Bash - compare two files and extract the line numbers where lines are identical

Question

I have two files:

File1:

water
doggy
moors

File2:

water1234
forge4432
moors5432

I need to compare the first 5 characters in each line in File2 with the characters in each line in File1 , in order to find out which lines in File1 are identical to the same lines in File2

So, to illustrate, water and moors are shared by both File1 and File2 , so my expected output should be:

1
3

Meaning line 1 and line 3 are shared by both files.

This is my attempt using awk , but it does not work:

awk '/substr($1,1,5)/{ print NR; exit }' File2 File1

My logic was to extract the first 5 characters of ever line in File2 and then print the line where it exists in File1 .

For this question, it is assumed that every line in File1 is 5 characters long.

Answer 1

Your approachi of using substr in awk seems the way for this. Note, though, thath you have to "play" with FNR and NR and also store the values for a further comparison:

$ awk 'FNR==NR{a[NR]=substr($0,0,5); next} a[FNR]==$1 {print FNR}' f2 f1
1
3

Explanation

This reads the file2 and then file1. When reading the first, it stores the 5 first characters into the array a[] using the line number as index. Then, it keeps comparing these values with the second file and printing the line when it matches.

• FNR==NR {} when reading the first file, do {} .
• In this case, {a[NR]=substr($0,0,5); next} {a[NR]=substr($0,0,5); next} : get the 5 first characters and store in the a[] array. Then, move to the next line.
• a[FNR]==$1 {print FNR} when reading the second file, compare the value of the line with what was stored in the array a[] for this line number. If it matches, print the line number.

Answer 2

sort <(cat -n <(cut -b 1-5 file1)) <(cat -n <(cut -b 1-5 file2)) | uniq -d | cut -b 1-6

Output:

1
     3