| 1 |
//===--- ClangCommentHTMLNamedCharacterReferenceEmitter.cpp -----------------=// |
| 2 |
// |
| 3 |
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. |
| 4 |
// See https://llvm.org/LICENSE.txt for license information. |
| 5 |
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception |
| 6 |
// |
| 7 |
//===----------------------------------------------------------------------===// |
| 8 |
// |
| 9 |
// This tablegen backend emits an efficient function to translate HTML named |
| 10 |
// character references to UTF-8 sequences. |
| 11 |
// |
| 12 |
//===----------------------------------------------------------------------===// |
| 13 |
|
| 14 |
#include "TableGenBackends.h" |
| 15 |
#include "llvm/ADT/SmallString.h" |
| 16 |
#include "llvm/Support/ConvertUTF.h" |
| 17 |
#include "llvm/TableGen/Error.h" |
| 18 |
#include "llvm/TableGen/Record.h" |
| 19 |
#include "llvm/TableGen/StringMatcher.h" |
| 20 |
#include "llvm/TableGen/TableGenBackend.h" |
| 21 |
#include <vector> |
| 22 |
|
| 23 |
using namespace llvm; |
| 24 |
|
| 25 |
/// Convert a code point to the corresponding UTF-8 sequence represented |
| 26 |
/// as a C string literal. |
| 27 |
/// |
| 28 |
/// \returns true on success. |
| 29 |
static bool translateCodePointToUTF8(unsigned CodePoint, |
| 30 |
SmallVectorImpl<char> &CLiteral) { |
| 31 |
char Translated[UNI_MAX_UTF8_BYTES_PER_CODE_POINT]; |
| 32 |
char *TranslatedPtr = Translated; |
| 33 |
if (!ConvertCodePointToUTF8(CodePoint, TranslatedPtr)) |
| 34 |
return false; |
| 35 |
|
| 36 |
StringRef UTF8(Translated, TranslatedPtr - Translated); |
| 37 |
|
| 38 |
raw_svector_ostream OS(CLiteral); |
| 39 |
OS << "\""; |
| 40 |
for (size_t i = 0, e = UTF8.size(); i != e; ++i) { |
| 41 |
OS << "\\x"; |
| 42 |
OS.write_hex(static_cast<unsigned char>(UTF8[i])); |
| 43 |
} |
| 44 |
OS << "\""; |
| 45 |
|
| 46 |
return true; |
| 47 |
} |
| 48 |
|
| 49 |
void clang::EmitClangCommentHTMLNamedCharacterReferences(RecordKeeper &Records, |
| 50 |
raw_ostream &OS) { |
| 51 |
std::vector<Record *> Tags = Records.getAllDerivedDefinitions("NCR"); |
| 52 |
std::vector<StringMatcher::StringPair> NameToUTF8; |
| 53 |
SmallString<32> CLiteral; |
| 54 |
for (std::vector<Record *>::iterator I = Tags.begin(), E = Tags.end(); |
| 55 |
I != E; ++I) { |
| 56 |
Record &Tag = **I; |
| 57 |
std::string Spelling = std::string(Tag.getValueAsString("Spelling")); |
| 58 |
uint64_t CodePoint = Tag.getValueAsInt("CodePoint"); |
| 59 |
CLiteral.clear(); |
| 60 |
CLiteral.append("return "); |
| 61 |
if (!translateCodePointToUTF8(CodePoint, CLiteral)) { |
| 62 |
SrcMgr.PrintMessage(Tag.getLoc().front(), |
| 63 |
SourceMgr::DK_Error, |
| 64 |
Twine("invalid code point")); |
| 65 |
continue; |
| 66 |
} |
| 67 |
CLiteral.append(";"); |
| 68 |
|
| 69 |
StringMatcher::StringPair Match(Spelling, std::string(CLiteral.str())); |
| 70 |
NameToUTF8.push_back(Match); |
| 71 |
} |
| 72 |
|
| 73 |
emitSourceFileHeader("HTML named character reference to UTF-8 " |
| 74 |
"translation", OS); |
| 75 |
|
| 76 |
OS << "StringRef translateHTMLNamedCharacterReferenceToUTF8(\n" |
| 77 |
" StringRef Name) {\n"; |
| 78 |
StringMatcher("Name", NameToUTF8, OS).Emit(); |
| 79 |
OS << " return StringRef();\n" |
| 80 |
<< "}\n\n"; |
| 81 |
} |