I have a text file with lines saying:
File fileA and fileB differ
File fileX and fileY differ
I need a bash script that goes through the whole file, for each line parses the file names and executes the command diff fileA fileB > fileA.diff
-
2Good on you! Why does it have to be a bash script? Let us know which part of the solution you get stuck with and we'll endeavour to help. We won't do your work for you.johnsyweb– johnsyweb2011-09-20 09:49:49 +00:00Commented Sep 20, 2011 at 9:49
Add a comment
|
2 Answers
The following command:
awk '/ differ$/ {print "diff "$2" "$4" >"$2".diff"}{}'
will give you a script you can run to do this.
See the following transcript:
pax$ echo 'File fileA and fileB differ
hello
File fileX and fileY differ' | awk '
/ differ$/ {print "diff "$2" "$4" >"$2".diff"}{}'
diff fileA fileB >fileA.diff
diff fileX fileY >fileX.diff
Capture the output, then run it with bash and you'll have what you need.
Note that this won't work well with filenames that have spaces in them - if you have such heinous beasts, you will need to do a little more intelligent parsing.
1 Comment
Michael
Thanks. I'm now using this: while read line; do awk '/ differ$/ {print "diff "$2" "$4" > "$2".diff"}{}'; done < diffs.txt Problem is awk only prints the command... how do I tell it to execute it?
This command generates a list of diff commands:
[me@home]$ sed -n 's/File \(.*\) and \(.*\) differ/diff "\1" "\2" > "\1.diff"/p' infile
diff "fileA" "fileB" > "fileA.diff"
diff "fileX" "fileY" > "fileX.diff"
You can execute the generated commands by passing them to bash using process substitution.
[me@home]$ bash <(sed -n 's/File \(.*\) and \(.*\) differ/diff "\1" "\2" > "\1.diff"/p' infile )
Comments
