Ticket #12536 (closed bug: fixed)
Make .offset() calc less wrong on browsers w/o getBoundingClientRect
| Reported by: | gabriel.schulhof@… | Owned by: | |
|---|---|---|---|
| Priority: | blocker | Milestone: | 1.8.2 |
| Component: | offset | Version: | git |
| Keywords: | Cc: | ||
| Blocking: | Blocked by: |
Description
We had agreed that the fix was not for the whole offset function to return { left: 0, top: 0 }, but to have something like
if ( typeof elem.getBoundingClientRect === "undefined" ) {
box = { left: 0, top: 0 };
} else {
box = elem.getBoundingClientRect();
}
... and then proceed to the calculation below, which will give us an offset that may be incorrect, but it's closer than 0, 0, because it takes into account the scrollLeft and scrollTop.
This is related to ticket #12534 ...
Change History
comment:2 Changed 8 months ago by dmethvin
- Priority changed from undecided to blocker
- Status changed from new to open
- Summary changed from Please re-open ticket #12534 to Make .offset() calc less wrong on browsers w/o getBoundingClientRect
- Component changed from unfiled to offset
- Milestone changed from None to 1.8.2
Sounds reasonable, I didn't get that from the IRC discussion yesterday. Before I make further changes though, let's discuss.
comment:3 Changed 8 months ago by Dave Methvin
- Status changed from open to closed
- Resolution set to fixed
Fix #12536. Start at .offset() 0,0 if no getBoundingClientRect.
This lets us still add the other offset components to the number so they're less wrong. Affects BlackBerry 5 and iOS 3, everyone else has gBCR.
Changeset: 560c178c82da95b2f88ae518552463d87fe5adbf
Please follow the bug reporting guidlines and use jsFiddle when providing test cases and demonstrations instead of pasting the code in the ticket.
I've also noticed that on BB5 elements have properties called clientLeft, clientTop, clientWidth, and clientHeight ... I think these could be used as well.