Notes |
|
|
Interestingly, nullptr keyword has been used for a long time. It's a C++11 feature. Wonderful if it's been supported by some older Microsoft compilers, before the switch to MinGW. |
|
|
(0021387)
|
Pol M
|
2020-06-07 20:33
|
|
We are already using the C++11 standard |
|
|
|
Quote from Pol M We are already using the C++11 standard
Yes, but as seen from child issues this can be further improved. |
|
|
(0021398)
|
Pol M
|
2020-06-07 22:51
|
|
Quote from Wub
Yes, but as seen from child issues this can be further improved.
C++ introduces new ways of doing the same things but better each version. That does not mean that necessarily all the code that did things the old way needs to be changed :)
That said, sure, let's adapt your proposed changes :D |
|
|
(0021401)
|
Blzut3
|
2020-06-08 00:31
|
|
Generally speaking old code bases don't find and replace everything when a new standard comes out since 1) it's usually fixing non-broken code. 2) Needlessly points git blame to an unhelpful commit.
Almost everything you point out isn't wrong, and is indeed how code should be written going forward. But is this solving something? Now I've been busy with other stuff so I'm not going to tell you all not to do this if it's something that motivates you all, but do consider that such refactors are unusual. |
|
|
(0021411)
|
Pol M
|
2020-06-08 02:06
|
|
I completely agree with Blzut, but if Wub is committed to doing these changes I won't be stopping nobody since right now there is no merge issues that could appear. I'll state that 0003801, 0003807, 0003813 and 0003800 are clear cases of what Blzut explains, then there's 0003811 that theoretically allows for optimizations but any mothern compiler can see through these so it's a bit on the limbo and then there's most (not saying all) of the others that do have some value either by making the code safer, objectively more readable, or are simply better. |
|