[rel=preload] Better granularity for preload/speculative load deduplication
Categories
(Core :: DOM: Core & HTML, enhancement, P2)
Tracking
()
People
(Reporter: mayhemer, Unassigned)
References
(Blocks 2 open bugs)
Details
Attachments
(4 files, 1 obsolete file)
Part of Design documentation for rel=preload
Please see the implementation notes for Better deduplication of requests in HTML5 parser.
We want to make speculative request deduplication in the HTML5 parser be more fine grained. From mere URL to the full "key" as defined in the design document above.
Reporter | ||
Comment 1•4 years ago
|
||
I more tend to move the deduplication checks inside the loaders themselves. If left in the parser we would need to build the (expensive) key for a resource load twice: in the parser and in the loader. Moving this inside loaders would also make the code more clear about where the coalescing happens. Right now the loaders depend on the caller (the HTML5 parser) to do the coalescing job. Loaders count on their respective "Preload*" methods being called NO MORE than once for exactly the same resource. This should change.
Reporter | ||
Updated•4 years ago
|
Reporter | ||
Comment 2•4 years ago
|
||
Reporter | ||
Comment 3•4 years ago
|
||
Depends on D73311
Reporter | ||
Comment 4•4 years ago
|
||
Depends on D73312
Reporter | ||
Comment 5•4 years ago
|
||
Depends on D73313
Reporter | ||
Updated•4 years ago
|
Reporter | ||
Updated•4 years ago
|
Updated•2 years ago
|
Comment 6•1 year ago
|
||
Updated•1 year ago
|
Updated•1 year ago
|
Updated•1 year ago
|
Description
•