This will make window.open a lot easier to implement. As written, the
implementation of Navigable::choose_a_navigable now looks a lot closer
to the old BrowsingContext::choose_a_browsing_context. With the notable
exception that we still crash in many cases, and don't properly handle
multiple top-level traversables in the same WebContent process.
With this change, we now have ~1200 CellAllocators across both LibJS and
LibWeb in a normal WebContent instance.
This gives us a minimum heap size of 4.7 MiB in the scenario where we
only have one cell allocated per type. Of course, in practice there will
be many more of each type, so the effective overhead is quite a bit
smaller than that in practice.
I left a few types unconverted to this mechanism because I got tired of
doing this. :^)
It is a mistake to add this early return in fragment navigation because
"ongoing navigation id" is not used during this kind of navigation.
With this change we no longer crash in Acid2 test after reload button
is pressed.
Window.h is a rather heavy file, so let's try not to include it in
header files when we can!
Element.h now also includes LibWeb/Bindings/Intrinsics.h, but that's
just out of my laziness. Most if not all objects call
`Bindings::ensure_web_prototype<>()` anyway, so I don't think we would
gain much by sticking the header to source files instead.
And remove assorted spec FIXMEs along the way. Also align
populate_session_history_entry_document to the spec, with a bonus spec
bug to be filed.
This involves creating a new NonFetchSchemeNavigationParams spec, and
having the associated AOs take a Variant rather than Optional to
accomodate the fact that this extra struct could be returned by the
algorithm. We don't actually *do* anything with these params, but the
scaffolding is there now, with less TODOs.
If `load_document()` is called with a response that has a mime type we
can't use to build a document, we should return nullptr as the spec
says, instead of crashing. Also we should not crash if error happened
during parsing.
This is not in the spec, but we need to make sure that "apply the
history step" for initial navigation to about:blank in iframe is
applied before subsequent navigations. Otherwise, "set ongoing
navigation" call during "about:blank" traversal might abort subsequent
ongoing navigation which is not expected to happen.
If a navigable has been destroyed during a navigation process, we
should early return from it. The introduced checks are not in
the spec because, as explained in
https://github.com/whatwg/html/issues/9690 the spec is not written
with such a level of detail.
Replaces direct "apply the history step" calls with new functions from
the spec:
- "update for navigable creation/destruction"
- "apply the push/replace history step"
- "apply the reload history step"
Fixes a crash in `get_session_history_entries()` that happens when
it attempts to find entries for a navigable that hasn't been fully
initialized and therefore doesn't have a nested history entry yet.
Callback running on the session history queue should capture necessary
pointers by value instead of reference, because navigate_to_a_fragment
stack will have been destroyed by the time it will be executed.
Instead of having a nested enum within a struct, use the macro
AK_ENUM_BITWISE_OPERATORS to add all the convienent has_flag free
functions and such for ease of use.
This is used by both Navigable and Navigation, so let's put it in
Navigable. Also add a missing AK/String include to make clangd happier
with the Navigable file.
Fixes stack-use-after-return bug found by ASAN that happens when
`response` reference captured by `process_response` is modified
after navigation has been canceled.
