r304956 - [ODRHash] Change the fall-back diagnostic error.

[Richard Trieu via cfe-commits]

Yes, I will go correct the diagnostic text.

In theory, there is nothing that is supposed to reach this diagnostic or
the one below it.  Except that the hasher isn't complete yet and some
things slip through.  Once things are more stable, these should be replaced
with llvm_unreachable instead.

On Wed, Jun 7, 2017 at 6:05 PM, Richard Smith <richard at metafoo.co.uk> wrote:

> On 7 June 2017 at 17:56, Richard Trieu via cfe-commits <
> cfe-commits at lists.llvm.org> wrote:
>
>> Author: rtrieu
>> Date: Wed Jun  7 19:56:21 2017
>> New Revision: 304956
>>
>> URL: http://llvm.org/viewvc/llvm-project?rev=304956&view=rev
>> Log:
>> [ODRHash] Change the fall-back diagnostic error.
>>
>> Provide a little more information when a ODR violation is detected, but
>> the
>> specific error could not be diagnosed.
>>
>> Modified:
>>     cfe/trunk/include/clang/Basic/DiagnosticSerializationKinds.td
>>     cfe/trunk/lib/Serialization/ASTReader.cpp
>>
>> Modified: cfe/trunk/include/clang/Basic/DiagnosticSerializationKinds.td
>> URL: http://llvm.org/viewvc/llvm-project/cfe/trunk/include/clang/
>> Basic/DiagnosticSerializationKinds.td?rev=304956&r1=304955&
>> r2=304956&view=diff
>> ============================================================
>> ==================
>> --- cfe/trunk/include/clang/Basic/DiagnosticSerializationKinds.td
>> (original)
>> +++ cfe/trunk/include/clang/Basic/DiagnosticSerializationKinds.td Wed
>> Jun  7 19:56:21 2017
>> @@ -174,6 +174,13 @@ def note_module_odr_violation_mismatch_d
>>    "method %2 with %ordinal3 parameter of type %4%select{| decayed from
>> %6}5|"
>>    "method %2 with %ordinal3 parameter named %4}1">;
>>
>> +def err_module_odr_violation_mismatch_decl_unknown : Error<
>> +  "%q0 has different definitions in different modules; first difference
>> is "
>> +  "%select{definition in module '%2'|defined here}1 found different "
>> +  "%select{||||static assert|field|method|other}3">;
>> +def note_module_odr_violation_mismatch_decl_unknown : Note<
>> +  "but in '%0' found different %select{||||static
>> assert|field|method|other}1">;
>>
>
> This will produce messages like
>
> "first difference is defined here found different static assert"
> "first difference is definition in module 'X' found different other"
>
> ... which seem pretty confusing. Can this be rephrased so that it forms
> easier-to-read sentences? (Also, can we get some test coverage for this
> diagnostic?)
>
>
>> +
>>  def warn_duplicate_module_file_extension : Warning<
>>    "duplicate module file extension block name '%0'">,
>>    InGroup<ModuleFileExtension>;
>>
>> Modified: cfe/trunk/lib/Serialization/ASTReader.cpp
>> URL: http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/Serializat
>> ion/ASTReader.cpp?rev=304956&r1=304955&r2=304956&view=diff
>> ============================================================
>> ==================
>> --- cfe/trunk/lib/Serialization/ASTReader.cpp (original)
>> +++ cfe/trunk/lib/Serialization/ASTReader.cpp Wed Jun  7 19:56:21 2017
>> @@ -9753,13 +9753,13 @@ void ASTReader::diagnoseOdrViolations()
>>        if (Diagnosed == true)
>>          continue;
>>
>> -      Diag(FirstRecord->getLocation(),
>> -           diag::err_module_odr_violation_different_definitions)
>> -          << FirstRecord << FirstModule.empty() << FirstModule;
>> -
>> -      Diag(SecondRecord->getLocation(),
>> -           diag::note_module_odr_violation_different_definitions)
>> -          << SecondModule;
>> +      Diag(FirstDecl->getLocation(),
>> +           diag::err_module_odr_violation_mismatch_decl_unknown)
>> +          << FirstRecord << FirstModule.empty() << FirstModule <<
>> FirstDiffType
>> +          << FirstDecl->getSourceRange();
>> +      Diag(SecondDecl->getLocation(),
>> +           diag::note_module_odr_violation_mismatch_decl_unknown)
>> +          << SecondModule << FirstDiffType <<
>> SecondDecl->getSourceRange();
>>        Diagnosed = true;
>>      }
>>
>>
>>
>> _______________________________________________
>> cfe-commits mailing list
>> cfe-commits at lists.llvm.org
>> http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/cfe-commits/attachments/20170607/d2c29a7d/attachment.html>