1 ;;;; -*- Mode: Lisp; Syntax: ANSI-Common-Lisp; Base: 10; Package: UFFI -*-
2 ;;;; *************************************************************************
3 ;;;; FILE IDENTIFICATION
6 ;;;; Purpose: non-ASCII character support
7 ;;;; Programmer: Kevin M. Rosenberg
8 ;;;; Date Started: Feb 2010
10 ;;;; This file, part of UFFI, is Copyright (c) 2002-2010 by Kevin M. Rosenberg
12 ;;;; *************************************************************************
16 #-(or (and lispworks unicode) (and sbcl sb-unicode)
17 (and allegro ics) (and clisp i18n)
18 (and openmcl openmcl-unicode-strings))
19 (pushnew 'no-i18n cl:*features*)
21 (defvar *default-foreign-encoding*
23 "Normalized name of default external character format to use
24 for foreign string conversions. nil means use implementation default
27 (defvar *foreign-encoding-mapping*
28 #+(and lispworks unicode)
29 '((:ascii . :ascii) (:latin-1 . :latin-1) (:ucs-2 . :unicode)
30 (:utf-8 . :utf-8) (:jis . :jis) (:sjis . :sjis) (:gbk . :gbk))
31 #+(and sbcl sb-unicode)
32 '((:ascii . :ascii) (:latin-1 . :latin-1) (:utf-8 . :utf-8)
33 (:ucs-2 . :ucs-2) (:sjis . :sjis) (:gbk . :gbk))
35 '((:ascii . :ascii) (:latin-1 . :latin1) (:utf-8 . :utf-8)
36 (:sjis . :shift-jis) (:euc-jp . :euc) (:gbk . :gb2313)
39 '((:ascii . charset:ascii) (:ucs-2 . charset:ucs-2)
40 (:utf-8 . charset:utf-8) (:latin-1 . charset:iso-8859-1)
41 (:jis . charset:jis_x0201) (:jis . charset:shift-jis)
42 (:gbk . charset:gbk) (:euc-jp . charset:euc-jp))
43 #+(and openmcl openmcl-unicode-strings)
44 '((:ascii . :ascii) (:latin-1 . :iso-8859-1) (:utf-8 . :utf-8)
46 #+nil (:euc-jp . :euc-jp)
48 #-(or (and lispworks unicode) (and sbcl sb-unicode)
49 (and allegro ics) (and clisp unicode)
50 (and openmcl openmcl-unicode-strings))
52 "Mapping between normalized external format name and implementation name.")
54 (defvar *foreign-encodings*
55 (mapcar 'car *foreign-encoding-mapping*)
56 "List of normalized names of external formats support by underlying implementation.")
58 (defun lookup-foreign-encoding (normalized)
59 (cdr (assoc normalized *foreign-encoding-mapping* :test 'eql)))
61 (defmacro string-to-octets (str &key (encoding *default-foreign-encoding*))
62 "Converts a Lisp string to a vector of octets."
63 #-(or allegro lispworks openmcl sbcl)
64 (declare (ignore encoding))
65 #-(or allegro lispworks openmcl sbcl)
66 (map-into (make-array (length str) :element-type '(unsigned-byte 8))
70 (let ((fe (gensym "FE-"))
73 `(let* ((,fe ,encoding)
74 (,ife (when ,fe (lookup-foreign-encoding ,fe)))
78 (excl:string-to-octets ,s :external-format ,ife :null-terminate nil)
79 (excl:string-to-octets ,s :null-terminate nil)))))
81 #+(or lispworks openmcl)
82 ;; simply reading each char-code from the LENGTH of string handles multibyte characters
83 ;; just fine in testing LW 6.0 and CCL 1.4
84 (map-into (make-array (length str) :element-type '(unsigned-byte 8))
88 (let ((fe (gensym "FE-"))
91 `(let* ((,fe ,encoding)
92 (,ife (when ,fe (lookup-foreign-encoding ,fe)))
95 (sb-ext:string-to-octets ,s :external-format ,ife)
96 (sb-ext:string-to-octets ,s))))
100 (defmacro octets-to-string (octets &key (encoding *default-foreign-encoding*))
101 "Converts a vector of octets to a Lisp string."
102 #-(or allegro lispworks openmcl sbcl)
103 (declare (ignore encoding))
104 #-(or allegro lispworks openmcl sbcl)
105 (let ((out (gensym "OUT-"))
106 (code (gensym "CODE-")))
107 `(with-output-to-string (,out)
108 (loop for ,code across ,octets
109 do (write-char (code-char ,code) ,out))))
112 (let ((fe (gensym "FE-"))
113 (ife (gensym "IFE-"))
114 (oct (gensym "OCTETS-")))
115 `(let* ((,fe ,encoding)
116 (,ife (when ,fe (lookup-foreign-encoding ,fe)))
120 (excl:octets-to-string ,oct :external-format ,ife)
121 (excl:octets-to-string ,oct)))))
123 #+(or lispworks openmcl)
124 ;; With LW 6.0 and CCL 1.4, writing multibyte character just one octet at a time tests fine
125 (let ((out (gensym "OUT-"))
126 (code (gensym "CODE-")))
127 `(with-output-to-string (,out)
128 (loop for ,code across ,octets
129 do (write-char (code-char ,code) ,out))))
132 (let ((fe (gensym "FE-"))
133 (ife (gensym "IFE-"))
134 (oct (gensym "OCTETS-")))
135 `(let* ((,fe ,encoding)
136 (,ife (when ,fe (lookup-foreign-encoding ,fe)))
139 (sb-ext:octets-to-string ,oct :external-format ,ife)
140 (sb-ext:octets-to-string ,oct))))
144 (defun foreign-encoded-octet-count (str &key (encoding *default-foreign-encoding*))
145 "Returns the octets required to represent the string when passed to a ~
147 ;; AllegroCL 8-bit, CCL, and Lispworks give correct value without converting
148 ;; to external-format. AllegroCL 16-bit, SBCL, and CLISP requires conversion
149 ;; with external-format
151 #+(or (and allegro ics) (and sbcl sb-unicode) (and clisp i18n))
152 (length (string-to-octets str :encoding encoding))
154 #-(or (and allegro ics) (and sbcl sb-unicode) (and clisp i18n))
155 (declare (ignore encoding))
156 #-(or (and allegro ics) (and sbcl sb-unicode) (and clisp i18n))