multiple variable definition and assignments - linux

Question

I have a piece of script that basically calculates the amount of space the directories in the current directory use but I want help understanding some of the syntax and language etiquette.
Here is the script:
#!/bin/bash
# This script prints a little histogram of how much space
# the directories in the current working directory use
error () {
echo "Error: $1"
exit $2
} >&2
# Create a tempfile (in a BSD- and Linux-friendly way)
my_mktemp () {
mktemp || mktemp -t hist
} 2> /dev/null
# check we are using bash 4
(( BASH_VERSINFO[0] < 4 )) && error "This script can only be run by bash 4 or higher" 1
# An array to keep all the file sizes
declare -A file_sizes
declare -r tempfile=$(my_mktemp) || error "Cannot create tempfile" 2
# How wide is the terminal?
declare -ir term_cols=$(tput cols)
# Longest file name, Largest file, total file size
declare -i max_name_len=0 max_size=0 total_size=0
# A function to draw a line
drawline () {
declare line=""
declare char="-"
for (( i=0; i<$1; ++i )); do
line="${line}${char}"
done
printf "%s" "$line"
}
# This reads the output from du into an array
# And calculates total size and maximum size, max filename length
read_filesizes () {
while read -r size name; do
file_sizes["$name"]="$size"
(( total_size += size ))
(( max_size < size )) && (( max_size=size ))
(( max_file_len < ${#name} )) && (( max_file_len=${#name} ))
done
}
# run du to get filesizes
# Using a temporary file for output from du
{ du -d 0 */ || du --max-depth 0 *; } 2>/dev/null > "$tempfile"
read_filesizes < "$tempfile"
# The length for each line and percentage for each file
declare -i length percentage
# How many columns may the lines take up?
declare -i cols="term_cols - max_file_len - 10"
for k in "${!file_sizes[#]}"; do
(( length=cols * file_sizes[$k] / max_size ))
(( percentage=100 * file_sizes[$k] / total_size ))
printf "%-${max_file_len}s | %3d%% | %s\n" "$k" "$percentage" $(drawline $length)
done
printf "%d Directories\n" "${#file_sizes[#]}"
printf "Total size: %d blocks\n" "$total_size"
# clean up
rm "$tempfile"
exit 0
In the first and second line of the read_filesizes() function that I highlighted in bold, why are two variables (size name) being created if the name is being assigned to size in the array?
In the same function, (( max_size < size )) && (( max_size=size )) this line seems odd to me because how can the two expressions both be true?
Then in the first line of the for loop, (( **length=cols** * file_sizes[$k] / max_size )) I don't understand why the variable length is assigned to cols..why were they defined separately to begin with?

Answer 1

While I'm not 100% sure of the syntax, it seems clear enough to answer your questions :
First Question
why are two variables (size name) being created if the name is being assigned to size in the array?
It looks like name holds the file name and size holds the file size. Then the assignment file_sizes["$name"]="$size" stores the file sizes indexed by the file names.
Second Question
(( max_size < size )) && (( max_size=size ))
I believe this line assigns size to max_size if the previous value of max_size is smaller than size. The goal is that at the end max_size would hold the size of the largest file.
Third Question
(( length=cols * file_sizes[$k] / max_size ))
This calculates the length of the line that would be displayed for each file (whose goal is probably to illustrate the relative size of the file compared to the largest file). The length of the line is relative to the size of the file. cols is the length of the line that would be displayed for the largest file (the one whose size is max_size). cols = the lengh of the terminal - the length of the longest file name - 10.