1.1 --- /dev/null Thu Jan 01 00:00:00 1970 +0000
1.2 +++ b/intl/icu/source/i18n/unesctrn.cpp Wed Dec 31 06:09:35 2014 +0100
1.3 @@ -0,0 +1,291 @@
1.4 +/*
1.5 + **********************************************************************
1.6 + * Copyright (c) 2001-2011, International Business Machines
1.7 + * Corporation and others. All Rights Reserved.
1.8 + **********************************************************************
1.9 + * Date Name Description
1.10 + * 11/19/2001 aliu Creation.
1.11 + **********************************************************************
1.12 + */
1.13 +
1.14 +#include "unicode/utypes.h"
1.15 +
1.16 +#if !UCONFIG_NO_TRANSLITERATION
1.17 +
1.18 +#include "unicode/uchar.h"
1.19 +#include "unicode/utf16.h"
1.20 +#include "unesctrn.h"
1.21 +#include "util.h"
1.22 +
1.23 +#include "cmemory.h"
1.24 +
1.25 +U_NAMESPACE_BEGIN
1.26 +
1.27 +/**
1.28 + * Special character marking the end of the spec[] array.
1.29 + */
1.30 +static const UChar END = 0xFFFF;
1.31 +
1.32 +// Unicode: "U+10FFFF" hex, min=4, max=6
1.33 +static const UChar SPEC_Unicode[] = {
1.34 + 2, 0, 16, 4, 6, 85/*U*/, 43/*+*/,
1.35 + END
1.36 +};
1.37 +
1.38 +// Java: "\\uFFFF" hex, min=4, max=4
1.39 +static const UChar SPEC_Java[] = {
1.40 + 2, 0, 16, 4, 4, 92/*\*/, 117/*u*/,
1.41 + END
1.42 +};
1.43 +
1.44 +// C: "\\uFFFF" hex, min=4, max=4; \\U0010FFFF hex, min=8, max=8
1.45 +static const UChar SPEC_C[] = {
1.46 + 2, 0, 16, 4, 4, 92/*\*/, 117/*u*/,
1.47 + 2, 0, 16, 8, 8, 92/*\*/, 85/*U*/,
1.48 + END
1.49 +};
1.50 +
1.51 +// XML: "" hex, min=1, max=6
1.52 +static const UChar SPEC_XML[] = {
1.53 + 3, 1, 16, 1, 6, 38/*&*/, 35/*#*/, 120/*x*/, 59/*;*/,
1.54 + END
1.55 +};
1.56 +
1.57 +// XML10: "" dec, min=1, max=7 (not really "Hex-Any")
1.58 +static const UChar SPEC_XML10[] = {
1.59 + 2, 1, 10, 1, 7, 38/*&*/, 35/*#*/, 59/*;*/,
1.60 + END
1.61 +};
1.62 +
1.63 +// Perl: "\\x{263A}" hex, min=1, max=6
1.64 +static const UChar SPEC_Perl[] = {
1.65 + 3, 1, 16, 1, 6, 92/*\*/, 120/*x*/, 123/*{*/, 125/*}*/,
1.66 + END
1.67 +};
1.68 +
1.69 +// All: Java, C, Perl, XML, XML10, Unicode
1.70 +static const UChar SPEC_Any[] = {
1.71 + 2, 0, 16, 4, 6, 85/*U*/, 43/*+*/, // Unicode
1.72 + 2, 0, 16, 4, 4, 92/*\*/, 117/*u*/, // Java
1.73 + 2, 0, 16, 8, 8, 92/*\*/, 85/*U*/, // C (surrogates)
1.74 + 3, 1, 16, 1, 6, 38/*&*/, 35/*#*/, 120/*x*/, 59/*;*/, // XML
1.75 + 2, 1, 10, 1, 7, 38/*&*/, 35/*#*/, 59/*;*/, // XML10
1.76 + 3, 1, 16, 1, 6, 92/*\*/, 120/*x*/, 123/*{*/, 125/*}*/, // Perl
1.77 + END
1.78 +};
1.79 +
1.80 +UOBJECT_DEFINE_RTTI_IMPLEMENTATION(UnescapeTransliterator)
1.81 +
1.82 +static UChar* copySpec(const UChar* spec) {
1.83 + int32_t len = 0;
1.84 + while (spec[len] != END) {
1.85 + ++len;
1.86 + }
1.87 + ++len;
1.88 + UChar *result = (UChar *)uprv_malloc(len*sizeof(UChar));
1.89 + // Check for memory allocation error.
1.90 + if (result != NULL) {
1.91 + uprv_memcpy(result, spec, len*sizeof(result[0]));
1.92 + }
1.93 + return result;
1.94 +}
1.95 +
1.96 +/**
1.97 + * Factory methods. Ignore the context.
1.98 + */
1.99 +static Transliterator* _createUnicode(const UnicodeString& ID, Transliterator::Token /*context*/) {
1.100 + return new UnescapeTransliterator(ID, SPEC_Unicode);
1.101 +}
1.102 +static Transliterator* _createJava(const UnicodeString& ID, Transliterator::Token /*context*/) {
1.103 + return new UnescapeTransliterator(ID, SPEC_Java);
1.104 +}
1.105 +static Transliterator* _createC(const UnicodeString& ID, Transliterator::Token /*context*/) {
1.106 + return new UnescapeTransliterator(ID, SPEC_C);
1.107 +}
1.108 +static Transliterator* _createXML(const UnicodeString& ID, Transliterator::Token /*context*/) {
1.109 + return new UnescapeTransliterator(ID, SPEC_XML);
1.110 +}
1.111 +static Transliterator* _createXML10(const UnicodeString& ID, Transliterator::Token /*context*/) {
1.112 + return new UnescapeTransliterator(ID, SPEC_XML10);
1.113 +}
1.114 +static Transliterator* _createPerl(const UnicodeString& ID, Transliterator::Token /*context*/) {
1.115 + return new UnescapeTransliterator(ID, SPEC_Perl);
1.116 +}
1.117 +static Transliterator* _createAny(const UnicodeString& ID, Transliterator::Token /*context*/) {
1.118 + return new UnescapeTransliterator(ID, SPEC_Any);
1.119 +}
1.120 +
1.121 +/**
1.122 + * Registers standard variants with the system. Called by
1.123 + * Transliterator during initialization.
1.124 + */
1.125 +void UnescapeTransliterator::registerIDs() {
1.126 + Token t = integerToken(0);
1.127 +
1.128 + Transliterator::_registerFactory(UNICODE_STRING_SIMPLE("Hex-Any/Unicode"), _createUnicode, t);
1.129 +
1.130 + Transliterator::_registerFactory(UNICODE_STRING_SIMPLE("Hex-Any/Java"), _createJava, t);
1.131 +
1.132 + Transliterator::_registerFactory(UNICODE_STRING_SIMPLE("Hex-Any/C"), _createC, t);
1.133 +
1.134 + Transliterator::_registerFactory(UNICODE_STRING_SIMPLE("Hex-Any/XML"), _createXML, t);
1.135 +
1.136 + Transliterator::_registerFactory(UNICODE_STRING_SIMPLE("Hex-Any/XML10"), _createXML10, t);
1.137 +
1.138 + Transliterator::_registerFactory(UNICODE_STRING_SIMPLE("Hex-Any/Perl"), _createPerl, t);
1.139 +
1.140 + Transliterator::_registerFactory(UNICODE_STRING_SIMPLE("Hex-Any"), _createAny, t);
1.141 +}
1.142 +
1.143 +/**
1.144 + * Constructor. Takes the encoded spec array.
1.145 + */
1.146 +UnescapeTransliterator::UnescapeTransliterator(const UnicodeString& newID,
1.147 + const UChar *newSpec) :
1.148 + Transliterator(newID, NULL)
1.149 +{
1.150 + this->spec = copySpec(newSpec);
1.151 +}
1.152 +
1.153 +/**
1.154 + * Copy constructor.
1.155 + */
1.156 +UnescapeTransliterator::UnescapeTransliterator(const UnescapeTransliterator& o) :
1.157 + Transliterator(o) {
1.158 + this->spec = copySpec(o.spec);
1.159 +}
1.160 +
1.161 +UnescapeTransliterator::~UnescapeTransliterator() {
1.162 + uprv_free(spec);
1.163 +}
1.164 +
1.165 +/**
1.166 + * Transliterator API.
1.167 + */
1.168 +Transliterator* UnescapeTransliterator::clone() const {
1.169 + return new UnescapeTransliterator(*this);
1.170 +}
1.171 +
1.172 +/**
1.173 + * Implements {@link Transliterator#handleTransliterate}.
1.174 + */
1.175 +void UnescapeTransliterator::handleTransliterate(Replaceable& text, UTransPosition& pos,
1.176 + UBool isIncremental) const {
1.177 + int32_t start = pos.start;
1.178 + int32_t limit = pos.limit;
1.179 + int32_t i, j, ipat;
1.180 +
1.181 + while (start < limit) {
1.182 + // Loop over the forms in spec[]. Exit this loop when we
1.183 + // match one of the specs. Exit the outer loop if a
1.184 + // partial match is detected and isIncremental is true.
1.185 + for (j=0, ipat=0; spec[ipat] != END; ++j) {
1.186 +
1.187 + // Read the header
1.188 + int32_t prefixLen = spec[ipat++];
1.189 + int32_t suffixLen = spec[ipat++];
1.190 + int8_t radix = (int8_t) spec[ipat++];
1.191 + int32_t minDigits = spec[ipat++];
1.192 + int32_t maxDigits = spec[ipat++];
1.193 +
1.194 + // s is a copy of start that is advanced over the
1.195 + // characters as we parse them.
1.196 + int32_t s = start;
1.197 + UBool match = TRUE;
1.198 +
1.199 + for (i=0; i<prefixLen; ++i) {
1.200 + if (s >= limit) {
1.201 + if (i > 0) {
1.202 + // We've already matched a character. This is
1.203 + // a partial match, so we return if in
1.204 + // incremental mode. In non-incremental mode,
1.205 + // go to the next spec.
1.206 + if (isIncremental) {
1.207 + goto exit;
1.208 + }
1.209 + match = FALSE;
1.210 + break;
1.211 + }
1.212 + }
1.213 + UChar c = text.charAt(s++);
1.214 + if (c != spec[ipat + i]) {
1.215 + match = FALSE;
1.216 + break;
1.217 + }
1.218 + }
1.219 +
1.220 + if (match) {
1.221 + UChar32 u = 0;
1.222 + int32_t digitCount = 0;
1.223 + for (;;) {
1.224 + if (s >= limit) {
1.225 + // Check for partial match in incremental mode.
1.226 + if (s > start && isIncremental) {
1.227 + goto exit;
1.228 + }
1.229 + break;
1.230 + }
1.231 + UChar32 ch = text.char32At(s);
1.232 + int32_t digit = u_digit(ch, radix);
1.233 + if (digit < 0) {
1.234 + break;
1.235 + }
1.236 + s += U16_LENGTH(ch);
1.237 + u = (u * radix) + digit;
1.238 + if (++digitCount == maxDigits) {
1.239 + break;
1.240 + }
1.241 + }
1.242 +
1.243 + match = (digitCount >= minDigits);
1.244 +
1.245 + if (match) {
1.246 + for (i=0; i<suffixLen; ++i) {
1.247 + if (s >= limit) {
1.248 + // Check for partial match in incremental mode.
1.249 + if (s > start && isIncremental) {
1.250 + goto exit;
1.251 + }
1.252 + match = FALSE;
1.253 + break;
1.254 + }
1.255 + UChar c = text.charAt(s++);
1.256 + if (c != spec[ipat + prefixLen + i]) {
1.257 + match = FALSE;
1.258 + break;
1.259 + }
1.260 + }
1.261 +
1.262 + if (match) {
1.263 + // At this point, we have a match
1.264 + UnicodeString str(u);
1.265 + text.handleReplaceBetween(start, s, str);
1.266 + limit -= s - start - str.length();
1.267 + // The following break statement leaves the
1.268 + // loop that is traversing the forms in
1.269 + // spec[]. We then parse the next input
1.270 + // character.
1.271 + break;
1.272 + }
1.273 + }
1.274 + }
1.275 +
1.276 + ipat += prefixLen + suffixLen;
1.277 + }
1.278 +
1.279 + if (start < limit) {
1.280 + start += U16_LENGTH(text.char32At(start));
1.281 + }
1.282 + }
1.283 +
1.284 + exit:
1.285 + pos.contextLimit += limit - pos.limit;
1.286 + pos.limit = limit;
1.287 + pos.start = start;
1.288 +}
1.289 +
1.290 +U_NAMESPACE_END
1.291 +
1.292 +#endif /* #if !UCONFIG_NO_TRANSLITERATION */
1.293 +
1.294 +//eof
