Mercurial > hg > Members > tobaru > cbc > CbC_llvm

annotate lib/Support/ConvertUTFWrapper.cpp @ 100:7d135dc70f03

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
LLVM 3.9
author Miyagi Mitsuki <e135756@ie.u-ryukyu.ac.jp>
date Tue, 26 Jan 2016 22:53:40 +0900
parents 60c9769439b8
children 1172e4bd9c6f
Ignore whitespace changes - Everywhere: Within whitespace: At end of lines:
rev   line source
0
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
1 //===-- ConvertUTFWrapper.cpp - Wrap ConvertUTF.h with clang data types -----===
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
2 //
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
3 // The LLVM Compiler Infrastructure
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
4 //
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
5 // This file is distributed under the University of Illinois Open Source
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
6 // License. See LICENSE.TXT for details.
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
7 //
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
8 //===----------------------------------------------------------------------===//
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
9
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
10 #include "llvm/Support/ConvertUTF.h"
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
11 #include "llvm/Support/SwapByteOrder.h"
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
12 #include <string>
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
13 #include <vector>
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
14
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
15 namespace llvm {
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
16
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
17 bool ConvertUTF8toWide(unsigned WideCharWidth, llvm::StringRef Source,
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
18 char *&ResultPtr, const UTF8 *&ErrorPtr) {
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
19 assert(WideCharWidth == 1 || WideCharWidth == 2 || WideCharWidth == 4);
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
20 ConversionResult result = conversionOK;
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
21 // Copy the character span over.
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
22 if (WideCharWidth == 1) {
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
23 const UTF8 *Pos = reinterpret_cast<const UTF8*>(Source.begin());
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
24 if (!isLegalUTF8String(&Pos, reinterpret_cast<const UTF8*>(Source.end()))) {
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
25 result = sourceIllegal;
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
26 ErrorPtr = Pos;
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
27 } else {
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
28 memcpy(ResultPtr, Source.data(), Source.size());
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
29 ResultPtr += Source.size();
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
30 }
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
31 } else if (WideCharWidth == 2) {
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
32 const UTF8 *sourceStart = (const UTF8*)Source.data();
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
33 // FIXME: Make the type of the result buffer correct instead of
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
34 // using reinterpret_cast.
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
35 UTF16 *targetStart = reinterpret_cast<UTF16*>(ResultPtr);
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
36 ConversionFlags flags = strictConversion;
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
37 result = ConvertUTF8toUTF16(
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
38 &sourceStart, sourceStart + Source.size(),
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
39 &targetStart, targetStart + 2*Source.size(), flags);
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
40 if (result == conversionOK)
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
41 ResultPtr = reinterpret_cast<char*>(targetStart);
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
42 else
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
43 ErrorPtr = sourceStart;
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
44 } else if (WideCharWidth == 4) {
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
45 const UTF8 *sourceStart = (const UTF8*)Source.data();
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
46 // FIXME: Make the type of the result buffer correct instead of
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
47 // using reinterpret_cast.
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
48 UTF32 *targetStart = reinterpret_cast<UTF32*>(ResultPtr);
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
49 ConversionFlags flags = strictConversion;
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
50 result = ConvertUTF8toUTF32(
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
51 &sourceStart, sourceStart + Source.size(),
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
52 &targetStart, targetStart + 4*Source.size(), flags);
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
53 if (result == conversionOK)
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
54 ResultPtr = reinterpret_cast<char*>(targetStart);
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
55 else
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
56 ErrorPtr = sourceStart;
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
57 }
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
58 assert((result != targetExhausted)
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
59 && "ConvertUTF8toUTFXX exhausted target buffer");
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
60 return result == conversionOK;
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
61 }
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
62
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
63 bool ConvertCodePointToUTF8(unsigned Source, char *&ResultPtr) {
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
64 const UTF32 *SourceStart = &Source;
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
65 const UTF32 *SourceEnd = SourceStart + 1;
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
66 UTF8 *TargetStart = reinterpret_cast<UTF8 *>(ResultPtr);
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
67 UTF8 *TargetEnd = TargetStart + 4;
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
68 ConversionResult CR = ConvertUTF32toUTF8(&SourceStart, SourceEnd,
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
69 &TargetStart, TargetEnd,
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
70 strictConversion);
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
71 if (CR != conversionOK)
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
72 return false;
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
73
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
74 ResultPtr = reinterpret_cast<char*>(TargetStart);
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
75 return true;
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
76 }
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
77
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
78 bool hasUTF16ByteOrderMark(ArrayRef<char> S) {
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
79 return (S.size() >= 2 &&
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
80 ((S[0] == '\xff' && S[1] == '\xfe') ||
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
81 (S[0] == '\xfe' && S[1] == '\xff')));
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
82 }
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
83
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
84 bool convertUTF16ToUTF8String(ArrayRef<char> SrcBytes, std::string &Out) {
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
85 assert(Out.empty());
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
86
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
87 // Error out on an uneven byte count.
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
88 if (SrcBytes.size() % 2)
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
89 return false;
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
90
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
91 // Avoid OOB by returning early on empty input.
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
92 if (SrcBytes.empty())
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
93 return true;
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
94
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
95 const UTF16 *Src = reinterpret_cast<const UTF16 *>(SrcBytes.begin());
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
96 const UTF16 *SrcEnd = reinterpret_cast<const UTF16 *>(SrcBytes.end());
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
97
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
98 // Byteswap if necessary.
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
99 std::vector<UTF16> ByteSwapped;
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
100 if (Src[0] == UNI_UTF16_BYTE_ORDER_MARK_SWAPPED) {
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
101 ByteSwapped.insert(ByteSwapped.end(), Src, SrcEnd);
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
102 for (unsigned I = 0, E = ByteSwapped.size(); I != E; ++I)
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
103 ByteSwapped[I] = llvm::sys::SwapByteOrder_16(ByteSwapped[I]);
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
104 Src = &ByteSwapped[0];
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
105 SrcEnd = &ByteSwapped[ByteSwapped.size() - 1] + 1;
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
106 }
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
107
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
108 // Skip the BOM for conversion.
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
109 if (Src[0] == UNI_UTF16_BYTE_ORDER_MARK_NATIVE)
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
110 Src++;
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
111
83
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
112 // Just allocate enough space up front. We'll shrink it later. Allocate
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
113 // enough that we can fit a null terminator without reallocating.
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
114 Out.resize(SrcBytes.size() * UNI_MAX_UTF8_BYTES_PER_CODE_POINT + 1);
0
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
115 UTF8 *Dst = reinterpret_cast<UTF8 *>(&Out[0]);
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
116 UTF8 *DstEnd = Dst + Out.size();
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
117
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
118 ConversionResult CR =
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
119 ConvertUTF16toUTF8(&Src, SrcEnd, &Dst, DstEnd, strictConversion);
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
120 assert(CR != targetExhausted);
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
121
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
122 if (CR != conversionOK) {
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
123 Out.clear();
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
124 return false;
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
125 }
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
126
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
127 Out.resize(reinterpret_cast<char *>(Dst) - &Out[0]);
83
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
128 Out.push_back(0);
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
129 Out.pop_back();
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
130 return true;
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
131 }
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
132
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
133 bool convertUTF8ToUTF16String(StringRef SrcUTF8,
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
134 SmallVectorImpl<UTF16> &DstUTF16) {
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
135 assert(DstUTF16.empty());
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
136
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
137 // Avoid OOB by returning early on empty input.
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
138 if (SrcUTF8.empty()) {
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
139 DstUTF16.push_back(0);
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
140 DstUTF16.pop_back();
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
141 return true;
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
142 }
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
143
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
144 const UTF8 *Src = reinterpret_cast<const UTF8 *>(SrcUTF8.begin());
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
145 const UTF8 *SrcEnd = reinterpret_cast<const UTF8 *>(SrcUTF8.end());
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
146
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
147 // Allocate the same number of UTF-16 code units as UTF-8 code units. Encoding
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
148 // as UTF-16 should always require the same amount or less code units than the
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
149 // UTF-8 encoding. Allocate one extra byte for the null terminator though,
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
150 // so that someone calling DstUTF16.data() gets a null terminated string.
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
151 // We resize down later so we don't have to worry that this over allocates.
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
152 DstUTF16.resize(SrcUTF8.size()+1);
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
153 UTF16 *Dst = &DstUTF16[0];
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
154 UTF16 *DstEnd = Dst + DstUTF16.size();
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
155
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
156 ConversionResult CR =
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
157 ConvertUTF8toUTF16(&Src, SrcEnd, &Dst, DstEnd, strictConversion);
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
158 assert(CR != targetExhausted);
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
159
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
160 if (CR != conversionOK) {
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
161 DstUTF16.clear();
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
162 return false;
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
163 }
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
164
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
165 DstUTF16.resize(Dst - &DstUTF16[0]);
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
166 DstUTF16.push_back(0);
60c9769439b8 LLVM 3.7
Tatsuki IHA <e125716@ie.u-ryukyu.ac.jp>
parents: 0
diff changeset
167 DstUTF16.pop_back();
0
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
168 return true;
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
169 }
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
170
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
171 } // end namespace llvm
95c75e76d11b LLVM 3.4
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff changeset
172