[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Optimal shell scripting (vs. archive generated by shar -V...)
From: |
Bruce Korb |
Subject: |
Re: Optimal shell scripting (vs. archive generated by shar -V...) |
Date: |
Sun, 29 Aug 2010 10:31:26 -0700 |
For next release:
On Sat, Aug 28, 2010 at 10:52 PM, Paul Eggert <address@hidden> wrote:
> For this particular example I'd suggest replacing this:
>
> test "X$1" = "X-c" && keep_file=false || keep_file=true
>
> with this:
>
> option=$1
>
> and this:
>
> if test -f %s && ${keep_file}; then
>
> with this:
>
> if test "X$option" = "X-c" && test -f %s; then
>
> This is a tad simpler and will save a file access when -c is absent.
Since we're optimizing, I still think it appropriate to do the
is-the-first-arg-"-c"
test once at the beginning, using Steve Bourne's original if/then/else style:
if test "X$1" = "X-c"
then keep_file=''
else keep_file=true
fi
instead of Steve Bourne's oft-used ``x && y || z'' construct, then:
if test -n "${keep_file}" && test -f %s
then
This way we don't run the "true" executable in order to check the "-c"
condition.
I should probably do a "shift" where the "-c" is recognized, too. Another day.
How many micro (nano?) seconds have we saved, for all those unshars that
get run all the time? :) Cheers - Bruce
- Re: Optimal shell scripting (vs. archive generated by shar -V...),
Bruce Korb <=