Update: Check member expressions with this in operator-assignment

[mdjermanovic]

What is the purpose of this pull request? (put an "X" next to item)

[X] Bug fix
[X] Changes an existing rule

I believe this is a bug fix, but it produces more warnings.

This PR fixes two things in the operator-assignment rule:

• The default always option was ignoring this member expressions, such as this.x = this.x + y.
• The never option was reporting this member expressions, but didn't auto-fix in some cases where it seems to be safe, such as this.x += y.

What rule do you want to change?

operator-assignment

Does this change cause the rule to produce more or fewer warnings?

more

How will the change be implemented? (New option, new default behavior, etc.)?

new default behavior

Please provide some example code that this change will affect:

/* eslint operator-assignment: ["error", "always"] */

this.foo = this.foo + bar

What does the rule currently do for this code?

No errors

What will the rule do after it's changed?

Error and autofix to:

/* eslint operator-assignment: ["error", "always"] */

this.foo += bar

What changes did you make? (Give an overview)

Account for ThisExpression in the function that compares two expressions. Also in the function that determines whether the autofix is safe.

Is there anything you'd like reviewers to focus on?

All added tests in valid were valid before, those are just regression tests.

The following tests with the never option had the same behavior (which is an error without autofix) before this PR, and that isn't changed.

this.foo.bar += baz
this[foo] >>>= bar

The following test with the never option was already an error, but wasn't being auto-fixed. That's changed now.

this.foo += bar

The following tests with the default always options are new warnings:

this.foo = this.foo + bar
this.foo.bar = this.foo.bar + baz
this[foo] = this[foo] + bar
this[5] = this[5] / foo

Can this be treated as a semver-minor bug fix that produces more warnings? There was a similar case in #12279.

[kaicataldo]

LGTM, thanks!

[btmills]

LGTM, thanks @mdjermanovic! Agree this is a semver-minor bugfix.