r177725 - More precisely recognize ObjC method declarations.

Jordan Rose jordan_rose at apple.com
Fri Mar 22 09:37:06 PDT 2013


I would hope to be consistent, though: if you're not going to have a space for the first example, the second shouldn't have one either.

On Mar 22, 2013, at 9:36 , Jordan Rose <jordan_rose at apple.com> wrote:

> I feel like I've seen it in old framework headers, but yeah, maybe we just don't care anymore / for now. I just see us getting a bug about it somewhere down the line...
> 
> 
> On Mar 22, 2013, at 9:33 , Daniel Jasper <djasper at google.com> wrote:
> 
>> The second case is handled (and we have tests for that). Is the first case realistically used somewhere (where people would mind us removing the space)?
>> 
>> 
>> On Fri, Mar 22, 2013 at 5:16 PM, Jordan Rose <jordan_rose at apple.com> wrote:
>> These are valid Objective-C method declarations.
>> 
>> - methodReturningID;
>> + methodWithArgument:first context:second;
>> 
>> The types are assumed to be 'id', just like C90 return types are assumed to be 'int'.
>> 
>> Most people don't write code like this today, but it is legal. It makes sense for clang-format to not handle these well in macros, though.
>> 
>> 
>> On Mar 22, 2013, at 3:44 , Daniel Jasper <djasper at google.com> wrote:
>> 
>> > Author: djasper
>> > Date: Fri Mar 22 05:44:43 2013
>> > New Revision: 177725
>> >
>> > URL: http://llvm.org/viewvc/llvm-project?rev=177725&view=rev
>> > Log:
>> > More precisely recognize ObjC method declarations.
>> >
>> > Otherwise, +/- and the beginning of constants can be recognized
>> > incorrectly.
>> >
>> > Before:  #define A - 1
>> > After:   #define A -1
>> >
>> > Modified:
>> >    cfe/trunk/lib/Format/TokenAnnotator.cpp
>> >    cfe/trunk/unittests/Format/FormatTest.cpp
>> >
>> > Modified: cfe/trunk/lib/Format/TokenAnnotator.cpp
>> > URL: http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/Format/TokenAnnotator.cpp?rev=177725&r1=177724&r2=177725&view=diff
>> > ==============================================================================
>> > --- cfe/trunk/lib/Format/TokenAnnotator.cpp (original)
>> > +++ cfe/trunk/lib/Format/TokenAnnotator.cpp Fri Mar 22 05:44:43 2013
>> > @@ -313,9 +313,11 @@ private:
>> >     switch (Tok->FormatTok.Tok.getKind()) {
>> >     case tok::plus:
>> >     case tok::minus:
>> > -      // At the start of the line, +/- specific ObjectiveC method
>> > -      // declarations.
>> > -      if (Tok->Parent == NULL)
>> > +      // At the start of the line, +/- specify ObjectiveC method declarations.
>> > +      if (Tok->Children.empty() || Tok->Children[0].Children.empty())
>> > +        break; // Can't be an ObjectiveC method declaration.
>> > +      if (Tok->Parent == NULL && (Tok->Children[0].is(tok::l_paren) ||
>> > +                                  Tok->Children[0].Children[0].is(tok::colon)))
>> >         Tok->Type = TT_ObjCMethodSpecifier;
>> >       break;
>> >     case tok::colon:
>> >
>> > Modified: cfe/trunk/unittests/Format/FormatTest.cpp
>> > URL: http://llvm.org/viewvc/llvm-project/cfe/trunk/unittests/Format/FormatTest.cpp?rev=177725&r1=177724&r2=177725&view=diff
>> > ==============================================================================
>> > --- cfe/trunk/unittests/Format/FormatTest.cpp (original)
>> > +++ cfe/trunk/unittests/Format/FormatTest.cpp Fri Mar 22 05:44:43 2013
>> > @@ -2067,6 +2067,8 @@ TEST_F(FormatTest, UnderstandsUnaryOpera
>> >                "case -1:\n"
>> >                "  break;\n"
>> >                "}");
>> > +  verifyFormat("#define X -1");
>> > +  verifyFormat("#define X -kConstant");
>> >
>> >   verifyFormat("const NSPoint kBrowserFrameViewPatternOffset = { -5, +3 };");
>> >   verifyFormat("const NSPoint kBrowserFrameViewPatternOffset = { +5, -3 };");
>> >
>> >
>> > _______________________________________________
>> > cfe-commits mailing list
>> > cfe-commits at cs.uiuc.edu
>> > http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
>> 
>> 
> 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/cfe-commits/attachments/20130322/2d9cc307/attachment.html>


More information about the cfe-commits mailing list