Re: a question about foreach construct
- Posted by ghaberek (admin) Oct 17, 2015
- 1428 views
You can add this by creating a preprocessor.
That would be great, but without a way to easily chew up the language and spit it back out, the preprocessor is effectively broken.
I've attempted write my own preprocessors for a few things and the smaller ones work well enough, but once it reaches a certain level of complexity I basically end up trying to write my own recursive-descent parser based on the included Euphoria tokenizer.
I don't think I should have to go through that much effort to write a preprocessor. I don't think the preprocessor should rely on ad hoc string parsing solutions to do its job. We should have a way to cleanly encounter bits of the language and interrupt or override the token stream with our own pieces.
I'm not trying to poo-poo your idea. It's a great idea. We just don't have a less-than-painfully-difficult way to implement it right now.
-Greg
