Need to used perl in k shell

[vmallya]

I need to used some perl code in k shell. But it is giving me bad substitution. Same code works fine in bash. Any help would be appreciated.

perl -i -e '$re=q~'${1//~/\\~}'~; print "$re"; while (<>) { /$re/ ? $c++ : print;}exit($c ? 0 : 1)' /home/vmallya/testing || echo " Delete of lines containing."}
ksh: "\$re=q~"${1//~/\\~}"~; print \"\$re\"; while (<>) { /\$re/ ? \$c++ : print;}exit(\$c ? 0 : 1)": bad substitution

[Cabhan]

Dare I ask why you have to use ksh? It's a scary world out there.

In any event, I don't use it myself, but I checked out the ksh manual at:
man/man1/ksh.html man page

In the "Quoting" section, it discusses how even in single-quoted strings, the '$' will still cause variable interpolation. Therefore, I suggest replacing your single-quoted string with a double-quoted one and escaping everything as necessary. This will allow you to backslash-escape your $ and your internal double-quoted string.

[drl]

Hi.

The passage in the ksh online manual is:

All characters enclosed between a pair of single quote marks ( ' ' ) that is not preceded by a $ are quoted. A single quote cannot appear within the single quotes. A single quoted string preceded by an unquoted $ is processed as an ANSI-C string except for the following ...

-- excerpt from man/man1/ksh.html man page

I read that as differentiating between something like:

'this is a line that uses \e as an escape'

and

$'this is a line that uses \e as an escape'

where the second line would cause an insertion of an ASCII ESCAPE for "\e"

I ran this:

#!/usr/bin/env ksh

# @(#) s1	Demonstrate quoting in ksh.

echo
set +o nounset
LC_ALL=C ; LANG=C ; export LC_ALL LANG
echo "Environment: LC_ALL = $LC_ALL, LANG = $LANG"
echo "(Versions displayed with local utility \"version\")"
version >/dev/null 2>&1 && version "=o" $(_eat $0 $1)
set -o nounset
echo

FILE=${1-data1}
cp data-orig $FILE

echo " Data file $FILE:"
cat $FILE

echo
echo " Results:"
# perl -i -e '$re=q~'${1//~/\\~}'~; print "$re"; while (<>) { /$re/ ? $c++ : print;}exit($c ? 0 : 1)' /home/vmallya/testing || echo " Delete of lines containing."}
perl -i -e '$re=q~'${1//~/\\~}'~; print "$re"; while (<>) { /$re/ ? $c++ : print;}exit($c ? 0 : 1)' $FILE || echo " Delete of lines containing."}

exit 0

(substituting my file for your data file) which produced:

% ./s1

Environment: LC_ALL = C, LANG = C
(Versions displayed with local utility "version")
OS, ker|rel, machine: Linux, 2.6.26-2-amd64, x86_64
Distribution        : Debian GNU/Linux 5.0 
ksh 93s+

 Data file data1:
Now is the time
for all good men
to come to the aid
of their country.

 Results:

which produced no printable results, but no error message either. Running with bash shell produced the same (non-)results. The data file was empty after both runs.

The trailing "}" on the echo seems unmatched.

My recommendation would be to place the perl code in a separate file so that the quoting required by perl is really protected from the shell, and requires no mind-bending, alternate, additional quoting.

Posting a very short data file and expected output also would be useful.

Best wishes ... cheers, drl