OLD | NEW |
(Empty) | |
| 1 /* |
| 2 ******************************************************************************* |
| 3 * |
| 4 * Copyright (C) 2001-2010, International Business Machines |
| 5 * Corporation and others. All Rights Reserved. |
| 6 * |
| 7 ******************************************************************************* |
| 8 * file name: casetrn.cpp |
| 9 * encoding: US-ASCII |
| 10 * tab size: 8 (not used) |
| 11 * indentation:4 |
| 12 * |
| 13 * created on: 2004sep03 |
| 14 * created by: Markus W. Scherer |
| 15 * |
| 16 * Implementation class for lower-/upper-/title-casing transliterators. |
| 17 */ |
| 18 |
| 19 #include "unicode/utypes.h" |
| 20 |
| 21 #if !UCONFIG_NO_TRANSLITERATION |
| 22 |
| 23 #include "unicode/uchar.h" |
| 24 #include "unicode/ustring.h" |
| 25 #include "tolowtrn.h" |
| 26 #include "ucase.h" |
| 27 #include "cpputils.h" |
| 28 |
| 29 /* case context iterator using a Replaceable */ |
| 30 U_CFUNC UChar32 U_CALLCONV |
| 31 utrans_rep_caseContextIterator(void *context, int8_t dir) |
| 32 { |
| 33 U_NAMESPACE_USE |
| 34 |
| 35 UCaseContext *csc=(UCaseContext *)context; |
| 36 Replaceable *rep=(Replaceable *)csc->p; |
| 37 UChar32 c; |
| 38 |
| 39 if(dir<0) { |
| 40 /* reset for backward iteration */ |
| 41 csc->index=csc->cpStart; |
| 42 csc->dir=dir; |
| 43 } else if(dir>0) { |
| 44 /* reset for forward iteration */ |
| 45 csc->index=csc->cpLimit; |
| 46 csc->dir=dir; |
| 47 } else { |
| 48 /* continue current iteration direction */ |
| 49 dir=csc->dir; |
| 50 } |
| 51 |
| 52 // automatically adjust start and limit if the Replaceable disagrees |
| 53 // with the original values |
| 54 if(dir<0) { |
| 55 if(csc->start<csc->index) { |
| 56 c=rep->char32At(csc->index-1); |
| 57 if(c<0) { |
| 58 csc->start=csc->index; |
| 59 } else { |
| 60 csc->index-=U16_LENGTH(c); |
| 61 return c; |
| 62 } |
| 63 } |
| 64 } else { |
| 65 // detect, and store in csc->b1, if we hit the limit |
| 66 if(csc->index<csc->limit) { |
| 67 c=rep->char32At(csc->index); |
| 68 if(c<0) { |
| 69 csc->limit=csc->index; |
| 70 csc->b1=TRUE; |
| 71 } else { |
| 72 csc->index+=U16_LENGTH(c); |
| 73 return c; |
| 74 } |
| 75 } else { |
| 76 csc->b1=TRUE; |
| 77 } |
| 78 } |
| 79 return U_SENTINEL; |
| 80 } |
| 81 |
| 82 U_NAMESPACE_BEGIN |
| 83 |
| 84 UOBJECT_DEFINE_ABSTRACT_RTTI_IMPLEMENTATION(CaseMapTransliterator) |
| 85 |
| 86 /** |
| 87 * Constructs a transliterator. |
| 88 */ |
| 89 CaseMapTransliterator::CaseMapTransliterator(const UnicodeString &id, UCaseMapFu
ll *map) : |
| 90 Transliterator(id, 0), |
| 91 fCsp(ucase_getSingleton()), |
| 92 fMap(map) |
| 93 { |
| 94 // TODO test incremental mode with context-sensitive text (e.g. greek sigma) |
| 95 // TODO need to call setMaximumContextLength()?! |
| 96 } |
| 97 |
| 98 /** |
| 99 * Destructor. |
| 100 */ |
| 101 CaseMapTransliterator::~CaseMapTransliterator() { |
| 102 } |
| 103 |
| 104 /** |
| 105 * Copy constructor. |
| 106 */ |
| 107 CaseMapTransliterator::CaseMapTransliterator(const CaseMapTransliterator& o) : |
| 108 Transliterator(o), |
| 109 fCsp(o.fCsp), fMap(o.fMap) |
| 110 { |
| 111 } |
| 112 |
| 113 /** |
| 114 * Assignment operator. |
| 115 */ |
| 116 /*CaseMapTransliterator& CaseMapTransliterator::operator=(const CaseMapTranslite
rator& o) { |
| 117 Transliterator::operator=(o); |
| 118 fCsp = o.fCsp; |
| 119 fMap = o.fMap; |
| 120 return *this; |
| 121 }*/ |
| 122 |
| 123 /** |
| 124 * Transliterator API. |
| 125 */ |
| 126 /*Transliterator* CaseMapTransliterator::clone(void) const { |
| 127 return new CaseMapTransliterator(*this); |
| 128 }*/ |
| 129 |
| 130 /** |
| 131 * Implements {@link Transliterator#handleTransliterate}. |
| 132 */ |
| 133 void CaseMapTransliterator::handleTransliterate(Replaceable& text, |
| 134 UTransPosition& offsets, |
| 135 UBool isIncremental) const |
| 136 { |
| 137 if (offsets.start >= offsets.limit) { |
| 138 return; |
| 139 } |
| 140 |
| 141 UCaseContext csc; |
| 142 uprv_memset(&csc, 0, sizeof(csc)); |
| 143 csc.p = &text; |
| 144 csc.start = offsets.contextStart; |
| 145 csc.limit = offsets.contextLimit; |
| 146 |
| 147 UnicodeString tmp; |
| 148 const UChar *s; |
| 149 UChar32 c; |
| 150 int32_t textPos, delta, result, locCache=0; |
| 151 |
| 152 for(textPos=offsets.start; textPos<offsets.limit;) { |
| 153 csc.cpStart=textPos; |
| 154 c=text.char32At(textPos); |
| 155 csc.cpLimit=textPos+=U16_LENGTH(c); |
| 156 |
| 157 result=fMap(fCsp, c, utrans_rep_caseContextIterator, &csc, &s, "", &locC
ache); |
| 158 |
| 159 if(csc.b1 && isIncremental) { |
| 160 // fMap() tried to look beyond the context limit |
| 161 // wait for more input |
| 162 offsets.start=csc.cpStart; |
| 163 return; |
| 164 } |
| 165 |
| 166 if(result>=0) { |
| 167 // replace the current code point with its full case mapping result |
| 168 // see UCASE_MAX_STRING_LENGTH |
| 169 if(result<=UCASE_MAX_STRING_LENGTH) { |
| 170 // string s[result] |
| 171 tmp.setTo(FALSE, s, result); |
| 172 delta=result-U16_LENGTH(c); |
| 173 } else { |
| 174 // single code point |
| 175 tmp.setTo(result); |
| 176 delta=tmp.length()-U16_LENGTH(c); |
| 177 } |
| 178 text.handleReplaceBetween(csc.cpStart, textPos, tmp); |
| 179 if(delta!=0) { |
| 180 textPos+=delta; |
| 181 csc.limit=offsets.contextLimit+=delta; |
| 182 offsets.limit+=delta; |
| 183 } |
| 184 } |
| 185 } |
| 186 offsets.start=textPos; |
| 187 } |
| 188 |
| 189 U_NAMESPACE_END |
| 190 |
| 191 #endif /* #if !UCONFIG_NO_TRANSLITERATION */ |
OLD | NEW |