Hello brethern and sisthren,

Sorry about the ropey title - I couldn't think of anything snappier.

I have a set of test cases defined as blocks of paramater-value pairs. There are (as yet) two types of block, say, A and B:

# A type A block
block_type=A
block_id=1
magic=true
unlikeliness=2

# Another type A block
block_type=A
block_id=2
magic=false
unlikeliness=9

# A type B block
block_type=B
block_id=1
colour=green
warty=false
horns=1

# Another type B block
block_type=B
block_id=2
colour=red
warty=false
horns=7

# A further type B block
block_type=B
block_id=3
colour=red
warty=true
horns=1
[download]

A test case is defined by a combination of the parameters from an A and a B block. I would like to be able to compare test cases.

My approach would be to read the file line for line, work out if the line belongs to the current block, and store the data for a block in a hash, thus producing two arrays of hashes. For each B-block I could obtain the appropriate A-block and create a hash for the test case. Then I could compare test cases by comparing hash-values.

However, I sure there is a smarter way to do it. If anyone has any ideas, they would be greatly appreciated.

Thanks,

loris