1 | // © 2018 and later: Unicode, Inc. and others. |
2 | // License & terms of use: http://www.unicode.org/copyright.html |
3 | |
4 | #include "unicode/utypes.h" |
5 | |
6 | #if !UCONFIG_NO_FORMATTING |
7 | |
8 | // Allow implicit conversion from char16_t* to UnicodeString for this file: |
9 | // Helpful in toString methods and elsewhere. |
10 | #define UNISTR_FROM_STRING_EXPLICIT |
11 | |
12 | #include "numparse_types.h" |
13 | #include "string_segment.h" |
14 | #include "putilimp.h" |
15 | #include "unicode/utf16.h" |
16 | #include "unicode/uniset.h" |
17 | |
18 | U_NAMESPACE_BEGIN |
19 | |
20 | |
21 | StringSegment::StringSegment(const UnicodeString& str, bool ignoreCase) |
22 | : fStr(str), fStart(0), fEnd(str.length()), |
23 | fFoldCase(ignoreCase) {} |
24 | |
25 | int32_t StringSegment::getOffset() const { |
26 | return fStart; |
27 | } |
28 | |
29 | void StringSegment::setOffset(int32_t start) { |
30 | fStart = start; |
31 | } |
32 | |
33 | void StringSegment::adjustOffset(int32_t delta) { |
34 | fStart += delta; |
35 | } |
36 | |
37 | void StringSegment::adjustOffsetByCodePoint() { |
38 | fStart += U16_LENGTH(getCodePoint()); |
39 | } |
40 | |
41 | void StringSegment::setLength(int32_t length) { |
42 | fEnd = fStart + length; |
43 | } |
44 | |
45 | void StringSegment::resetLength() { |
46 | fEnd = fStr.length(); |
47 | } |
48 | |
49 | int32_t StringSegment::length() const { |
50 | return fEnd - fStart; |
51 | } |
52 | |
53 | char16_t StringSegment::charAt(int32_t index) const { |
54 | return fStr.charAt(index + fStart); |
55 | } |
56 | |
57 | UChar32 StringSegment::codePointAt(int32_t index) const { |
58 | return fStr.char32At(index + fStart); |
59 | } |
60 | |
61 | UnicodeString StringSegment::toUnicodeString() const { |
62 | return UnicodeString(fStr.getBuffer() + fStart, fEnd - fStart); |
63 | } |
64 | |
65 | const UnicodeString StringSegment::toTempUnicodeString() const { |
66 | // Use the readonly-aliasing constructor for efficiency. |
67 | return UnicodeString(FALSE, fStr.getBuffer() + fStart, fEnd - fStart); |
68 | } |
69 | |
70 | UChar32 StringSegment::getCodePoint() const { |
71 | char16_t lead = fStr.charAt(fStart); |
72 | if (U16_IS_LEAD(lead) && fStart + 1 < fEnd) { |
73 | return fStr.char32At(fStart); |
74 | } else if (U16_IS_SURROGATE(lead)) { |
75 | return -1; |
76 | } else { |
77 | return lead; |
78 | } |
79 | } |
80 | |
81 | bool StringSegment::startsWith(UChar32 otherCp) const { |
82 | return codePointsEqual(getCodePoint(), otherCp, fFoldCase); |
83 | } |
84 | |
85 | bool StringSegment::startsWith(const UnicodeSet& uniset) const { |
86 | // TODO: Move UnicodeSet case-folding logic here. |
87 | // TODO: Handle string matches here instead of separately. |
88 | UChar32 cp = getCodePoint(); |
89 | if (cp == -1) { |
90 | return false; |
91 | } |
92 | return uniset.contains(cp); |
93 | } |
94 | |
95 | bool StringSegment::startsWith(const UnicodeString& other) const { |
96 | if (other.isBogus() || other.length() == 0 || length() == 0) { |
97 | return false; |
98 | } |
99 | int cp1 = getCodePoint(); |
100 | int cp2 = other.char32At(0); |
101 | return codePointsEqual(cp1, cp2, fFoldCase); |
102 | } |
103 | |
104 | int32_t StringSegment::getCommonPrefixLength(const UnicodeString& other) { |
105 | return getPrefixLengthInternal(other, fFoldCase); |
106 | } |
107 | |
108 | int32_t StringSegment::getCaseSensitivePrefixLength(const UnicodeString& other) { |
109 | return getPrefixLengthInternal(other, false); |
110 | } |
111 | |
112 | int32_t StringSegment::getPrefixLengthInternal(const UnicodeString& other, bool foldCase) { |
113 | U_ASSERT(other.length() > 0); |
114 | int32_t offset = 0; |
115 | for (; offset < uprv_min(length(), other.length());) { |
116 | // TODO: case-fold code points, not chars |
117 | char16_t c1 = charAt(offset); |
118 | char16_t c2 = other.charAt(offset); |
119 | if (!codePointsEqual(c1, c2, foldCase)) { |
120 | break; |
121 | } |
122 | offset++; |
123 | } |
124 | return offset; |
125 | } |
126 | |
127 | bool StringSegment::codePointsEqual(UChar32 cp1, UChar32 cp2, bool foldCase) { |
128 | if (cp1 == cp2) { |
129 | return true; |
130 | } |
131 | if (!foldCase) { |
132 | return false; |
133 | } |
134 | cp1 = u_foldCase(cp1, TRUE); |
135 | cp2 = u_foldCase(cp2, TRUE); |
136 | return cp1 == cp2; |
137 | } |
138 | |
139 | bool StringSegment::operator==(const UnicodeString& other) const { |
140 | return toTempUnicodeString() == other; |
141 | } |
142 | |
143 | |
144 | U_NAMESPACE_END |
145 | #endif /* #if !UCONFIG_NO_FORMATTING */ |
146 | |