-
Notifications
You must be signed in to change notification settings - Fork 36
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Combobox: Several issues in IE11 #382
Comments
You're again right, I reproduce these three IE11-specific issues on my IE11/Win7. Honestly, on desktop, so far it was tested with latest Chrome and FF, and IE10 on Windows 7, not with IE11. We'll work on fixing it. Thanks again. |
No problem. Delite(ful) is going to make my life much easier one day (once you stop renaming everything :-)) so this is just way to show gratitude for your work! |
:-) (I hope we'll soon stabilize the API, for now we're still at 0.x version numbers...) |
Besides your suggested solution .fill {
-ms-flex-preferred-size: auto;
flex-basis: auto;
} (@dmandrioli, any lights about it?), in my testing it also started working if I give up on embedding the List inside a LinearLayout here: I'll choose the final solution after seeing @dmandrioli feedback on your suggested addition, and after the merge of https://github.com/dmandrioli/deliteful/compare/listdirect-ci.
If combo.autoFilter is
Just to be sure, you only got that when combo.selectionMode is "multiple", right? |
These two css properties are not used in LinearLayout and 'auto' is the default value, so I'm not surprised this is not a solution.
This CSS in Combobox.less You should remove the CSS workaround and remove the LinearLayout wrapper. I continue to work on listdirect branch and will produce a PR when ready (Combobox tested, build running, etc) |
Nope, that happens with the default selectionMode "single" as well. |
Okay, I now understand it better. With my IE 11.0.9600.17358 / Win7, in samples/ComboBox.html of deliteful v. 0.4.0 I only get it on the combo with selectionMode = "multiple", not on the one with "single". But the thing which matters is whether the input element is readonly: it only hurts on readonly inputs, which in this sample is only the case for the "multiple" one. I hope you can confirm that it only hurts with either autoFilter = false or selectionMode = "multiple" (which forces autoFilter at false, which means the input element is readonly). This is important because the workaround I know (prevent-defaulting the backspace keydown event) is only applicable for readonly inputs, otherwise it would forbid from deleting characters on editable inputs. |
Well, I'm not sure why you say it is not a solution. Not that it's necessarily the best solution, but as @mythic2 said, adding these settings to the
Yes, I already said that I found removing the LinearLayout wrapper being a workaround for the issue currently hurting IE11, even without removing the CSS workaround. Now, you say this CSS workaround "is anticipating the merging of #341" - its initial motivation was precisely to bring to the combo a (useful) bit of #341, as long as it isn't merged yet. So to be sure I can safely remove it, I'd need to test for non-regression on all browsers. Not sure why you think I should remove it (as long as #341 or a variation around it isn't merged yet), maybe there's a clear reason that I don't see?
Yes, unfortunately.
Okay, thanks for the info. |
Yep, that seems to be correct: autoFilter = false is the root problem. |
Good, thanks, so I do have an usable workaround in hands. And a clarification: the browser still navigates to the previous page when pressing backspace while the focus is anywhere else on the page (say, on some non-editable text). This has nothing to do with the Combobox widget; if an app wants to prevent it it would apply the same kind of workaround at the level of the page itself. The purpose of the workaround (prevent-defaulting for readonly inputs) at the level of Combo is to avoid the misleading situation where the cursor blinks in a readonly input, thus motivating the user to possibly hit backspace... |
I don't understand, It would have been useful if #341 had been merged? |
What I was saying is that it was initially thought useful (by me, possibly wrongly) as long as #341 is NOT merged yet. Okay, so you do see a logical reason to remove it. I'll go for the removal after testing for non-regression (as I said) on all supported browsers.
Again, what I found is even worse than that: not only it's useless but it's even harmful on IE 11. So, yes, I remove it. That said, I guess it normally should also work with the List inside a LinearLayout, maybe this will be the case once a fix for #341 will be on board. |
…opdown in IE, and prevent-default backspace key events on readonly inputs to avoid unwanted browser navigation to previous page. Refs #382
5620cdc fixes two of the points (invisible dropdown, and backspace issue). |
2 out or 3 points being fixed, I transferred "it is not possible to click on the dropdown arrow, one has to click on the text" to a distinct issue: #397 and I close the present one. |
…opdown in IE, and prevent-default backspace key events on readonly inputs to avoid unwanted browser navigation to previous page. Refs #382
…opdown in IE, and prevent-default backspace key events on readonly inputs to avoid unwanted browser navigation to previous page. Refs ibm-js#382
I know the widget is not ready for prime time yet, here are some experiences when using IE11. Note, that none of those happens while testing in Chrome:
-ms-flex-preferred-size: 0px; flex-basis: 0px
. If I addto my local css file, popup renders correctly in both IE and Chrome.
The text was updated successfully, but these errors were encountered: