| 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 | #ifndef __NUMPARSE_STRINGSEGMENT_H__ | 
|---|
| 8 | #define __NUMPARSE_STRINGSEGMENT_H__ | 
|---|
| 9 |  | 
|---|
| 10 | #include "unicode/unistr.h" | 
|---|
| 11 | #include "unicode/uniset.h" | 
|---|
| 12 |  | 
|---|
| 13 | U_NAMESPACE_BEGIN | 
|---|
| 14 |  | 
|---|
| 15 |  | 
|---|
| 16 | /** | 
|---|
| 17 | * A mutable UnicodeString wrapper with a variable offset and length and | 
|---|
| 18 | * support for case folding. The charAt, length, and subSequence methods all | 
|---|
| 19 | * operate relative to the fixed offset into the UnicodeString. | 
|---|
| 20 | * | 
|---|
| 21 | * Intended to be useful for parsing. | 
|---|
| 22 | * | 
|---|
| 23 | * CAUTION: Since this class is mutable, it must not be used anywhere that an | 
|---|
| 24 | * immutable object is required, like in a cache or as the key of a hash map. | 
|---|
| 25 | * | 
|---|
| 26 | * @author sffc (Shane Carr) | 
|---|
| 27 | */ | 
|---|
| 28 | // Exported as U_I18N_API for tests | 
|---|
| 29 | class U_I18N_API StringSegment : public UMemory { | 
|---|
| 30 | public: | 
|---|
| 31 | StringSegment(const UnicodeString& str, bool ignoreCase); | 
|---|
| 32 |  | 
|---|
| 33 | int32_t getOffset() const; | 
|---|
| 34 |  | 
|---|
| 35 | void setOffset(int32_t start); | 
|---|
| 36 |  | 
|---|
| 37 | /** | 
|---|
| 38 | * Equivalent to <code>setOffset(getOffset()+delta)</code>. | 
|---|
| 39 | * | 
|---|
| 40 | * <p> | 
|---|
| 41 | * This method is usually called by a Matcher to register that a char was consumed. If the char is | 
|---|
| 42 | * strong (it usually is, except for things like whitespace), follow this with a call to | 
|---|
| 43 | * {@link ParsedNumber#setCharsConsumed}. For more information on strong chars, see that method. | 
|---|
| 44 | */ | 
|---|
| 45 | void adjustOffset(int32_t delta); | 
|---|
| 46 |  | 
|---|
| 47 | /** | 
|---|
| 48 | * Adjusts the offset by the width of the current code point, either 1 or 2 chars. | 
|---|
| 49 | */ | 
|---|
| 50 | void adjustOffsetByCodePoint(); | 
|---|
| 51 |  | 
|---|
| 52 | void setLength(int32_t length); | 
|---|
| 53 |  | 
|---|
| 54 | void resetLength(); | 
|---|
| 55 |  | 
|---|
| 56 | int32_t length() const; | 
|---|
| 57 |  | 
|---|
| 58 | char16_t charAt(int32_t index) const; | 
|---|
| 59 |  | 
|---|
| 60 | UChar32 codePointAt(int32_t index) const; | 
|---|
| 61 |  | 
|---|
| 62 | UnicodeString toUnicodeString() const; | 
|---|
| 63 |  | 
|---|
| 64 | const UnicodeString toTempUnicodeString() const; | 
|---|
| 65 |  | 
|---|
| 66 | /** | 
|---|
| 67 | * Returns the first code point in the string segment, or -1 if the string starts with an invalid | 
|---|
| 68 | * code point. | 
|---|
| 69 | * | 
|---|
| 70 | * <p> | 
|---|
| 71 | * <strong>Important:</strong> Most of the time, you should use {@link #startsWith}, which handles case | 
|---|
| 72 | * folding logic, instead of this method. | 
|---|
| 73 | */ | 
|---|
| 74 | UChar32 getCodePoint() const; | 
|---|
| 75 |  | 
|---|
| 76 | /** | 
|---|
| 77 | * Returns true if the first code point of this StringSegment equals the given code point. | 
|---|
| 78 | * | 
|---|
| 79 | * <p> | 
|---|
| 80 | * This method will perform case folding if case folding is enabled for the parser. | 
|---|
| 81 | */ | 
|---|
| 82 | bool startsWith(UChar32 otherCp) const; | 
|---|
| 83 |  | 
|---|
| 84 | /** | 
|---|
| 85 | * Returns true if the first code point of this StringSegment is in the given UnicodeSet. | 
|---|
| 86 | */ | 
|---|
| 87 | bool startsWith(const UnicodeSet& uniset) const; | 
|---|
| 88 |  | 
|---|
| 89 | /** | 
|---|
| 90 | * Returns true if there is at least one code point of overlap between this StringSegment and the | 
|---|
| 91 | * given UnicodeString. | 
|---|
| 92 | */ | 
|---|
| 93 | bool startsWith(const UnicodeString& other) const; | 
|---|
| 94 |  | 
|---|
| 95 | /** | 
|---|
| 96 | * Returns the length of the prefix shared by this StringSegment and the given UnicodeString. For | 
|---|
| 97 | * example, if this string segment is "aab", and the char sequence is "aac", this method returns 2, | 
|---|
| 98 | * since the first 2 characters are the same. | 
|---|
| 99 | * | 
|---|
| 100 | * <p> | 
|---|
| 101 | * This method only returns offsets along code point boundaries. | 
|---|
| 102 | * | 
|---|
| 103 | * <p> | 
|---|
| 104 | * This method will perform case folding if case folding was enabled in the constructor. | 
|---|
| 105 | * | 
|---|
| 106 | * <p> | 
|---|
| 107 | * IMPORTANT: The given UnicodeString must not be empty! It is the caller's responsibility to check. | 
|---|
| 108 | */ | 
|---|
| 109 | int32_t getCommonPrefixLength(const UnicodeString& other); | 
|---|
| 110 |  | 
|---|
| 111 | /** | 
|---|
| 112 | * Like {@link #getCommonPrefixLength}, but never performs case folding, even if case folding is | 
|---|
| 113 | * enabled for the parser. | 
|---|
| 114 | */ | 
|---|
| 115 | int32_t getCaseSensitivePrefixLength(const UnicodeString& other); | 
|---|
| 116 |  | 
|---|
| 117 | bool operator==(const UnicodeString& other) const; | 
|---|
| 118 |  | 
|---|
| 119 | private: | 
|---|
| 120 | const UnicodeString& fStr; | 
|---|
| 121 | int32_t fStart; | 
|---|
| 122 | int32_t fEnd; | 
|---|
| 123 | bool fFoldCase; | 
|---|
| 124 |  | 
|---|
| 125 | int32_t getPrefixLengthInternal(const UnicodeString& other, bool foldCase); | 
|---|
| 126 |  | 
|---|
| 127 | static bool codePointsEqual(UChar32 cp1, UChar32 cp2, bool foldCase); | 
|---|
| 128 | }; | 
|---|
| 129 |  | 
|---|
| 130 |  | 
|---|
| 131 | U_NAMESPACE_END | 
|---|
| 132 |  | 
|---|
| 133 | #endif //__NUMPARSE_STRINGSEGMENT_H__ | 
|---|
| 134 | #endif /* #if !UCONFIG_NO_FORMATTING */ | 
|---|
| 135 |  | 
|---|