Why does one code give different results?

Question

linux

I'm trying to count the number of lines of code in *.js files:

#!/bin/bash
# файл wc.sh
set -e
find ./src -type f -name "*.js" | xargs cat | sed '/^\s*$/d' | grep -i -v -e '^\s*(\/\/|\/\*|\s*\*)\s*' | wc -l

Result:

$ ./wc.sh
4585

And if you run it directly in the terminal:

$ find ./src -type f -name "*.js" | xargs cat | sed '/^\s*$/d' | grep -i -v -e '^\s*(\/\/|\/\*|\s*\*)\s*' | wc -l
2906

Why is the correct amount in the terminal? That is, the command in the script does not take into account grep processing?

Answer 1

In the end it worked the same way:

$ find ./src -type f -name "*.js" | xargs cat | sed '/^\s*$/d' | grep -i -v -E '^\s*(\/\/|\/\*|\s*\*)\s*' -c
2469

Answer 2

Escaping in scripts works... a bit different than in the command line :) Additional slash escaping is required. Or replace double quotes with single quotes.