Dan McGee wrote:
On Thu, Oct 22, 2009 at 7:41 PM, Allan McRae <[email protected]> wrote:
Cedric Staniewski wrote:
Suggested-by: Dan McGee <[email protected]>
Signed-off-by: Cedric Staniewski <[email protected]>
---
I do not know if this patch is usefull at all, because I do not notice any
change. Just grabbed this from Dan's TODO list and wanted to play a bit
with tput.
It looks fine to me...
I just have no idea on the advantage we are achieving with this change apart
from the apparent terminal safeness. Was the old version not safe?
Dan: comments?
First comment- that TODO list is still huge, wow. :)
Anyway, this seems pretty reasonable to me, but not sure it is worth
it. At the least, we should capture these sequences once on script
startup, and then use the global variable in each function. And does
$(tput offbold) make more sense for the reset?
There does not appear to be such a thing as "tput offbold".
Anyway, I like the idea of setting these all at the start.
if [ $COLORMSG -eq 1 ]; then
BOLD_ON=$(tput bold)
BOLD_OFF=${tput srg0)
GREEN_ON...
...
fi
then the messages can just be
plain() {
local mesg=$1; shift
printf "${BOLD_ON} ${mesg}${BOLD_OFF}\n" "$@" >&2
fi
}
with none of the tests needed.
Allan
