On Sat, Jan 21, 2012 at 10:43:20AM -0700, Bob Proulx wrote: > Tony Baldwin wrote: > > John Hasler wrote: > > > Bob writes: > > Sort of OT, but mentioned in this thread, > > Why use $(command) instead of `command`? > > Generally, I've always used $(command), but see scripts all over using > > `command`, and wondered if there were advantages of one over the other. > > It seems they give the same result, no? > > The `...` is the original Bourne shell syntax for executing a command > and replacing the output from it onto the command line. It is the > oldest method and in the simple case works everywhere. However the > quoting rules are difficult. Quoting includes both double and single > quotes and also includes nested `...` statements within `...` > statements. There is always a quoting solution but it can be tedious > to create correctly. > > The $(...) is newer syntax from ksh (newer from 1986 or so) which was > standardized into POSIX and therefore all POSIX shells will support > it. The use of matching parenthesis means that nesting of commands is > simpler. Most importantly the quoting rules are regular throughout. > You can even nest $(...) inside of $(...). For example: $(... $(...) ) > > The simpler and regular quoting rules wins the choice for me. > > Bob Thanks for this thorough and enlightening explanation.
./tony -- http://www.tonybaldwin.me all tony, all the time! -- To UNSUBSCRIBE, email to [email protected] with a subject of "unsubscribe". Trouble? Contact [email protected] Archive: http://lists.debian.org/[email protected]
