#!/bin/sh -u
# Answer file: Exercise #6 - exercise06script1.sh
# -IAN! idallen@idallen.ca

# Use the standard PATH, umask, character sorting sequence.
# Set the language and character set to be ASCII/C standard.
#
PATH=/bin:/usr/bin ; export PATH
LC_COLLATE=C ; export LC_COLLATE
LANG=C ; export LANG
umask 022

rm -rf jobs6
mkdir jobs6
cd jobs6
pwd

cp ~alleni/cst8129/jobs.txt.bz2 .
sum jobs.txt.bz2

bunzip2 jobs.txt.bz2
sum jobs.txt

cut -f 2 jobs.txt > subjects.txt
sum subjects.txt

tr A-Z a-z < subjects.txt > tmp
mv tmp subjects.txt
sum subjects.txt

sort subjects.txt | uniq > uniquesubjects.txt
#  OR: sort -u subjects.txt > uniquesubjects.txt
wc uniquesubjects.txt

# file is all lower-case; no need to specify A-Z here
#
tr -c 'a-z0-9' '\n' <uniquesubjects.txt \
    | sort \
    | uniq -c \
    | sort -nr >keywords.txt
wc keywords.txt
sum keywords.txt

rm jobs.txt subjects.txt uniquesubjects.txt

head -15 keywords.txt