Shell Programming and Scripting

I'm new at this script stuff... only have minor exposure to java.

My problem is largely syntax and being unable to figure out what the manuals are telling me what each option does. Basically I have a hard time understanding the documentation and need help with what awk is capable of on the shell command line.

I have a file and if the line does not start with a ">" I want to count all the characters but not spaces. Just [A-Z]. In this case I know the characters are going to be A E or Z.

My thinking has got me this far so far..

If myfile is "ATHROUGHZ SPACES"

grep -v ">" myfile.tab | awk if nextChar == "A || E || Z" {++count} END {print count}'


I want it to return 4, because there are two A, one E and one Z.



The grep part was to select for the lines that did not contain ">".



The {++count} END {print count}' part is to count and then print the count


However I am stuck in the middle part which is highlighted in red.



I haven't found (or understood how to use) anything that lets me go through the line character by character to compare it with if it is a desired letter and count it if it is a match.


I found the following threads but don't understand them enough to apply them to my situation.



https://www.unix.com/shell-programming-scripting/9721-counting-characters.html
From this one it appears that -F won't work to seperate the characters because it is used to seperate strings?



https://www.unix.com/unix-dummies-questions-answers/58760-counting-occurence-particular-characters.html
From this one, I don't have seperators between the characters


https://www.unix.com/shell-programming-scripting/39765-counting-number-occurances-all-characters-z-string.html
This thread is in reference to PEARL which I am not familar with, and not sure how is to be applied to just the command line?

If anyone has any suggestions or good directions to point me for what feels like should be incredibly easy to do, it would be awesome!

Thanks a bunch!

-Diana

Don't know about awk, but in Perl (if you only want it only on the command line) you would write it like this

Code:

$ perl -e '$count=0; while(<>){next if /^>/; @line=split //; foreach(@line){$count++ if /[AEZ]/;}} print $count."\n";' myfile.tab

Thank you for the reply.

I don't know anything about PEARL yet, but I am sure your example there will be useful once I can read up on it more.

I'm about 30 hours in to my first Linux style command line experience so am still a n00b.

To count chars in the A-Z range

Code:

$ echo 'ATHROUGHZ SPACES' | nawk '!/^>/ {print gsub(/[A-Z]/, "")}'
15

$ echo '>ATHROUGHZ SPACES' | nawk '!/^>/ {print gsub(/[A-Z]/, "")}'

To count ALL the chars, but spaces ' '

Code:

$ echo 'ATHROUGHZ SPACES' | nawk '!/^>/ {print gsub(/[^ ]/, "")}'
15

$ echo '>ATHROUGHZ SPACES' | nawk '!/^>/ {print gsub(/[^ ]/, "")}'

Pick your poison.

Thank you for the help.

I will play with that for a bit and see what I can do!