On Mon, 22 Aug 2011, TJ Luoma wrote:

> Can anyone explain why 'wc' adds leading spaces to its output? ^1
>
> for example:
>
> $ ls | wc -l | sed 's# #~#g'
> ~~~~1299
>
> or in another dir:
>
> $ ls | wc -l | sed 's# #~#g'
> ~~~~~~47
>
> I don't understand why:
>
> a) anyone would want leading spaces
>
> b) why they add enough spaces so that the numbers are "right" 
> justified (that might not be the proper term, but you get the idea)

I don't get leading spaces² when any of '-c/--bytes', '-m/--chars', or 
'-l/--lines' is added.  Without one of those options, the output of `wc` 
isn't useful for machine consumption without processing, so they made it 
"pretty" for human consumption.


> Zsh question: Is there a way to get rid of the spaces without using 
> either "| awk '{print $1}'" or "| sed 's#^ *##g'"?

var=${=$(ls | wc -l)}

A possibly better way to count files: (pretty sure that was just an 
example... but either way...):

set -- *(N)
print $#

A possibly better way to count lines (keeping with the ls example):

${#${(f):-"$(ls)"}}


> ^1 — well, GNU's 'wc' does not seem to add leading spaces, but my 
> standard 'wc' in Mac OS X does…

²: with GNU `wc`... --version
wc (GNU coreutils) 8.12

-- 
Best,
Ben