Want to search for two consecutive lines with at least n characters

Question

I would like to use sed to search for the first part of a text file with consecutive lines with at least n non-space characters. I would like to print from the first of those lines to the end of the file.

What's the best way to formulate this?

Answer 1

With sed, this should work:

n=5
sed -ne "/\([^[:blank:]].*\)\{$n\}/!d;h;n;//!d;x;p;x;:1" -e 'p;n;b1'

Answer 2

awk -v n=$n ' 
    !p {line = $0; gsub(/[[:space:]]/, "")}  
    !p && length($0) >= n && prev_is_long {p = 1; print prev}  
    !p {prev = line; prev_is_long = (length($0) >= n)}  
    p {print} 
' file1 

Answer 3

I know you said sed (no pun intended) but if you can live with PERL, the following should do what you want (n=20):

#!/usr/bin/perl -w
my $n=20;    ## The minimum length of the line
my $prev=""; ## This holds the number of chars in the previous line
my $pline;   ## This holds the previous line
my $pp=0;    ## Counter, lines will be printed if set to 1
while(<>){
    ## Skip line processing if we have already 
    ## found our lines of interest
    $pp==1 && do {print; next};
    ## Get non-space chars
    my $a=join("",/[^\s]+/g);
    ## Print if requirements are met.
    if (length($prev)> $n && length($a)> $n){
    print $pline,$_;
    $pp=1;
    }
    $prev=$a;
    $pline=$_;
}

Save it as foo.pl and run like so:

$ perl foo.pl infile.txt