[css-text] Percentage of 'text-indent' against intrinsically sized box?

[kojiishi]

Percentage of text-indent defines:

Gives the amount of the indent as a percentage of the containing block’s logical width.

A discussion arose what to do when the containing block’s logical width is intrinsically sized.

Intrinsic sizing in CSS Sizing defer the definition to CSS2.

text-indent in CSS2 does not define either, but width in CSS2 defines it's undefined.

Should text-indent follow the same definition as width?

[FremyCompany]

https://wptest.center/#/0jcvq1

Chrome and Edge do match in both cases.

• Use parent size in min-content if available
• Use parent size in actual layout if available (might results in rewrap/overflows)

Firefox doesn't seem to ever consider the parent in min-content.

I'd argue we should spec what Edge+Chrome are doing, if we want to spec something?

[css-meeting-bot]

The Working Group just discussed intrinsically-sized box with text-indent.

The full IRC log of that discussion

<fantasai> Topic: intrinsically-sized box with text-indent
<fantasai> https://drafts.csswg.org/css-text-3/issues-lc-2013#issue-156
<fantasai> https://github.com//issues/1597
<dael> github: https://github.com//issues/1597
<dael> fantasai: What impact does % have on intrinisic sizing?
<dael> myles: If there's no web compat risk can we use a model for what happens in other situations?
<dael> fantasai: Treat as 0
<dael> florian: Risk is if you want to treat it as -999% to get it out.
<dael> fantasai: You resolve, but for computed size calculation we treat as 0.
<dael> [everyone looks at test case]
<dael> fantasai: It treats it as 0.
<dael> myles: Everyone does 0?
<dael> astearns: No, Chrome and Edge are using parent size according to fremy comment.
<dael> florian: Going to the parent?
<dael> myles: Parent can also be shrink wrapped.
<dael> fremy: If available is if you're not sizing hte parent as min-content.
<dael> myles: Else it's 0?
<dael> fremy: Yes.
<dael> myles: WE should just always 0. This is just a corner case.
<dael> fantasai: If we have 2 impl we might as well align. I'm not sure what this comment means.
<dael> fantasai: Chrome and edge use parent size in min-content if available.
<dael> florian: if parent has definite size work from that if not 0?
<dael> fantasai: We're resolving against element itself.
<dael> fremy: New version of edge is completely different. We use 0 all the time.
<dael> fantasai: We'll cover this one on a telecon.

[frivoal]

The last line in the previous minutes is: "We'll cover this one on a telecon.". We seemed closed to a resolution, so let's shoot for it. Agenda+

[MatsPalmgren]

The CSSWG has already resolved "Make text indent percentage relative to the content box of the element". Using the parent's size contradicts that. It makes more sense to me to always use the content box size of the element itself and thus resolve against zero when calculating its intrinsic size and then resolve against that during layout, like we do for other percentages.

CC @dbaron