CSV Column Counts via Command Line Alias - bash

Question

I have a Bash scripting question that's no longer important to me but has still been driving me crazy because I can't figure it out.
I was trying to compare column counts of CSV files via the command line, and found some instructions here that looked promising. The original command worked perfectly, but every attempt to create an alias failed.
Original command line - WORKS!
head -n 1 FILENAME.CSV | awk '{split($0, pieces, ","); for (i=1; i<=length(pieces); i++) print pieces[i];}' | wc -l
Ex. molsen#molsen-M14xR2:~/projects/ee-export$ head -n 1 ee-content.csv | awk '{split($0, pieces, ","); for (i=1; i<=length(pieces); i++) print pieces[i];}' | wc -l
7
Aliased command line - FAILS!
alias csvcc="head -n 1 $1 | awk '{split($0, pieces, ","); for (i=1; i<=length(pieces); i++) print pieces[i];}' | wc -l"
Ex. molsen#molsen-M14xR2:~/projects/ee-export$ csvcc ee-content.csv
awk: cmd. line:1: {split(bash, pieces, ,); for (i=1; i<=length(pieces); i++) print pieces[i];}
awk: cmd. line:1: ^ syntax error
awk: cmd. line:1: {split(bash, pieces, ,); for (i=1; i<=length(pieces); i++) print pieces[i];}
awk: cmd. line:1: ^ syntax error
awk: cmd. line:1: {split(bash, pieces, ,); for (i=1; i<=length(pieces); i++) print pieces[i];}
awk: cmd. line:1: ^ 0 is invalid as number of arguments for split
1023 ee-content.csv
I've tried all sorts of variations on this command with different quoting and escaping, but they all produce errors. I even looked into strong vs. weak quoting, but didn't find anything useful. Also, I came up with this simpler command, but it fails in the same way:
COMMAND LINE USING SED - WORKS!
head -n 1 file.csv | sed 's/\,/\n/g' | wc -l
molsen#molsen-M14xR2:~/projects/ee-export$ head -n 1 ee-content.csv | sed 's/\,/\n/g' | wc -l
7
ALIAS USING SED - FAILS (COUNTS LINES, NOT COLUMNS)
alias csvcc="head -n 1 $1 | sed 's/\,/\n/g' | wc -l"
molsen#molsen-M14xR2:~/projects/ee-export$ csvcc ee-content.csv
1023 ee-content.csv
Can anyone can tell me what I'm missing here? Thanks for your help!

Answer 1

rzymek's Solution 1 doesn't work because csvcc <ee-content.csv expands to
head -n 1 | sed 's/\,/\n/g' | wc -l <ee-content.csv
i. e., the input file is not fed to head, but to wc -l; this explains why it gives the line count.
Another solution with alias and awk:
alias csvcc="awk -F, 'NR==1 { print NF }'"

Answer 2

Aliases don't use $1, scripts do.
Solution 1.
alias csvcc="head -n 1 | sed 's/\,/\n/g' | wc -l"
^--- no $1, input will be read from stdin
Usage:
csvcs < ee-content.csv
Solution 2.
Put the line in a bash script file in $HOME/bin/csvcc:
#!/bin/bash
head -n 1 | sed 's/\,/\n/g' | wc -l
Then make the script executable:
chmod +x ~/bin/csvcc
On ubuntu $HOME/bin is added to $PATH by default. If it's not on your system, than add this line to your ~/.profile:
export PATH="$HOME/bin:$PATH"
Usage:
csvcs ee-content.csv