| Age | Commit message (Collapse) | Author |
|---|
|
|
|
|
|
|
|
|
|
|
|
This is essentially the best we can get without taking advantage
of literally brand new navigation APIs, if we don't want to get
reeeeeeeeeeally futzy.
This approach is strictly oriented around the back/forward cache
and does not store anything in history state (replaceState etc).
|
|
Alas, it goes completely unused.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
and use domEvent more everywhere
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Discussion at #code-quarantine
https://discord.com/channels/749042497610842152/854020929113423924/1356984473396187364
|
|
|
|
|
|
|
|
|
|
|
|
Sigh
|
|
https://discord.com/channels/749042497610842152/854020929113423924/1358446868811485426
Not too smart for our own good, though.
https://discord.com/channels/749042497610842152/854020929113423924/1358454755566944286
|
|
Since we're checking for [inert] ancestor anyway, we don't need
to depend on BS "first matched by querySelector()" logic anymore
|
|
...rather than a static mock, which wasn't *quite* matching
the real deal's world wrap every time
|
|
Not sure how we missed this!!
|
|
During hit detection. We fudge a typical height for the
sticky subheading "instead", although it's not exactly clear
the adjusting function of this +20 (now +40) in the first
place.
|
