1 // Copyright 2018 Google LLC.
2 // Use of this source code is governed by a BSD-style license that can be found in the LICENSE file.
6 #include "include/core/SkTypes.h"
8 typedef int32_t SkUnichar;
12 /** Given a sequence of UTF-8 bytes, return the number of unicode codepoints.
13 If the sequence is invalid UTF-8, return -1.
15 SK_SPI int CountUTF8(const char* utf8, size_t byteLength);
17 /** Given a sequence of aligned UTF-16 characters in machine-endian form,
18 return the number of unicode codepoints. If the sequence is invalid
21 SK_SPI int CountUTF16(const uint16_t* utf16, size_t byteLength);
23 /** Given a sequence of aligned UTF-32 characters in machine-endian form,
24 return the number of unicode codepoints. If the sequence is invalid
27 SK_SPI int CountUTF32(const int32_t* utf32, size_t byteLength);
29 /** Given a sequence of UTF-8 bytes, return the first unicode codepoint.
30 The pointer will be incremented to point at the next codepoint's start. If
31 invalid UTF-8 is encountered, set *ptr to end and return -1.
33 SK_SPI SkUnichar NextUTF8(const char** ptr, const char* end);
35 /** Given a sequence of aligned UTF-16 characters in machine-endian form,
36 return the first unicode codepoint. The pointer will be incremented to
37 point at the next codepoint's start. If invalid UTF-16 is encountered,
38 set *ptr to end and return -1.
40 SK_SPI SkUnichar NextUTF16(const uint16_t** ptr, const uint16_t* end);
42 /** Given a sequence of aligned UTF-32 characters in machine-endian form,
43 return the first unicode codepoint. The pointer will be incremented to
44 point at the next codepoint's start. If invalid UTF-32 is encountered,
45 set *ptr to end and return -1.
47 SK_SPI SkUnichar NextUTF32(const int32_t** ptr, const int32_t* end);
49 constexpr unsigned kMaxBytesInUTF8Sequence = 4;
51 /** Convert the unicode codepoint into UTF-8. If `utf8` is non-null, place the
52 result in that array. Return the number of bytes in the result. If `utf8`
53 is null, simply return the number of bytes that would be used. For invalid
54 unicode codepoints, return 0.
56 SK_SPI size_t ToUTF8(SkUnichar uni, char utf8[kMaxBytesInUTF8Sequence] = nullptr);
58 /** Convert the unicode codepoint into UTF-16. If `utf16` is non-null, place
59 the result in that array. Return the number of UTF-16 code units in the
60 result (1 or 2). If `utf16` is null, simply return the number of code
61 units that would be used. For invalid unicode codepoints, return 0.
63 SK_SPI size_t ToUTF16(SkUnichar uni, uint16_t utf16[2] = nullptr);
65 /** Returns the number of resulting UTF16 values needed to convert the src utf8 sequence.
66 * If dst is not null, it is filled with the corresponding values up to its capacity.
67 * If there is an error, -1 is returned and the dst[] buffer is undefined.
69 SK_SPI int UTF8ToUTF16(uint16_t dst[], int dstCapacity, const char src[], size_t srcByteLength);
71 /** Returns the number of resulting UTF8 values needed to convert the src utf16 sequence.
72 * If dst is not null, it is filled with the corresponding values up to its capacity.
73 * If there is an error, -1 is returned and the dst[] buffer is undefined.
75 SK_SPI int UTF16ToUTF8(char dst[], int dstCapacity, const uint16_t src[], size_t srcLength);
78 * Given a UTF-16 code point, returns true iff it is a leading surrogate.
79 * https://unicode.org/faq/utf_bom.html#utf16-2
81 static inline bool IsLeadingSurrogateUTF16(uint16_t c) { return ((c) & 0xFC00) == 0xD800; }
84 * Given a UTF-16 code point, returns true iff it is a trailing surrogate.
85 * https://unicode.org/faq/utf_bom.html#utf16-2
87 static inline bool IsTrailingSurrogateUTF16(uint16_t c) { return ((c) & 0xFC00) == 0xDC00; }
92 #endif // SkUTF_DEFINED