path: root/lib/AST/CommentLexer.cpp

#include "clang/AST/CommentLexer.h"
#include "clang/Basic/ConvertUTF.h"
#include "llvm/ADT/StringSwitch.h"
#include "llvm/Support/ErrorHandling.h"

namespace clang {
namespace comments {

void Token::dump(const Lexer &L, const SourceManager &SM) const {
  llvm::errs() << "comments::Token Kind=" << Kind << " ";
  Loc.dump(SM);
  llvm::errs() << " " << Length << " \"" << L.getSpelling(*this, SM) << "\"\n";
}

bool Lexer::isVerbatimBlockCommand(StringRef BeginName,
                                  StringRef &EndName) const {
  const char *Result = llvm::StringSwitch<const char *>(BeginName)
    .Case("code", "endcode")
    .Case("verbatim", "endverbatim")
    .Case("htmlonly", "endhtmlonly")
    .Case("latexonly", "endlatexonly")
    .Case("xmlonly", "endxmlonly")
    .Case("manonly", "endmanonly")
    .Case("rtfonly", "endrtfonly")

    .Case("dot", "enddot")
    .Case("msc", "endmsc")

    .Case("f$", "f$") // Inline LaTeX formula
    .Case("f[", "f]") // Displayed LaTeX formula
    .Case("f{", "f}") // LaTeX environment

    .Default(NULL);

  if (Result) {
    EndName = Result;
    return true;
  }

  for (VerbatimBlockCommandVector::const_iterator
           I = VerbatimBlockCommands.begin(),
           E = VerbatimBlockCommands.end();
       I != E; ++I)
    if (I->BeginName == BeginName) {
      EndName = I->EndName;
      return true;
    }

  return false;
}

bool Lexer::isVerbatimLineCommand(StringRef Name) const {
  bool Result = llvm::StringSwitch<bool>(Name)
  .Case("fn", true)
  .Case("var", true)
  .Case("property", true)
  .Case("typedef", true)

  .Case("overload", true)

  .Case("defgroup", true)
  .Case("ingroup", true)
  .Case("addtogroup", true)
  .Case("weakgroup", true)
  .Case("name", true)

  .Case("section", true)
  .Case("subsection", true)
  .Case("subsubsection", true)
  .Case("paragraph", true)

  .Case("mainpage", true)
  .Case("subpage", true)
  .Case("ref", true)

  .Default(false);

  if (Result)
    return true;

  for (VerbatimLineCommandVector::const_iterator
           I = VerbatimLineCommands.begin(),
           E = VerbatimLineCommands.end();
       I != E; ++I)
    if (I->Name == Name)
      return true;

  return false;
}

namespace {
bool isHTMLNamedCharacterReferenceCharacter(char C) {
  return (C >= 'a' && C <= 'z') ||
         (C >= 'A' && C <= 'Z');
}

bool isHTMLDecimalCharacterReferenceCharacter(char C) {
  return C >= '0' && C <= '9';
}

bool isHTMLHexCharacterReferenceCharacter(char C) {
  return (C >= '0' && C <= '9') ||
         (C >= 'a' && C <= 'f') ||
         (C >= 'A' && C <= 'F');
}
} // unnamed namespace

StringRef Lexer::resolveHTMLNamedCharacterReference(StringRef Name) const {
  return llvm::StringSwitch<StringRef>(Name)
      .Case("amp", "&")
      .Case("lt", "<")
      .Case("gt", ">")
      .Case("quot", "\"")
      .Case("apos", "\'")
      .Default("");
}

StringRef Lexer::resolveHTMLDecimalCharacterReference(StringRef Name) const {
  unsigned CodePoint = 0;
  for (unsigned i = 0, e = Name.size(); i != e; ++i) {
    assert(isHTMLDecimalCharacterReferenceCharacter(Name[i]));
    CodePoint *= 10;
    CodePoint += Name[i] - '0';
  }

  char *Resolved = Allocator.Allocate<char>(UNI_MAX_UTF8_BYTES_PER_CODE_POINT);
  char *ResolvedPtr = Resolved;