r181458 - documentation parsing. Patch to do typo correction for

[Dmitri Gribenko]

On Wed, May 8, 2013 at 10:21 PM, Fariborz Jahanian <fjahanian at apple.com> wrote:
> Author: fjahanian
> Date: Wed May  8 14:21:00 2013
> New Revision: 181458
>
> URL: http://llvm.org/viewvc/llvm-project?rev=181458&view=rev
> Log:
> documentation parsing. Patch to do typo correction for
> documentation commands. Patch was reviewed, along with
> great suggestions for improvement, by Doug.
> // rdar://12381408
>
> Modified:
>     cfe/trunk/include/clang/AST/CommentCommandTraits.h
>     cfe/trunk/include/clang/Basic/DiagnosticCommentKinds.td
>     cfe/trunk/lib/AST/CommentCommandTraits.cpp
>     cfe/trunk/lib/AST/CommentLexer.cpp
>     cfe/trunk/test/Sema/warn-documentation-fixits.cpp
>
> Modified: cfe/trunk/include/clang/AST/CommentCommandTraits.h
> URL: http://llvm.org/viewvc/llvm-project/cfe/trunk/include/clang/AST/CommentCommandTraits.h?rev=181458&r1=181457&r2=181458&view=diff
> ==============================================================================
> --- cfe/trunk/include/clang/AST/CommentCommandTraits.h (original)
> +++ cfe/trunk/include/clang/AST/CommentCommandTraits.h Wed May  8 14:21:00 2013
> @@ -142,6 +142,8 @@ public:
>      llvm_unreachable("the command should be known");
>    }
>
> +  const CommandInfo *getTypoCorrectCommandInfo(StringRef Typo) const;
> +
>    const CommandInfo *getCommandInfo(unsigned CommandID) const;
>
>    const CommandInfo *registerUnknownCommand(StringRef CommandName);
>
> Modified: cfe/trunk/include/clang/Basic/DiagnosticCommentKinds.td
> URL: http://llvm.org/viewvc/llvm-project/cfe/trunk/include/clang/Basic/DiagnosticCommentKinds.td?rev=181458&r1=181457&r2=181458&view=diff
> ==============================================================================
> --- cfe/trunk/include/clang/Basic/DiagnosticCommentKinds.td (original)
> +++ cfe/trunk/include/clang/Basic/DiagnosticCommentKinds.td Wed May  8 14:21:00 2013
> @@ -159,5 +159,9 @@ def warn_verbatim_block_end_without_star
>  def warn_unknown_comment_command_name : Warning<
>    "unknown command tag name">, InGroup<Documentation>, DefaultIgnore;
>
> +def warn_correct_comment_command_name : Warning<
> +  "unknown command tag name '%0'; did you mean '%1'?">,
> +  InGroup<Documentation>;
> +
>  } // end of documentation issue category
>  } // end of AST component
>
> Modified: cfe/trunk/lib/AST/CommentCommandTraits.cpp
> URL: http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/AST/CommentCommandTraits.cpp?rev=181458&r1=181457&r2=181458&view=diff
> ==============================================================================
> --- cfe/trunk/lib/AST/CommentCommandTraits.cpp (original)
> +++ cfe/trunk/lib/AST/CommentCommandTraits.cpp Wed May  8 14:21:00 2013
> @@ -43,6 +43,33 @@ const CommandInfo *CommandTraits::getCom
>    return getRegisteredCommandInfo(CommandID);
>  }
>
> +const CommandInfo *
> +CommandTraits::getTypoCorrectCommandInfo(StringRef Typo) const {
> +  const unsigned MaxEditDistance = 1;
> +  unsigned BestEditDistance = MaxEditDistance + 1;
> +  SmallVector<const CommandInfo *, 2> BestCommand;
> +
> +  int NumOfCommands = sizeof(Commands) / sizeof(CommandInfo);
> +  for (int i = 0; i < NumOfCommands; i++) {
> +    StringRef Name = Commands[i].Name;

This loop should also check dynamically registered commands.

> +    unsigned MinPossibleEditDistance = abs((int)Name.size() - (int)Typo.size());
> +    if (MinPossibleEditDistance > 0 &&
> +        Typo.size() / MinPossibleEditDistance < 1)
> +      continue;
> +    unsigned EditDistance = Typo.edit_distance(Name, true, MaxEditDistance);
> +    if (EditDistance > MaxEditDistance)
> +      continue;
> +    if (EditDistance == BestEditDistance)
> +      BestCommand.push_back(&Commands[i]);
> +    else if (EditDistance < BestEditDistance) {
> +      BestCommand.clear();
> +      BestCommand.push_back(&Commands[i]);
> +      BestEditDistance = EditDistance;
> +    }
> +  }
> +  return (BestCommand.size() != 1) ? NULL : BestCommand[0];
> +}
> +
>  CommandInfo *CommandTraits::createCommandInfoWithName(StringRef CommandName) {
>    char *Name = Allocator.Allocate<char>(CommandName.size() + 1);
>    memcpy(Name, CommandName.data(), CommandName.size());
>
> Modified: cfe/trunk/lib/AST/CommentLexer.cpp
> URL: http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/AST/CommentLexer.cpp?rev=181458&r1=181457&r2=181458&view=diff
> ==============================================================================
> --- cfe/trunk/lib/AST/CommentLexer.cpp (original)
> +++ cfe/trunk/lib/AST/CommentLexer.cpp Wed May  8 14:21:00 2013
> @@ -265,6 +265,7 @@ const char *findCCommentEnd(const char *
>    }
>    llvm_unreachable("buffer end hit before '*/' was seen");
>  }
> +
>  } // unnamed namespace
>
>  void Lexer::lexCommentText(Token &T) {
> @@ -354,8 +355,17 @@ void Lexer::lexCommentText(Token &T) {
>          if (!Info) {
>            formTokenWithChars(T, TokenPtr, tok::unknown_command);
>            T.setUnknownCommandName(CommandName);
> -          Diag(T.getLocation(), diag::warn_unknown_comment_command_name);
> -          return;
> +          if (Info = Traits.getTypoCorrectCommandInfo(CommandName)) {
> +            StringRef CorrectedName = Info->Name;
> +            SourceRange CommandRange(T.getLocation().getLocWithOffset(1),
> +                                     T.getEndLocation());
> +            Diag(T.getLocation(), diag::warn_correct_comment_command_name)
> +              << CommandName << CorrectedName
> +              << FixItHint::CreateReplacement(CommandRange, CorrectedName);

We recover by assuming that the user wanted this correction.  What if
they did not?..  (But I can only imagine such situation only for
external users who use comment parsing for something non-Doxygen.)

Dmitri

--
main(i,j){for(i=2;;i++){for(j=2;j<i;j++){if(!(i%j)){j=0;break;}}if
(j){printf("%d\n",i);}}} /*Dmitri Gribenko <gribozavr at gmail.com>*/