Parsing multi-line fixed-width files

Question

I have a fixed-width flat file. To make matters worse, each line can either be a new record or a subrecord of the line above, identified by the first character on each line:

A0020SOME DESCRIPTION   MORE DESCRIPTION 922 2321      # Separate
A0021ANOTHER DESCRIPTIONMORE DESCRIPTION 23111442      # records
B0021ANOTHER DESCRIPTION   THIS TIME IN ANOTHER FORMAT # sub-record of record "0021"

I've tried using Flatworm which seems to be an excellent library for parsing fixed-width data. It's documentation, unfortunately, states:

"Repeating segments are supported only for delimited files"

(ibid, "Repeating segments").

I'd rather not write a custom parser for this. Is it (1) possible to do this in Flatworm or (2) is there a library providing such (multi-line, multi-sub-record) capabilities?

Answer 1

Have you looked at JRecordBind?

http://jrecordbind.dev.java.net/

"JRecordBind supports hierarchical fixed length files: records of some type that are 'sons' of other record types."

Answer 2

You can import the data into a relational database.

Answer 3

Check Preon. Although Preon is targeting bitstream compressed data, you might be able to twist its arm and use it for the file format you identified as well. The benefit of using Preon would be that it will generate human-readable documentation as well.