Intent to Implement and Ship: history.index
Domenic Denicola
Contact emails
dom...@chromium.org (spec), pani...@chromium.org, jap...@chromium.org (implementation)
Spec
https://html.spec.whatwg.org/multipage/history.html#dom-history-index-2
(but, https://html.spec.whatwg.org/multipage/history.html#dom-history-index once we fix a spec tooling bug)
TAG Review: skipped as this is a very minor and natural addition.
Summary
Adds history.index, a counterpart to history.length, giving the index of the current entry in the joint session history.
Motivation
This allows web content to more simply track its location within the history, especially relative to previous states, by just comparing indices.
More motivation can be found in the AMP team's description of the problem and current workarounds in the spec issue.
Interoperability and Compatibility Risk
This feature should be simple to implement interoperably; it is a natural exposure of data that already exists in all browsers. As such, we hope to see it implemented everywhere.
Edge: No signals (bug)
Firefox: No signals (bug)
Safari: Public support (bug)
Web developers: Positive
Ongoing technical constraints
None
Will this feature be supported on all six Blink platforms (Windows, Mac, Linux, Chrome OS, Android, and Android WebView)?
Yes.
Is this feature fully tested by web-platform-tests?
Yes: see the test additions at https://chromium-review.googlesource.com/c/chromium/src/+/596716.
OWP launch tracking bug
Entry on the feature dashboard
https://www.chromestatus.com/feature/5733738818306048
Requesting approval to ship?
Yes
TAMURA, Kent
--
You received this message because you are subscribed to the Google Groups "blink-dev" group.
To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAM0wra-oUzodqiTwn6FqDWiTbQOq6wCT5YAH%2BUQARfGoNzp2Bg%40mail.gmail.com.
Software Engineer, Google
Jochen Eisinger
To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAGH7WqGDXFWPGVFdJopsbT%2ByxENzWMk3p7%3D%2BnCa3b05fn-Qp4A%40mail.gmail.com.
Dimitri Glazkov
To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CALjhuidpLOhvSnOTOiP%2B%2BYU57CQ0QFK844j22g206i1BtS2rVA%40mail.gmail.com.
Boris Zbarsky
> Firefox: No signals (bug
> <https://bugzilla.mozilla.org/show_bug.cgi?id=1392897>)
Reading that bug, the Firefox signal is negative, though that postdates
the intent being sent.
In particular, per that bug indices are not stable at least in the
Firefox implementation; as document with subframes are discarded and
their subframes session histories are removed from the joint session
history, the indexing changes.
The spec around this stuff seems pretty broken to me, for what it's
worth. In particular, it doesn't make it clear whether indices are
stable or not, and is broken either way. If they're unstable, then
.index is impossible to use sanely, as pointed out in the Firefox bug.
If they're stable, then history.back() as specified doesn't actually
work, and neither do all sorts of other things; history.back() was just
the simplest one to construct an example for. I filed
https://github.com/whatwg/html/issues/2985 describing a simple version
of this problem.
Speaking for Mozilla, I don't think .index should be shipped until there
is some indication that it can actually be used sanely somehow. At the
moment, it's not obvious to me how one would do that.
-Boris
Philip Jägenstedt
--
You received this message because you are subscribed to the Google Groups "blink-dev" group.
To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/blink-dev/fede3405-4998-a4f3-7bb6-f500cfebf318%40mit.edu.