Closed dwhly closed 9 years ago
Short checklist here to be done.
I have revised the list above.
After debugging the h/angularjs problem it seems that IE doesn't have the document.baseURI
property, as described here: https://github.com/hypothesis/h/issues/1191
I'll modify this list element to reflect the problem
We now have a fix(workaround) for hypothesis/h#1191
Moving this to in progress now since it appears there are outstanding issues for Webplatform. If that's in error, go ahead and remove it, or downgrade to backlog.
If it's in progress, who's assigned?
Not sure... but I added it because of this IRC convo:
•tilgovi> IE support for robust anchoring may require significant rewriting of our fuzzy anchoring support 10:40 <•tilgovi> if you insist on having IE "work" before then, without fuzzy anchoring, so that IE users can at least make annotations, then it should be possible today 10:40 <•tilgovi> we've not verified that for a while, but it should be possible 10:41 <•tilgovi> i'm looking up the option 10:43 <•tilgovi> crap 10:43 <•tilgovi> we may have lost that ability 10:43 <•tilgovi> I'll have to look into it
I've moved it back to the backlog, and whoever is appropriate can assign it.
Yes, without the fuzzy anchoring our product has to be IE11 compatible. (This was working in the past)
I glanced in our code and couldn't figure out where to turn that off, though. Do we just not load FuzzyTextAnchors plugin in the guest? Can we do that now using our new window.hypothesisConfig
function?
Or is there something more to do? @csillag?
Can someone toss toss me a trace from ie11?
Without that, I can only guy guess.
Try not to load the DTM plugin. Other plugins should react accordingly. 2014.10.09. 22:07 ezt írta ("Randall Leeds" notifications@github.com):
Or is there something more to do? @csillag https://github.com/csillag?
— Reply to this email directly or view it on GitHub https://github.com/hypothesis/vision/issues/24#issuecomment-58569291.
Try not to load the DTM plugin. Other plugins should react accordingly.
If that's the case, then it should be easy, I think. I'll give it a try later.
I was just confused because annotator still calls ._scan()
and expects to have a dom mapper and everything. But I guess there's a dummy / default one when DTM isn't loaded?
On 2014-10-10 00:19, Randall Leeds wrote:
I was just confused because annotator still calls |._scan()| and expects to have a dom mapper and everything. But I guess there's a dummy / default one when DTM isn't loaded?
Exactly.
I'm fairly positive I tested via.hypothes.is with IE11 last week. However, at present, I cannot get the adder to show. This would be great to fix before we send emails next week. Bumping this higher.
@gergely-ujvari did you notice an adder issue previously or is that a regression?
No, I haven't seen this error in the past.
Oh wait.. I have! It means that the d-t-m library is probably throwing an exception.
We are not using the d-t-m with IE. We are supposed to deploy on IE with a specific request parameter, turning off d-t-m.
(I proposed to make that automatic, but the decision was to make this the responsibility of the integrator, so currently it needs to be done manually.)
I can't remember if the last time we discussed this was before the hypothesisConfig
embed hooks. I think I may be okay with having a default hook try to feature detect for IE.
I may be okay with having a default hook try to feature detect for IE.
Great! See PR https://github.com/hypothesis/h/pull/2047
However, we have another, non d-t-m related bug; see https://github.com/hypothesis/h/issues/2048
In both https://ie.dokku.hypothes.is/ (if that's relevant) and the current live product, IE 11 does not display annotations.