[Caml-list] ocaml-pcre and UTF-8

[Philippe Strauss <philippe@strauss-acoustics.ch>]

Hello caml'ers,

How do I convince PCRE to be UTF-8 friendly? example:

--
open Pcre

external show : 'a -> string = "%show"

let recomp = regexp ~flags:[`UTF8; `CASELESS]

let res_w = "(*UTF8)^(\w+)$"
let rec_w = recomp res_w

let accents = ["blurb"; "toxicité"; "velléités"; "à"; "où"; "über"; "marie-jeanne"]

let () =
	Printf.printf "config_utf8=%b\n" Pcre.config_utf8 ;
	List.iter (fun word ->
		try
        	let sub = Pcre.extract_opt ~full_match:false ~rex:rec_w word in
        	print_endline (show sub)
        with Not_found -> Printf.eprintf "Not_found was raised on \"%s\" :-(\n%!" word
	) accents
--

at least on my setup it gives:

--
philou@air:~/mysrc/web/myco$ ./pcre_utf 
config_utf8=true
[|Some ("blurb")|]
Not_found was raised on "toxicité" :-(
Not_found was raised on "velléités" :-(
Not_found was raised on "à" :-(
Not_found was raised on "où" :-(
Not_found was raised on "über" :-(
Not_found was raised on "marie-jeanne" :-(
--

A php user which happen to be a nice guy got it working on PHP :-( how lame.

--
Philippe Strauss
http://www.strauss-acoustics.ch/