[illumos-Developer] BEADM [WEBREV] 418 replace beadm by tbeadm with gettext support

[Albert Lee]

Is this able to use the existing translations for the Python version,
or does it use the tbeadm messages? The usage message still says
"tbeadm" on the first line.

be_command_tbl probably should use array syntax in run_be_cmd, even.

usage() in main on line 263 is called before setlocale(), etc. Maybe
the argc check should just be moved to run_be_cmd.

The strftime calls on 492, 548 and 622 can exceed the target array bounds.

A literal "6" is used to refer to the col_info array size in many
places, I would suggest a #define or a sizeof(cols)/sizeof(col_info).
cols should be indexed using array syntax instead of "cols + i".

Why does line 868 use a new variable instead of snap_name as  in 852?
snap_name is guaranteed to be NULL at that point.

Is ZFS_MAXNAMELEN+32 the actual maximum length for a  concatenated BE
and snapshot name?

Is there a consistent distinction between printing to stderr and
stdout? I see a lot of printf()s for error messages.

snprintf(), malloc(), and retry on 1408 could be replaced with asprintf().

-Albert