Multiple Commands ( |'ed) and $?

[jda123]

Hello,

I run a executable and then use the stream editor on the output before redirecting it into a file. The $? appears to have the completion/exit code of the stream editor and not the executable. Is there any way to get the exit code of the first command?

Example:

executable this this etc parameters | sed -e 'whatever' > file.out.somefile

$? returns 0 unless sed fails. What to use to check the return code of the executable?

Thanks.

[drl]

Hi.

A few ways:

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

...
       The return status of a pipeline is the exit status of the last command,
       unless the pipefail option is enabled.  If  pipefail  is  enabled,  the
       pipeline's  return  status is the value of the last (rightmost) command
       to exit with a non-zero status, or zero if all commands  exit  success-
       fully.  If the reserved word !  precedes a pipeline, the exit status of
       that pipeline is the logical negation of the exit status  as  described
       above.   The  shell waits for all commands in the pipeline to terminate
       before returning a value.

-- excerpts form man bash

The man pages, Google, and Wikipedia are your friends ... cheers, drl

[alf55]

Don't use this...

SORRY ... I failed here was the "stdout" bypassed the second command.

Hello,

I run a executable and then use the stream editor on the output before redirecting it into a file. The $? appears to have the completion/exit code of the stream editor and not the executable. Is there any way to get the exit code of the first command?

Example:

executable this this etc parameters | sed -e 'whatever' > file.out.somefile

$? returns 0 unless sed fails. What to use to check the return code of the executable?

Thanks.

#!/bin/bash
#

# make file descriptor 3 goto wherever stdout is going...
exec 3>&1

# make file descriptor 4 goto wherever stderr is going...
exec 4>&1

# Run whatever but:
#        reroute stdout through file descriptor 3
#        reroute stderr through file descriptor 4
#   Then echo the status into file descriptor 5 
#   After the second command out the status saved in file descriptor 5 to
#   output.
status=`((first_cmd 2>&4 1>&3 3>&- 4>&- 5>&-; echo $? >&5) | second_cmd) 5>&1`
exit $status

[alf55]

SORRY ... I failed here was the "stdout" bypassed the second command.

[clowenstein]

If the object is to avoid the "sed" command if the first "executable" fails, you really can't get there from here. Failure or success of the first part of the pipeline is not determined until it finishes, and there is an indefinite amount of data in the pipe, some of which will have already been processed by "sed".

[unlimitedscolobb]

executable this this etc parameters | sed -e 'whatever' > file.out.somefile

While $PIPESTATUS seems to be the really general purpose solution, you might also want to consider joining the commands with the ``&&'' operator. I am not sure what you are trying to achieve, but if you are just interested in whether the ``executable'' failed, using ``&&'' would be simpler.