Shell Programming and Scripting

doesn't my awk script produces the correct output based on your sample above?
Is there still an issue?

another awk suggestion:

Code:

awk -v i=200 '
/>/ {if (length(s)>i)printf h s; c=0; s=""; h=$0"\n" ; next;}
{ c+=length ; s=s $0 "\n";}
END {if (length(s)>i)printf h s}
' infile.fasta

Thanks! All worked like charm!
Can I ask what was wrong with my own script? I thought it should work. Actually it did sometime, yet sometime did not!
By the way, I believe the scripts will be used by a lot of "bioinformaticians" for this routine job with DNA sequences!
Thank you all again!

I couldn't tell what your original script was even trying to do.

getline profoundly doesn't do what what you thought. It reads lines, but doesn't squeeze them together.

Your control logic was also kind of odd. That 'next' guarantees no lines containing > would ever be printed unless you read them manually with getline. And when you didn't see a >, 'getline' ran every single line -- so you were reading double-speed, then throwing half your lines away. Doing a manual getline bypasses your /^>S/ completely, which is the only way any of those lines could have been printed at all.

I can only guess that some of your input had very long lines, and that work only meant "work". Or perhaps you weren't reading the output you thought you were.

Thanks Corona!
That's the part I was afraid of, which can lead to a totally wrong result! I am trying to catch more details about awk and hope use it correctly.
Thank you again!

Understand that awk has a built-in while loop. This code:

Code:

awk '/asdf/'

amounts to:

Code:

while( (getline) > 0)
{
        if($0 ~ /asdf/) print
}

Putting a getline after that 'if' would mean it wouldn't get checked by the if-statement.