[cfe-dev] Getting fully qualified names of random qualtypes.

[Manuel Klimek via cfe-dev]

The right next step is to send a patch / code review where we can discuss
the change. Is there one that I have missed?

On Thu, Oct 22, 2015, 10:57 PM Sterling Augustine <saugustine at google.com>
wrote:

> I copied these file directly from Cling to Tooling/Core/, and renamed them
> QualTypeNames.[cpp|h].
>
> It compiles and works just fine as a direct copy.
>
> I have made a couple of very minor changes, such as moving the enclosing
> namespace from cling to clang. Shall I do a more substantial clang-style
> edit?
>
>
>
> On Thu, Oct 22, 2015 at 5:29 AM, Manuel Klimek <klimek at google.com> wrote:
>
>> Specifially, Benjamin has a patch out that adds a Tooling/Core/Lookup.h
>> where this would fit in quite nicely.
>>
>> On Wed, Oct 21, 2015 at 5:31 PM Daniel Berlin <dannyb at google.com> wrote:
>>
>>> The license itself is fine.
>>>
>>>
>>> On Wed, Oct 21, 2015 at 1:04 AM, Manuel Klimek <klimek at google.com>
>>> wrote:
>>>
>>>> On Wed, Oct 21, 2015 at 8:08 AM Richard Smith <richard at metafoo.co.uk>
>>>> wrote:
>>>>
>>>>> ASTContext isn't appropriate for utility code for tooling support;
>>>>> that should go in libTooling instead. The question is whether any of this
>>>>> functionality is useful for the compiler itself or just for tools that use
>>>>> it as a library, and I suspect it's only really useful for the latter.
>>>>>
>>>>
>>>> Good point, in that case I'd put it somewhere in Tooling/Core. cc'ing
>>>> Benjamin, who had (planned) a patch to open up a new source file there
>>>> anyway.
>>>>
>>>>
>>>>>
>>>>> Danny, can you comment on the licensing question?
>>>>>
>>>>> On Tue, Oct 20, 2015 at 9:13 PM, Manuel Klimek via cfe-dev <
>>>>> cfe-dev at lists.llvm.org> wrote:
>>>>>
>>>>>> I think it should probably be part of ASTContext. I have no idea
>>>>>> about the license. Richard?
>>>>>>
>>>>>> On Wed, Oct 21, 2015, 2:27 AM Sterling Augustine via cfe-dev <
>>>>>> cfe-dev at lists.llvm.org> wrote:
>>>>>>
>>>>>>> So clang folks, where should this new code to calculate the
>>>>>>> complete, fully qualified name of arbitrary qualtypes go?
>>>>>>>
>>>>>>> My best guess is as a new file in clang/lib/AST, but it's not a
>>>>>>> great fit.
>>>>>>>
>>>>>>> Does the license as listed in the file (reproduced below) create any
>>>>>>> issues?
>>>>>>>
>>>>>>> // This file is dual-licensed: you can choose to license it under
>>>>>>> the University
>>>>>>> // of Illinois Open Source License or the GNU Lesser General Public
>>>>>>> License. See
>>>>>>> // LICENSE.TXT for details.
>>>>>>>
>>>>>>> On Thu, Oct 15, 2015 at 8:47 AM, Philippe Canal <pcanal at fnal.gov>
>>>>>>> wrote:
>>>>>>>
>>>>>>>> Hi Sterling,
>>>>>>>>
>>>>>>>> > Would you all be amenable to contributing cling/lib/Utils/AST.cpp
>>>>>>>> to clang?
>>>>>>>>
>>>>>>>> Of course :).  I agree that this feature belongs in clang proper.
>>>>>>>>
>>>>>>>> AST.cpp is pretty much standalone (depends only on clang and llvm)
>>>>>>>>
>>>>>>>> Thanks,
>>>>>>>> Philippe.
>>>>>>>>
>>>>>>>> PS.  A part that has not yet been integrated in AST.cpp is the
>>>>>>>> adding of default
>>>>>>>> template parameters (for user classes where that parameter might
>>>>>>>> matter for I/O)
>>>>>>>> and can be found in:
>>>>>>>>
>>>>>>>> https://root.cern.ch/gitweb?p=root.git;a=blob;f=core/metautils/src/TMetaUtils.cxx
>>>>>>>> in the routines
>>>>>>>>     ROOT::TMetaUtils::GetNormalizedType
>>>>>>>>     ROOT::TMetaUtils::AddDefaultParameters
>>>>>>>>     ROOT::TMetaUtils::KeepNParams
>>>>>>>>
>>>>>>>>
>>>>>>>> On 10/14/15 11:53 AM, Sterling Augustine wrote:
>>>>>>>>
>>>>>>>> On Tue, Oct 13, 2015 at 2:48 PM, Philippe Canal < <pcanal at fnal.gov>
>>>>>>>> pcanal at fnal.gov> wrote:
>>>>>>>>
>>>>>>>>> Hi,
>>>>>>>>>
>>>>>>>>> We also discovered that we could not find in clang a routine that
>>>>>>>>> reliably generate the name (or the equivalent
>>>>>>>>> decl/type) corresponding to a type as it would need to be typed
>>>>>>>>> from the global namespace (i.e. a unique identifier
>>>>>>>>> of the class);
>>>>>>>>>
>>>>>>>>
>>>>>>>> Philippe,
>>>>>>>>
>>>>>>>> Would you all be amenable to contributing cling/lib/Utils/AST.cpp
>>>>>>>> to clang?
>>>>>>>>
>>>>>>>> I would be willing to write-up the patch if that would be helpful.
>>>>>>>> I'm not sure how deep the dependencies go.
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> cfe-dev mailing list
>>>>>>> cfe-dev at lists.llvm.org
>>>>>>> http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-dev
>>>>>>>
>>>>>>
>>>>>> _______________________________________________
>>>>>> cfe-dev mailing list
>>>>>> cfe-dev at lists.llvm.org
>>>>>> http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-dev
>>>>>>
>>>>>>
>>>>>
>>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/cfe-dev/attachments/20151023/03a4ec9d/attachment.html>