Hello, I have a pre-script that I use to pre-generate some files. I would like to use the C preprocessor to pre-process my input file (json) so that I can benefit from the #include and #ifdef directives before acutally processing the input json file. I checked env, and I could not see anything use…

o_O? [env:esp32dev] platform = espressif32 board = esp32dev framework = arduino extra_scripts = test.py Import("env") def search_compiler(env): from pathlib import Path, PurePath # Use any item in $PATH corresponding to a platformio toolchain bin folder path_separator = ':' gcc_ex…

[image] fariouche: I checked env, and I could not see anything usefull. I just need the cross compiler gcc path. env[“CC”] is equal to “cc” Correct. If you execute a command within the environment (PATH) that PlatformIO has created (env.Execute()) it should be fine. Otherwise, duplicate of C…

thank you for the quick reply. I checked again the link “duplicate” and I still fail to see how I can do it. My point is that I need the xtensa or arm gcc, and env is returning me the host gcc. I’ve searched all the links in the link you provided, and I see this: MarlinFirmware/Marlin/blo…

And after that something like cmd = [ '"%s"' % compiler, "--version" ] env.Execute(" ".join(cmd)) outputs "/home/max/.platformio/packages/toolchain-xtensa-esp32@8.4.0+2021r2-patch5/bin/xtensa-esp32-elf-g++" --version xtensa-esp32-elf-g++ (crosstool-NG esp-2021r2-patch5) 8.4.0 Copyright (C) 20…

And just to prove my point that indeed “CXX” or “CC” does 100% suffice: Import("env") cmd = [ "$CXX", "--version" ] env.Execute(" ".join(cmd)) outputs xtensa-esp32-elf-g++ --version xtensa-esp32-elf-g++ (crosstool-NG esp-2021r2-patch5) 8.4.0 Copyright (C) 2018 Free Software Foundation, Inc. …

Thanks! I think I know what is the problem: in my case $CC is just cc, $CXX is g++ etc… and my script is a pre-script… Seems like it is run before all the variables are set! And I need to run before the compilation is started. I’ve used in the past something like env.AddPreAction("buildprog", be…

[image] maxgerhardt: extra_scripts = test.py If you reference the script like that and directly execute the preprocessing steps you need, not as a a pre or post action to something, it should work fine. Maybe check that the current target is “build” (and not Intellisense discovery) for safety…

ok, got it. My issue was that the AddPreAction(“buildprog”… is only called if there is something to build! This is neat, but since I generate header files, I cannot rely on that mechanism. So I’m facing a dilemma… If I use pre:extra_script.py, I got executed before the LDF and anything in platformi…

Hello, Just for information, it seems like I need to go back using a pre-extra script… When doing a clean build (ie, no generated source files yet), I can see the file being generated after the dependency graph being displayed and before the first compilation starts… however it does not pick up my …

[image] fariouche: however it does not pick up my generated C files for compilation You can put them in a non-default build folder (like <root>/generated) and call env.BuildSources() on them, for example .

PlatformIO Community

Call target C preprocessor from a script

PlatformIO Core

maxgerhardt March 13, 2023, 1:10pm 2

Correct. If you execute a command within the environment (PATH) that PlatformIO has created (env.Execute()) it should be fine.

Otherwise, duplicate of Change firmware name based on main.cpp - #2 by maxgerhardt.