I ran across a couple of questions/observations today while importing a
project. The project is enormous and consists of several top-level
directories all within one common directory, and each top-level directoy
has multiple nested dirs. Buried deep inside the nested dirs are all of
the outputs, libraries, supporting files and most importantly, executables
for the program.
This is all on a Linux system.
To build, one simply goes to the top level directory (in an xterm) and
executes "make rc init bin" to do the whole thing. There are layers and
layers of Makefiles buried in here, and they all must resolve to the
various executables.
I noticed that when I created an Eclipse CDT project for this, it not only
successfully read in all the files and folders, but also seemed to be able
to pick up the top-level Makefile just by me telling it what the top-level
directory is and which make command to use (even though I can just invoke
"make", Eclipse CDT wanted /usr/bin/make, which on my system is aliased to
gmake).
I guess my question/observation is, I don't have to tell it exactly which
top-level Makefile to build from, do I?
Also, based on watching the output in the console, it looks like (from the
amount of text it output as well as the time it took), I assume the build
process essentially worked.
My other question, based on a company Wiki page describing that I have to
"Add Make Targets", is, Do I really need to do this? I am not going to
try and find potentially upwards of 100 executables and add them all
individually as make targets. (I am basically trying to get Eclipse CDT
to recognize the huge pile of code and build it exactly the way I would do
it manually.)