I am trying to identify active accounts over dead accounts and was wondering if diff along with grep or sed or regex can be used, instead of writing a long program.

File1 (usernames)                  File2 (emails)
janedoe                            johndoe@email.com
johndoe                            janedoe@email.com

Each file contain about 1000 times and I need to do this frequently, once a week or so.

-Check if the usernames from the File1 exist in the File2. In the above sample data, they exist.
-If they do exist, comment them out in File1.

In the past, I have used diff to compare files and have used regex to ignore lines. But unfortunately, I cannot wrap my brains around to ignore or a consider only a part of the string from the emails. (Since @ and anything after that is not comparable)

Any assistance would be appreciated. :)


You can achieve that using any script language that knows hashes/dictionaries/associative arrays/whatever it calls the feature.

A very very very simple approach would be like this:

$> cat File1
$> cat File2
$> awk -F'@' 'FILENAME=="File2" { emails[$1]=$0; next}; { print ($1 in emails) ? $1 : "# "$1}' File2 File1
# nosuchkid

Probably, you can see that this one does not modify anything in the input files just writes stdout.

EDIT: Redirecting the output into a file and renaming that would appear as changes in the original file called File1 here (making a backup of the original file is always a good idea):

$> awk -F'@' 'FILENAME=="File2" { emails[$1]=$0; next}; { print ($1 in emails) ? $1 : "# "$1}' File2 File1 > File1.tmp ; cp File1 File1.old ; mv File1.tmp File1
$> cat File1
# nosuchkid

EDIT2: Let's be a little less literal:

$> export PERSONFILE=File1 EMAILFILE=File2; awk -F'@' 'FILENAME==ENVIRON["EMAILFILE"] { emails[$1]=$0; next}; { print ($1 in emails) ? $1 : "# "$1}' "$EMAILFILE" "$PERSONFILE" > "$PERSONFILE.tmp" ; cp "$PERSONFILE" "$PERSONFILE.old" ; mv "$PERSONFILE.tmp" "$PERSONFILE"
  • But awk cannot write the output to the input file. Which is a necessity. – Boring Loop Mar 3 '16 at 16:20
  • Nothings stops you from redirecting the output to a new file and renaming the new file to File1. Actually, that is what any program would do inside when working on text files where there is not a dedicated byte in the file for flag purposes. – Gombai Sándor Mar 3 '16 at 16:23
  • Can you provide a solution using your answer. As I am not familiar with awk and its properties. – Boring Loop Mar 3 '16 at 16:24
  • Okay, did so. Actually, for redirection and renaming, there is no other feature required than what the basic OS provides. – Gombai Sándor Mar 3 '16 at 16:31
  • Can I assign the variable File1 & File2, to actual filenames or I have to type the filenames in the command. – Boring Loop Mar 3 '16 at 16:38

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Not the answer you're looking for? Browse other questions tagged or ask your own question.