LuaJIT implementation is faster than Lua, but it's compatible with Lua 5.1 with some extensions from 5.2 instead of Lua 5.4

C++ API should be mostly the same, but some lua code in OD2 may need to be updated

Describe the bug Reaching the end of build process this message pop out

[100%] Linking CXX executable abyss_test
mold: error: undefined symbol: absl::lts_20220623::container_internal::ConvertDeletedToEmptyAndFullToDeleted(absl::lts_20220623::container_internal::ctrl_t*, unsigned long)
>>> referenced by inifile.cpp

Would be great to skip the tests with an appropriate flag using cmake, for example launching:

cmake ../abyss-sources \
    -DBUILD_TESTING=off \

Thanks for your amazing efforts, I follow this project since the early days.

Additional note: mold (referenced in the log) is a drop-in replacement for gold (the GNU linker). same issue also with gold anyway

Workaround For now I'm using this workaround

# temporary workaround to disable tests
  pacman -S gtest
  rm -rf tests/*
  touch tests/CMakeLists.txt

When compiling on Fedora 35, there's an error during linking:

[[email protected] AbyssEngine-1]$ make [ 27%] Built target libabyss [ 33%] Linking C executable "Abyss Engine" /usr/bin/ld: CMakeFiles/abyssengine.dir/src/abyssengine.c.o: relocation R_X86_64_32 against.rodata' can not be used when making a PIE object; recompile with -fPIE /usr/bin/ld: CMakeFiles/abyssengine.dir/src/engine/engine.c.o: relocation R_X86_64_32 against .rodata' can not be used when making a PIE object; recompile with -fPIE /usr/bin/ld: CMakeFiles/abyssengine.dir/src/scripting/scripting.c.o: relocation R_X86_64_32 against.rodata' can not be used when making a PIE object; recompile with -fPIE /usr/bin/ld: CMakeFiles/abyssengine.dir/src/misc/util.c.o: relocation R_X86_64_32 against .rodata' can not be used when making a PIE object; recompile with -fPIE /usr/bin/ld: CMakeFiles/abyssengine.dir/src/misc/ini.c.o: relocation R_X86_64_32 against.rodata' can not be used when making a PIE object; recompile with -fPIE /usr/bin/ld: CMakeFiles/abyssengine.dir/src/loader/filesystemloader.c.o: relocation R_X86_64_32S against symbol filesystem_loader_load' can not be used when making a PIE object; recompile with -fPIE /usr/bin/ld: CMakeFiles/abyssengine.dir/src/loader/loader.c.o: relocation R_X86_64_32 against.rodata' can not be used when making a PIE object; recompile with -fPIE /usr/bin/ld: ../libabyss/liblibabyss.a(log.c.o): relocation R_X86_64_32 against .rodata' can not be used when making a PIE object; recompile with -fPIE collect2: error: ld returned 1 exit status make[2]: *** [abyssengine/CMakeFiles/abyssengine.dir/build.make:217: abyssengine/Abyss Engine] Error 1 make[1]: *** [CMakeFiles/Makefile2:160: abyssengine/CMakeFiles/abyssengine.dir/all] Error 2 make: *** [Makefile:136: all] Error 2

This requires adding a line to CMakeLists.txt

set(CMAKE_C_FLAGS "-fPIE")

$ gcc --version gcc (GCC) 11.2.1 20210728 (Red Hat 11.2.1-1) Copyright (C) 2021 Free Software Foundation, Inc. This is free software; see the source for copying conditions. There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

Describe the bug Hack font does not render in console in-game also

(process:18748): Pango-WARNING **: 19:28:44.748: failed to create cairo scaled font, expect ugly output. the offending font is 'Hack Bold 9'

(process:18748): Pango-WARNING **: 19:28:44.749: font_face status is: out of memory

(process:18748): Pango-WARNING **: 19:28:44.749: scaled_font status is: out of memory

(process:18748): Pango-WARNING **: 19:28:44.752: failed to create cairo scaled font, expect ugly output. the offending font is 'Hack Bold 9'

(process:18748): Pango-WARNING **: 19:28:44.752: font_face status is: out of memory

(process:18748): Pango-WARNING **: 19:28:44.752: scaled_font status is: out of memory

(process:18748): Pango-WARNING **: 19:28:44.753: failed to create cairo scaled font, expect ugly output. the offending font is 'Hack Bold 9'

(process:18748): Pango-WARNING **: 19:28:44.753: font_face status is: out of memory

(process:18748): Pango-WARNING **: 19:28:44.753: scaled_font status is: out of memory

To Reproduce Steps to reproduce the behavior: Run Engine on Windows

Expected behavior All text to render correctly

Screenshots

Hello, libavresample is completely removed from ffmpeg 5.0

• https://github.com/FFmpeg/FFmpeg/commit/f3f9041302203e458d720028e6f82fb4fb51d8e1
• https://ffmpeg.org/
• https://archlinux.org/todo/libavresample-drop/

It is not checked by AbyssEngine at build time while libswresample is

-- /usr/lib64/cmake/sol2
-- Checking for module 'libavcodec'
--   Found libavcodec, version 59.18.100
-- Checking for module 'libavformat'
--   Found libavformat, version 59.16.100
-- Checking for module 'libavdevice'
--   Found libavdevice, version 59.4.100
-- Checking for module 'libavutil'
--   Found libavutil, version 57.17.100
-- Checking for module 'libavfilter'
--   Found libavfilter, version 8.24.100
-- Checking for module 'libswscale'
--   Found libswscale, version 6.4.100
-- Checking for module 'libswresample'
--   Found libswresample, version 4.3.100
-- Found FFMPEG: /usr/lib/libavcodec.so;/usr/lib/libavformat.so;/usr/lib/libavutil.so;/usr/lib/libswscale.so;/usr/lib/libswresample.so  
-- Found Lua: /usr/lib/liblua5.4.so;/usr/lib/libm.so (found suitable exact version "5.4.4")

Seems safe to be removed

Normally, the frame data starts at the beginning, and has Length bytes, and after that there are several (3 or 4) bytes filled with Termination e.g. EE EE EE, before the next frame header starts

But in the new buysellbtn.dc6 there's no termination, and instead the data is aligned to the right: first several bytes are 00 00 00, which are noop, but if to just take the first Length bytes, the last bytes (which were previously just before the termination area) are now outside the Length range

Old file (from MPQ):

1F 04 00 00 20 AE AD AD AD AE AE AE ... 16 C6 C7 19 BC 80 80 EE EE EE 00 00 00 00
~~~~~       ~~~~                                        ~~~~ ~~~~~~~~ ~~~~
length      data starts                          data ends    termi-    next frame header
                                   (according to "length")    nation

New file:

1F 04 00 00 00 00 00 20 AE AD AD AD AE AE AE ... 16 C6 C7 19 BC 80 80 00 00 00 00
                     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~
                                      same data as before              next frame header

Describe the bug shows warning

To Reproduce Steps to reproduce the behavior:

1. use osx
2. run the engine
3. See error

Expected behavior no errors

Screenshots If applicable, add screenshots to help explain your problem.

Desktop (please complete the following information):

• OS: [e.g. iOS] osx

Additional context Add any other context about the problem here.

please assign this bug to me

Describe the bug

AbyssEngine.exe - System Error

The code execution cannot proceed because MSVCP140D.dll was not found. Reinstalling the program may fix this problem.

OK

Other similar errors about missing dlls follow this one.

To Reproduce Steps to reproduce the behavior: Run AbyssEngine.exe

Expected behavior The game starts

Desktop (please complete the following information):

• OS: Windows 10
• Browser Chrome
• Version 98.0.4758.102

Additional context

• I couldn't find the runtime for the D variant
• I also couldn't install VS2015 with the web installer, at which point I gave up.

• [ ] Animation speed: e.g. the two balls (health/mana) have 24 and 20 frames per second, to avoid being in sync.
• [ ] Partially render the sprite: the ball's image is just a ball, but when the health or mana is not full, only part of it should be drawn
• [ ] Change color: mana is blue, poison is green, but the image is the same, and it's red. HD images do color changes in HSV space.

• [ ] Scaling of map render node
• [x] label:setVerticalSpacing(pixels)
• [x] button.disabled
• [x] label:textAlign()
• [ ] add color mod to image type, sec
• [x] node:getType() so you can also do node:getType().toString()

If I didn't start this now even with one thing I never would.