[css-gcpm][css-content] "string-set" on elements with no boxes: How to determine which page the assignment is on #8404
Comments
|
Agree this is complex, and definitely agree going back through the DOM makes more sense rather than to try and guess what-might-have-been based on layout. I created a test to see what current implementations do: <!DOCTYPE HTML>
<html>
<head>
<style>
@page {
margin: 72px;
@top-center {
content: "String is " string(foo, first);
}
}
.set {
display: none;
string-set: foo attr(string);
}
h1 {
break-after: page;
}
</style>
</head>
<body>
<h1>Page 1</h1>
<div class="set" string="string1">string1</div>
<h1>Page 2</h1>
<div class="set" string="string2">string2</div>
<h1>Page 3</h1>
<div class="set" string="string3">string3</div>
</body>
</html>If I understand your proposal correctly you'd expect to see "string1, string2, string3" at the top of the three pages, correct? Testing with the engines at https://printcss.live gives:
other findings
So of the three tools trying to do this properly (Prince seems a bit lost here), the only ambiguity is where the string-set is immediately followed by a page break. In this situation, both Vivliostyle and BFO are moving the string-set to the top of the next page, rather than the bottom of the previous one as in your proposal. So that's the data. Based on this I think there are four options.
I'm obviously leaning towards (2), not just because that's what we've done but because if string-set is used, it seems like this would be done at the start of any section relating to that string: it should stick to content that follows it. <section>
<div style="display:none; string-set: chapter "Widget"></div>
... content relating to widgets ...
</section>
<section>
... content relating to widgets ...
<div style="display:none; string-set: chapter "Widget"></div>
</section>The first of those two seems more likely to me, and the existing css-gcpm test at https://github.com/web-platform-tests/wpt/blob/master/css/css-gcpm/string-set-011.html seems to agree. cc @MurakamiShinyu for Vivliostyle's opinion. Related question: The definition for |
|
I agree with @faceless2, (2) is most preferable. Or (4) also makes sense, because page breaking occurs between elements with boxes, and the element with no boxes at the page boundary comes with the previous or the next element, that would depend on implementation.
In HTML documents, the root |
|
The wording "If the element is the first element on the page," may be misleading. In my understanding, "the first element on the page" here should not be a single element. It was "if the element begins the page" in the old draft: CSS GCPM, 24 September 2013 Editor’s Draft:
The wording was rewritten from "begins the page" to "is the first element on the page" but no spec change would be intended. See the "Changes" section of the CSS GCPM: https://drafts.csswg.org/css-gcpm-3/#changes
Also, the wording about the
to
I found a related discussion on the www-style ML when the first wording was written: Does string-set work on elements or on boxes? It may be helpful to understand the intent of the spec. At that time there was probably no consideration for the case that the element with no boxes falls on a page boundary. |
|
We agree that option 2 is probably more useful to authors than option 1. It would be the same as our previous suggestion, but searching forward and picking the first box of the found element, right? We don't consider options 3 or 4 viable. However, we would like to propose one, that would cover both examples by combing options 1 and 2:
This way well-structured documents should work as expected in a vast majority of cases. Also it would implicitly eliminate edge-cases of options 1 and 2, where the assignment could end up before the first or after the last page, assuming that one of the two ifs is automatically satisfied if one of the two pages does not exist. On the topic of |
The
string-setproperty applies to "all elements, but not pseudo-elements", which includes elements with no boxes (e.g. viadisplay: none).Regarding such elements both specifications say ( after https://drafts.csswg.org/css-content-3/#valdef-string-first-except ):
"The content values of named strings are assigned at the point when the content box of the element is first created (or would have been created if the element’s display value is none)."
Determining where a box "would have been" is rather complex, as that could be influenced by it's height (near a page break) which in turn could be influenced by it's content.
Adding that the hypothetical box should be considered to have a block-size of 0 would solve this for simple cases.
However in more complex contexts this would still not be easy to determine, e.g. when the
orderproperty is involved.We would propose to go backwards through the DOM from the element until an element is found that has boxes in the content flow.
The page that the last of those boxes is on is the page of the assignment.
If coordinates on the page are also required they should be at the block-end of that box.
However, we would prefer that inside the page the order of assignments is strictly the DOM order.
The text was updated successfully, but these errors were encountered: