CODEWITHSUNDEEP
awksedgrep
Caterina
Caterina

Reputation: 987

Print name of file if second line is empty

the file looks like this:

>5BW0 B
CHHHHHHHHHHHHHHHHHHHHHHHCCCCCCEEEEEEEEEECCEEEEEEEEEEECCCCCCEEEEEEEEECCCCCCCHHHHCCEEEEEEEC

However the second line could be empty.

I'm trying to find the files with empty second lines.

So far I came up with this in a for loop:

cat file1 | sed -n '2p' | grep '^$'

the problem is that I'm not saving the filename so that I can print it afterwards. How can I fix this?

Upvotes: 3

Views: 229

Answers (3)

potong
potong

Reputation: 58381

This might work for you (GNU sed):

sed -n '2{/\S/!F}' file

If the second line of a file does not contain a non-whitespace character, output the name of the file.

Or you may prefer:

sed '2!d;/^$/F;d' file

Upvotes: 0

RavinderSingh13
RavinderSingh13

Reputation: 133458

With single GNU awk you could try following awk program. Written and tested in GNU awk. I have passed 2 Input_files named file1 and file2 in this program you can pass number of files as per your requirement also.

awk 'FNR==3{nextfile} FNR==2 && !NF{print FILENAME;nextfile}' file1 file2

OR as per Ed sir's recommendation above could be shorten to:

awk 'FNR==2{ if (!NF) print FILENAME; nextfile }' file1 file2

Explanation: Simply checking if its 3rd line of a file then move to next file anyways(which means 2nd line is NOT empty). Then checking 2 more conditions if line is 2nd line and is empty then print file name and move to nextfile.

NOTE: In case you need to find files which are having 2nd line as an empty and print their names then make this above code as a script.awk code and pass it to run in find command too.

Upvotes: 3

Jens
Jens

Reputation: 72639

What about iterating over file names and looking if the second line is empty as follows. The ;2q part quits sed after line 2 to avoid reading the rest of possibly many gigalines of data for nothing.

for file in file1 file2; do
  if test -z "$(sed -n "2p;2q" "$file")"; then
     echo "2nd line of $file is empty"
  fi
done

You could also pipe a list of file names into a loop:

find . -name '*.txt' |
while read -r file; do
  if test -z "$(sed -n "2p;2q" "$file")"; then
    echo "2nd line of $file is empty"
  fi
done

Upvotes: 1

Related Questions