992249 - [WebComponents] ::content pseudo does not seems to work in http://www.html5rocks.com/en/tutorials/webcomponents/shadowdom-201/#toc-style-disbtributed-nodes

Status: RESOLVED INVALID

(Core :: DOM: Core & HTML, defect, P3)

Description

[Vivien Nicolas (:vingtetun) (:21) - (NOT reading bugmails, needinfo? please)]

Just looking at some tutos in both Chrome and Firefox and it seems like the ::content pseudo demo at http://www.html5rocks.com/en/tutorials/webcomponents/shadowdom-201/#toc-style-disbtributed-nodes does not work as expected in Firefox.

(This is with dom.webcomponents.enabled = true).

Comment 1

[Pavel Ivanov [:ivanovpavel][:pivanov] UX]

The demo works now with `dom.webcomponents.enabled = true`
but I can't override css rules for webcomponents with `html[dir="rtl"] webcomponent`. Vivien do you have idea?

Comment 2

[Wilson Page [:wilsonpage]]

(In reply to Pavel Ivanov [:ivanovpavel] from comment #1)
> The demo works now with `dom.webcomponents.enabled = true`
> but I can't override css rules for webcomponents with `html[dir="rtl"]
> webcomponent`. Vivien do you have idea?

The demo vingtetun linked to does not appear to work for me. Viewing the 'Live Demo' in Chrome Canary shows how is should look.

Comment 3

[Vivien Nicolas (:vingtetun) (:21) - (NOT reading bugmails, needinfo? please)]

(In reply to Pavel Ivanov [:ivanovpavel] from comment #1)
> The demo works now with `dom.webcomponents.enabled = true`
> but I can't override css rules for webcomponents with `html[dir="rtl"]
> webcomponent`. Vivien do you have idea?

Sorry Pavel I don't know. Can you open a separate bug with a testcase ?

Comment 4

[Stephany Wilkes]

Sorry for the annoying question but...

Is this the same bug to address to ensure we have RTL and bidi language support as needed? For example, in changing what the directions of arrows mean for languages like Arabic, Farsi, Urdu etc? Those are on deck for 2.1 and 2.2 so I'm trying to find the best bug to track that sort of work or support.

Comment 5

[Olli Pettay [:smaug][[e-mail]]]

This is not the same bug. Please file a new one.

Comment 6

[Olli Pettay [:smaug][[e-mail]]]

And when filing a bug, please attach a minimal testcase which can be run on desktop FF, please.

Comment 7

[Pavel Ivanov [:ivanovpavel][:pivanov] UX]

Hey Olli,

you are right, sorry about this. I opened a Bug 1051763 for this.

Comment 8

[William Chen [:wchen]]

Attachment: Implement shadow DOM ::content pseudo-element

In this patch:

- Add code to nsBindingManager::WalkRules to walk additional rules in shadow trees that contain an insertion point where the element or any of its ancestors are distributed because the ::content pseudo-element allows selectors to step out of a shadow root.
- Adjust the AutoAncestorPusher to keep track of ancestors that have been distributed to insertion points.
- Fix specificity calculation for selectors that include a pseudo-element per spec. [1]
- Add a precise exception for the CSS parser to allow ::content to be followed by more selectors. Spec bug filed. [2]
- Adjust SelectorMatchesTree to handle walking into the ::content pseudo-element.

[1] http://www.w3.org/TR/selectors/#specificity
[2] https://www.w3.org/Bugs/Public/show_bug.cgi?id=27884

Comment 9

[Boris Zbarsky [:bzbarsky]]

Looking at that list, this should be multiple patches, which may in fact want review from different people.

Comment 10

[Boris Zbarsky [:bzbarsky]]

Comment on attachment 8553462 [details] [diff] [review]
Implement shadow DOM ::content pseudo-element

And most of those people should probably not be me, but some combination of dbaron and heycam.

Comment 11

[Tyler]

http://stackoverflow.com/a/27629265/2756409 This answer on the subject at Stack Overflow might help give more info on the pseudo-element and its behavior for those who end up taking this bug for review/implementation.

Comment 12

[Philipp Kewisch [:Fallen] [:📆][:🧩]]

I just stumbled upon this today in playing around with web components. Looks like there is a patch but review missing? It would be great to dig this up and get it ready. William, are you still interested in getting this fixed?

Comment 13

[Andrew Overholt [:overholt]]

William's focused on his slots work but once that's landed can get back to this (the patch would need rebasing at a minimum).

Comment 14

[Olli Pettay [:smaug][[e-mail]]]

There is no ::content anymore.

Comment 15

[Tyler]

::content was simply replaced with ::slotted; are we sure this isnt an issue with slotted, too?

Comment 16

[Olli Pettay [:smaug][[e-mail]]]

Not simply. Rather different model. But if there is an issue, a new bug should be filed.