seq | head: embedded vs desktop

[walter harms]

hello
i could not find any official documentation but based on this sentence:
"`seq' prints the numbers from FIRST to LAST by INCREMENT."
i would say that is what to espect. Since bb tries to emulate the behavior
of the original and seq (coreutils) does the same it looks like a bug closed.

re,
 wh


busybox at eehouse.org schrieb:
> I work for a company that uses busybox and am upgrading us to 1.17.3.
> 'seq' has been fixed since our old version so that 'seq <n> 0 <n+1>'
> now returns an infinite sequence of <n> rather than exiting
> immediately.  As a result, the following testcase in seq.tests now
> fails:
> 
> testing "seq count by zero" "seq 4 0 8 | head -n 10" "4\n4\n4\n4\n4\n4\n4\n4\n4\n4\n" "" ""
> 
> The device is running an OMAP 3430 and linux 2.6.24.  When I build
> from the same source tree and .config file for my Ubuntu
> 2.6.28-19-generic x86 laptop, the test case *passes*.  Running strace
> against busybox's seq on the two systems doesn't show much difference;
> certainly it doesn't tell me why seq keeps running on the device once
> write(stdout,...) starts failing.
> 
> (Yes, I'm certain I'm not picking up coreutils' seq by mistake! :-)
> 
> Should I file a bug about this?
> 
> Thanks,
> 
> --Eric