Shell Programming and Scripting

Hello Everyone,

I'm currently have a requirement where I've generated a list of files with specific attributes and I need to know what lines are similar between the two files. For example:

-File 1-
line1
line2
line3

-File 2-
line1
line2
line4
line5

-Desires Output-
line1
line2

Is there a simple command that will let be view the similarities between files? Basically something the opposite of diff.

Below is a copy of the bash script I've been working on so far. The problem with the script so far is that it doesn't work when there "grep -v" excludes the first line from output when scanning the target file the first time, but the same line would be included again when grep does the same for the second line. So in the end, after sort and uniq, the output is basically identical to the original target file.

I'm eager to know what you guys think!

Code:

#Obtain diff between two files
diff $1 $2 | grep ">" | awk '{ print $2 }' >> exclusions;
diff $1 $2 | grep ">" | awk '{ print $2 }' > ./Records/Exclusions_-_"$1"_-_"$2"

#Read exclusions and inverse grep against target file
while IFS= read line; do grep -v "$line" $2; done < exclusions >> out

#clean up output
cat out | sort | uniq > ./Output/Output_-_"$1"_-_"$2"

#clean up files
cp exclusions exclusions.tmp; cat exclusions | sort | uniq > exclusions; rm -f exclusions.tmp out

Code:

awk 'FNR==NR{a[$0]=1;next}$0 in a' file1 file2

If the files are sorted:

Code:

comm -12 file1 file2

Regards,
Alister

Try...

Code:

grep -f file1 file2

If the goal is to match identical lines, I would recommend:

Code:

grep -xFf file1 file2

Using fixed string, whole-line matching will prevent false positives (fixed substrings or regexp wildcard matches), false negatives (some regexps don't match themselves, e.g. [a]), and syntax errors from lines that aren't valid regular expressions.

Regards,
Alister

Awesome! I got it working! Thanks a lot guys.