You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently, per spec, if an config-generating API generates a FencedFrameConfig, and the embedder of a fenced frame or even the fenced frame itself "guesses" the urn:uuid backing it, it could navigate the fenced frame to that config.
This doesn't match implementations, and is bad because the intention of a FencedFrameConfig IDL object is to allow only the embedder to navigate the fenced frame with the config object. We should modify the navigation algorithm monkeypatches to be able to accept a FencedFrameConfig object, and only check the fenced frame config mapping if a FencedFrameConfig is supplied to the algorithm, as opposed to if a urn is supplied.
domfarolino
changed the title
[Spec] Have navigate patch be able to take in a FencedFrameConfig IDL object.
[Spec] Have navigate algorithm be able to take in a FencedFrameConfig IDL object.
Oct 4, 2024
Currently, per spec, if an config-generating API generates a FencedFrameConfig, and the embedder of a fenced frame or even the fenced frame itself "guesses" the
urn:uuid
backing it, it could navigate the fenced frame to that config.This doesn't match implementations, and is bad because the intention of a FencedFrameConfig IDL object is to allow only the embedder to navigate the fenced frame with the config object. We should modify the navigation algorithm monkeypatches to be able to accept a FencedFrameConfig object, and only check the fenced frame config mapping if a FencedFrameConfig is supplied to the algorithm, as opposed to if a urn is supplied.
See this comment thread: #183 (comment)
The text was updated successfully, but these errors were encountered: