On Tue, Dec 18, 2012 at 11:47:59AM +0100, Alexander Hall wrote: > > > Andres Perera <[email protected]> wrote: > > >On Mon, Dec 17, 2012 at 7:07 PM, Alexander Hall <[email protected]> > >wrote: > >> > >> For scripting, echo is one of the commands I tend to avoid unless I > >know the > >> data is "safe", because of it's horrific argument parsing. > >> > >> I've yet to find a way to echo a single '-n' using the sh/ksh > >builtin. When > >> printing unknown data, I usually end up using 'print -r -- "$var"' > >(or > >> 'printf "%s" "$var"' if I care about portability). > >> > >> /Alexander > >> > > > >the worrysome part is what happens with make: > > > >andres@pote:~ $ echo '/nonexsistent:;@echo -e hello' | make -f- | vis > >-e hello\$ > >andres@pote:~ $ echo '/nonexsistent:;@echo -e hello;' | make -f- | vis > >hello\$ > > > >this is due to the optimization to fork+exec instead of shell when > >there are no meta characters. the second makefile has `;', so the > >optimization doesn't get triggered > > > >what is the problem? 2 echos that disagree or the optimization itself? > >are the calls coming from *INSIDE* the house? > > I don't think make should optimize this, but recognize echo as a shell > builtin. Is this theoretical, or is there an actual Makefile where this is a problem ?
I'd rather NOT go through extra shell forking just for the sake of it. Also, echo -e is specific to ksh and NOT a posix option, so you're going to end up with something that is deeply non-portable. Either way, this is fucked up.
