How to find the combination of files in unix directory? - shell

Question

I have a set of files at a directory. I need to exit out of my script if i don't find the pairs of files at a given time.
Let's say i have these 3 files at directory $SRC_DIR
file 1: apple_iphone_file.zip
file 2: apple_ipad_file.zip
file 3: apple_mac_file.zip
If these 3 set of files are present i am doing some post processing.
There can be multiple pairs like 2,3, OR N set of these 3 files (file1,file2,file3).
I should exit the script if the same set are not present for all 3 files.
I am planing to count file 1 and if it gives me 2 , i will check if the other two files (file 2 and file 3) also gives me same count , else i will exit.
Do you think , we can do in any other way too?
Any input is highly appreciated.
Code Tried
#!/usr/bin/ksh
file1_count=$(ls ${SRC_DIR}/apple_iphone_file.zip | wc -l)
file2_count=$(ls ${SRC_DIR}/apple_ipad_file.zip | wc -l)
file3_count=$(ls ${SRC_DIR}/apple_mac_file.zip | wc -l)
if [ "$file1_count" == "$file2_count" -a "$file2_count" == "$file3_count" ]; then
echo "Files count match"
else
echo "Files count don't match"
exit 1
fi
This is giving me the results. However, if the files aren't present (none of them) it still shows me "Count Match".

Answer 1

Two scripts, check and pre-process. In check, if the variables corresponding with all three files equal 1, run pre-process. The default action (which runs next if that fails) is to exit.
Pre-process finds all files in the directory, puts their names in an input stack, and then uses each name as input for the main function. The code between the two ed lines is an example; replace it with your own. After that, it exits.
check.sh:-
#!/bin/sh
find apple_iphone_file.zip && iphone=1
find apple_ipad_file.zip && ipad=1
find apple_mac_file.zip && mac=1
[ "${iphone}" -eq 1 ] && [ "${ipad}" -eq 1 ] && \
[ "${mac}" -eq 1 ] && ./pre-process.sh
exit 0
pre-process.sh:-
#!/bin/sh
next() {
[ -s stack ] && main
end
}
main() {
line=$(ed -s stack < edprint+.txt)
echo "${line}" | tr '[a-z]' '[A-Z]'
ed -s stack < edpop+.txt
next
}
end() {
rm -v ./stack
rm -v ./edprint+.txt
rm -v ./edpop+.txt
exit 0
}
find . -type -f > stack
cat >> edprint+.txt << EOF
1
q
EOF
cat >> edpop+.txt << EOF
1d
wq
EOF

Answer 2

There are so many possibilities.
One option:
my_counter=""
[ -f "file_name_1" ] && my_counter="x$my_counter"
[ -f "file_name_2" ] && my_counter="x$my_counter"
[ -f "file_name_3" ] && my_counter="x$my_counter"
if [ "${#my_counter}" -lt 2 ]; then
echo "Error"
else
echo "doing stuffs"
fi
You can easily add check for file, change lower and upper limit, and ev. doing more tests (in similar way) in series. Note: for shell scripts, I tend to copy paste, because the command are easy (low probability of refactoring), and often I find the need to add extra tests on specific cases.