few warnings/errors in copy assignment

[richardlalancetteyoui]

Hi!
I have integrated your library in our code.
Can you advise regarding the following 3 problems?

Static analysis found 3 issues in your PR (xcode 11's clang-analyzer):
• Use of identical expressions in json.hpp:17964 (in function operator=)
• Use of identical expressions in json.hpp:21647 (in function swap)
• Use of identical expressions in json.hpp:24444 (in function swap)

What do you want to achieve?

Integration without warning or issue if possible.

What have you tried?

I'm starting to investigation removing the warnings in our own copy of the code.

Which compiler and operating system are you using?

Model Identifier: iMac14,2
Processor Speed: 3.70 GHz
Number of Processors: 1
Total Number of Cores: 6
Memory: 64 GB

• Compiler: xcode
• Operating system: 10.13.6

Which version of the library did you use?

• latest release version 3.7.3
• other release - please state the version: ___
• the develop branch
  commit 9212590

[nlohmann]

These are the lines in question:

17964 std::is_nothrow_move_constructible<value_t>::value
21647 std::is_nothrow_move_constructible<value_t>::value
24444 is_nothrow_move_constructible<nlohmann::json>::value

I am not sure what this warning means. Can you describe how you executed the static anaylsis?

[richardlalancetteyoui]

This is part of our build system on jenkins.
Let me dig the tool that generated this.

[richardlalancetteyoui]

That's odd.
AppCode also sees an issue. But, might be red herrings.

[richardlalancetteyoui]

Unless you see an issue in that code. I am not familiar with noexcept.

[richardlalancetteyoui]

Analyser used is xcode 11's clang-analyzer

https://clang-analyzer.llvm.org/xcode.html ?

[richardlalancetteyoui]

Got it to show up in xcode as well.

[FrancoisChabot]

The only way this warning can make any sense at all would be if the static analyzer has reached the conclusion that value_t and json_value are 100% guaranteed to be the same type in every template instantiation.

[richardlalancetteyoui]

The quality of those analyzers may vary for sure!

I'm wrapping up this code with preprocessor for now.

Feel free to close this issue.

[nlohmann]

I think this is a false positive. Closing.