[MemberExpression] Long object access doesn't wrap about prettier

[vjeux]

@despairblue thank you! This is great! Shouldn't be too hard to make work.

[vjeux]

Repro case

https://jlongster.github.io/prettier/#%7B%22content%22%3A%22someVeryLongMethodObjectName%5Cn%20%20.withSomeVeryLongObjectKey%5Cn%20%20.withSomeVeryLongObjectKey%5Cn%20%20.andAnotherLongKey%3B%22%2C%22options%22%3A%7B%22printWidth%22%3A80%2C%22tabWidth%22%3A2%2C%22singleQuote%22%3Afalse%2C%22trailingComma%22%3Afalse%2C%22bracketSpacing%22%3Afalse%7D%7D

[despairblue]

@JakeCoxon If you want to give it a shot you might be interested in the discussion in #212 and my take on this (which created a lot of other issues) in #217.

I'd still like to see a solution for this. Chai assertions are another thing that suffer from this.

I wonder if there is a way to only add an softspace after a dot if the currently printed line would exceed the max line length with the next token. But I think that's not easy at all and not performant.

[vjeux]

@despairblue do you mind giving a real-world example of chai assertions that do not look good? It's usually better for this kind of things to look at real code rather than make up examples.

[despairblue]

@vjeux sure this would be a small one. I'd rather see the left case instead of the right one.

Or this one which is a little longer.

[vjeux]

With #462 it looks like

if (testConfig.ENABLE_ONLINE_TESTS === "true") {
  describe("POST /users/me/pet", function() {
    it("saves pet", function() {
      function assert(pet) {
        expect(pet).to.have.property("OwnerAddress").that.deep.equals({
          AddressLine1: "Alexanderstrasse",
          AddressLine2: "",
          PostalCode: "10999",
          Region: "Berlin",
          City: "Berlin",
          Country: "DE"
        });
      }
    });
  });
}

which is better. But we should also be able to break the .to.have.property()