Separate out OpenSearch initialisation functions to a separate object
Categories
(Firefox :: Search, task, P2)
Tracking
()
| Tracking | Status | |
|---|---|---|
| firefox80 | --- | fixed |
People
(Reporter: standard8, Assigned: standard8)
References
(Regressed 1 open bug)
Details
Attachments
(2 files)
Spitting a couple of patches out from bug 1637744 as that is getting quite big.
The first step here is to define a separate OpenSearchEngine class and move the OpenSearch related initialisation functions across to it. That's actually quite a bit of code, so I think it is worth it being separate.
Note that until we get some more cache rework done, we won't be able to load engines from the cache as OpenSearchEngine objects, we'll still be using the old SearchEngine. However, as this only affects initialisation routes, we should be good for that. Once we get further down the reload engines / init / cache tidy up, we should be able to load these as the 'correct' object.
Also including here a patch which starts to move things out of getEngineParams - namely the loadPath which doesn't actually need to go through getEngineParams.
|
Assignee | |
Comment 1•4 years ago
|
||
For now this only handles the initialisation routines. Loading from cache still creates a plain SearchEngine class, and as a result updates handling is still in the main SearchEngine class.
Depends on D80498
|
|
Assignee | |
Comment 2•4 years ago
|
||
Depends on D81696
Pushed by mbanner@mozilla.com: https://hg.mozilla.org/integration/autoland/rev/91b79a854e78 Create a separate OpenSearchEngine class to handle OpenSearch engines. r=daleharvey. https://hg.mozilla.org/integration/autoland/rev/fa7ca9bd0ba9 initEngine doesn't need to be in the search engine parameters, and pass the loadPath to the constructor of SearchEngine. r=daleharvey
|
||
Comment 4•4 years ago
|
||
| bugherder | ||
https://hg.mozilla.org/mozilla-central/rev/91b79a854e78
https://hg.mozilla.org/mozilla-central/rev/fa7ca9bd0ba9
Description
•