[Caml-list] Encodings for ocaml sources

[Caml-list] Encodings for ocaml sources

[Андрей Бергман]

Hello!

Do I understand it right, that currently Ocaml compiler can compile programs, written in
Latin1, various 8-bit encodings with Latin1 base (character code < 128) and UTF-8,
all the identifiers should be within Latin1 base, while string literals can contain any characters?

And UTF-16 is not supported - am I correct?

And this is pretty much the same for all Windows/Linux/OSX?

Sincerely,
   Andrey.

Re: [Caml-list] Encodings for ocaml sources

[Daniel Bünzli]

On 22 May 2019 at 03:16:46, Андрей Бергман (vkni@yandex.ru) wrote:

> Latin1, various 8-bit encodings with Latin1 base (character code < 128) and UTF-8,
> all the identifiers should be within Latin1 base, 

Latin1 identifiers are deprecated, see [1].

> while string literals can contain any characters?

Yes IIRC they can contain arbitrary *bytes*. But a forward looking habit would be to stick to UTF-8 encoded literals rather use arbitrary bytes and avoid US-ASCII control characters except for white space.

> And UTF-16 is not supported - am I correct?

Yes. Strictly speaking because of the preceeding you can UTF-16 encode the string literals, but for the rest of the program that wouldn't work, the lexer wouldn't recognize the keywords. 

> And this is pretty much the same for all Windows/Linux/OSX?

Yes. 

Best, 

Daniel

[1]: https://caml.inria.fr/pub/docs/manual-ocaml/lex.html#sec77