1 | // © 2016 and later: Unicode, Inc. and others. |
2 | // License & terms of use: http://www.unicode.org/copyright.html |
3 | /* |
4 | ********************************************************************** |
5 | * Copyright (c) 2001-2011, International Business Machines |
6 | * Corporation and others. All Rights Reserved. |
7 | ********************************************************************** |
8 | * Date Name Description |
9 | * 11/19/2001 aliu Creation. |
10 | ********************************************************************** |
11 | */ |
12 | |
13 | #include "unicode/utypes.h" |
14 | |
15 | #if !UCONFIG_NO_TRANSLITERATION |
16 | |
17 | #include "unicode/uchar.h" |
18 | #include "unicode/utf16.h" |
19 | #include "unesctrn.h" |
20 | #include "util.h" |
21 | |
22 | #include "cmemory.h" |
23 | |
24 | U_NAMESPACE_BEGIN |
25 | |
26 | /** |
27 | * Special character marking the end of the spec[] array. |
28 | */ |
29 | static const UChar END = 0xFFFF; |
30 | |
31 | // Unicode: "U+10FFFF" hex, min=4, max=6 |
32 | static const UChar SPEC_Unicode[] = { |
33 | 2, 0, 16, 4, 6, 85/*U*/, 43/*+*/, |
34 | END |
35 | }; |
36 | |
37 | // Java: "\\uFFFF" hex, min=4, max=4 |
38 | static const UChar SPEC_Java[] = { |
39 | 2, 0, 16, 4, 4, 92/*\*/, 117/*u*/, |
40 | END |
41 | }; |
42 | |
43 | // C: "\\uFFFF" hex, min=4, max=4; \\U0010FFFF hex, min=8, max=8 |
44 | static const UChar SPEC_C[] = { |
45 | 2, 0, 16, 4, 4, 92/*\*/, 117/*u*/, |
46 | 2, 0, 16, 8, 8, 92/*\*/, 85/*U*/, |
47 | END |
48 | }; |
49 | |
50 | // XML: "" hex, min=1, max=6 |
51 | static const UChar SPEC_XML[] = { |
52 | 3, 1, 16, 1, 6, 38/*&*/, 35/*#*/, 120/*x*/, 59/*;*/, |
53 | END |
54 | }; |
55 | |
56 | // XML10: "" dec, min=1, max=7 (not really "Hex-Any") |
57 | static const UChar SPEC_XML10[] = { |
58 | 2, 1, 10, 1, 7, 38/*&*/, 35/*#*/, 59/*;*/, |
59 | END |
60 | }; |
61 | |
62 | // Perl: "\\x{263A}" hex, min=1, max=6 |
63 | static const UChar SPEC_Perl[] = { |
64 | 3, 1, 16, 1, 6, 92/*\*/, 120/*x*/, 123/*{*/, 125/*}*/, |
65 | END |
66 | }; |
67 | |
68 | // All: Java, C, Perl, XML, XML10, Unicode |
69 | static const UChar SPEC_Any[] = { |
70 | 2, 0, 16, 4, 6, 85/*U*/, 43/*+*/, // Unicode |
71 | 2, 0, 16, 4, 4, 92/*\*/, 117/*u*/, // Java |
72 | 2, 0, 16, 8, 8, 92/*\*/, 85/*U*/, // C (surrogates) |
73 | 3, 1, 16, 1, 6, 38/*&*/, 35/*#*/, 120/*x*/, 59/*;*/, // XML |
74 | 2, 1, 10, 1, 7, 38/*&*/, 35/*#*/, 59/*;*/, // XML10 |
75 | 3, 1, 16, 1, 6, 92/*\*/, 120/*x*/, 123/*{*/, 125/*}*/, // Perl |
76 | END |
77 | }; |
78 | |
79 | UOBJECT_DEFINE_RTTI_IMPLEMENTATION(UnescapeTransliterator) |
80 | |
81 | static UChar* copySpec(const UChar* spec) { |
82 | int32_t len = 0; |
83 | while (spec[len] != END) { |
84 | ++len; |
85 | } |
86 | ++len; |
87 | UChar *result = (UChar *)uprv_malloc(len*sizeof(UChar)); |
88 | // Check for memory allocation error. |
89 | if (result != NULL) { |
90 | uprv_memcpy(result, spec, (size_t)len*sizeof(result[0])); |
91 | } |
92 | return result; |
93 | } |
94 | |
95 | /** |
96 | * Factory methods. Ignore the context. |
97 | */ |
98 | static Transliterator* _createUnicode(const UnicodeString& ID, Transliterator::Token /*context*/) { |
99 | return new UnescapeTransliterator(ID, SPEC_Unicode); |
100 | } |
101 | static Transliterator* _createJava(const UnicodeString& ID, Transliterator::Token /*context*/) { |
102 | return new UnescapeTransliterator(ID, SPEC_Java); |
103 | } |
104 | static Transliterator* _createC(const UnicodeString& ID, Transliterator::Token /*context*/) { |
105 | return new UnescapeTransliterator(ID, SPEC_C); |
106 | } |
107 | static Transliterator* _createXML(const UnicodeString& ID, Transliterator::Token /*context*/) { |
108 | return new UnescapeTransliterator(ID, SPEC_XML); |
109 | } |
110 | static Transliterator* _createXML10(const UnicodeString& ID, Transliterator::Token /*context*/) { |
111 | return new UnescapeTransliterator(ID, SPEC_XML10); |
112 | } |
113 | static Transliterator* _createPerl(const UnicodeString& ID, Transliterator::Token /*context*/) { |
114 | return new UnescapeTransliterator(ID, SPEC_Perl); |
115 | } |
116 | static Transliterator* _createAny(const UnicodeString& ID, Transliterator::Token /*context*/) { |
117 | return new UnescapeTransliterator(ID, SPEC_Any); |
118 | } |
119 | |
120 | /** |
121 | * Registers standard variants with the system. Called by |
122 | * Transliterator during initialization. |
123 | */ |
124 | void UnescapeTransliterator::registerIDs() { |
125 | Token t = integerToken(0); |
126 | |
127 | Transliterator::_registerFactory(UNICODE_STRING_SIMPLE("Hex-Any/Unicode" ), _createUnicode, t); |
128 | |
129 | Transliterator::_registerFactory(UNICODE_STRING_SIMPLE("Hex-Any/Java" ), _createJava, t); |
130 | |
131 | Transliterator::_registerFactory(UNICODE_STRING_SIMPLE("Hex-Any/C" ), _createC, t); |
132 | |
133 | Transliterator::_registerFactory(UNICODE_STRING_SIMPLE("Hex-Any/XML" ), _createXML, t); |
134 | |
135 | Transliterator::_registerFactory(UNICODE_STRING_SIMPLE("Hex-Any/XML10" ), _createXML10, t); |
136 | |
137 | Transliterator::_registerFactory(UNICODE_STRING_SIMPLE("Hex-Any/Perl" ), _createPerl, t); |
138 | |
139 | Transliterator::_registerFactory(UNICODE_STRING_SIMPLE("Hex-Any" ), _createAny, t); |
140 | } |
141 | |
142 | /** |
143 | * Constructor. Takes the encoded spec array. |
144 | */ |
145 | UnescapeTransliterator::UnescapeTransliterator(const UnicodeString& newID, |
146 | const UChar *newSpec) : |
147 | Transliterator(newID, NULL) |
148 | { |
149 | this->spec = copySpec(newSpec); |
150 | } |
151 | |
152 | /** |
153 | * Copy constructor. |
154 | */ |
155 | UnescapeTransliterator::UnescapeTransliterator(const UnescapeTransliterator& o) : |
156 | Transliterator(o) { |
157 | this->spec = copySpec(o.spec); |
158 | } |
159 | |
160 | UnescapeTransliterator::~UnescapeTransliterator() { |
161 | uprv_free(spec); |
162 | } |
163 | |
164 | /** |
165 | * Transliterator API. |
166 | */ |
167 | UnescapeTransliterator* UnescapeTransliterator::clone() const { |
168 | return new UnescapeTransliterator(*this); |
169 | } |
170 | |
171 | /** |
172 | * Implements {@link Transliterator#handleTransliterate}. |
173 | */ |
174 | void UnescapeTransliterator::handleTransliterate(Replaceable& text, UTransPosition& pos, |
175 | UBool isIncremental) const { |
176 | int32_t start = pos.start; |
177 | int32_t limit = pos.limit; |
178 | int32_t i, j, ipat; |
179 | |
180 | while (start < limit) { |
181 | // Loop over the forms in spec[]. Exit this loop when we |
182 | // match one of the specs. Exit the outer loop if a |
183 | // partial match is detected and isIncremental is true. |
184 | for (j=0, ipat=0; spec[ipat] != END; ++j) { |
185 | |
186 | // Read the header |
187 | int32_t prefixLen = spec[ipat++]; |
188 | int32_t suffixLen = spec[ipat++]; |
189 | int8_t radix = (int8_t) spec[ipat++]; |
190 | int32_t minDigits = spec[ipat++]; |
191 | int32_t maxDigits = spec[ipat++]; |
192 | |
193 | // s is a copy of start that is advanced over the |
194 | // characters as we parse them. |
195 | int32_t s = start; |
196 | UBool match = TRUE; |
197 | |
198 | for (i=0; i<prefixLen; ++i) { |
199 | if (s >= limit) { |
200 | if (i > 0) { |
201 | // We've already matched a character. This is |
202 | // a partial match, so we return if in |
203 | // incremental mode. In non-incremental mode, |
204 | // go to the next spec. |
205 | if (isIncremental) { |
206 | goto exit; |
207 | } |
208 | match = FALSE; |
209 | break; |
210 | } |
211 | } |
212 | UChar c = text.charAt(s++); |
213 | if (c != spec[ipat + i]) { |
214 | match = FALSE; |
215 | break; |
216 | } |
217 | } |
218 | |
219 | if (match) { |
220 | UChar32 u = 0; |
221 | int32_t digitCount = 0; |
222 | for (;;) { |
223 | if (s >= limit) { |
224 | // Check for partial match in incremental mode. |
225 | if (s > start && isIncremental) { |
226 | goto exit; |
227 | } |
228 | break; |
229 | } |
230 | UChar32 ch = text.char32At(s); |
231 | int32_t digit = u_digit(ch, radix); |
232 | if (digit < 0) { |
233 | break; |
234 | } |
235 | s += U16_LENGTH(ch); |
236 | u = (u * radix) + digit; |
237 | if (++digitCount == maxDigits) { |
238 | break; |
239 | } |
240 | } |
241 | |
242 | match = (digitCount >= minDigits); |
243 | |
244 | if (match) { |
245 | for (i=0; i<suffixLen; ++i) { |
246 | if (s >= limit) { |
247 | // Check for partial match in incremental mode. |
248 | if (s > start && isIncremental) { |
249 | goto exit; |
250 | } |
251 | match = FALSE; |
252 | break; |
253 | } |
254 | UChar c = text.charAt(s++); |
255 | if (c != spec[ipat + prefixLen + i]) { |
256 | match = FALSE; |
257 | break; |
258 | } |
259 | } |
260 | |
261 | if (match) { |
262 | // At this point, we have a match |
263 | UnicodeString str(u); |
264 | text.handleReplaceBetween(start, s, str); |
265 | limit -= s - start - str.length(); |
266 | // The following break statement leaves the |
267 | // loop that is traversing the forms in |
268 | // spec[]. We then parse the next input |
269 | // character. |
270 | break; |
271 | } |
272 | } |
273 | } |
274 | |
275 | ipat += prefixLen + suffixLen; |
276 | } |
277 | |
278 | if (start < limit) { |
279 | start += U16_LENGTH(text.char32At(start)); |
280 | } |
281 | } |
282 | |
283 | exit: |
284 | pos.contextLimit += limit - pos.limit; |
285 | pos.limit = limit; |
286 | pos.start = start; |
287 | } |
288 | |
289 | U_NAMESPACE_END |
290 | |
291 | #endif /* #if !UCONFIG_NO_TRANSLITERATION */ |
292 | |
293 | //eof |
294 | |