* Parse flow nested array type annotations like number[][]
Fixes#217.
* Do not parse a newline separated array statement into a flow type annotation
* Get rid of lineBreak dependency in flow.js in favor of higher level method
* Add code coverage from Babel tests
* Configure nyc correctly for Babel test coverage
* Guard against nyc config conflicts if Babel switches to nyc in the future
* Move .nyc_output from Babel build dir to root
* Use .gitattributes to ignore files with weird newlines
* [Flow] Make parameter names in function types optional
* [Flow] Anonymous function types with single params can omit parens
* [Flow] Optional names for object indexer keys
* Add noAnonFunctionType explicitly to state
* Adjust gitattributes as files have been fixed
* Increase test coverage
* Test for error when binding `this` in destructuring pattern
* Ignore coverage of inAsync check in parseAwait - already checked externally
* Ignore coverage of default case in checkPropClash
* Remove unused parameter isAsync from parseParenAndDistinguishExpression
* Ignore coverage of an `else` branch in flowParseTypeParameterDeclaration
* Flow: remove unused parameters to flowParseTypeAnnotatableIdentifier
* Flow: ignore coverage of pass-through throw statement in parseConditional
* Flow: Add test for error on property with type param
* Flow: ignore coverage of pass-through throw statements in parseMaybeAssign, parseArrow
* Add test for error on XML-style comment in module code
* Update test for error on method in object pattern
* Test for error: "Only '=' operator can be used for specifying default value"
* Property variance type annotations for Flow plugin
Non-method properties and indexers of object types, declare class, and
interfaces can be "positive" or "negative." Class fields, but again not
methods, can also have variance.
This PR generalizes the variance annotations for type parameters into a
new node type, and reuses that node for those properties.
The code for object types is reused for interfaces and declare classes.
The changes there are straightfoward.
The code for class fields is reused for object literals, which do not
support variance annotations (currently). This code is a bit sketchy,
because we always parse variance annotations in the `parsePropertyName`
extension, then error in a the subsequent parse phase for object
literals (`parseObjPropValue`) or class methods (`parseClassMethod`).
* Remove bogus unreachable code, clarify variance parsing conditional
* Don't use a new node type for variance annotations
Adding a new node type, specifically changing the TypeParameter node's
variance property to be node-valued, is a breaking change. We might
choose to make this breaking change in a later version.
* s/start/variancePos
Expanding on #150, this allows `unexpected()` to accept the expected token type instead of a message string.
This overload is then used in a couple more places (that independently implement a logic similar to `expect()`'s) to construct an `Unexpected token, expected FOO` message.
I also noticed that there was missing test coverage for builtin types,
so I added a test for that, which includes the newly added type as well
as the other existing types.
