Re: strange pattern replacing in parameter expansion

[Chet Ramey]

Bernard VAUQUELIN wrote:
> Configuration Information [Automatically generated, do not change]:
> Machine: i586
> OS: linux-gnu
> Compiler: i586-mandrake-linux-gnu-gcc
> Compilation CFLAGS:  -DPROGRAM='bash' -DCONF_HOSTTYPE='i586' 
> -DCONF_OSTYPE='linux-gnu' -DCONF_MACHTYPE='i586-mandrake-linux-gnu' 
> -DCONF_VENDOR='mandrake' -DSHELL -DHAVE_CONFIG_H  -I.  -I. -I./include 
> -I./lib -D_GNU_SOURCE  -O2 -fomit-frame-pointer -pipe -march=i586 
> -mcpu=pentiumpro
> uname output: Linux philtre.labri.fr 2.6.3-4mdk #1 Tue Mar 2 07:26:13 
> CET 2004 i686 unknown unknown GNU/Linux
> Machine Type: i586-mandrake-linux-gnu
>
> Bash Version: 2.05b
> Patch Level: 0
> Release Status: release
>
> Description:
>        strange pattern replacing in parameter expansion
>
>                x="*.c" ; echo $x ; echo ${x//.c/.o}
>        gives
>                toto.c titi.c
>                *.o

I'll bet there aren't actually any .o files in that directory.  In
this example, you're relying on filename generation to produce the
corresponding .o names.  If no files matching "*.o" exist, the
pattern is left unchanged.

>        which seems strange to me,
>        but
>                x=`echo *.c` ; echo $x ; echo ${x//.c/.o}
>        gives
>                toto.c titi.c
>                toto.o titi.o
>        as expected

In this example, you're not relying on filename generation to produce
the .o names, since $x is `toto.c titi.c' instead of `*.c'.

Chet
--
``The lyf so short, the craft so long to lerne.'' - Chaucer
( ``Discere est Dolere'' -- chet )
                                                Live...Laugh...Love
Chet Ramey, ITS, CWRU    chet@case.edu    http://tiswww.tis.cwru.edu/~chet/