> On Dec. 22, 2015, 4:52 a.m., Benjamin Hindman wrote:
> > 3rdparty/libprocess/3rdparty/stout/include/stout/jsonify.hpp, lines 58-63
> > <https://reviews.apache.org/r/41593/diff/2/?file=1173648#file1173648line58>
> >
> >     An implicit cast operator that is non-const is always code smell 
> > because the compiler is what is often implicitly invoking the function! In 
> > this case I know why it's safe for you to do this, but anyone else reading 
> > the code would have no idea. It would be great to have a comment here which 
> > explains why it's safe to `std::move(*this)`, otherwise, this looks 
> > dangerous!
> >     
> >     Also, if anyone ever changes this implementation is there anyway to 
> > check this invariant that you have so that they make sure they change this 
> > code too?
> 
> Michael Park wrote:
>     Right. I initially had rvalue ref-qualifier on this operator like this 
> `operator std::string() && { ... }`, which makes it so that it is only 
> invocable on a rvalue `JsonifyProxy`. I took it out because I thought maybe 
> it would have been controversial reactions towards it.
>     
>     How about we just call `write_` directly instead? It would have been nice 
> to leverage `operator<<` but without the ref-qualifier it's not quite 
> accurate.
>     
>     ```cpp
>     operator std::string() const {
>       std::ostringstream stream;
>       that.write_(&stream);
>       return stream.str();
>     }
>     ```
>     
>     What do you think?

Synced with BenH offline, we agreed to introduce the ref-qualifiers.


- Michael


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41593/#review111575
-----------------------------------------------------------


On Jan. 4, 2016, 7:14 p.m., Michael Park wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41593/
> -----------------------------------------------------------
> 
> (Updated Jan. 4, 2016, 7:14 p.m.)
> 
> 
> Review request for mesos and Benjamin Hindman.
> 
> 
> Repository: mesos
> 
> 
> Description
> -------
> 
> See summary.
> 
> 
> Diffs
> -----
> 
>   3rdparty/libprocess/3rdparty/stout/README.md 
> a2a3dc6120b10c8fe82088b9f0ebecfa5642f945 
>   3rdparty/libprocess/3rdparty/stout/include/Makefile.am 
> b2dea9b93adfa3ea0415a0b5c81a369dd29b6cfe 
>   3rdparty/libprocess/3rdparty/stout/include/stout/jsonify.hpp PRE-CREATION 
> 
> Diff: https://reviews.apache.org/r/41593/diff/
> 
> 
> Testing
> -------
> 
> 
> Thanks,
> 
> Michael Park
> 
>

Reply via email to