Contextual css selector

[dustinsanders]

Issue #23.

Added support for descendent, >, ~, and +.

[ljharb]

this seems like a pretty sketchy way to avoid let predicate; let selectSiblings;

[dustinsanders]

I usually do this when defining multiple variables to undefined, but let predicate; let selectSiblings; is also fine

[blainekasten]

Is there a good argument for this to be a class instead of just pure functions?

[dustinsanders]

Originally I was currying it but it didn't seem as clean.

complexSelector(buildInstPredicate, findWhereUnwrapped, childrenOfInst)(selector, this);

I guess it could all be one function, but I wanted to separate initialization from the logic.

[blainekasten]

Ultimately it's up to the maintainers. They might see this argument as moot. But it could make maintenance easier when all selector engines follow a similar flow. Previously that was the case until here.

Additionally, it is heavier to instantiate an object for every selector. But I don't they worry too much about the perf for this lib.

[blainekasten]

Does this only support 1 level of nested selectors? Will it be extra hard to add infinite depth to nested selectors later on if/when wanted?

e.g.,

find('.foo + div > span')

[dustinsanders]

@blainekasten I just ran the following

it.only('.foo + div > span', () => {
    const wrapper = shallow(
      <div>
        <div className="foo" />
        <div>
          <span />
        </div>
      </div>
    );

    expect(wrapper.find('.foo + div > span').length).to.equal(1);
  });

and it passed. Based on how it is written it should support nesting, but there should be more testing around the nested selectors.

[dustinsanders]

See changes made based off of suggestions

[borisyankov]

Array.prototype.reduce is available in ES5

[dustinsanders]

@borisyankov good call, using native reduce in latest commit

[koba04]

any progress on this?

[dustinsanders]

I think I've addressed all the concerns, but am open to other changes

[ljharb]

@dustinsanders would you mind rebasing this on top of latest master, and down to a smaller number of conceptually atomic commits? It's got merge conflicts atm.

[dustinsanders]

@ljharb I have rebased and squashed to one commit

[ljharb]

@dustinsanders looks like this needs a fresh rebase - sorry this has taken so long.

[dustinsanders]

@ljharb this has been freshly rebased

[lelandrichardson]

@dustinsanders this looks awesome. great work. (Sorry this took so long for me to get to it)