Re: [lmi] Switch to using C++11 uniform initialization in the ctor initializer lists?

[Greg Chicares]

On 2018-08-30 11:55, Vadim Zeitlin wrote:
[...]
>  I've finally pushed d8c4d89c72821b79486abb18c6d7a51580854537 to the main
> repository. It's mostly just your commit (and I kept its original authoring
> information for this reason) with a couple of small amendments from mine,

Okay, your amendments were in these two files only:
  multidimgrid_safe.tpp
  multidimgrid_tools.hpp
One is just initializing a vector<string> by value, the same as
many changes we'd already made elsewhere. The other initializes
bounds of type 'Integral', which is a template parameter, and
there's no easy way to see the actual type of that parameter in
every case, and no assertion that it's an integer type, so let's
add one: okay, it compiles. (I'll commit that assertion soon.)

Because this is the only one of these hundreds of changes that I
feel I don't understand, I'd better test it. Reading the inline
documentation leads me to think this is for the "Duration" limit
in the product editor, so I edit a '.database' file and hammer
at it with all the Krash-Fu I can muster, and I can't break it.
So there's a probabilistic proof that it's correct.

> the whitespace changes got committed together with the more significant
> ones, but I don't think it was worth untangling them at this stage.

Agreed. Sometimes, commingled whitespace changes are bad, e.g.,
if there's only one substantive code change mixed with ten
whitespace-only changes. But these changes aren't in that bad
category: the lines where whitespace was changed are almost
exactly the lines where code changed substantively.

>  I reviewed the changes several times and ran all the tests I could think
> of, so I sincerely hope that I hadn't introduced any errors, but please let
> me know if I still did.

I'm confident that you didn't. But we'll test everything anyway.

>  And thanks for resolving this whole issue so quickly!

Thank you for merging your work and mine together, and then
rebasing the whole thing.