Parsing the Infamous Japanese Postal CSV
Summary
The article discusses posuto, a Python library for parsing Japan Post's ken_all.csv, highlighting the data quality and formatting quirks that make the CSV difficult to parse. It covers issues like long-field line folding, non-quoted lines, and locale-specific notes, and points to use of pre-processed JSON or a romaji converter as practical alternatives. The piece also situates these challenges within broader considerations for handling postal data and invites reader feedback and contributions.