Repository navigation
[css-display-3] Clarify whether display: contents applies to ::before and ::after #1345
Description
I have a (probably bitrotted already) implementation for this in Blink, but I talked about this with @dbaron a while ago and asked me to file this in order to clarify it, given Gecko doesn't implement it.
Gecko always treats display: contents in ::before and ::after as display: inline. It's not clear to me whether display: contents in pseudo-elements is really useful at all, and the implementation is complex.
It's pretty much non-observable, except for cases like https://github.com/w3c/web-platform-tests/blob/master/css/css-display-3/display-contents-before-after-002.html (where you specify reset properties on the pseudo, like border).
I think it's nice for consistency, but I wouldn't be too sad if it didn't apply to these pseudo-elements.
Activity
Loirooriol commentedon May 10, 2017
It's not much clear, but browsers seem to believe that
display: contentsshould join different texts into a single run of text (#1281). Then, in the following example, "FooBar" would be a single flex item:So in this case whether
::beforeand::aftersupportdisplay: contentsor not is observable by other means than box-related properties on the pseudo-elements.I think
display: contentsshould apply to these pseudo-elements.tabatkins commentedon May 10, 2017
As @Loirooriol points out, it definitely has an observable effect, and is straightforward to define. Basically, the following two bits of markup:
and
...should be completely indistinguishable as far as CSS is concerned. There is no reason for the two to diverge in any way.
That said, I also wouldn't be too sad if we defined that it didn't work. It's just inconsistent and breaks a constant that has nicely applied to ::before and ::after previously.
css-meeting-bot commentedon May 24, 2017
The CSS Working Group just discussed
https://github.com/w3c/csswg-drafts/issues/1345, and agreed to the following resolutions:RESOLVED: display:contents applies to ::after and ::beforeThe full IRC log of that discussion
<myles> TOPIC: https://github.com//issues/1345<myles> Github topic: https://github.com//issues/1345
<myles> s/write/right/
<myles> fantasai: <restates the topic>
<myles> fantasai: TabAtkins and i are leaning toward "yes" but we aren't particularly resolute
<myles> fantasai: any opinions/
<myles> fantasai: ?
<myles> Rossen: it is unclear, if we say "no," what is observable?
<myles> fantasai: you can put a border on ::after, and the border will disappear if you say display:contents
<myles> fantasai: because the box will disappear
<myles> Rossen: was this just an oversight?
<myles> fantasai: just needed clarification
<myles> Rossen: this works for us
<myles> Rossen: objections?
<myles> RESOLVED: display:contents applies to ::after and ::before
tabatkins commentedon Jul 5, 2017
Closing as no change, because the WG decided to stick with it applying, and the effects of that just fall out of the definition.