[selectors-4] circular references in section 3.1

[therealglazou]

I see a few issues in that section.

Issue 1:

The subject of a selector is the element(s) that selector is defined to be about:

...
For a complex selector, it is any element that matches the last compound selector in the selector.

The above is wrong: it says the element "matches" the last compound
selector. Which means "only". A more correct prose would say it matches
the last compound selector while its context matches the other compound
selector in the complex selector AND the relationships expressed by the
combinators...

Issue 2:

The definition of "subject of selector" relies on a mix of
"represented" and "matches". I think it's wrong because below "match"
is itself defined in reference to "subject"... We then have an
circular ref there.

Proposal:

I think we have too many circular definitions right now in that section
3.1. Here's a different proposal:

The term selector can refer to a simple selector, compound selector,
complex selector, or selector list.

A simple selector is a single condition on an element. (...) A given
element is said to match a simple selector when it makes the
corresponding condition true.

A compound selector is a sequence of simple selectors that are not
separated by a combinator. It represents a set of simultaneous
conditions on a single element. (...) A given element is said to
match a compound selector when it matches all simple selectors
in the compound selector.

A combinator is a condition of relationship between two elements
represented by the compound selectors on either side of the
combinator. (...) Two given elements are said to match a combinator
when the condition of relationship between these elements is true.

A complex selector is a sequence of one or more compound selectors
separated by combinators. It represents a set of simultaneous
conditions on a list of elements in particular relationships. (...)
A given element is said to match a complex selector when it exists
a list of elements individually matching each corresponding compound
selector in the complex selector, each pair of consecutive elements in
that list matches the combinator separating the two corresponding
compound selectors, and the last element in that list is the given
element.

A list of simple/compound/(...)

For simple, compound or complex selectors, the subject of that
selector is any element matching that selector. For a selector
list, the subject of that selector list is any element matching any
selector in the list.

[fantasai]

Daniel, thank you so much for this! This is a subtle and fiddly section that's never been quite right. I've gone ahead and cleaned up this section following your advice, with some additional wording changes. Please have a look at the resulting changes and let me know if it works, and if I've messed anything up in translation?

[Loirooriol]

I prefer Daniel's proposal here:

A given element is said to match a complex selector when it exists
a list of elements individually matching each corresponding compound
selector in the complex selector, each pair of consecutive elements in
that list matches the combinator separating the two corresponding
compound selectors

better than

A given element is said to match a complex selector when there exists
a list of elements, each matching a corresponding compound selector
in the complex selector, with their relationships matching the combinators
between them, and with the given element matching the last compound selector.

The latter is less precise and can seem over-constrained, e.g. if you have a + b c, one may think that the elements matched by a and c must match both combinators between them (i.e. the element matched by c should be both the next sibling and a descendant of a, which is not possible).

And I would also say "matching the corresponding" instead of "matching a corresponding".

[fantasai]

Thanks for the review, @Loirooriol ! I pushed some edits to bring it closer to the former phrasing.

And I would also say "matching the corresponding" instead of "matching a corresponding".

I definitely debated that back and forth (between "a", "the", and "its") for awhile, but ended up resolving on "a" as the best option there. It's definitely a little more awkward than I would like ideally, but I suspect fixing it would involve some more substantial rephrasing which I cannot figure out.