I am trying to understand the "cdargs-bash.sh" script with cdargs packages. And I have a question about in the following function:
function _cdargs_get_dir ()
{
local bookmark extrapath
# if there is one exact match (possibly with extra path info after it),
# then just use that match without calling cdargs
if [ -e "$HOME/.cdargs" ]; then
dir=`/bin/grep "^$1 " "$HOME/.cdargs"`
if [ -z "$dir" ]; then
bookmark="${1/\/*/}"
if [ "$bookmark" != "$1" ]; then
dir=`/bin/grep "^$bookmark " "$HOME/.cdargs"`
extrapath=`echo "$1" | /bin/sed 's#^[^/]*/#/#'`
fi
fi
[ -n "$dir" ] && dir=`echo "$dir" | /bin/sed 's/^[^ ]* //'`
fi
if [ -z "$dir" -o "$dir" != "${dir/
/}" ]; then
# okay, we need cdargs to resolve this one.
# note: intentionally retain any extra path to add back to selection.
dir=
if cdargs --noresolve "${1/\/*/}"; then
dir=`cat "$HOME/.cdargsresult"`
/bin/rm -f "$HOME/.cdargsresult";
fi
fi
if [ -z "$dir" ]; then
echo "Aborted: no directory selected" >&2
return 1
fi
[ -n "$extrapath" ] && dir="$dir$extrapath"
if [ ! -d "$dir" ]; then
echo "Failed: no such directory '$dir'" >&2
return 2
fi
}
What's the purpose of the testing:
"$dir" != "${dir/
/}"
Here the testing span over two lines; does it want to remove the newline character in $dir
or maybe for some other reason? I am just starting to learn bash scripting and I have Googled some time but couldn't find any usage like this.
Yes you are right, it removes the newline character. I think the purpose of the test is to make sure $dir
doesn't contain multiple lines.
Alternatively, you can remove \newline
by
${dir/$'\n'/}
This doesn't require two lines so I think it looks better.