Grep all files in a directory and print matches with filename

Question

I have a list of log files that I create as part of a video encoding script that I wrote.

I would like to search all of them and print out certain statistics from the encode - how fast they were encoded, what settings were used, etc.

I can search for the average framerate in one file via this 1 liner:

cat ${filename} | grep average

which outputs:

 work: average encoding speed for job is 23.211176 fps

and search for the ratefactor:

cat ${filename} | grep RF

I would like to search all files in the directory and print off one, or prefereably both pieces of information along with the filename. Is there any way I can use find or grep to get this in a one-liner, or do I need to write a script?

I would like output like this:

/home/javanix/filename.log
    <RF line>
    <average line>

I would like this to either work using FreeBSD 9 or Ubuntu 12.04.

Answer 1

Using awk with the desired output format:

awk '/average|RF/{ if(f!=FILENAME){f=FILENAME;print f} print"\t"$0}' *.log

For recursive search,

find path -name '*.log' -type f -exec \
  awk '/average|RF/{if(f!=FILENAME){f=FILENAME;print f}print"\t"$0}' {} \;

Answer 2

Do this:

egrep 'RF|average' *

The output won't match the output you show above. It will be like this:

filename1: ...
filename1: ...
filename2: ...
filename3: ...

Some file names will appear just once and some more, depending on how many matches there were.

Answer 3

I'd say you can do this with grep only:

grep -srHE 'average|RF' path/

s to suppress errors, r for recursive, H to display the filename and E (and the pipe) to match any of both patterns.

Answer 4

Try this command:

grep -P "average|RF" *

grep will search in all files in the directory (*) and show all matches for the using the Perl Compatible Regular Expression (PCRE) average|RF, which matches both average and RF.

The output format will be the following:

filename.log:<RF line>
filename.log:<average line>

Answer 5

Try this one-liner:

find -exec sh -c 'x=$(grep average "$0";grep RF "$0") && echo "$0\n$x"' {} \;

It will produce:

<file 1>
<RF line>
<average line>
[...]

As you can see there's no indentation to keep the code short, an alternative would be placing a mark on the file line, something like:

[...] echo ">>> $0\n$x"' {} \;