LLVM  8.0.1
ConvertUTF.h
Go to the documentation of this file.
1 /*===--- ConvertUTF.h - Universal Character Names conversions ---------------===
2  *
3  * The LLVM Compiler Infrastructure
4  *
5  * This file is distributed under the University of Illinois Open Source
6  * License. See LICENSE.TXT for details.
7  *
8  *==------------------------------------------------------------------------==*/
9 /*
10  * Copyright 2001-2004 Unicode, Inc.
11  *
12  * Disclaimer
13  *
14  * This source code is provided as is by Unicode, Inc. No claims are
15  * made as to fitness for any particular purpose. No warranties of any
16  * kind are expressed or implied. The recipient agrees to determine
17  * applicability of information provided. If this file has been
18  * purchased on magnetic or optical media from Unicode, Inc., the
19  * sole remedy for any claim will be exchange of defective media
20  * within 90 days of receipt.
21  *
22  * Limitations on Rights to Redistribute This Code
23  *
24  * Unicode, Inc. hereby grants the right to freely use the information
25  * supplied in this file in the creation of products supporting the
26  * Unicode Standard, and to make copies of this file in any form
27  * for internal or external distribution as long as this notice
28  * remains attached.
29  */
30 
31 /* ---------------------------------------------------------------------
32 
33  Conversions between UTF32, UTF-16, and UTF-8. Header file.
34 
35  Several funtions are included here, forming a complete set of
36  conversions between the three formats. UTF-7 is not included
37  here, but is handled in a separate source file.
38 
39  Each of these routines takes pointers to input buffers and output
40  buffers. The input buffers are const.
41 
42  Each routine converts the text between *sourceStart and sourceEnd,
43  putting the result into the buffer between *targetStart and
44  targetEnd. Note: the end pointers are *after* the last item: e.g.
45  *(sourceEnd - 1) is the last item.
46 
47  The return result indicates whether the conversion was successful,
48  and if not, whether the problem was in the source or target buffers.
49  (Only the first encountered problem is indicated.)
50 
51  After the conversion, *sourceStart and *targetStart are both
52  updated to point to the end of last text successfully converted in
53  the respective buffers.
54 
55  Input parameters:
56  sourceStart - pointer to a pointer to the source buffer.
57  The contents of this are modified on return so that
58  it points at the next thing to be converted.
59  targetStart - similarly, pointer to pointer to the target buffer.
60  sourceEnd, targetEnd - respectively pointers to the ends of the
61  two buffers, for overflow checking only.
62 
63  These conversion functions take a ConversionFlags argument. When this
64  flag is set to strict, both irregular sequences and isolated surrogates
65  will cause an error. When the flag is set to lenient, both irregular
66  sequences and isolated surrogates are converted.
67 
68  Whether the flag is strict or lenient, all illegal sequences will cause
69  an error return. This includes sequences such as: <F4 90 80 80>, <C0 80>,
70  or <A0> in UTF-8, and values above 0x10FFFF in UTF-32. Conformant code
71  must check for illegal sequences.
72 
73  When the flag is set to lenient, characters over 0x10FFFF are converted
74  to the replacement character; otherwise (when the flag is set to strict)
75  they constitute an error.
76 
77  Output parameters:
78  The value "sourceIllegal" is returned from some routines if the input
79  sequence is malformed. When "sourceIllegal" is returned, the source
80  value will point to the illegal value that caused the problem. E.g.,
81  in UTF-8 when a sequence is malformed, it points to the start of the
82  malformed sequence.
83 
84  Author: Mark E. Davis, 1994.
85  Rev History: Rick McGowan, fixes & updates May 2001.
86  Fixes & updates, Sept 2001.
87 
88 ------------------------------------------------------------------------ */
89 
90 #ifndef LLVM_SUPPORT_CONVERTUTF_H
91 #define LLVM_SUPPORT_CONVERTUTF_H
92 
93 #include <cstddef>
94 #include <string>
95 #include <system_error>
96 
97 // Wrap everything in namespace llvm so that programs can link with llvm and
98 // their own version of the unicode libraries.
99 
100 namespace llvm {
101 
102 /* ---------------------------------------------------------------------
103  The following 4 definitions are compiler-specific.
104  The C standard does not guarantee that wchar_t has at least
105  16 bits, so wchar_t is no less portable than unsigned short!
106  All should be unsigned values to avoid sign extension during
107  bit mask & shift operations.
108 ------------------------------------------------------------------------ */
109 
110 typedef unsigned int UTF32; /* at least 32 bits */
111 typedef unsigned short UTF16; /* at least 16 bits */
112 typedef unsigned char UTF8; /* typically 8 bits */
113 typedef unsigned char Boolean; /* 0 or 1 */
114 
115 /* Some fundamental constants */
116 #define UNI_REPLACEMENT_CHAR (UTF32)0x0000FFFD
117 #define UNI_MAX_BMP (UTF32)0x0000FFFF
118 #define UNI_MAX_UTF16 (UTF32)0x0010FFFF
119 #define UNI_MAX_UTF32 (UTF32)0x7FFFFFFF
120 #define UNI_MAX_LEGAL_UTF32 (UTF32)0x0010FFFF
121 
122 #define UNI_MAX_UTF8_BYTES_PER_CODE_POINT 4
123 
124 #define UNI_UTF16_BYTE_ORDER_MARK_NATIVE 0xFEFF
125 #define UNI_UTF16_BYTE_ORDER_MARK_SWAPPED 0xFFFE
126 
127 typedef enum {
128  conversionOK, /* conversion successful */
129  sourceExhausted, /* partial character in source, but hit end */
130  targetExhausted, /* insuff. room in target for conversion */
131  sourceIllegal /* source sequence is illegal/malformed */
133 
134 typedef enum {
138 
139 ConversionResult ConvertUTF8toUTF16 (
140  const UTF8** sourceStart, const UTF8* sourceEnd,
141  UTF16** targetStart, UTF16* targetEnd, ConversionFlags flags);
142 
143 /**
144  * Convert a partial UTF8 sequence to UTF32. If the sequence ends in an
145  * incomplete code unit sequence, returns \c sourceExhausted.
146  */
147 ConversionResult ConvertUTF8toUTF32Partial(
148  const UTF8** sourceStart, const UTF8* sourceEnd,
149  UTF32** targetStart, UTF32* targetEnd, ConversionFlags flags);
150 
151 /**
152  * Convert a partial UTF8 sequence to UTF32. If the sequence ends in an
153  * incomplete code unit sequence, returns \c sourceIllegal.
154  */
155 ConversionResult ConvertUTF8toUTF32(
156  const UTF8** sourceStart, const UTF8* sourceEnd,
157  UTF32** targetStart, UTF32* targetEnd, ConversionFlags flags);
158 
159 ConversionResult ConvertUTF16toUTF8 (
160  const UTF16** sourceStart, const UTF16* sourceEnd,
161  UTF8** targetStart, UTF8* targetEnd, ConversionFlags flags);
162 
163 ConversionResult ConvertUTF32toUTF8 (
164  const UTF32** sourceStart, const UTF32* sourceEnd,
165  UTF8** targetStart, UTF8* targetEnd, ConversionFlags flags);
166 
167 ConversionResult ConvertUTF16toUTF32 (
168  const UTF16** sourceStart, const UTF16* sourceEnd,
169  UTF32** targetStart, UTF32* targetEnd, ConversionFlags flags);
170 
171 ConversionResult ConvertUTF32toUTF16 (
172  const UTF32** sourceStart, const UTF32* sourceEnd,
173  UTF16** targetStart, UTF16* targetEnd, ConversionFlags flags);
174 
175 Boolean isLegalUTF8Sequence(const UTF8 *source, const UTF8 *sourceEnd);
176 
177 Boolean isLegalUTF8String(const UTF8 **source, const UTF8 *sourceEnd);
178 
179 unsigned getNumBytesForUTF8(UTF8 firstByte);
180 
181 /*************************************************************************/
182 /* Below are LLVM-specific wrappers of the functions above. */
183 
184 template <typename T> class ArrayRef;
185 template <typename T> class SmallVectorImpl;
186 class StringRef;
187 
188 /**
189  * Convert an UTF8 StringRef to UTF8, UTF16, or UTF32 depending on
190  * WideCharWidth. The converted data is written to ResultPtr, which needs to
191  * point to at least WideCharWidth * (Source.Size() + 1) bytes. On success,
192  * ResultPtr will point one after the end of the copied string. On failure,
193  * ResultPtr will not be changed, and ErrorPtr will be set to the location of
194  * the first character which could not be converted.
195  * \return true on success.
196  */
197 bool ConvertUTF8toWide(unsigned WideCharWidth, llvm::StringRef Source,
198  char *&ResultPtr, const UTF8 *&ErrorPtr);
199 
200 /**
201 * Converts a UTF-8 StringRef to a std::wstring.
202 * \return true on success.
203 */
204 bool ConvertUTF8toWide(llvm::StringRef Source, std::wstring &Result);
205 
206 /**
207 * Converts a UTF-8 C-string to a std::wstring.
208 * \return true on success.
209 */
210 bool ConvertUTF8toWide(const char *Source, std::wstring &Result);
211 
212 /**
213 * Converts a std::wstring to a UTF-8 encoded std::string.
214 * \return true on success.
215 */
216 bool convertWideToUTF8(const std::wstring &Source, std::string &Result);
217 
218 
219 /**
220  * Convert an Unicode code point to UTF8 sequence.
221  *
222  * \param Source a Unicode code point.
223  * \param [in,out] ResultPtr pointer to the output buffer, needs to be at least
224  * \c UNI_MAX_UTF8_BYTES_PER_CODE_POINT bytes. On success \c ResultPtr is
225  * updated one past end of the converted sequence.
226  *
227  * \returns true on success.
228  */
229 bool ConvertCodePointToUTF8(unsigned Source, char *&ResultPtr);
230 
231 /**
232  * Convert the first UTF8 sequence in the given source buffer to a UTF32
233  * code point.
234  *
235  * \param [in,out] source A pointer to the source buffer. If the conversion
236  * succeeds, this pointer will be updated to point to the byte just past the
237  * end of the converted sequence.
238  * \param sourceEnd A pointer just past the end of the source buffer.
239  * \param [out] target The converted code
240  * \param flags Whether the conversion is strict or lenient.
241  *
242  * \returns conversionOK on success
243  *
244  * \sa ConvertUTF8toUTF32
245  */
246 inline ConversionResult convertUTF8Sequence(const UTF8 **source,
247  const UTF8 *sourceEnd,
248  UTF32 *target,
249  ConversionFlags flags) {
250  if (*source == sourceEnd)
251  return sourceExhausted;
252  unsigned size = getNumBytesForUTF8(**source);
253  if ((ptrdiff_t)size > sourceEnd - *source)
254  return sourceExhausted;
255  return ConvertUTF8toUTF32(source, *source + size, &target, target + 1, flags);
256 }
257 
258 /**
259  * Returns true if a blob of text starts with a UTF-16 big or little endian byte
260  * order mark.
261  */
263 
264 /**
265  * Converts a stream of raw bytes assumed to be UTF16 into a UTF8 std::string.
266  *
267  * \param [in] SrcBytes A buffer of what is assumed to be UTF-16 encoded text.
268  * \param [out] Out Converted UTF-8 is stored here on success.
269  * \returns true on success
270  */
271 bool convertUTF16ToUTF8String(ArrayRef<char> SrcBytes, std::string &Out);
272 
273 /**
274 * Converts a UTF16 string into a UTF8 std::string.
275 *
276 * \param [in] Src A buffer of UTF-16 encoded text.
277 * \param [out] Out Converted UTF-8 is stored here on success.
278 * \returns true on success
279 */
280 bool convertUTF16ToUTF8String(ArrayRef<UTF16> Src, std::string &Out);
281 
282 /**
283  * Converts a UTF-8 string into a UTF-16 string with native endianness.
284  *
285  * \returns true on success
286  */
288  SmallVectorImpl<UTF16> &DstUTF16);
289 
290 #if defined(_WIN32)
291 namespace sys {
292 namespace windows {
293 std::error_code UTF8ToUTF16(StringRef utf8, SmallVectorImpl<wchar_t> &utf16);
294 /// Convert to UTF16 from the current code page used in the system
295 std::error_code CurCPToUTF16(StringRef utf8, SmallVectorImpl<wchar_t> &utf16);
296 std::error_code UTF16ToUTF8(const wchar_t *utf16, size_t utf16_len,
297  SmallVectorImpl<char> &utf8);
298 /// Convert from UTF16 to the current code page used in the system
299 std::error_code UTF16ToCurCP(const wchar_t *utf16, size_t utf16_len,
300  SmallVectorImpl<char> &utf8);
301 } // namespace windows
302 } // namespace sys
303 #endif
304 
305 } /* end namespace llvm */
306 
307 #endif
unsigned int UTF32
Definition: ConvertUTF.h:110
bool hasUTF16ByteOrderMark(ArrayRef< char > SrcBytes)
Returns true if a blob of text starts with a UTF-16 big or little endian byte order mark...
This class represents lattice values for constants.
Definition: AllocatorList.h:24
ConversionResult
Definition: ConvertUTF.h:127
unsigned short UTF16
Definition: ConvertUTF.h:111
ConversionResult ConvertUTF8toUTF32(const UTF8 **sourceStart, const UTF8 *sourceEnd, UTF32 **targetStart, UTF32 *targetEnd, ConversionFlags flags)
Convert a partial UTF8 sequence to UTF32.
Definition: ConvertUTF.cpp:711
ConversionResult ConvertUTF16toUTF8(const UTF16 **sourceStart, const UTF16 *sourceEnd, UTF8 **targetStart, UTF8 *targetEnd, ConversionFlags flags)
Definition: ConvertUTF.cpp:248
This class consists of common code factored out of the SmallVector class to reduce code duplication b...
Definition: APFloat.h:42
ConversionResult ConvertUTF32toUTF16(const UTF32 **sourceStart, const UTF32 *sourceEnd, UTF16 **targetStart, UTF16 *targetEnd, ConversionFlags flags)
Definition: ConvertUTF.cpp:145
unsigned char UTF8
Definition: ConvertUTF.h:112
bool convertWideToUTF8(const std::wstring &Source, std::string &Result)
Converts a std::wstring to a UTF-8 encoded std::string.
unsigned getNumBytesForUTF8(UTF8 firstByte)
Definition: ConvertUTF.cpp:520
ArrayRef - Represent a constant reference to an array (0 or more elements consecutively in memory)...
Definition: APInt.h:33
bool convertUTF8ToUTF16String(StringRef SrcUTF8, SmallVectorImpl< UTF16 > &DstUTF16)
Converts a UTF-8 string into a UTF-16 string with native endianness.
ConversionResult ConvertUTF8toUTF16(const UTF8 **sourceStart, const UTF8 *sourceEnd, UTF16 **targetStart, UTF16 *targetEnd, ConversionFlags flags)
Definition: ConvertUTF.cpp:542
ConversionResult ConvertUTF32toUTF8(const UTF32 **sourceStart, const UTF32 *sourceEnd, UTF8 **targetStart, UTF8 *targetEnd, ConversionFlags flags)
Definition: ConvertUTF.cpp:318
bool ConvertUTF8toWide(unsigned WideCharWidth, llvm::StringRef Source, char *&ResultPtr, const UTF8 *&ErrorPtr)
Convert an UTF8 StringRef to UTF8, UTF16, or UTF32 depending on WideCharWidth.
Boolean isLegalUTF8Sequence(const UTF8 *source, const UTF8 *sourceEnd)
Definition: ConvertUTF.cpp:413
bool ConvertCodePointToUTF8(unsigned Source, char *&ResultPtr)
Convert an Unicode code point to UTF8 sequence.
Boolean isLegalUTF8String(const UTF8 **source, const UTF8 *sourceEnd)
Definition: ConvertUTF.cpp:530
ConversionResult ConvertUTF8toUTF32Partial(const UTF8 **sourceStart, const UTF8 *sourceEnd, UTF32 **targetStart, UTF32 *targetEnd, ConversionFlags flags)
Convert a partial UTF8 sequence to UTF32.
Definition: ConvertUTF.cpp:702
unsigned char Boolean
Definition: ConvertUTF.h:113
ConversionResult convertUTF8Sequence(const UTF8 **source, const UTF8 *sourceEnd, UTF32 *target, ConversionFlags flags)
Convert the first UTF8 sequence in the given source buffer to a UTF32 code point. ...
Definition: ConvertUTF.h:246
auto size(R &&Range, typename std::enable_if< std::is_same< typename std::iterator_traits< decltype(Range.begin())>::iterator_category, std::random_access_iterator_tag >::value, void >::type *=nullptr) -> decltype(std::distance(Range.begin(), Range.end()))
Get the size of a range.
Definition: STLExtras.h:1167
ConversionFlags
Definition: ConvertUTF.h:134
ConversionResult ConvertUTF16toUTF32(const UTF16 **sourceStart, const UTF16 *sourceEnd, UTF32 **targetStart, UTF32 *targetEnd, ConversionFlags flags)
Definition: ConvertUTF.cpp:194
bool convertUTF16ToUTF8String(ArrayRef< char > SrcBytes, std::string &Out)
Converts a stream of raw bytes assumed to be UTF16 into a UTF8 std::string.
StringRef - Represent a constant reference to a string, i.e.
Definition: StringRef.h:49