On Sunday, 30 April 2023 at 17:51:15 UTC, Eric P626 wrote:
The title of this thread might be weird, but I am currently
reconsidering the language and tools I am using before
returning to production again.
# Objective and projects
Make simple turn based (no animation) video games using Allegro
4, maybe 5, and eventually SDL on multiple platforms: Linux and
Windows for now, but could expand, if possible, to other OS or
hardware like MacOS, Android, Raspberry Pi, Arduino, etc. I
will also be using Sqlite databases in most of my projects.
I have a set of projects I am decomposing and refactoring. I am
wondering if I should continue in the same direction or if I
should change my tools and languages. I am considering D or
betterC as an alternative.
# Language
I am currently using plain C and I love it. I am allergic to
C++ and object oriented programming. Having read a bit about
the features of the D language makes me doubt continuing using
the C language. Especially the betterC option that allows
making C compatible libraries and software using some of the D
syntax. Unlocking for example the use of foreach statements.
The D syntax is interesting, trying to extend the C language
the same way would require macros which is not recommended for
many reasons. The C language evolves very slowly, so I'll
probably be dead by the time they implements foreachs (bools
are not even native to the language yet(requires stdbool.h).
With D, I have 2 options: 1) create a D software with the D
runtime 2) Create a betterC software that create C compatible
code. I am making both games and libraries, so making C
compatible libraries is almost essential.
The objective is to have a simple language, without object
oriented stuff, that is relatively productive and allow low
level manipulations. This explains the interest in getting
foreachs for more control, security, less bugs therefore less
debugging and more production.
So again betterC could be a good solution even though it's not
well documented what features is exactly part of betterC. I
wonder if it's worth learning the entire D language since a
majority of the features might not make it to betterC. Also I
have no interest in OOP, which is relatively present in the D
language.
# Cross-Compiling and Interoperability
This is where the troubles are starting. Now I am trying to
figure out what can work together in a cross compiling context.
Some of my projects are currently compiling windows binaries
from linux OS using Mingw on gitlab.
Allegro seems to support both C and D, but if I only use the
better C switch, I should be able to use any C library without
any issue. Sqlite is also a plain C library, it should be
usable with betterC and D.
But how is this possible in a cross-compiling context. I am not
sure if I can do that with the D language either as pure D or
better C. DMD does not seem to offer cross compiling. GDC can
compile better C, but not sure mingw can compile D/betterC code.
So I am not really sure what can be done in cross-compiling,
and if it works for Linux and Windows, it might not work for
other platforms like MacOS, Rpie, Arduino if one day, I decide
to take this route.
So what language do you recommend:
* Keep everything in plain C
* Use C patched with macros to gain some language features like
Foreach
* Use BetterC for everything
* Use D for the games, and better C or C for the libraries(To
keep some compatibilities)
* Use D for everything, no C compatibility.
If you have other suggestions or questions let me know.
I recommend using D with -betterC
D is better at being a better C than it is at being a better high
level language
DMD doesn't support cross compilation but LDC do!
The consensus is to use DMD for fast compiles and then use LDC
for your optimized release builds (LLVM is unmatched for
optimizations, LDC is our savior)
So since you are making games, that's the perfect combo, and
since game devs don't want any kind of GC or exception handling
in their way, -betterC will suit your needs
-betterC and DMD is still painful to use for scenarios like this
so be warned: https://issues.dlang.org/show_bug.cgi?id=20737, it
is sad that it still is not fixed in 2023..
I suggest to maintain C compatibility since you are also making
libraries, it would be sad for your users to not be able to
consume your libs because it requires bunch of needs they do not
use, C works everywhere and is one of the few languages that
didn't got infested with OOP, screw redhat/gnome people
