Create a bash array from a flat file of whitespaces only.

UNIX macOS
1

Hi guys and gals...
MacBook Pro.
OSX 10.13.2, default bash terminal.
I have a flat file 1920 bytes in size of whitespaces only. I need to put every single whitespace character into a bash array cell.
Below are two methods that work, but both are seriously ugly.
The first one requires that I have newlines after every whitespace, not at all practical.
The second requires hexdump to be called 1920 times.
Speed is not important so hexdump would be a way to go but I am asking you guys if there is a more elegant solution...
(Ignore the 'printf' formatting, this was done for quickness only.)
TIA.

#!/bin/bash

ifsstr="$IFS"

# Method 1.
IFS=$'\n'

printf "%b" " \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n" > /tmp/ARRAY
echo "Method 1..."
INDEX=0
while read -r line
do
	ARRAY[$INDEX]="$line"
	INDEX=$(( $INDEX + 1 ))
done < /tmp/ARRAY

# Test...
for n in {0..15}
do
	printf "${ARRAY[$n]}$n\n"
done

# Method 2.
IFS="$ifsstr"

printf "                " > /tmp/ARRAY
echo "Method 2..."
for INDEX in {0..15}
do
	ARRAY[$INDEX]=$( hexdump -n1 -s$INDEX -v -e '1/1 "%c"' /tmp/ARRAY )
done

# Test...
for n in {0..15}
do
	printf "${ARRAY[$n]}$n\n"
done

# End of methods...

Results...
Note the leading spaces before the numbers...

Last login: Mon Feb  5 14:52:58 on ttys000
AMIGA:barrywalker~> cd Desktop/Code/Shell
AMIGA:barrywalker~/Desktop/Code/Shell> ./Space_Array.sh
Method 1...
 0
 1
 2
 3
 4
 5
 6
 7
 8
 9
 10
 11
 12
 13
 14
 15
Method 2...
 0
 1
 2
 3
 4
 5
 6
 7
 8
 9
 10
 11
 12
 13
 14
 15
AMIGA:barrywalker~/Desktop/Code/Shell> _
2

With "whitespace" you mean just spaces (" ", 0x20), or what is generally in the white space character class? man isspace :

Does your system offer the stat command? Would

for i in $(seq $(stat -c%s /tmp/ARRAY)); do ARRAY=" "; done

come close to what you need?

3

Hi RudiC...
Yes I do mean ASCII character '0x20'...
'stat' does not have the '-c' option and 'man stat' does not show any alternative.
'seq' is forced into error too:-
(/tmp/ARRAY does exist in case you ask.)

Last login: Mon Feb  5 17:34:32 on ttys000
AMIGA:barrywalker~> for i in $(seq $(stat -c%s /tmp/ARRAY)); do ARRAY=" "; done
stat: illegal option -- c
usage: stat [-FlLnqrsx] [-f format] [-t timefmt] [file ...]
usage: seq [-w] [-f format] [-s string] [-t string] [first [incr]] last
AMIGA:barrywalker~> man stat
AMIGA:barrywalker~> _

Thanks for trying...

4

Try

IFS="" read  TMP < /tmp/ARRAY 
for ((i=0; i<${#TMP}; i++)); do ARRAY[${#ARRAY[@]}]=${TMP:i:1}; done
1 Like
5

Hi RudiC...
You are a star, works with spaces and in this test case too.

#!/bin/bash

# Only 14 elements here.
printf "ABCDEFGHIJKLMN" > /tmp/ARRAY

IFS="" read TMP < /tmp/ARRAY 
for ((i=0; i<${#TMP}; i++))
do
	ARRAY[${#ARRAY[@]}]=${TMP:i:1}
done

# Note that array element 14 and above should be nothing followed by a number.
for n in {0..15}
do
	printf "${ARRAY[$n]}$n\n"
done

Results...

Last login: Mon Feb  5 18:52:10 on ttys000
AMIGA:barrywalker~> cd Desktop/Code/Shell
AMIGA:barrywalker~/Desktop/Code/Shell> ./whitespace2.sh
A0
B1
C2
D3
E4
F5
G6
H7
I8
J9
K10
L11
M12
N13
14
15
AMIGA:barrywalker~/Desktop/Code/Shell> _

Perfect...
This bit is clever, (love it) ARRAY[${#ARRAY[@]}] .
It took me a short while to understand it, but I got there in the end...

Many thanks...
(Thanks button pressed.)

6

I assume that you're also aware that if you want an array containing 16 elements each of which has the value <space> you cant just use:

#!/bin/bash
ARRAY=(" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " ")
echo "Number of elements in ARRAY: ${#ARRAY[@]}"
for ((i=0; i<${#ARRAY[@]}; i++))
do      printf "'%s'%d\n" "${ARRAY[$i]}" "$i"
done

which produces the output:

Number of elements in ARRAY: 16
' '0
' '1
' '2
' '3
' '4
' '5
' '6
' '7
' '8
' '9
' '10
' '11
' '12
' '13
' '14
' '15

and also produces exactly the same results if run by a 1993 or later version of ksh instead of bash .

7

Yes, that was why I used my original 'hexdump' method as that placed just a single <space> character into each cell.
However RudiC's solution is the one I will use.
The way he creates the array indexing is cool. No amount of Googling gave me that kind of result.
Even my method didn't show up in Google and that is pretty much bullet proof, plus, I knew it would work, but 1920 calls to it would be well OTT.
There is a pointer now though for Google... ;o)
I won't mark as solved yet as others might have bash or maybe even dash solutions too.

8

Try also

IFS=,
ARRAY=( $(sed 's/./&,/g;' /tmp/ARRAY) )

You may want to save the old IFS value beforehand.