Bash arithmetic expression with variable

Question

I'm writing a small bash script to treat my data.

In part of the script, I need to use the data, which exists in my raw data, to generate a new value.

However, the script I wrote does not work.

The terminal gives me the following error

")0.192500rror: invalid arithmetic operator (error token is "

the value of each variable is: inc= 0.0200000, str= -0.192500, pts= 1024,

My script is as follows:

for i in *.TXT; do

    inc=$(sed -n '/^#XPERCHAN/ p' $i | sed 's/[A-Z:# ]//g')
    str=$(sed -n '/^#OFFSET/ p' $i | sed 's/[A-Z:# ]//g')
    pts=$(sed -n '/^#NPOINTS/ p' $i | sed 's/[A-Z:# .]//g')
    lst=$(( ($pts-1)*$inc+$str))
    echo $lst

done

Please help me out here. Thanks!

Amadan · Accepted Answer · 2015-08-05 00:35:55Z

2

bash only knows integers. Use bc or a higher programming language for floats.

lst=$(echo "($pts - 1) * $inc + $str" | bc)

EDIT: If your file is formatted like I think it is (#XPERCHAN: 0.0200000), you can do all of that in one awk:

lst=$(awk '/^#XPERCHAN/ { inc=$2 } /^#OFFSET/ { str=$2 } /^#NPOINTS/ { pts=$2 } END { print ((pts + 1) * inc + str) }' $i)

edited Aug 5, 2015 at 0:35

answered Aug 5, 2015 at 0:29

Amadan

200k23 gold badges252 silver badges321 bronze badges

Sign up to request clarification or add additional context in comments.

5 Comments

user4336241 Over a year ago

I used your code, it returns me the following: (standard_in) 1: illegal character: ^M (standard_in) 1: illegal character: ^M (standard_in) 1: illegal character: ^M

Amadan Over a year ago

That means your file is made on Windows, and your sed didn't clean it up well enough. Either add \r to the list of the characters to filter out, or use the awk method (which will not pick it up in the first place).

user4336241 Over a year ago

I used your EDIT, it works perfectly! Thanks! I still would like to know why the 1st way gave me that error.

user4336241 Over a year ago

Thanks again! I got it!

Amadan Over a year ago

As I said, Windows. In Windows (and MSDOS), line ends with \r\n (CR LF, ^M^J, lots of ways to write it.) In Linux, a line ends with \n (LF, ^J), and \r (^M) is just another character. So when you delete all A-Z:# and spaces from it, ^M is still there, at the end; and bc doesn't like it.

Collectives™ on Stack Overflow

Bash arithmetic expression with variable

1 Answer 1

5 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

5 Comments

Your Answer

Sign up or log in

Post as a guest

Related