECD
Component: Client (show other bugs)
Version: 0.2   Edit
Hardware: PC Windows 7
: P3 normal (vote)
Target Milestone: 0.3 RC3   Edit
Assignee: Simon Kaegi CLA
QA Contact:
URL:
Whiteboard:
Keywords:
: 334218 (view as bug list)
Depends on:
Blocks: 334198 337740
  Show dependency tree
 
Reported: 2011-01-12 22:11 EST by Susan McCourt CLA
Modified: 2011-10-17 18:42 EDT (History)
4 users (show)

See Also:


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Susan McCourt CLA 2011-01-12 22:11:54 EST
We have a bunch of copied code right now across the various pages' "glue" code that initializes the service registry and defines the services.  

We need to figure out how this should really work.
Comment 1 Simon Kaegi CLA 2011-01-12 22:26:15 EST
Sorry Susan, I feel responsible for not cleaning up this mess. I want to target end of next week to sort this stuff out better.
Comment 2 John Arthorne CLA 2011-03-07 11:01:49 EST
Moving bugs targeting past milestones...
Comment 3 Susan McCourt CLA 2011-03-30 19:42:47 EDT
This is a slight tangent, but is something that should be done as part of cleaning up our page load story.

From http://dev.eclipse.org/mhonarc/lists/orion-dev/msg00366.html
JJB says:
>My embedding of coding.html suffers from timing problems.
>I wire up to the page on web page "load", but I guess that coding.html 
>is using 'load' to configure the page. If I get called first I lose.

We need to have a page load sequence so that we know we do stuff like:
1 - initialize the registry and services
2 - load the stuff specific to the page
3 - signal that we are done doing all of our work so that an embedding page can know everything is there

I'm not sure that #1 even requires the DOM to be loaded.
I know there is plenty of garden-variety stuff in #2 that needs the DOM loaded, so we probably do need to define some common event for the end of #2.
Comment 4 John Arthorne CLA 2011-04-27 17:31:32 EDT
*** Bug 334218 has been marked as a duplicate of this bug. ***
Comment 5 Simon Kaegi CLA 2011-04-28 11:54:34 EDT
We cover the basic first time load case now. For m8 I'd like to use this bug to further formalize the approach.
Comment 6 Simon Kaegi CLA 2011-10-17 18:42:47 EDT
Fixed.

We've moved all the plugin load / service registry initialize code to bootstrap.js and fixed inconsistencies so now all pages will load properly the first time and pick up changes correctly.

The bootstrap still needs some work around preferences initialization but that is captured in another bug.

--
Mark reviewed and tested this change with me.