1// © 2016 and later: Unicode, Inc. and others.
2// License & terms of use: http://www.unicode.org/copyright.html
3/*
4**********************************************************************
5* Copyright (C) 1997-2014, International Business Machines
6* Corporation and others. All Rights Reserved.
7**********************************************************************
8*
9* File USCRIPT.C
10*
11* Modification History:
12*
13* Date Name Description
14* 07/06/2001 Ram Creation.
15******************************************************************************
16*/
17
18#include "unicode/uchar.h"
19#include "unicode/uscript.h"
20#include "unicode/uloc.h"
21#include "bytesinkutil.h"
22#include "charstr.h"
23#include "cmemory.h"
24#include "cstring.h"
25#include "ulocimp.h"
26
27static const UScriptCode JAPANESE[3] = { USCRIPT_KATAKANA, USCRIPT_HIRAGANA, USCRIPT_HAN };
28static const UScriptCode KOREAN[2] = { USCRIPT_HANGUL, USCRIPT_HAN };
29static const UScriptCode HAN_BOPO[2] = { USCRIPT_HAN, USCRIPT_BOPOMOFO };
30
31static int32_t
32setCodes(const UScriptCode *src, int32_t length,
33 UScriptCode *dest, int32_t capacity, UErrorCode *err) {
34 int32_t i;
35 if(U_FAILURE(*err)) { return 0; }
36 if(length > capacity) {
37 *err = U_BUFFER_OVERFLOW_ERROR;
38 return length;
39 }
40 for(i = 0; i < length; ++i) {
41 dest[i] = src[i];
42 }
43 return length;
44}
45
46static int32_t
47setOneCode(UScriptCode script, UScriptCode *scripts, int32_t capacity, UErrorCode *err) {
48 if(U_FAILURE(*err)) { return 0; }
49 if(1 > capacity) {
50 *err = U_BUFFER_OVERFLOW_ERROR;
51 return 1;
52 }
53 scripts[0] = script;
54 return 1;
55}
56
57static int32_t
58getCodesFromLocale(const char *locale,
59 UScriptCode *scripts, int32_t capacity, UErrorCode *err) {
60 UErrorCode internalErrorCode = U_ZERO_ERROR;
61 char lang[8] = {0};
62 char script[8] = {0};
63 int32_t scriptLength;
64 if(U_FAILURE(*err)) { return 0; }
65 // Multi-script languages, equivalent to the LocaleScript data
66 // that we used to load from locale resource bundles.
67 /*length = */ uloc_getLanguage(locale, lang, UPRV_LENGTHOF(lang), &internalErrorCode);
68 if(U_FAILURE(internalErrorCode) || internalErrorCode == U_STRING_NOT_TERMINATED_WARNING) {
69 return 0;
70 }
71 if(0 == uprv_strcmp(lang, "ja")) {
72 return setCodes(JAPANESE, UPRV_LENGTHOF(JAPANESE), scripts, capacity, err);
73 }
74 if(0 == uprv_strcmp(lang, "ko")) {
75 return setCodes(KOREAN, UPRV_LENGTHOF(KOREAN), scripts, capacity, err);
76 }
77 scriptLength = uloc_getScript(locale, script, UPRV_LENGTHOF(script), &internalErrorCode);
78 if(U_FAILURE(internalErrorCode) || internalErrorCode == U_STRING_NOT_TERMINATED_WARNING) {
79 return 0;
80 }
81 if(0 == uprv_strcmp(lang, "zh") && 0 == uprv_strcmp(script, "Hant")) {
82 return setCodes(HAN_BOPO, UPRV_LENGTHOF(HAN_BOPO), scripts, capacity, err);
83 }
84 // Explicit script code.
85 if(scriptLength != 0) {
86 UScriptCode scriptCode = (UScriptCode)u_getPropertyValueEnum(UCHAR_SCRIPT, script);
87 if(scriptCode != USCRIPT_INVALID_CODE) {
88 if(scriptCode == USCRIPT_SIMPLIFIED_HAN || scriptCode == USCRIPT_TRADITIONAL_HAN) {
89 scriptCode = USCRIPT_HAN;
90 }
91 return setOneCode(scriptCode, scripts, capacity, err);
92 }
93 }
94 return 0;
95}
96
97/* TODO: this is a bad API and should be deprecated, ticket #11141 */
98U_CAPI int32_t U_EXPORT2
99uscript_getCode(const char* nameOrAbbrOrLocale,
100 UScriptCode* fillIn,
101 int32_t capacity,
102 UErrorCode* err){
103 UBool triedCode;
104 UErrorCode internalErrorCode;
105 int32_t length;
106
107 if(U_FAILURE(*err)) {
108 return 0;
109 }
110 if(nameOrAbbrOrLocale==nullptr ||
111 (fillIn == nullptr ? capacity != 0 : capacity < 0)) {
112 *err = U_ILLEGAL_ARGUMENT_ERROR;
113 return 0;
114 }
115
116 triedCode = false;
117 const char* lastSepPtr = uprv_strrchr(nameOrAbbrOrLocale, '-');
118 if (lastSepPtr==nullptr) {
119 lastSepPtr = uprv_strrchr(nameOrAbbrOrLocale, '_');
120 }
121 // Favor interpretation of nameOrAbbrOrLocale as a script alias if either
122 // 1. nameOrAbbrOrLocale does not contain -/_. Handles Han, Mro, Nko, etc.
123 // 2. The last instance of -/_ is at offset 3, and the portion after that is
124 // longer than 4 characters (i.e. not a script or region code). This handles
125 // Old_Hungarian, Old_Italic, etc. ("old" is a valid language code)
126 // 3. The last instance of -/_ is at offset 7, and the portion after that is
127 // 3 characters. This handles New_Tai_Lue ("new" is a valid language code).
128 if (lastSepPtr==nullptr
129 || (lastSepPtr-nameOrAbbrOrLocale == 3 && uprv_strlen(nameOrAbbrOrLocale) > 8)
130 || (lastSepPtr-nameOrAbbrOrLocale == 7 && uprv_strlen(nameOrAbbrOrLocale) == 11) ) {
131 /* try long and abbreviated script names first */
132 UScriptCode code = (UScriptCode) u_getPropertyValueEnum(UCHAR_SCRIPT, nameOrAbbrOrLocale);
133 if(code!=USCRIPT_INVALID_CODE) {
134 return setOneCode(code, fillIn, capacity, err);
135 }
136 triedCode = true;
137 }
138 internalErrorCode = U_ZERO_ERROR;
139 length = getCodesFromLocale(nameOrAbbrOrLocale, fillIn, capacity, err);
140 if(U_FAILURE(*err) || length != 0) {
141 return length;
142 }
143 icu::CharString likely;
144 {
145 icu::CharStringByteSink sink(&likely);
146 ulocimp_addLikelySubtags(nameOrAbbrOrLocale, sink, &internalErrorCode);
147 }
148 if(U_SUCCESS(internalErrorCode) && internalErrorCode != U_STRING_NOT_TERMINATED_WARNING) {
149 length = getCodesFromLocale(likely.data(), fillIn, capacity, err);
150 if(U_FAILURE(*err) || length != 0) {
151 return length;
152 }
153 }
154 if(!triedCode) {
155 /* still not found .. try long and abbreviated script names again */
156 UScriptCode code = (UScriptCode) u_getPropertyValueEnum(UCHAR_SCRIPT, nameOrAbbrOrLocale);
157 if(code!=USCRIPT_INVALID_CODE) {
158 return setOneCode(code, fillIn, capacity, err);
159 }
160 }
161 return 0;
162}
163