On Wed, Apr 02, 2014 at 09:58:54PM +0100, Peter Stephenson wrote:
>On Wed, 2 Apr 2014 14:06:21 -0500
>Erik Johnson <palehose@gmail.com> wrote:
>> Python os.getlogin() does indeed show "erik" instead of "root". So this
>> may just be a difference between glibc and BSD libc. Either way, it
>> "just works" in bash, sh, csh, tcsh, and ksh, as can be seen below, so
>> I believe there is an argument for making it work properly in FreeBSD as
>> well, or at the very least adding some wording to the zshparam manpage
>> which makes this difference in behavior clear.
>
>Hmm...  to me, having LOGNAME *not* report the same as getlogin() is
>"just not working".  Presumably there's some reason why getlogin() does
>what it does and it doesn't seem to me to be the shell's job to second
>guess system calls.  POSIX seems to agree with me --- to be fair, this
>isn't for the variable, it's for the command "logname", but it would be
>confusing if they were different.
>
>  The logname utility shall write the user's login name to standard
>  output. The login name shall be the string that would be returned by
>  the getlogin() function defined in the System Interfaces volume of
>  POSIX.1-2008. Under the conditions where the getlogin() function would
>  fail, the logname utility shall write a diagnostic message to standard
>  error and exit with a non-zero exit status.
>
>However, there may be some documented prior art for LOGNAME that I'm
>missing --- the history of shell development isn't necessarily particularly
>rational.
>
>It certainly makes sense to document it in any case.
>

Yeah, and for what it's worth it seems that not all of the shells are
exporting LOGNAME. But the ones that do are apparently doing so
differently than zsh does it. So, documentation is probably the best
route to go here. Thanks!


-- 

-Erik

"For me, it is far better to grasp the universe as it really is than to
persist in delusion, however satisfying and reassuring."  --Carl Sagan