Get the return code of a C program in my shell program

You can use "set -o pipefail" option.

[root@myserver Test]# set -o pipefail[root@myserver Test]# ./a.out | tail -l[root@myserver Test]# echo $?100

Here my program a.out returns 100.

Or another options is to use pipestatus environment variable. You can read about it here.http://www.linuxnix.com/2011/03/pipestatus-internal-variable.html

If you are using bash shell, you can use PIPESTATUS array variable to get the status of the pipe process.

$ tail sat | wc -ltail: cannot open ‘sat’ for reading: No such file or directory0$ echo "${PIPESTATUS[0]} ${PIPESTATUS[1]}"1 0$

From man bash:

PIPESTATUS

An array variable containing a list of exit status values from the processes in the most-recently-executed foreground pipeline (which may contain only a single command).

This assigns the last line of the output of Foo to foobar and Foo's exit code is assigned to code:

{ read -r foobar; read code; } < <( (Foo; echo $? ) | tail -2)

The <(...) construct is called process substitution. In the code above, the read commands receive their stdin from the process substitution. Because of the tail -2, the process substitution produces a total of two lines. The first line is the last line produced by Foo and it is assigned to foobar. The second is assigned to code.

The space between the first and second < is essential.

Example

After creating a function Foo, the above can be tested:

$ Foo() { echo "Testing"; false; }$ { read -r foobar; read code; } < <( (echo "Testing"; false; echo $? ) | tail -2)$ echo "foobar=$foobar code=$code"foobar=Testing code=1

And:

$ Foo() { echo "2nd Test"; true; }$ { read -r foobar; read code; } < <( (Foo; echo $? ) | tail -2)$ echo "foobar=$foobar code=$code"foobar=2nd Test code=0