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...
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 $(stat -c%s /tmp/ARRAY)); do ARRAY=" "; done
stat: illegal option -- c
usage: stat [-FlLnqrsx] [-f format] [-t timefmt] [file ...]
usage: seq [-w] [-f format] [-s string] [-t string] [first [incr]] last
AMIGA:barrywalker~> man stat
AMIGA:barrywalker~> _
Hi RudiC...
You are a star, works with spaces and in this test case too.
#!/bin/bash
# Only 14 elements here.
printf "ABCDEFGHIJKLMN" > /tmp/ARRAY
IFS="" read TMP < /tmp/ARRAY
for ((i=0; i<${#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...
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.