In D3075#137876, @wraitii wrote:

This is essentially what this diff does though. Except there's no "passing" since the server can 'peek' via being in the same process. Doesn't work for dedicated servers but I don't think it's a huge concern.

That sounds a bit unsafe. Since two parallel object creations depend on the global state. When you create NetClient and NetServer you know that they're on the same machine/in the same process. Why do you need to hide it and delay it?

Still get this when running the game:

Build failure - The Moirai have given mortals hearts that can endure.

Build failure - The Moirai have given mortals hearts that can endure.

Rebase

Build failure - The Moirai have given mortals hearts that can endure.

Build failure - The Moirai have given mortals hearts that can endure.

Fix notes

Successful build - Chance fights ever on the side of the prudent.

Fix the boost compilation. The autolink tries to use the vc141 libraries otherwise, and they don't exist. Since the vc140 ought to work as they are ABI compatible, we can just force using those.
(Coincidentally, the vc120 libs are still there and should be removed).

I'm not a super big fan of what seems to be warning against implicit bool conversions here and there (the nullptr one in particular), I'm kind of wondering what the warnings are.

This fixes the assertion. Can't tell if it's proper though.

Build failure - The Moirai have given mortals hearts that can endure.

Build failure - The Moirai have given mortals hearts that can endure.

Restore the complete patch

Tweaks from @Gallaecio 's comments.

Missing #include <map> in LongPathfinder.h

Successful build - Chance fights ever on the side of the prudent.

In D3075#137875, @vladislavbelov wrote:

So it does auth correctly on if ps_generate_guid() will be the same for client and server? So it might mean that someone can relatively easily generate it?

So it does auth correctly on if ps_generate_guid() will be the same for client and server? So it might mean that someone can relatively easily generate it? Why to not generate a GUID with some randomness before and pass it to both?

I'm not convinced TBH. If this is hardcoded at compile-time, either we drop support or it's pretty much useless for releases. SIMD-capable compilers seem able to vectorise this functions, so custom versions don't seem particularly useful.
If there was a runtime switch that actually increased performance, might be more interesting.

grep -r '"units/"' suggest you should also have a look at the maps/scripts/Regicide.js and maps/random/rmgen-common/player.js files.
There might be more, I don't know what to look for, and I can't say I understand map scripts.

• Refactor the pkgconfig premake5 module
• Attempt to place Windows includes in premake5 into the correct place in the include search order
• Sort the new aftersysincludedirs premake5 global to a better location (alphabetically) in its file

Broke my patches 😠

We gained a big 30 seconds of 23 minutes compilation time on the CI with is like 2% :p

Build failure - The Moirai have given mortals hearts that can endure.

Successful build - Chance fights ever on the side of the prudent.

Build failure - The Moirai have given mortals hearts that can endure.

Needed a rebase ofc.

Build failure - The Moirai have given mortals hearts that can endure.

Build failure - The Moirai have given mortals hearts that can endure.

Build failure - The Moirai have given mortals hearts that can endure.

Fix MSVC warnings, ready for merge.

Also copyright headers.

In D2689#137770, @Nescio wrote:

Maybe it has a different name on Debian? E.g. fmt-dev, fmtlib, libfmt.

Seems no real benefit to compilation time found on CI (I suspect our precompiled headers are too good), though the incremental one remains, particularly on the simulation side with the pathfinding cleanup.

D2901/rP23890 added the champion crossbowman to the arsenal. Moreover, various crossbowman units are available in the Han China mod: https://github.com/0ADMods/han_china

Maybe it has a different name on Debian? E.g. fmt-dev, fmtlib, libfmt.

We have a minor pickle: our GCC CI runs from the gcc:6 docker image, which is built on Debian 8, and I can't apt-get install fmt on that (there's just no package). It's also no longer on Buster, which ships gcc7 minimum.

This looks fairly good to me, and seems a nice solution for clang and gcc.

That is exactly what I meant also :) But I understand my wording was wrong.

(They _are_ in the game already.)

What I meant is available to be trained by a civilization, so we can see how they actually behave in a real game.

Build failure - The Moirai have given mortals hearts that can endure.

Build failure - The Moirai have given mortals hearts that can endure.

Build failure - The Moirai have given mortals hearts that can endure.

Fix notes, more warnings

why are you removing size_t ?

Successful build - Chance fights ever on the side of the prudent.

As noted in D3126 there are a few warnings on MSVC2015 C++14

Committing this over the WE.

Successful build - Chance fights ever on the side of the prudent.

Working fine here (GCC 7.5, Lubuntu 18.04).

Rebased before committing.

Build failure - The Moirai have given mortals hearts that can endure.

Build failure - The Moirai have given mortals hearts that can endure.

Build failure - The Moirai have given mortals hearts that can endure.