Regular Expression, bash

Question

I need to read an ASCII file coordinates.ascii that have one line only:

-I0.00130000258937/0.000899999864241

inside a bash script, and attribute 0.00130000258937 to the variable x_inc and 0.000899999864241 to y_inc. guess that the regexp for x_inc is:

(\d.\d+)(?=/)

but I do not know the regexp for y_inc and also the sed or grep commands/syntax to implement the regexp inside the bash ..

x_inc=$(sed -n '(\d.\d+)(?=/)' coordinates.ascii )     # does not work!!!

Answer 1

Just with bash:

$ IFS="I/" read _ x_inc y_inc < coordinates.ascii
$ echo $x_inc
0.00130000258937
$ echo $y_inc
0.000899999864241

Answer 2

Using awk

x_inc=$(awk -F"[I/]" '{print $2}' coordinates.ascii)
echo $x_inc
0.00130000258937

y_inc=$(awk -F"[I/]" '{print $3}' coordinates.ascii)

Answer 3

Here is another version using sed and read. It reads both variables at once:

read -r x_inc y_inc <<<$(sed 's@[-I/]@ @g' < coordinates.ascii); 

# $x_inc is now: 0.00130000258937
# $y_inc is now: 0.000899999864241

Edit: or with process substitution as suggested in the comments

read -r x_inc y_inc < <(sed 's@[-I/]@ @g' < coordinates.ascii);