Hello
I'm new to this. I'm using ksh93 version Version JM 93u 2011-02-08
This little example will create an *array of hashes* ( perl speaking)
Created a little script like this :
#!/usr/bin/ksh
typeset -a ALL
. ./arr.txt # see below what's in this file.
# array of hashes in ksh93
print ${#ALL[*]} # show the number of entries in the ${arr[*]}
for (( x=0; $x <= ((${#ALL[*]}-1)) ; x++ ))
do
echo Index : $x , keys : ${!ALL[$x][*]}
for z in ${!ALL[$x][*]}
do
print key $z '->' value : ${ALL[$x][$z]}
# print "${!ALL[$x][$z]} '->' ${ALL[$x][$z]}"
done
done
The arr.txt file that is imported looks like this :
ALL[0]=(
[TEST1]=t1
[TEST2]=t2
[TEST3]=t3
)
ALL[1]=(
[PROEF1]=proef1
[proef2]=proef2
[proef3]=proef3
)
ALL[2]=(
[TEST1]=bla1
[TEST2]=bla1.1
[TEST3]=bla1.2
)
ALL[3]=(
[TEST1]=simac1
[TEST2]=simac2
[TEST3]=simac3
)
ALL[4]=(
[Machine1]="192.168.2.136"
[Machine2]="192.168.2.137"
[Machine3]="192.168.2.138"
)
output of the script :
5
Index : 0 , keys : TEST1 TEST2 TEST3
key TEST1 -> value : t1
key TEST2 -> value : t2
key TEST3 -> value : t3
Index : 1 , keys : PROEF1 proef2 proef3
key PROEF1 -> value : proef1
key proef2 -> value : proef2
key proef3 -> value : proef3
Index : 2 , keys : TEST1 TEST2 TEST3
key TEST1 -> value : bla1
key TEST2 -> value : bla1.1
key TEST3 -> value : bla1.2
Index : 3 , keys : TEST1 TEST2 TEST3
key TEST1 -> value : simac1
key TEST2 -> value : simac2
key TEST3 -> value : simac3
Index : 4 , keys : Machine1 Machine2 Machine3
key Machine1 -> value : 192.168.2.136
key Machine2 -> value : 192.168.2.137
key Machine3 -> value : 192.168.2.138
I've used the same concept for the hash of hashes ..
This is the script.
#!/usr/bin/ksh
# please notice this is ksh93
typeset -A systems
. ./systems1.txt
print "showing ALL\n";
# Getting the keys and values
for i in ${!systems[@]}
do
print $i :
for z in ${!systems[$i][@]}
do
print $z : ${systems[$i][$z]}
done
print "\n"
done
# just showing one system
print "see the data for system : \c"
read system
print "Getting data for $system\n"
if [[ ! $systems[$system] ]]; then
print "could not find system : \"$system\""
else
for z in ${!systems[$system][@]}
do
print $z : ${systems[$system][$z]}
done
fi
In the file systems1.txtwe see :
systems=(
[system1]=(
[description]="rc1_750_boven"
[omgeving]="IBM P750"
[status]="active"
[beheer]="FB"
[beheer_omgeving]="FAO"
[ip_beheer]=""
[ip_applicatie]=""
[pnode]="pnode0165"
[location]="Beursplein 1"
[servicelevel]="Just testing"
[type]="IBM 8233-E8B"
[contract]="gold 7x24"
[klantnummer]="Fake1"
[telefoon]="123-345345"
[serial]="Fake1"
[record]="40"
)
[system2]=(
[description]="IB-nieuw , rc1_750_onder"
[omgeving]="IBM P750"
[status]="active"
[beheer]="FB"
[beheer_omgeving]="FAO"
[ip_beheer]=""
[ip_applicatie]=""
[pnode]="pnode0076"
[location]="Beursplein 1"
[servicelevel]="Gold 7x24"
[type]="IBM 8233-E8B"
[contract]="Fake2"
[klantnummer]="Fake2"
[telefoon]="Fake.1"
[serial]="Serial-Fake"
[record]="32"
)
)
You can add add many [systems1], [system2] .... systems101 blocks as
you like. It's just an example. Ofcouse these items must be unique (
it's a hash )
One question about compound variables :
# uname -a on my linux give me back :
#Linux arnot-VirtualBox 3.0.0-12-generic #20-Ubuntu SMP Fri Oct 7
14:56:25 UTC 2011 x86_64 x86_64 x86_64 GNU/Linux
# So I use the following code to store the items I want into a compound
variable:
str=$(uname -a)
typeset -C uname # declare a compound variable
uname.os=${str:0:5} # ${str:start:length} is substr
( perl speaking )
uname.virtualMachine=${str:6:16}
uname.linuxKernel=${str:23:16}
uname.osName=${str:40:10}
uname.date=${str:55:27}
uname.arch=${str:84:5}
uname.end=${str:104:9}
# trying , in a loop to get at the values :
for i in ${!uname.*}
do
print $i '->' ${i} VALUE??
done
output of the script :
uname.arch -> uname.arch VALUE??
uname.date -> uname.date VALUE??
uname.end -> uname.end VALUE??
uname.linuxKernel -> uname.linuxKernel VALUE??
uname.os -> uname.os VALUE??
uname.osName -> uname.osName VALUE??
uname.virtualMachine -> uname.virtualMachine VALUE??
I do not get back the value of the compound variable : What am I doing
wrong ???
kind regards Arno Teunisse
_______________________________________________
ast-users mailing list
[email protected]
https://mailman.research.att.com/mailman/listinfo/ast-users
