[cfe-commits] r171732 - in /cfe/trunk: include/clang/Format/Format.h lib/Format/Format.cpp lib/Format/UnwrappedLineParser.cpp lib/Format/UnwrappedLineParser.h unittests/Format/FormatTest.cpp

[Daniel Jasper]

I see. I first thought the same. But switching forward and back between the
two, I actually think that the new one is at least as readable. Or at least
I can't say that one is strictly better then the other. Maybe we'll find
other examples where the decision is more obvious..


On Mon, Jan 7, 2013 at 2:33 PM, Chandler Carruth <chandlerc at google.com>wrote:

> FYI
>
> On Mon, Jan 7, 2013 at 5:26 AM, Daniel Jasper <djasper at google.com> wrote:
>
>> -      FormatTok.HasUnescapedNewline =
>> -          Text.count("\\\n") != FormatTok.NewlinesBefore;
>> +      FormatTok.HasUnescapedNewline = Text.count("\\\n") !=
>> +                                      FormatTok.NewlinesBefore;
>>
>
> This doesn't seem as good as the rest of the changes. Here I agree with
> your strategy of breaking the outer most binary operator. That's why I was
> trying to differentiate between function arguments being broken after an
> '=' and breaking a binary operator...
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/cfe-commits/attachments/20130107/af8e2a75/attachment.html>