Shell script to extract specific data from a file - regex

Question

Given a text file containing records of the following form:
.....
feGroup1Person1 Person ::= {
id 1011,
uniquename "name1",
data 40,
moredata 100
}
feGroup1Person2 Person ::= {
id 5223,
uniquename "name2",
data 40,
moredata 200
}
.......
In a shell script, how could I go about extracting the Group and Person IDs for a particular uniquename?
For Example: Given "name2", I want to extract "feGroup1Person2".
I'm assuming some regular expressions will be required, but I'm not having any luck with it.
Any help appreciated

Answer 1

> awk '$0~/Person ::= \{/{x=$1; print x}' file
feGroup1Person1
feGroup1Person2
>
If you just want the group id you can use below:
for example you want the group is of person whose name is "name2",then:
awk '/name2/{print x2}{x2=x1;x1=x;x=$1}' file
feGroup1Person2
if name is "name1"
awk '/name1/{print x2}{x2=x1;x1=x;x=$1}' file
feGroup1Person1

Answer 2

You don't want to use shell scripting for this. You need to use something like Perl, VBScript, PowerShell or one of the many other more sophisticated scripting languages.
Which you use will depend primarily on your platform. On Windows try VBScript as a first choice. On Linux, try Perl first.

Answer 3

Don't attempt to formulate a solution entirely in terms of regular expressions. Your problem is sufficiently complex that regexes alone are not a wise choice of tool.
With a bit of manipulation, you could make this look like data in the JSON format and then parse it using a JSON parser. Any decent programming language (Python, Perl, Ruby...) should come with a JSON parser.
{
"feGroup1Person1" : {
"id" = 1011,
"uniquename" = "name1",
"data" = 40,
"moredata" = 100
}
"feGroup1Person2" :
{
"id" : 5223,
"uniquename" : "name2",
"data" : 40,
"moredata" : 200
}
}