Case study
Contact record extractor
Parses several thousand irregular contact records out of one text file into a clean spreadsheet, keyed on a repeating delimiter rather than on absent fields.
Overview
A site had been collecting contact details for years and the result was a single text file of several thousand entries, none of them laid out the same way. The client wanted it as a spreadsheet.
What I own
The parser and the output format.
Complexities tackled
There were no fields, only a repeating boundary. Entries did not have consistent labels, ordering or line counts. The one reliable signal was that each new record began with an opening bracket. So the parse keys on that boundary and works out the shape of each record from what falls between them, rather than expecting a format the data never actually had.
Failing on the sample is the point. A parser written against eight visible entries will break on the other few thousand. The useful behaviour is to surface the records it cannot make sense of rather than silently emitting blank columns for them, because a spreadsheet with quiet gaps is worse than one with an obvious problem.
Stack
Python, with spreadsheet output.
Working on something similar?
Tell me what you are building and what is in the way. I will tell you honestly whether I am the right person for it.