utf8proc
view ruby/utf8proc_native.c @ 17:47b467f4c128
Contribution from libmojibake fork (missing file "normtest.c")
| author | Jiahao Chen, Steven G. Johnson, Anthony David Kelman |
|---|---|
| date | Mon Dec 01 14:32:19 2014 -0500 (2014-12-01) |
| parents | d79da2302625 |
| children |
line source
1 /*
2 * Copyright (c) 2009 Public Software Group e. V., Berlin, Germany
3 *
4 * Permission is hereby granted, free of charge, to any person obtaining a
5 * copy of this software and associated documentation files (the "Software"),
6 * to deal in the Software without restriction, including without limitation
7 * the rights to use, copy, modify, merge, publish, distribute, sublicense,
8 * and/or sell copies of the Software, and to permit persons to whom the
9 * Software is furnished to do so, subject to the following conditions:
10 *
11 * The above copyright notice and this permission notice shall be included in
12 * all copies or substantial portions of the Software.
13 *
14 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
15 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
16 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
17 * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
18 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
19 * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
20 * DEALINGS IN THE SOFTWARE.
21 */
24 /*
25 * File name: ruby/utf8proc_native.c
26 *
27 * Description:
28 * Native part of the ruby wrapper for libutf8proc.
29 */
32 #include "../utf8proc.c"
33 #include "ruby.h"
35 #ifndef RSTRING_PTR
36 #define RSTRING_PTR(s) (RSTRING(s)->ptr)
37 #endif
38 #ifndef RSTRING_LEN
39 #define RSTRING_LEN(s) (RSTRING(s)->len)
40 #endif
42 typedef struct utf8proc_ruby_mapenv_struct {
43 int32_t *buffer;
44 } utf8proc_ruby_mapenv_t;
46 void utf8proc_ruby_mapenv_free(utf8proc_ruby_mapenv_t *env) {
47 free(env->buffer);
48 free(env);
49 }
51 VALUE utf8proc_ruby_module;
52 VALUE utf8proc_ruby_options;
53 VALUE utf8proc_ruby_eUnicodeError;
54 VALUE utf8proc_ruby_eInvalidUtf8Error;
55 VALUE utf8proc_ruby_eCodeNotAssignedError;
57 VALUE utf8proc_ruby_map_error(ssize_t result) {
58 VALUE excpt_class;
59 switch (result) {
60 case UTF8PROC_ERROR_NOMEM:
61 excpt_class = rb_eNoMemError; break;
62 case UTF8PROC_ERROR_OVERFLOW:
63 case UTF8PROC_ERROR_INVALIDOPTS:
64 excpt_class = rb_eArgError; break;
65 case UTF8PROC_ERROR_INVALIDUTF8:
66 excpt_class = utf8proc_ruby_eInvalidUtf8Error; break;
67 case UTF8PROC_ERROR_NOTASSIGNED:
68 excpt_class = utf8proc_ruby_eCodeNotAssignedError; break;
69 default:
70 excpt_class = rb_eRuntimeError;
71 }
72 rb_raise(excpt_class, "%s", utf8proc_errmsg(result));
73 return Qnil;
74 }
76 VALUE utf8proc_ruby_map(VALUE self, VALUE str_param, VALUE options_param) {
77 VALUE str;
78 int options;
79 VALUE env_obj;
80 utf8proc_ruby_mapenv_t *env;
81 ssize_t result;
82 VALUE retval;
83 str = StringValue(str_param);
84 options = NUM2INT(options_param) & ~UTF8PROC_NULLTERM;
85 env_obj = Data_Make_Struct(rb_cObject, utf8proc_ruby_mapenv_t, NULL,
86 utf8proc_ruby_mapenv_free, env);
87 result = utf8proc_decompose(RSTRING_PTR(str), RSTRING_LEN(str),
88 NULL, 0, options);
89 if (result < 0) {
90 utf8proc_ruby_map_error(result);
91 return Qnil; /* needed to prevent problems with optimization */
92 }
93 env->buffer = ALLOC_N(int32_t, result+1);
94 result = utf8proc_decompose(RSTRING_PTR(str), RSTRING_LEN(str),
95 env->buffer, result, options);
96 if (result < 0) {
97 free(env->buffer);
98 env->buffer = 0;
99 utf8proc_ruby_map_error(result);
100 return Qnil; /* needed to prevent problems with optimization */
101 }
102 result = utf8proc_reencode(env->buffer, result, options);
103 if (result < 0) {
104 free(env->buffer);
105 env->buffer = 0;
106 utf8proc_ruby_map_error(result);
107 return Qnil; /* needed to prevent problems with optimization */
108 }
109 retval = rb_str_new((char *)env->buffer, result);
110 free(env->buffer);
111 env->buffer = 0;
112 return retval;
113 }
115 static VALUE utf8proc_ruby_char(VALUE self, VALUE code_param) {
116 char buffer[4];
117 ssize_t result;
118 int uc;
119 uc = NUM2INT(code_param);
120 if (!utf8proc_codepoint_valid(uc))
121 rb_raise(rb_eArgError, "Invalid Unicode code point");
122 result = utf8proc_encode_char(uc, buffer);
123 return rb_str_new(buffer, result);
124 }
126 #define register_utf8proc_option(sym, field) \
127 rb_hash_aset(utf8proc_ruby_options, ID2SYM(rb_intern(sym)), INT2FIX(field))
129 void Init_utf8proc_native() {
130 utf8proc_ruby_module = rb_define_module("Utf8Proc");
131 rb_define_module_function(utf8proc_ruby_module, "utf8map",
132 utf8proc_ruby_map, 2);
133 rb_define_module_function(utf8proc_ruby_module, "utf8char",
134 utf8proc_ruby_char, 1);
135 utf8proc_ruby_eUnicodeError = rb_define_class_under(utf8proc_ruby_module,
136 "UnicodeError", rb_eStandardError);
137 utf8proc_ruby_eInvalidUtf8Error = rb_define_class_under(
138 utf8proc_ruby_module, "InvalidUtf8Error", utf8proc_ruby_eUnicodeError);
139 utf8proc_ruby_eCodeNotAssignedError = rb_define_class_under(
140 utf8proc_ruby_module, "CodeNotAssignedError",
141 utf8proc_ruby_eUnicodeError);
142 utf8proc_ruby_options = rb_hash_new();
143 register_utf8proc_option("stable", UTF8PROC_STABLE);
144 register_utf8proc_option("compat", UTF8PROC_COMPAT);
145 register_utf8proc_option("compose", UTF8PROC_COMPOSE);
146 register_utf8proc_option("decompose", UTF8PROC_DECOMPOSE);
147 register_utf8proc_option("ignore", UTF8PROC_IGNORE);
148 register_utf8proc_option("rejectna", UTF8PROC_REJECTNA);
149 register_utf8proc_option("nlf2ls", UTF8PROC_NLF2LS);
150 register_utf8proc_option("nlf2ps", UTF8PROC_NLF2PS);
151 register_utf8proc_option("nlf2lf", UTF8PROC_NLF2LF);
152 register_utf8proc_option("stripcc", UTF8PROC_STRIPCC);
153 register_utf8proc_option("casefold", UTF8PROC_CASEFOLD);
154 register_utf8proc_option("charbound", UTF8PROC_CHARBOUND);
155 register_utf8proc_option("lump", UTF8PROC_LUMP);
156 register_utf8proc_option("stripmark", UTF8PROC_STRIPMARK);
157 OBJ_FREEZE(utf8proc_ruby_options);
158 rb_define_const(utf8proc_ruby_module, "Options", utf8proc_ruby_options);
159 }