Shell Programming and Scripting

I am trying to parse a csv file in the below 'name-value pair' format and then use the values corresponding to the name.

Type:G,Instance:instance1,FunctionalID:funcid,Env:dev,AppName:appname
Type:A,AppName:appname,ProcessNamerocess1.ear,Machine:machine1|Mode:FT|Setting:P,Machine:machine2| Mode:FT|Setting:B
Type:A,AppName:appname,ProcessNamerocess2.ear,Machine:machine1|Mode:FT|Setting:P,Machine:machine2| Mode:FT|Setting:B

I have to get the the value instance1(from Instance) dynamically by looking for Type:G and similarly other variables corresponding to their name pair. And then from Type:A, I have to compare AppName with Type:G and get the variables corresponding to the name pairs. Please note that if there are more than two values, there is an other delimitter '|' - Machine:machine1|Mode:FT|Setting:P

I am trying to use sed , cut etc., but couldn't hit the right logic. I really appreciate if I get a solution here.

Thanks,
Chiru

Can you provide the expected/desired output?
Also, to help with exception & rule-handling, perhaps more input lines.

The output is that I should get each value corresponding to the name-value pair from the csv file. I am using these variables in another scripts. For example, from the line Type:G,Instance:instance1,FunctionalID:funcid,Env:dev,AppName:appname
I should get the value instance1 corresponding to Instance. I am trying to look for Instance and get the value next to Instance:
In the script I am exporting the value of instance1 as export INSTANCE=instance1

I hope it is clear and please let me know how can I do that.

Thanks,
Chiru

Simply, here is what I am looking for:
from the csv file that has below line:
Type:G,Instance:instance1,FunctionalID:funcid,Env:dev,AppName:appname

The output may be as a file or be a variable:

INSTANCE=instance1
FUNCID=funcid
ENV=dev
APPNAME=appname

Thanks,
Ravi

Hi,
Actually i am not sure about your requirements, could you please specific on it?

According to your last thread, i think below code can address your issue.

Code:

echo "Type:G,Instance:instance1,FunctionalID:funcid,Env:dev,AppName:appname" | sed 's/:/=/g' | awk 'BEGIN{FS=","}
{
print $2
print $3
print $4
print $5
}'

Well, maybe I've put my problem in a complicated way - excuse my English

There is an input file(csv file)which has the following lines:

cat csvfile.csv
Type:G,Instance:instance1,FunctionalID:funcid,Env:dev,AppName:appname
Type:A,AppName:appname,ProcessNamerocess1.ear,Machine:machine1|Setting:P,Machine:machine2|Setting:B
Type:A,AppName:appname,ProcessNamerocess2.ear,Machine:machine1|Setting:P,Machine:machine2|Setting:B

This file is created as a name-value pair like Instance:instance1.

From this file, I need to extract 'instance1' variable, and similarly all other variables. I am using these variables and using in another script to export those variables.

The one challenge I am seeing is that these name-value pairs in the csv file can be in different order. For example,
cat csvfile.csv
Type:G,Instance:instance1,FunctionalID:funcid,Env:dev,AppName:appname
....

cat csvfile.csv
FunctionalID:funcid,Env:dev,Type:G,Instance:instance1,AppName:appname

So I cannot just say $2 or $3 to get instance1. I have to look for corresponding name 'Instance' and then from that, derive its value-instance1.

Any ideas??