Skip to content

Release the codes as only 1 CSV

Right now, the releases have this file structure:

2025-1 UNLOCODE CodeList.mdb
loc242csv/
├── 2025-1 SubdivisionCodes.csv
├── 2025-1 UNLOCODE CodeListPart1.csv
├── 2025-1 UNLOCODE CodeListPart2.csv
└── 2025-1 UNLOCODE CodeListPart3.csv
loc242txt/
├── 2025-1 SubdivisionCodes.txt
├── 2025-1 UNLOCODE CodeList.txt
└── 2025-1 UNLOCODE Codelist.xml

Since we're doing things different anyway, I suggest to replace all this with just 1 CSV.

TXT There is no reason for anyone to use TXT instead of CSV: it's pretty much the same.

CSV I don't see why the CodeList should be 3 files, while the TXT is only 1 file. It complicates things when reading these CSV's programmatically. I also don't see why 3? One per country (like the repo does) makes sense, or all in 1 file, but why 3 specifically?

MDB The only reason to use this old file format in the previous release, was that this was the only way to get reliable diacritics (special characters). This will be fixed (is already fixed apart from 2 edge cases), so there will be no reason to have to have this MDB anymore.

Other file structure for the CSV Right now, the CodeListPartX.csv files are basically 3 files in 1:

  • The UN/LOCODE entries
  • The country codes + names
  • Aliases

My suggestion is also to split these up, but there are separate issues for this: #12 to split up or get rid of the aliases, and #10 (closed) talks about country codes (among other things)