Concept ACK.

• [Qt] implements concept for different disk sizes on intro #13216 ([Qt] implements concept for different disk sizes on intro by marcoagner)
• [WIP][NET] Add NATPMP support. #12288 ([WIP][NET] Add NATPMP support. by annanay25)

If you consider this pull request important, please also help to review the conflicting pull requests. Ideally, start with the one that should be merged first.

If this is going to be done, shouldn't -Wzero-as-null-pointer-constant or a similar sort of lint script be added to the build system to try and stop it's usage re-occurring? Otherwise this seems like it could end up similar to -Wshadow.

@fanquake Yes that would be the best way to make sure we don't regress! Unfortunately -Wzero-as-null-pointer-constant generates some false positives in our code and some true positives in our dependencies.

Unfortunately -Wzero-as-null-pointer-constant generates some false positives in our code and some true positives in our dependencies.

In which case I suggest to do the same as with -Wshadow: Don't enforce it and don't change existing code. So I tend to Concept NACK here.

@MarcoFalke Oh, I'd love enabling that warning but in contrast to shadowing I think the probably of regressions is much lower – hopefully making this a one time "switch to C++11" change. New code is unlikely to use 0 instead of the proper pointer literal nullptr, no?

I'd assume the instances changed here are from the pre-C++11 era.

Fine, nullptr is type safe, so at least it is an improvement and also doesn't hurt. There shouldn't be any issues given that this compiles, right?

@MarcoFalke Correct! Type safety is our friend.

This change is safe:

• Changing from 0 (or NULL) to nullptr where the zero is used as a null pointer constant is always safe.
• Changing from 0 (or NULL) to nullptr where the zero is NOT used as a null pointer constant will result in a compiler error.

So yes, if it compiles it is safe :-)

Weak concept NAK, but I'm okay with it if others are (I understand it's safe if it compiles).

Too many files changed (100 lines over 58 files) for a change that isn't really a fix (the reason stated in C++ guidelines is 'readability', which is rather weak for a 58 file change IMO). I feel this is one of those things that can be fixed gradually over time as people update the transgressing code.

Closing this PR due to lack of consensus ACK :-)