r240185 - Allow the cf_returns_[not_]retained attributes to appear on out-parameters.

[Aaron Ballman]

On Mon, Jun 22, 2015 at 11:15 PM, Jordan Rose <jordan_rose at apple.com> wrote:
>
> On Jun 22, 2015, at 12:06 , Douglas Gregor <dgregor at apple.com> wrote:
>
>
> On Jun 20, 2015, at 12:55 PM, Aaron Ballman <aaron at aaronballman.com> wrote:
>
> On Fri, Jun 19, 2015 at 7:17 PM, Douglas Gregor <dgregor at apple.com> wrote:
>
> Author: dgregor
> Date: Fri Jun 19 18:17:46 2015
> New Revision: 240185
>
> URL: http://llvm.org/viewvc/llvm-project?rev=240185&view=rev
> Log:
> Allow the cf_returns_[not_]retained attributes to appear on out-parameters.
>
> Includes a simple static analyzer check and not much else, but we'll also
> be able to take advantage of this in Swift.
>
> This feature can be tested for using
> __has_feature(cf_returns_on_parameters).
>
> This commit also contains two fixes:
> - Look through non-typedef sugar when deciding whether something is a CF
> type.
> - When (cf|ns)_returns(_not)?_retained is applied to invalid properties,
>  refer to "property" instead of "method" in the error message.
>
> rdar://problem/18742441
>
> Added:
>    cfe/trunk/test/SemaObjC/attr-cf_returns.m
> Modified:
>    cfe/trunk/include/clang/Basic/DiagnosticSemaKinds.td
>    cfe/trunk/include/clang/StaticAnalyzer/Checkers/ObjCRetainCount.h
>    cfe/trunk/lib/Analysis/CocoaConventions.cpp
>    cfe/trunk/lib/Lex/PPMacroExpansion.cpp
>    cfe/trunk/lib/Sema/SemaDeclAttr.cpp
>    cfe/trunk/lib/StaticAnalyzer/Checkers/RetainCountChecker.cpp
>    cfe/trunk/test/Analysis/retain-release.m
>
> Modified: cfe/trunk/include/clang/Basic/DiagnosticSemaKinds.td
> URL:
> http://llvm.org/viewvc/llvm-project/cfe/trunk/include/clang/Basic/DiagnosticSemaKinds.td?rev=240185&r1=240184&r2=240185&view=diff
> ==============================================================================
> --- cfe/trunk/include/clang/Basic/DiagnosticSemaKinds.td (original)
> +++ cfe/trunk/include/clang/Basic/DiagnosticSemaKinds.td Fri Jun 19 18:17:46
> 2015
> @@ -2751,8 +2751,8 @@ def warn_ns_attribute_wrong_return_type
>   "return %select{an Objective-C object|a pointer|a non-retainable
> pointer}2">,
>   InGroup<IgnoredAttributes>;
> def warn_ns_attribute_wrong_parameter_type : Warning<
> -  "%0 attribute only applies to %select{Objective-C object|pointer}1 "
> -  "parameters">,
> +  "%0 attribute only applies to "
> +  "%select{Objective-C object|pointer|pointer-to-CF-pointer}1 parameters">,
>
>
> CF-pointer doesn't appear to be a common term in the code base (this
> is the only diagnostic with it, and we have one comment on it), or on
> the web. Is there a better term we could use? If not, it's fine, I was
> mostly curious whether this is a common term of art that users would
> situationally understand.
>
>
> I don’t think we have a better term for this. Jordan, any ideas?
>
>
> I think the official name would be "CF object pointer" or "CF object
> reference", or "Core Foundation" spelled out in full, but given how it's a
> descriptor for "parameters" I'm concerned about making it much longer than
> what's already there. I think it will make sense in practice.

Okay! If folks will understand it in situ, then it's great by me. I
just wanted to double-check. :-)

Thanks!

~Aaron