I'm wondering how to declare a 2D array in bash and then initialize to 0.
In C it looks like this:
int a[4][5] = {0};
And how do I assign a value to an element? As in C:
a[2][3] = 3;
You can simulate them for example with hashes, but need care about the leading zeroes and many other things. The next demonstration works, but it is far from optimal solution.
#!/bin/bash
declare -A matrix
num_rows=4
num_columns=5
for ((i=1;i<=num_rows;i++)) do
for ((j=1;j<=num_columns;j++)) do
matrix[$i,$j]=$RANDOM
done
done
f1="%$((${#num_rows}+1))s"
f2=" %9s"
printf "$f1" ''
for ((i=1;i<=num_rows;i++)) do
printf "$f2" $i
done
echo
for ((j=1;j<=num_columns;j++)) do
printf "$f1" $j
for ((i=1;i<=num_rows;i++)) do
printf "$f2" ${matrix[$i,$j]}
done
echo
done
the above example creates a 4x5 matrix with random numbers and print it transposed, with the example result
1 2 3 4
1 18006 31193 16110 23297
2 26229 19869 1140 19837
3 8192 2181 25512 2318
4 3269 25516 18701 7977
5 31775 17358 4468 30345
The principle is: Creating one associative array where the index is an string like 3,4. The benefits:
it's possible to use for any-dimension arrays ;) like: 30,40,2 for 3 dimensional.
the syntax is close to "C" like arrays ${matrix[2,3]}
Bash doesn't have multi-dimensional array. But you can simulate a somewhat similar effect with associative arrays. The following is an example of associative array pretending to be used as multi-dimensional array:
declare -A arr
arr[0,0]=0
arr[0,1]=1
arr[1,0]=2
arr[1,1]=3
echo "${arr[0,0]} ${arr[0,1]}" # will print 0 1
If you don't declare the array as associative (with -A), the above won't work. For example, if you omit the declare -A arr line, the echo will print 2 3 instead of 0 1, because 0,0, 1,0 and such will be taken as arithmetic expression and evaluated to 0 (the value to the right of the comma operator).
Bash does not support multidimensional arrays.
You can simulate it though by using indirect expansion:
#!/bin/bash
declare -a a0=(1 2 3 4)
declare -a a1=(5 6 7 8)
var="a1[1]"
echo ${!var} # outputs 6
Assignments are also possible with this method:
let $var=55
echo ${a1[1]} # outputs 55
Edit 1: To read such an array from a file, with each row on a line, and values delimited by space, use this:
idx=0
while read -a a$idx; do
let idx++;
done </tmp/some_file
Edit 2: To declare and initialize a0..a3[0..4] to 0, you could run:
for i in {0..3}; do
eval "declare -a a$i=( $(for j in {0..4}; do echo 0; done) )"
done
Another approach is you can represent each row as a string, i.e. mapping the 2D array into an 1D array. Then, all you need to do is unpack and repack the row's string whenever you make an edit:
# Init a 4x5 matrix
a=("00 01 02 03 04" "10 11 12 13 14" "20 21 22 23 24" "30 31 32 33 34")
aset() {
row=$1
col=$2
value=$3
IFS=' ' read -r -a rowdata <<< "${a[$row]}"
rowdata[$col]=$value
a[$row]="${rowdata[#]}"
}
aget() {
row=$1
col=$2
IFS=' ' read -r -a rowdata <<< "${a[$row]}"
echo ${rowdata[$col]}
}
aprint() {
for rowdata in "${a[#]}"; do
echo $rowdata
done
}
echo "Matrix before change"
aprint
# Outputs: a[2][3] == 23
echo "a[2][3] == $( aget 2 3 )"
echo "a[2][3] = 9999"
aset 2 3 9999
# Show result
echo "Matrix after change"
aprint
Outputs:
Matrix before change
00 01 02 03 04
10 11 12 13 14
20 21 22 23 24
30 31 32 33 34
a[2][3] == 23
a[2][3] = 9999
Matrix after change
00 01 02 03 04
10 11 12 13 14
20 21 22 9999 24
30 31 32 33 34
You can also approach this in a much less smarter fashion
q=()
q+=( 1-2 )
q+=( a-b )
for set in ${q[#]};
do
echo ${set%%-*}
echo ${set##*-}
done
of course a 22 line solution or indirection is probably the better way to go and why not sprinkle eval every where to .
2D array can be achieved in bash by declaring 1D array and then elements can be accessed using (r * col_size) + c). Below logic delcares 1D array (str_2d_arr) and prints as 2D array.
col_size=3
str_2d_arr=()
str_2d_arr+=('abc' '200' 'xyz')
str_2d_arr+=('def' '300' 'ccc')
str_2d_arr+=('aaa' '400' 'ddd')
echo "Print 2D array"
col_count=0
for elem in ${str_2d_arr[#]}; do
if [ ${col_count} -eq ${col_size} ]; then
echo ""
col_count=0
fi
echo -e "$elem \c"
((col_count++))
done
echo ""
Output is
Print 2D array
abc 200 xyz
def 300 ccc
aaa 400 ddd
Below logic is very useful to get each row from the above declared 1D array str_2d_arr.
# Get nth row and update to 2nd arg
get_row_n()
{
row=$1
local -n a=$2
start_idx=$((row * col_size))
for ((i = 0; i < ${col_size}; i++)); do
idx=$((start_idx + i))
a+=(${str_2d_arr[${idx}]})
done
}
arr=()
get_row_n 0 arr
echo "Row 0"
for e in ${arr[#]}; do
echo -e "$e \c"
done
echo ""
Output is
Row 0
abc 200 xyz
A way to simulate arrays in bash (it can be adapted for any number of dimensions of an array):
#!/bin/bash
## The following functions implement vectors (arrays) operations in bash:
## Definition of a vector <v>:
## v_0 - variable that stores the number of elements of the vector
## v_1..v_n, where n=v_0 - variables that store the values of the vector elements
VectorAddElementNext () {
# Vector Add Element Next
# Adds the string contained in variable $2 in the next element position (vector length + 1) in vector $1
local elem_value
local vector_length
local elem_name
eval elem_value=\"\$$2\"
eval vector_length=\$$1\_0
if [ -z "$vector_length" ]; then
vector_length=$((0))
fi
vector_length=$(( vector_length + 1 ))
elem_name=$1_$vector_length
eval $elem_name=\"\$elem_value\"
eval $1_0=$vector_length
}
VectorAddElementDVNext () {
# Vector Add Element Direct Value Next
# Adds the string $2 in the next element position (vector length + 1) in vector $1
local elem_value
local vector_length
local elem_name
eval elem_value="$2"
eval vector_length=\$$1\_0
if [ -z "$vector_length" ]; then
vector_length=$((0))
fi
vector_length=$(( vector_length + 1 ))
elem_name=$1_$vector_length
eval $elem_name=\"\$elem_value\"
eval $1_0=$vector_length
}
VectorAddElement () {
# Vector Add Element
# Adds the string contained in the variable $3 in the position contained in $2 (variable or direct value) in the vector $1
local elem_value
local elem_position
local vector_length
local elem_name
eval elem_value=\"\$$3\"
elem_position=$(($2))
eval vector_length=\$$1\_0
if [ -z "$vector_length" ]; then
vector_length=$((0))
fi
if [ $elem_position -ge $vector_length ]; then
vector_length=$elem_position
fi
elem_name=$1_$elem_position
eval $elem_name=\"\$elem_value\"
if [ ! $elem_position -eq 0 ]; then
eval $1_0=$vector_length
fi
}
VectorAddElementDV () {
# Vector Add Element
# Adds the string $3 in the position $2 (variable or direct value) in the vector $1
local elem_value
local elem_position
local vector_length
local elem_name
eval elem_value="$3"
elem_position=$(($2))
eval vector_length=\$$1\_0
if [ -z "$vector_length" ]; then
vector_length=$((0))
fi
if [ $elem_position -ge $vector_length ]; then
vector_length=$elem_position
fi
elem_name=$1_$elem_position
eval $elem_name=\"\$elem_value\"
if [ ! $elem_position -eq 0 ]; then
eval $1_0=$vector_length
fi
}
VectorPrint () {
# Vector Print
# Prints all the elements names and values of the vector $1 on sepparate lines
local vector_length
vector_length=$(($1_0))
if [ "$vector_length" = "0" ]; then
echo "Vector \"$1\" is empty!"
else
echo "Vector \"$1\":"
for ((i=1; i<=$vector_length; i++)); do
eval echo \"[$i]: \\\"\$$1\_$i\\\"\"
###OR: eval printf \'\%s\\\n\' \"[\$i]: \\\"\$$1\_$i\\\"\"
done
fi
}
VectorDestroy () {
# Vector Destroy
# Empties all the elements values of the vector $1
local vector_length
vector_length=$(($1_0))
if [ ! "$vector_length" = "0" ]; then
for ((i=1; i<=$vector_length; i++)); do
unset $1_$i
done
unset $1_0
fi
}
##################
### MAIN START ###
##################
## Setting vector 'params' with all the parameters received by the script:
for ((i=1; i<=$#; i++)); do
eval param="\${$i}"
VectorAddElementNext params param
done
# Printing the vector 'params':
VectorPrint params
read temp
## Setting vector 'params2' with the elements of the vector 'params' in reversed order:
if [ -n "$params_0" ]; then
for ((i=1; i<=$params_0; i++)); do
count=$((params_0-i+1))
VectorAddElement params2 count params_$i
done
fi
# Printing the vector 'params2':
VectorPrint params2
read temp
## Getting the values of 'params2'`s elements and printing them:
if [ -n "$params2_0" ]; then
echo "Printing the elements of the vector 'params2':"
for ((i=1; i<=$params2_0; i++)); do
eval current_elem_value=\"\$params2\_$i\"
echo "params2_$i=\"$current_elem_value\""
done
else
echo "Vector 'params2' is empty!"
fi
read temp
## Creating a two dimensional array ('a'):
for ((i=1; i<=10; i++)); do
VectorAddElement a 0 i
for ((j=1; j<=8; j++)); do
value=$(( 8 * ( i - 1 ) + j ))
VectorAddElementDV a_$i $j $value
done
done
## Manually printing the two dimensional array ('a'):
echo "Printing the two-dimensional array 'a':"
if [ -n "$a_0" ]; then
for ((i=1; i<=$a_0; i++)); do
eval current_vector_lenght=\$a\_$i\_0
if [ -n "$current_vector_lenght" ]; then
for ((j=1; j<=$current_vector_lenght; j++)); do
eval value=\"\$a\_$i\_$j\"
printf "$value "
done
fi
printf "\n"
done
fi
################
### MAIN END ###
################
If each row of the matrix is the same size, then you can simply use a linear array and multiplication.
That is,
a=()
for (( i=0; i<4; ++i )); do
for (( j=0; j<5; ++j )); do
a[i*5+j]=0
done
done
Then your a[2][3] = 3 becomes
a[2*5+3] = 3
This approach might be worth turning into a set of functions, but since you can't pass arrays to or return arrays from functions, you would have to use pass-by-name and sometimes eval. So I tend to file multidimensional arrays under "things bash is simply Not Meant To Do".
One can simply define two functions to write ($4 is the assigned value) and read a matrix with arbitrary name ($1) and indexes ($2 and $3) exploiting eval and indirect referencing.
#!/bin/bash
matrix_write () {
eval $1"_"$2"_"$3=$4
# aux=$1"_"$2"_"$3 # Alternative way
# let $aux=$4 # ---
}
matrix_read () {
aux=$1"_"$2"_"$3
echo ${!aux}
}
for ((i=1;i<10;i=i+1)); do
for ((j=1;j<10;j=j+1)); do
matrix_write a $i $j $[$i*10+$j]
done
done
for ((i=1;i<10;i=i+1)); do
for ((j=1;j<10;j=j+1)); do
echo "a_"$i"_"$j"="$(matrix_read a $i $j)
done
done
Mark Reed suggested a very good solution for 2D arrays (matrix)! They always can be converted in a 1D array (vector). Although Bash doesn't have a native support for 2D arrays, it's not that hard to create a simple ADT around the mentioned principle.
Here is a barebone example with no argument checks, etc, just to keep the solution clear: the array's size is set as two first elements in the instance (documentation for the Bash module that implements a matrix ADT, https://github.com/vorakl/bash-libs/blob/master/src.docs/content/pages/matrix.rst )
#!/bin/bash
matrix_init() {
# matrix_init instance x y data ...
declare -n self=$1
declare -i width=$2 height=$3
shift 3;
self=(${width} ${height} "$#")
}
matrix_get() {
# matrix_get instance x y
declare -n self=$1
declare -i x=$2 y=$3
declare -i width=${self[0]} height=${self[1]}
echo "${self[2+y*width+x]}"
}
matrix_set() {
# matrix_set instance x y data
declare -n self=$1
declare -i x=$2 y=$3
declare data="$4"
declare -i width=${self[0]} height=${self[1]}
self[2+y*width+x]="${data}"
}
matrix_destroy() {
# matrix_destroy instance
declare -n self=$1
unset self
}
# my_matrix[3][2]=( (one, two, three), ("1 1" "2 2" "3 3") )
matrix_init my_matrix \
3 2 \
one two three \
"1 1" "2 2" "3 3"
# print my_matrix[2][0]
matrix_get my_matrix 2 0
# print my_matrix[1][1]
matrix_get my_matrix 1 1
# my_matrix[1][1]="4 4 4"
matrix_set my_matrix 1 1 "4 4 4"
# print my_matrix[1][1]
matrix_get my_matrix 1 1
# remove my_matrix
matrix_destroy my_matrix
For simulating a 2-dimensional array, I first load the first n-elements (the elements of the first column)
local pano_array=()
i=0
for line in $(grep "filename" "$file")
do
url=$(extract_url_from_xml $line)
pano_array[i]="$url"
i=$((i+1))
done
To add the second column, I define the size of the first column and calculate the values in an offset variable
array_len="${#pano_array[#]}"
i=0
while [[ $i -lt $array_len ]]
do
url="${pano_array[$i]}"
offset=$(($array_len+i))
found_file=$(get_file $url)
pano_array[$offset]=$found_file
i=$((i+1))
done
The below code will definitely work provided if you are working on a Mac you have bash version 4. Not only can you declare 0 but this is more of a universal approach to dynamically accepting values.
2D Array
declare -A arr
echo "Enter the row"
read r
echo "Enter the column"
read c
i=0
j=0
echo "Enter the elements"
while [ $i -lt $r ]
do
j=0
while [ $j -lt $c ]
do
echo $i $j
read m
arr[${i},${j}]=$m
j=`expr $j + 1`
done
i=`expr $i + 1`
done
i=0
j=0
while [ $i -lt $r ]
do
j=0
while [ $j -lt $c ]
do
echo -n ${arr[${i},${j}]} " "
j=`expr $j + 1`
done
echo ""
i=`expr $i + 1`
done
Situation: I have 2 lists.
list_all_process.ini
Listing all of my bash scripts
exclude_proccess.ini
listing all of my "excluding" bash scripts
These contain
list_all_process.ini
script-1.0.sh
script-2.0.sh
script-3.0.sh
script-4.0.sh
-
exclude_proccess.ini
script-3.0.sh
What I would like to have:
If the value doesn't correspond to the exclude proccess -> execute the script. If match -> nothing
My code:
except_proccess(){
readarray a -t < ${INC}/full_proccess_list.ini
readarray b -t < ${INC}/exclude_list.ini
for ACCEPT in ${a[#]} ; do
for DENNY in ${b[#]} ; do
if [[ $ACCEPT == $DENNY ]]; then
continue;
else
bash ${INC}/${ACCEPT}
fi
done
done
My problem:
False positive and ID contain in exclude_list is launched ...
I am wondering why you do it in such a complicated way. Try this:
for p in $(cat list_all_process.ini) ; do
grep -q "$p" exclude_proccess.ini || echo "execute $p"
done
I have a flat file which contains the following
INDIA USA SA NZ AUS ARG GER BRA
so there are eight columns altogether . Now I want to store the indexes of those columns only which starts with A into an array.
For that I tried the following statement
awk '{for(i=1;i<=NF;i++){if($i~/^A/){set -A newArray $i}}}' testUnix.txt
when I echo the file using
echo "${newArray[*]}"
it's printing 5 6 but whenever I am trying to get the length of that array
echo ${#newArray[#]}
its length is being shown as 1 only. Should not it be 2 ?
I also tried
awk '{y = 0;for(i=1;i<=NF;i++){if($i~/^A/){newArray[y] = $i ; y++}}}' testUnix.txt
but also it's producing the same result.
What am I missing ?Please explain. I intend to get the desired output 2.
No need for awk. You can loop through the elements and check if they start with A:
r="INDIA USA SA NZ AUS ARG GER BRA"
arr=()
for w in $r
do
[[ $w == A* ]] && arr+=("$w")
done
If you execute it then the arr array contains:
$ for i in "${arr[#]}"; do echo "$i"; done
AUS
ARG
And to confirm that is has two elements, let's count them:
$ echo "${#arr[#]}"
2
What is happening with your aproach?
awk '{for(i=1;i<=NF;i++){if($i~/^A/){set -A newArray $i}}}' testUnix.txt
this says set -A newArray but it is not really defining the variable in bash, because you are in awk.
What I would do to have a bash array :
bash_arr=( $(awk '{for(i=1;i<=NF;i++){if($i~/^A/){print $i}}}' file) )
echo "${bash_arr[#]}"
AUS ARG
And you don't even need awk in reality, bash is capable of doing regex :
for word in $(<file); do [[ $word =~ ^A ]] && basharr+=( "$word" ); done
I am storing a file list in an array . What I'd like to do is loop through a process that will read N elements of an array and delete all the elements it just read. The Exception is the last iteration , when you come to the last iteration of the loop - whatever remains in the array - just split that out .
th=$1
ar=('f1' 'f2' 'f3' 'f4' 'f5' 'f6')
for ((i=1; i<=$th; i++)); do
<stuff>
if [ "$i" -eq "$th" ]
then
# if its the last iteration of the loop.Whatever remains in the array - spit that out
echo " `echo ${ar[*]}`" >> somefile
# it its anything short of the last iteration.Read N elements of the array at a time
# and then delete them
else
echo " `echo ${ar[*]:0:$N}` " >> somefile
for ((x=0; x<=$N; x++)) ; do
unset ar[$x]
done
fi
The results are very erratic. Even when I use this approach and test if separately
for ((x=0; x<=$N; x++)) ; do
unset ar[$x]
done
It will delete the WHOLE array EXCEPT the $Nth element
I am new to arrays in shell. Any help is gladly appreciated
Try the following:
#! /bin/bash
th=3
N=2
ar=('f1 f' 'f2' 'f3' 'f4 x' 'f5' 'f6' 'f7')
for ((i=0; i<$th; i++)); do
if (( $i == $(($th - 1)) )) ; then
echo "${ar[*]}"
else
echo "${ar[*]:0:$N}"
ar=( "${ar[#]:$N}" )
fi
done
Output:
f1 f f2
f3 f4 x
f5 f6 f7
Note:
Arrays in bash are zero based.
Indices are not adjusted after unset ar[$x], therefore it would be easier to reconstruct the array as ar=( "${ar[#]:$N}" ) to force new indices to start at zero..
Update:
Or you could avoid the reconstruction of the array using:
#! /bin/bash
th=3
N=2
ar=('f1 f' 'f2' 'f3' 'f4 x' 'f5' 'f6' 'f7')
for ((i=0; i<$th; i++)); do
if (( $i == $(($th - 1)) )) ; then
echo "${ar[*]:$(($i * $N))}"
else
echo "${ar[*]:$(($i * $N)):$N}"
fi
done
I'm trying to convert the output of a command like echo -e "a b\nc\nd e" to an array.
X=( $(echo -e "a b\nc\nd e") )
Splits the input for every new line and whitespace character:
$ echo ${#X[#]}
> 5
for i in ${X[#]} ; do echo $i ; done
a
b
c
d
e
The result should be:
for i in ${X[#]} ; do echo $i ; done
a b
c
d e
You need to change your Internal Field Separator variable (IFS) to a newline first.
$ IFS=$'\n'; arr=( $(echo -e "a b\nc\nd e") ); for i in ${arr[#]} ; do echo $i ; done
a b
c
d e
readarray -t ARRAY < <(COMMAND)
Set the IFS to newline. By default, it is space.
[jaypal:~] while IFS=$'\n' read -a arry; do
echo ${arry[0]};
done < <(echo -e "a b\nc\nd e")
a b
c
d e