将字符串拆分为数组（以“\\”分隔）

Question

I'm currently writing a bash script. Let's assume I have the following string:

C:\Program Files\Folder1\ C:\Program Files\Folder2\ C:\Program Files\Folder3\ 

This string can contain an unknown amount of windows locations. Now I need to split the string into an array at the end of each location, so each location has its own key in the array. The main problem are the spaces in every path (Program Files).

What I've tried so far:

Using IFS variable:

oldIFS="$IFS"
IFS="\ "
myArray=( $myString )
IFS="$oldIFS"

The Problem here is, that it cuts the locations on every backslash, not just backslash and whitespace as it was intended. So C: is the first element of my array.

Using read command:

IFS="\ " read -ra myArray "$myString"

Same problem as before.

Any ideas on how to solve this?

Edit: I'm not using real bash but git bash on windows. So not every bash command is available.

Answer 1

You can use this script:

s='C:\Program Files\Folder1\ C:\Program Files\Folder2\ C:\Program Files\Folder3\'
IFS='^K' read -ra arr < <(sed 's/\\ /^K/g' <<< "$s")

OR Non-BASH

echo "$s" | sed 's/\\ /^K/g' | IFS='^K' read -ra arr

Testing:

declare -p arr
declare -a arr='([0]="C:\\Program Files\\Folder1" [1]="C:\\Program Files\\Folder2" [2]="C:\\Program Files\\Folder3\\")'

PS: Here ^K is actually Control V K

---

You can also use awk :

echo "$s" | awk -F '\\\\( |$)' '{for (i=1; i<=NF; i++) if ($i) print $i}'
C:\Program Files\Folder1
C:\Program Files\Folder2
C:\Program Files\Folder3

For array creation:

IFS=';' arr=($(echo "$s" | awk -F '\\\\( |$)' -v ORS=';' '{for (i=1; i<=NF; i++) 
     if ($i) print $i}'))

Answer 2

Probably no filename contains a tab, and the following will work:

IFS=$'\t' read -ra the_files <<<"${the_string//\\ /$'\\\t'}"

In the unlikely case that a filename does contain a tab, you could change the tab character above to something else, like \\1 .