The magic behind configure, make, make install - How it works in Unix
The magic behind configure, make, make install - How it works in Unix
thoughtbot.com
The magic behind configure, make, make install - How it works in Unix
12 comments
-
-
Autocrap is a bloated obsolete stuff that is mostly replaced with other tools like cmake or meson. There's no need to learn it today.
-
Hell maybe I do need to learn some shit, because I was under the impression that you cd into the folder after you untar it, then type
./configuremakesudo make install, but the last two packages I attempted to install from source like this just did nothing.-
The first step after you untar is always "open the README and look for build instructions."
-
Maybe. But maybe they did nothing because there was no
./configurescript and you had to use another tool, e. g. one of that I mentioned, so you need to learn another shit.BTW installing anything from source like this is the right way only in (B)LFS.
But you definitely don't need to learn this if you are a developer and starting a new project in 2024. You can use cmake or write plain makefiles, even shell scripts if you want, but as you value life or your reason keep away from the autotools. It is a nightmare to debug thousands lines of scripts they generate and put into your source tree.
-
-
This. If you're starting a new project, and are not GNU, just use Meson. It's the best widely-used build system.
-
Indeed - the general configure, build install steps are fairly universal and the configure script doesn't have to cover from autoconf. We still have that and Makefiles as a wrapper around a meson based setup to keep the process familiar.
-
12 comments
Make is simple, easy, and effective. It's just "configure" that's full of black magic.