Question

The following code works, but is quite slow:

#!/bin/bash
for f in *.xml; do
  for r in $(cat ../reports.txt); do
    grep -q "$r" $f

    if [ $? -eq 0 ]; then
      echo "$r,$f"
    fi
  done
done

This could be sped up using:

for f in *.xml; do
  grep -lif ../reports.txt $f
done

However, this does not display the lines in reports.txt that were found to match the file ($f).

How would you display the matching lines from reports.txt that were found in each file along with the name of the file that had 1 or more matches from reports.txt?

For example, if reports.txt contains:

aardvark
aardwolf
anteater

And one of the files (e.g., story.txt) contains:

This is a story about an aardvark and an aardwolf.

Then I’m trying to produce the following output:

story.txt,aardvark
story.txt,aardwolf

If it helps, the lines in reports.txt aren’t regular expressions: they are plain text strings.

Any ideas?

Answer 1

take a look this, does it fit your requirement?

kent$  head report.txt story.txt
==> report.txt <==
aardvark
aardwolf
anteater

==> story.txt <==
is a story about an aardvark and an aardwolf.

kent$  grep -Hof report.txt story.txt
story.txt:aardvark
story.txt:aardwolf

test was done with grep:

kent$  grep --version|head -1
grep (GNU grep) 2.14