On Wed, Sep 26, 2012 at 3:46 AM, Didier Verna didier@lrde.epita.fr wrote:
does anyone have a global view on the status of unicode support in the main CL implementations (SBCL, CMU-CL, ECL, CCL, CLISP, ABCL, ACL, LispWorks). By that, I mean mostly being able to encode source files in utf8 with the following concerns:
- do I need to do something special at the implementation-level ?
- do I need to do something special at the ASDF system level ?
- could this break the code of people using my libraries ?
So far as I can tell, all these implementations support Unicode, though some of them can be explicitly compiled without.
ASDF, since release 2.21 (April 2012) supports Unicode. The recommended, backwards-compatible, incantation is (in your defsystem, or any specific component): #+asdf-unicode :encoding #+asdf-unicode :utf-8.
I have the intention of making utf-8 the default eventually, but last we checked (in April this year), that would break 7 files out of all of quicklisp, the authors of which have never replied regarding fixing them.
Also, if you (asdf:load-system :asdf-encodings) explicitly and early, you can use :encoding :latin1, or :encoding euc-jp, or whichever encoding your implementation supports (asdf-encodings at this time won't transcode things for you).
An example system that explicitly uses this UTF-8 support is lambda-reader, which I published earlier this year (last edited in April also), based on an initial implementation by Brian Mastenbrook.
—♯ƒ • François-René ÐVB Rideau •Reflection&Cybernethics• http://fare.tunes.org To send men to the firing squad, judicial proof is unnecessary... These procedures are an archaic bourgeois detail. This is a revolution! And a revolutionary must become a cold killing machine motivated by pure hate. — Che Guevara