Probably a noob question, but I'm having a particular challenge where I want to dynamically generate variable names based on a value. Eg. If I'd require 3 variables, variable names to be incremented dynamically var_0,var_1,var_3 respectively.
The solution I have right now is barebones. I'm just using read -p get user input and save it to variable. So for 5 hosts, I've just duplicated this 5 times to get the job done, but not a clean solution. I was looking around and reading through declare and eval but haven't got anywhere
Thinking of a solution where I input no. of hosts first and this dynamically picks up and asks for user input based on number of hosts and saves to variables incremented dynamically
Any help is appreciated, cheers
Use arrays. However, you don't have to ask the user for the number of hosts. Let them enter as many hosts as they like and finish by entering an empty line:
echo "Please enter hosts, one per line."
echo "Press enter on an empty line to finish."
hosts=()
while read -p "host ${#hosts[#]}: " host && [ -n "$host" ]; do
hosts+=("$host")
done
echo "Finished"
Alternatively, let the user enter all hosts on the same line separated by whitespace:
read -p "Please enter hosts, separated by whitespace: " -a hosts
To access the individual hosts use ${hosts[0]}, ${hosts[1]}, ..., ${hosts[${#hosts[#]}]}.
To list them all at once use ${hosts[#]}.
Use an array instead!
But read can actualy create dynamic vars, just like this:
for i in {1..3}; do
read -p 'helo: ' var_$i
done
Related
i'm struggling with this topic: I've a file with a list of IDs, something like this:
34
23
478
12579
342356
On the other side, i've a command that i want to run, to show me the details for each object of the list (no much to say on this, is a specific command coming actually from a API). It looks like this, where "id" rappresents a single integer, which must exist in the previous list.
command-to-get-details str id
My idea was to loop in the first file with read, store the IDs in a variable through readarrayand pass them to the second command in a while loop. The problem with this solution is that i can't use readarray in this machine.
Is there any other solution to do this?
Thank you very much
From https://mywiki.wooledge.org/BashFAQ/001
#!/bin/bash
while IFS= read -r id
do
echo "command-o-get-details str $id"
done < "ids.txt"
Remove the echo to use your specific command.
I know something like this is possible with DOS but I am not sure how to do it within bash.
I am writing a script that takes some configuration data: source, name, and destination. There will be a variable number of these in the configuration. I need to iterate over each set.
So, for example:
#!/bin/bash
FOLDER_1_SOURCE="/path/one"
FOLDER_1_NAME="one"
FOLDER_1_DESTINATION="one"
FOLDER_2_SOURCE="/path/two two"
FOLDER_2_NAME="two"
FOLDER_2_DESTINATION="here"
FOLDER_3_SOURCE="/something/random"
FOLDER_3_NAME="bravo"
FOLDER_3_DESTINATION="there"
FOLDER_..._SOURCE="/something/random"
FOLDER_..._NAME="bravo"
FOLDER_..._DESTINATION=""
FOLDER_X_SOURCE="/something/random"
FOLDER_X_NAME="bravo"
FOLDER_X_DESTINATION=""
Then I want to iterate over each set and get the SOURCE and NAME values for each set.
I am not stuck on this format. I just don't know how else to do this. The end goal is that I have 1 or more set of variables with source, name, and destination and then I need to iterate over them.
The answer to this type of question is nearly always "use arrays".
declare -a folder_source folder_name folder_dest
folder_source[1]="/path/one"
folder_name[1]="one"
folder_dest[1]="one"
folder_source[2]="/path/two two"
folder_name[2]="two"
folder_dest[2]="here"
folder_source[3]="/something/random"
folder_name[3]="bravo"
folder_dest[3]="there"
folder_source[4]="/something/random"
folder_name[4]="bravo"
folder_dest[4]=""
for((i=1; i<=${#folder_source[#]}; ++i)); do
echo "$i source:" "${folder_source[$i]}"
echo "$i name:" "${folder_name[$i]}"
echo "$i destination:" "${folder_dest[$i]}"
done
Demo: https://ideone.com/gZn0wH
Bash array indices are zero-based, but we just leave the zeroth slot unused here for convenience.
Tangentially, avoid upper case for your private variables.
AFIK bash does not have a facility to list all variables. A workaround - which also would mimic what is going on in DOS - is to use environment variables and restrict your search to those. In this case, you could do a
printenv|grep ^FOLDER||cut -d = -f 1
This is the same as doing in Windows CMD shell a
SET FOLDER
I want to create a list of variables,so later i can do a prompts for users and if their input is matching any variable within that list, I want to use that variable to execute it within a command, for example:
var1=a
var2=b
...
read input
(user chooses var1) command $var1,rest of the command
The biggest problem is that this list will be huge, what would be the best solution? Thanks!
You are looking for the associative array feature in bash 4 or later.
declare -A foo
foo[us]="United States"
foo[uk]="Great Britain"
read -p "Region? " region
echo "${foo[$region]}"
If the value of $region is not a defined key, then ${foo[$region]} will be treated like any unset variable.
Is it possible to build an array from a parameterized jenkins build?
I have tried the https://wiki.jenkins-ci.org/display/JENKINS/Extended+Choice+Parameter+plugin which allows me to create a single title with multiple options within it. So I build a extended choice called services with 5 services listed as check boxes.
However when I try to do a loop over what I thought would be an array ${services[#]} I just get the single value of comma separated values. I tried setting IFS=',' and that does not work.
Any ideas?
This just doesn't work with check boxes. If you use a text field and specify each variable there it will loop as if it were a true array.
You can create an array first from Jenkins multiple choice variable:
IFS=',' read -r -a services <<< "$services"
for service in ${services[#]}; do
echo "$service"
done
I'm creating a *.deb package that transform your wireless card into an hotspot.
I'm stuck at the configurations:
I have to write a postinst file in which I ask to the user what ip address he likes for his hotspot and then use it to generate the range & the subnet addresses for the isc-dhcp-server.
Something like that:
10.10.0.01 + 0.0.0.9 = 10.10.0.10
I know how to assign strings and numbers to variables and how to ask to user his choosen IP, but how to modify a variable and assign the result to another one? expr thinks it's a floating number and won't work.
Hoping that everything it's clear enough,
waiting for a help,
thank you in advance
Avoid leading zeros.
IFS="." read -a a <<< 10.10.0.1
IFS="." read -a b <<< 0.0.0.9
s="$[a[0]+b[0]].$[a[1]+b[1]].$[a[2]+b[2]].$[a[3]+b[3]]"
echo $s
Output:
10.10.0.10
Ok, I found a workaround method:
when I ask to the user its choosen ip I use these:
IFS="." read -r a b c d
choosenip="$a.$b.$c.$d"
subnetip="$a.$b.$c.0"
rangeipmin="$a.$b.$c.20"
rangeipmax="$a.$b.$c.30"
IFS change the default "space" or "tab" to whatever you want.
So when I have to put these in the dhcpd.conf with "echo", I just have to call the variables.
If you have more elegant ways to do that, you're welcome.
Thank you