I\'m using map()
function of array in for-in loop like this:
let numbers = [2, 4, 6, 8, 10]
for doubled in numbers.map { $0 * 2 } // compile error
Trailing closure grammar generates a known ambiguity between a body of the statement (in your case, that's for
loop, but it applies to other statements as well) and the body of trailing closure. Although it is technically possible to solve this problem, compiler designers decided to prohibit trailing closure syntax in controlling portions of various high-level statements:
While it would be possible to tell what is intended in some cases by performing arbitrary lookahead or by performing type checking while parsing, these approaches have significant consequences for the architecture for the compiler. As such, we've opted keep the parser simple and disallow this.
To understand the nature of the conflict, consider this example:
for v in expr { /* code 1 */ } { /* code 2 */ }
The parser needs to make a choice regarding code 1
block. It could either use it as a trailing closure of expr
and treat code 2
as the body of the for
loop, or use code 1
as the body of the for
loop, while treating code 2
as an independent group of statements enclosed in curly braces - a classic shift-reduce conflict.
Solving such conflicts is very expensive. Essentially, your parser needs to continue looking ahead through more tokens, until only one interpretation makes sense, or the parser runs out of tokens (in which case it makes an arbitrary decision one way or the other, requiring the programmers to disambiguate their program when that choice is not what they wanted).
Adding parentheses removes the ambiguity. Proposals were considered to remove the ambiguity by adding a mandatory keyword to separate the control portion of the loop from its body, i.e.
// The syntax of rejected proposal
for doubled in numbers.map { $0 * 2 } do {
print(doubled) // ^^
}
Adding an extra do
keyword "attaches" the block on the left, if any, to the expression, and makes the block on the right the body of the loop statement. This approach has a major drawback, because it is a breaking change. That is why this proposal has been rejected.