[PATCH] Introduce llvm/ADT/OptionSet.h utility class

[Philip Reames via llvm-commits]

On 02/12/2016 01:19 PM, Chris Lattner wrote:
> On Feb 12, 2016, at 1:08 PM, Chandler Carruth <chandlerc at gmail.com 
> <mailto:chandlerc at gmail.com>> wrote:
>>
>>>         I got permission internally and contributed the header under
>>>         the LLVM license, I don’t think that whether it originated
>>>         from swift repo or not makes any difference here.
>>>         If you could elaborate more on your concerns it would be
>>>         helpful.
>>>
>>>
>>>     How do we know whether any other contributor to Swift authored
>>>     some of the code in this file? Does Apple get copyright
>>>     assignment for all Swift contributions? Only if Apple has
>>>     copyright assignment for all contributions to Swift or only
>>>     people from Apple have every contributed to this part of Swift
>>>     is asking internally enough.
>>>
>>>     And since this is an open source project, it would seem polite
>>>     (even if not necessary) to also ask the community rather than
>>>     just asking internally. We don't all work at Apple. =/
>>
>
> The code is 100% copyrighted by Apple.  The full history in the swift 
> repo is here:
> https://github.com/apple/swift/commits/master/include/swift/Basic/OptionSet.h
>
> There are two trivial patches by non-apple contributors.  Both are 
> both single line patches that adjust comments (one is the first line 
> of the file, one is the copyright date):
> https://github.com/apple/swift/commit/1339b5403bbaf6205abb2bfdf7fabadef1aacc70#diff-1e8f8df8addd9510deea20d4bea2eda2
> https://github.com/apple/swift/commit/e3a4147ac94e55fcab1d14e949f572b53d9eb638#diff-1e8f8df8addd9510deea20d4bea2eda2
>
> That said, both of those patches are irrelevant to the discussion, 
> because Argyrios removed the swift header and replaced it with the 
> LLVM header when he checked it in.
Just to be clear, you're saying that Apple owns the copyright on the 
entire file as posted for review and can thus relicense under the LLVM 
license?  If so, that would resolve the licensing concern.  In the 
future, let's make sure that gets mentioned in the review/commit thread 
to avoid confusion.

With that, we're back to "let's post a patch and get it reviewed" per 
the normal process.





>
>>     + Chris, for comment.
>>
>>
>> Also, just to be clear (since I think my original email wasn't 
>> terribly): I think something as simple as this code review thread 
>> would have been fine if you had waited for the code review, and made 
>> it clear why it was OK to contribute (only Apple folks have touched 
>> it, or whatever).
>>
>> I'm not suggesting we need to have some big or complex discussion. =] 
>> We have plenty of those on harder subjects.
>
> I agree that the header should get the normal patch review!
>
> -Chris

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20160212/c1cf96ab/attachment.html>