I would like to add support for editor placeholder to Clang’s lexer. The placeholder tokens will function similarly to the placeholders in the Swift compiler:
- An editor placeholder token starts with ‘<#’ and ends with ‘#>’.
- When lexing the source that contains placeholders:
- Clang will report an error when it finds the placeholder token.
- Then the lexer will treat the placeholders as if they were expanded, e.g. given:
void Class::method(<#int x#>)
The parser will see the tokens that correspond to the following source:
void Class::method(int x)
Do you have any feedback/requests? Should placeholders be recognized at all times or just when Clang is used by the editor?