Small cleanups 01

[vondele]

PR to collect small patches that are close to trivial.

No functional change

[adentong]

search.cpp line 1329, can get rid of bracket at the end around !PvNode && ttPv

[joergoster]

#include <numeric> in bitbase.cpp is not needed.

[joergoster]

In main.cpp, move #include "endgame.h" up and place it below #include "bitboard.h" to preserve alphabetical order.
And in endgame.h, move #include <unordered_map> down and place it above #include <utility>.
That's it.

[xoto10]

Remove redundant comment in evaluate.cpp

[TerjeKir]

Line 636 in position.cpp could use check_squares() instead of st->checkSquares[] directly

[snicolet]

The comment "Greater bonus when landing square is occupied" near RestictedPiece can also be removed.

[protonspring]

the comment for shift doesn't seem accurate since we may shift more than one step.

[protonspring]

movegen.cpp, line 55. This comment can probably be removed.

[gvreuls]

Changing the signature of the new void* aligned_ttmem_alloc(size_t allocSize, void** mem) function into the more C++ idiom void* aligned_ttmem_alloc(size_t allocSize, void*& mem) would be nice, or is that too large a change?
misc.h line 36
misc.cpp lines 306-318
tt.cpp line 69
(Changing it into std::pair<void*, void*> aligned_ttmem_alloc(size_t allocSize) probably diverts too far from our coding style.)

[vondele]

the comment for shift doesn't seem accurate since we may shift more than one step.

file and line number please ?

[protonspring]

Looks like I am too late, but the comment for shift is bitboard.h, line 357. it says we do one step, but we also do NORTH+NORTH and SOUTH+SOUTH.

…

On Wed, Feb 5, 2020 at 7:44 AM Joost VandeVondele ***@***.***> wrote: Closed #2532 <#2532> via 0c878ad <0c878ad> . — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#2532?email_source=notifications&email_token=AHCWOSG5TMQTJDA2623QT6LRBLGFTA5CNFSM4KM7QR3KYY3PNVWWK3TUL52HS4DFWZEXG43VMVCXMZLOORHG65DJMZUWGYLUNFXW5KTDN5WW2ZLOORPWSZGOWNWQSWA#event-3010267480>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AHCWOSBAT5XQCSBIAHJ5HUTRBLGFTANCNFSM4KM7QR3A> .

-- Michael T. Whiteley mobile: 801.707.6886

[vondele]

it is still one step in a direction D, but D happens to be two squares, so I think the comment is still right. Agree ?

[protonspring]

This is super minor, but to me, a KING move is probably one step. NORTH+NORTH is definitely two steps. I put a comment in the new PR so let's move this discussion there.