[raw-strings] Newline character handling

Remi Forax forax at univ-mlv.fr
Sat Jan 27 10:59:40 UTC 2018


Hi Tagir,
you have miss this line:
  CR (\u000D) and CRLF (\u000D\u000A) sequences are always translated to LF (\u000A). This translation provides least surprise behavior across platforms.

this is also the behavior of Perl, PHP, etc.

as a guy that had to write too many shaders in Java recently, thanks for resurrecting this discussion, i think we should not wait another 10 years to add raw strings in Java. 

regards,
Rémi

----- Mail original -----
> De: "Tagir Valeev" <amaembo at gmail.com>
> À: "amber-spec-experts" <amber-spec-experts at openjdk.java.net>
> Envoyé: Samedi 27 Janvier 2018 05:13:58
> Objet: [raw-strings] Newline character handling

> Hello!
> 
> I looked through Raw String Literals JEP draft [1] and did not find
> any explicit statement about CR/LF translation within multiline raw
> string. Usually in text files (and, I believe, Java source qualifies
> as a text file) it's assumed that changing \n to \r\n and vice versa
> would not change the semantics. Sometimes such changes are performed
> automatically, e.g. on Git checkout via core.autocrlf=true setting
> [2]. If multiline string literal is used, then such replacement may
> badly affect the semantics of the program. E.g.:
> 
> public class Hello {
>  public static void main(String[] args) {
>    System.out.println(`Hello
> World!`.length());
>  }
> }
> 
> The output of this program may change if its source text is converted
> from CR/LF to LF line endings or vice versa.
> 
> As far as I know, Kotlin forcibly replaces CR/LF to LF within
> multiline strings, though I did not find any explicit statement about
> this in the documentation. This looks a good compromise, though could
> be annoying for people who actually want to encode CR/LF inside a
> multiline string. Nevertheless, I feel, that the special handling of
> line terminators within multiline strings (or absence of such
> handling) should be explicitly mentioned in the JEP and the following
> specification.
> 
> With best regards,
> Tagir Valeev.
> 
> 
> [1] http://openjdk.java.net/jeps/8196004
> [2] https://help.github.com/articles/dealing-with-line-endings/


More information about the amber-spec-experts mailing list