Bug 1531289 Comment 7 Edit History

Note: The actual edited comment in the bug view page will always show the original commenter’s name and original timestamp.

OK, so there are at least two issues with the noopener codepath:

1) https://searchfox.org/mozilla-central/rev/92d11a33250a8e368c8ca3e962e15ca67117f765/docshell/base/nsDocShell.cpp#8677-8709 does not set the INTERNAL_LOAD_FLAGS_FIRST_LOAD flag on the loadState.  It should, maybe.

2) Even if I fix that, the flag gets lost when we get to the SendCreateWindowInDifferentProcess call at https://searchfox.org/mozilla-central/rev/3e0f1d95fcf8832413457e3bec802113bdd1f8e8/dom/ipc/ContentChild.cpp#878 because that doesn't get passed either the loadstate or any info about its flags.

Kyle, I guess we could pass the loadstate through SendCreateWindowInDifferentProcess instead of deconstructing it into a bunch of separate args, and then use it on the other side?  But also, it seems like all the code in ContentParent::CommonCreateWindow (which is where RecvCreateWindowInDifferentProcess lands) should be terminating in loads that set NTERNAL_LOAD_FLAGS_FIRST_LOAD somehow...  If true, we should just fix _that_.
OK, so there are at least two issues with the noopener codepath:

1) https://searchfox.org/mozilla-central/rev/92d11a33250a8e368c8ca3e962e15ca67117f765/docshell/base/nsDocShell.cpp#8677-8709 does not set the INTERNAL_LOAD_FLAGS_FIRST_LOAD flag on the loadState.  It should, maybe.

2) Even if I fix that, the flag gets lost when we get to the SendCreateWindowInDifferentProcess call at https://searchfox.org/mozilla-central/rev/3e0f1d95fcf8832413457e3bec802113bdd1f8e8/dom/ipc/ContentChild.cpp#878 because that doesn't get passed either the loadstate or any info about its flags.

Kyle, I guess we could pass the loadstate through SendCreateWindowInDifferentProcess instead of deconstructing it into a bunch of separate args, and then use it on the other side?  But also, it seems like all the code in ContentParent::CommonCreateWindow (which is where RecvCreateWindowInDifferentProcess lands) should be terminating in loads that set INTERNAL_LOAD_FLAGS_FIRST_LOAD somehow...  If true, we should just fix _that_.

Back to Bug 1531289 Comment 7