1 | // © 2016 and later: Unicode, Inc. and others. |
2 | // License & terms of use: http://www.unicode.org/copyright.html |
3 | /* |
4 | ********************************************************************** |
5 | * Copyright (c) 2003-2011, International Business Machines |
6 | * Corporation and others. All Rights Reserved. |
7 | ********************************************************************** |
8 | * Author: Alan Liu |
9 | * Created: September 24 2003 |
10 | * Since: ICU 2.8 |
11 | ********************************************************************** |
12 | */ |
13 | #include "ruleiter.h" |
14 | #include "unicode/parsepos.h" |
15 | #include "unicode/symtable.h" |
16 | #include "unicode/unistr.h" |
17 | #include "unicode/utf16.h" |
18 | #include "patternprops.h" |
19 | |
20 | /* \U87654321 or \ud800\udc00 */ |
21 | #define MAX_U_NOTATION_LEN 12 |
22 | |
23 | U_NAMESPACE_BEGIN |
24 | |
25 | RuleCharacterIterator::RuleCharacterIterator(const UnicodeString& theText, const SymbolTable* theSym, |
26 | ParsePosition& thePos) : |
27 | text(theText), |
28 | pos(thePos), |
29 | sym(theSym), |
30 | buf(0), |
31 | bufPos(0) |
32 | {} |
33 | |
34 | UBool RuleCharacterIterator::atEnd() const { |
35 | return buf == 0 && pos.getIndex() == text.length(); |
36 | } |
37 | |
38 | UChar32 RuleCharacterIterator::next(int32_t options, UBool& isEscaped, UErrorCode& ec) { |
39 | if (U_FAILURE(ec)) return DONE; |
40 | |
41 | UChar32 c = DONE; |
42 | isEscaped = false; |
43 | |
44 | for (;;) { |
45 | c = _current(); |
46 | _advance(U16_LENGTH(c)); |
47 | |
48 | if (c == SymbolTable::SYMBOL_REF && buf == 0 && |
49 | (options & PARSE_VARIABLES) != 0 && sym != 0) { |
50 | UnicodeString name = sym->parseReference(text, pos, text.length()); |
51 | // If name is empty there was an isolated SYMBOL_REF; |
52 | // return it. Caller must be prepared for this. |
53 | if (name.length() == 0) { |
54 | break; |
55 | } |
56 | bufPos = 0; |
57 | buf = sym->lookup(name); |
58 | if (buf == 0) { |
59 | ec = U_UNDEFINED_VARIABLE; |
60 | return DONE; |
61 | } |
62 | // Handle empty variable value |
63 | if (buf->length() == 0) { |
64 | buf = 0; |
65 | } |
66 | continue; |
67 | } |
68 | |
69 | if ((options & SKIP_WHITESPACE) != 0 && PatternProps::isWhiteSpace(c)) { |
70 | continue; |
71 | } |
72 | |
73 | if (c == 0x5C /*'\\'*/ && (options & PARSE_ESCAPES) != 0) { |
74 | UnicodeString tempEscape; |
75 | int32_t offset = 0; |
76 | c = lookahead(tempEscape, MAX_U_NOTATION_LEN).unescapeAt(offset); |
77 | jumpahead(offset); |
78 | isEscaped = true; |
79 | if (c < 0) { |
80 | ec = U_MALFORMED_UNICODE_ESCAPE; |
81 | return DONE; |
82 | } |
83 | } |
84 | |
85 | break; |
86 | } |
87 | |
88 | return c; |
89 | } |
90 | |
91 | void RuleCharacterIterator::getPos(RuleCharacterIterator::Pos& p) const { |
92 | p.buf = buf; |
93 | p.pos = pos.getIndex(); |
94 | p.bufPos = bufPos; |
95 | } |
96 | |
97 | void RuleCharacterIterator::setPos(const RuleCharacterIterator::Pos& p) { |
98 | buf = p.buf; |
99 | pos.setIndex(p.pos); |
100 | bufPos = p.bufPos; |
101 | } |
102 | |
103 | void RuleCharacterIterator::skipIgnored(int32_t options) { |
104 | if ((options & SKIP_WHITESPACE) != 0) { |
105 | for (;;) { |
106 | UChar32 a = _current(); |
107 | if (!PatternProps::isWhiteSpace(a)) break; |
108 | _advance(U16_LENGTH(a)); |
109 | } |
110 | } |
111 | } |
112 | |
113 | UnicodeString& RuleCharacterIterator::(UnicodeString& result, int32_t maxLookAhead) const { |
114 | if (maxLookAhead < 0) { |
115 | maxLookAhead = 0x7FFFFFFF; |
116 | } |
117 | if (buf != 0) { |
118 | buf->extract(bufPos, maxLookAhead, result); |
119 | } else { |
120 | text.extract(pos.getIndex(), maxLookAhead, result); |
121 | } |
122 | return result; |
123 | } |
124 | |
125 | void RuleCharacterIterator::jumpahead(int32_t count) { |
126 | _advance(count); |
127 | } |
128 | |
129 | /* |
130 | UnicodeString& RuleCharacterIterator::toString(UnicodeString& result) const { |
131 | int32_t b = pos.getIndex(); |
132 | text.extract(0, b, result); |
133 | return result.append((char16_t) 0x7C).append(text, b, 0x7FFFFFFF); // Insert '|' at index |
134 | } |
135 | */ |
136 | |
137 | UChar32 RuleCharacterIterator::_current() const { |
138 | if (buf != 0) { |
139 | return buf->char32At(bufPos); |
140 | } else { |
141 | int i = pos.getIndex(); |
142 | return (i < text.length()) ? text.char32At(i) : (UChar32)DONE; |
143 | } |
144 | } |
145 | |
146 | void RuleCharacterIterator::_advance(int32_t count) { |
147 | if (buf != 0) { |
148 | bufPos += count; |
149 | if (bufPos == buf->length()) { |
150 | buf = 0; |
151 | } |
152 | } else { |
153 | pos.setIndex(pos.getIndex() + count); |
154 | if (pos.getIndex() > text.length()) { |
155 | pos.setIndex(text.length()); |
156 | } |
157 | } |
158 | } |
159 | |
160 | U_NAMESPACE_END |
161 | |
162 | //eof |
163 | |