As I (believe I) start to understand the parser, I somehow think that doing
something like the following would
not violate (not take too much risk) the current Lexer/Parser
be somewhat reasonable to implement
address many of the reqs I read
leave some infrastructure in the Lexer/Parser to add more feature without
needing to go back to square 0 next time
Any thoughts?!
#!/usr/bin/env xcrun swift
let s1 = "{\"key1\": \"stringValue\"}"
let s2 = _"{"key1": "stringValue"}"_
let s3 =
/* this is the same template */
_"{"key1": "stringValue"}"_
let s4 =
/* this is (almost) the same template */
_"
{
"key1": "stringValue"
, "key2": "stringValue"
}
"_
let s5 = @literal(json)
/* this is exactly the same template as s4 */
_"
{
"key1": "stringValue"
, "key2": "stringValue"
}
"_
let s6 = @literal(json)
/* this is exactly the same template as s5&s4 */
_"
|{
| "key1": "stringValue"
| , "key2": "stringValue"
|}
"_
_______________________________________________
swift-evolution mailing list
swift-evolution@swift.org
https://lists.swift.org/mailman/listinfo/swift-evolution
