How to remove starting newlines or the starting new from a binary file?

Question

I see there are discussions about removing trailing newlines.

How can I delete a newline if it is the last character in a file?

But I don't find a discusion about removing starting newlines. Could anybody let me know what is the best way to delete starting newlines (one liner preferred)? Thanks.

Answer 1

The equivalent-opposite Perl code to chomp is s/^\n//. Instead of doing it on the last line (eof), do it on the first line. Even though it will only be an empty line, removing the newline will mean that line will print nothing in the output.

perl -pe 's/^\n// if $. == 1' filename >filename2

or in place:

perl -pi -e 's/^\n// if $. == 1' filename

Since starting newlines are by definition empty lines, you can also just skip printing them by using -n instead of -p (same behavior but without printing, so you can determine which lines to print).

perl -ni -e 'print unless $. == 1 and m/^\n/' filename

If you want to remove potentially multiple starting newlines, you could take another approach; advance the handle yourself in the beginning until you receive a non-empty line.

perl -pi -e 'if ($. == 1) { $_ = <> while m/^\n/ }' filename

It's all much easier if you don't mind reading the entire file into memory at once rather than line by line:

perl -0777 -pi -e 's/^\n+//' filename

To avoid doing any excess work editing the file unless it starts with newline characters, you could condition the edit by prefixing it with another command (reads first line of the file and causes a non-zero exit status if it doesn't start with a newline):

perl -e 'exit 1 unless <> =~ m/^\n/' filename && perl ...

Answer 2

In Python, start reading the file without writing in a loop until you get a non-empty line.

outdata = ""
with open(filename) as infile:
    while True:
        line = infile.readline()
        if line != "\n":
            break
    if line:
        outdata = line # save first non-empty line
    outdata += infile.read() # save the rest of the file
with open(filename, "w") as outfile:
    outfile.write(outdata)

Answer 3

Simple filter to skip leading empty lines

#!/usr/bin/perl

use strict;
use warnings;
use feature 'say';

my $begin = 1;

while( <> ) {
        next if /^$/ and $begin;
        $begin = 0;
        say;
}

One liner version perl -0777 -pe 's/^\n+//' filename

Answer 4

This is what I came up with, I'm sure it can still be improved a bit.

with open('../resources/temp_in.txt', 'r+') as file:
    overwrite = False
    for line in file:
        if line:
            overwrite = True
            first_line = line
            break
    if overwrite:
        contents = first_line + file.read()
        file.seek(0)
        file.write(contents)
        file.truncate()

---

Here is an alternative solution, which opens the file twice.

with open('../resources/temp_in.txt') as file:
    for line in file:
        if line.strip():
            contents = line + file.read()
            break
    else:
        contents = ''

with open('../resources/temp_in.txt', 'w') as file:
    file.write(contents)

Answer 5

Set a flag when you find a line that's not just a newline and print when that flag is set:

awk '/./{f=1}f' file

e.g.:

$ printf '\n\n\nfoo\nbar\n'



foo
bar

$ printf '\n\n\nfoo\nbar\n' | awk '/./{f=1}f'
foo
bar