Upgrade prefixer, don't handle multiple rules at once

[zgotsch]

Use the new inline-style-prefix-all, which gives us a nice array of rules which need to be prefixed. Only importantify one rule at a time, so we don't have to parse css in order to pick the right semicolons :)

fixes #70

[zgotsch]

I probably need to sign the CLA now, will do that in a bit.

[xymostech]

Maybe use Array.isArray?

[zgotsch]

Yeah I thought of this but I don't know what your polyfilling setup is. If it's safe, it's way better.

[xymostech]

This looks awesome!!

Could you add a test to see how this interacts with our stringHandlers stuff? For instance, our fontFamily string handler also accepts arrays as a valid value, and I want to make sure that this doesn't clobber that (we unfortunately don't have tests for stringHandlers yet, maybe that would answer my question).

I guess something that looks like

assertCSS('.foo', [{
    test: ['a', 'b'],
}], ".foo { test: a b !important; }", { test: (val) => val.join(" ") });

(I'm worried it would turn into ".foo { test: a !important; test: b !important; }", which isn't what we want)

[zgotsch]

If I understand correctly, there is already a test for this:

    it('supports custom string handlers', () => {
        assertCSS('.foo', [{
            fontFamily: ["Helvetica", "sans-serif"]
        }], '.foo{font-family:Helvetica, sans-serif !important;}', {
            fontFamily: (val) => val.join(", "),
        });
    });

which passes

[xymostech]

Oh yep, you're right! Sweet. Did you sign the CLA? (we should really hook our CLA bot up to this repo...)

[zgotsch]

I did sign the CLA!

[jlfwong]

I'm confused -- didn't we only need importantify in the first place because we weren't getting an array of values? Why do we still need the regex? Just to avoid double important'ing things?

[zgotsch]

Yeah, that's why I kept using it. There should probably be a test for that
so it doesn't get broken in future.
On Wed, May 4, 2016 at 4:37 AM Jamie Wong notifications@github.com wrote:

I'm confused -- didn't we only need importantify in the first place
because we weren't getting an array of values? Why do we still need the
regex? Just to avoid double important'ing things?

—
You are receiving this because you authored the thread.
Reply to this email directly or view it on GitHub
#71 (comment)

[zgotsch]

Oof, seems like prefixes are not ordered consistently across platforms. I'm going to sort them...

[zgotsch]

Is this safe? In the case of something like -webkit-box and -webkit-flex, the older prefix should probably go first. It's not trivial to know the rules for ordering, though. This should also probably be the responsibility of inline-prefix-all, not aphrodite.

[kentcdodds]

This should also probably be the responsibility of inline-prefix-all, not aphrodite.

Yes

[xymostech]

@zgotsch Okay, I'm gonna merge this!