Shell Programming and Scripting

Hi,

Can someone let me know how I can split a record when it contains a vairable length of fields.

Code:

Line1

field1,field101,field102,field 103,field104

Line 2

field1,field101,field102,field 103,field104,field201,field202,field 203,field204

Line 3
field1,field101,field102,field 103,field104,field201,field202,field 203,field204,field301,field 302,field303,field304

In the above scenario,I wanna blow up individual record for the number of times the field101..field104 or field201...field204 or field 301..304 exists.

So for the above example the expected output would be,

Line 1 would remain as such.

Line 2 would be splitted into 2 where field101,..field104(record1) and field201..field204(record 2)

Line 3 would be splitted into 3 where field101,..field104(record1) ,field201..field204(record 2) and field301..field304(record 3)

Appreciate your inputs.

Thx

For each record, would there be only 4 fields ? Or does that vary. If it varies, then you need to show us what the fields look like.

Hi,
There would only four fields for each record split.

Code:

ruby -ne 'scan(/([a-z]+ ?(\d)\d*,(?:[a-z]+ ?\2\d*,?)*)/i ).each{|x|
  puts x.first.chomp(",") }' myfile

Hi Futurelet,
could you tell me what is this command 'RUBY' ? Is this pearl ?????
Iam looking to acheive this in shell scripting.

Hi Braindrain,

this will work

where f1.txt is ur input file,

Gaurav