markinp — a free tool to validate & build MARK `.inp` files (feedback welcome)

5 views

Skip to first unread message

Leon Botzenhardt

unread,

3:43 PM (4 hours ago) 3:43 PM

to unmarked

Hi all,

I've built a small, free, open-source tool called markinp to take some of the pain out of preparing Program MARK input files, and I'd love this community's feedback before I call it stable.

The two moments it targets:

"MARK rejected my .inp and the message didn't tell me why."
"I have a capture spreadsheet and need to hand-build an .inp."

What it does (three commands, nothing more):

markinp validate file.inp — checks the file and reports line-numbered, plain-English problems with a suggested fix, e.g.:

MK001 ERROR line 12: record is not terminated by a semicolon

hint: Add a ';' at the end of this line to close the record

markinp inspect file.inp — summarises the inferred structure (occasions, groups, covariates, #individuals, data-type guess).

markinp build captures.csv -o out.inp — builds a valid, deterministic .inp from a tidy CSV (long or wide), handling groups, covariates, frequencies, and comments.

It catches the usual suspects: missing semicolons, ragged history lengths, the wrong number of frequency columns for the group count, non-integer/blank values, unterminated /* */ comments, all-zero histories, encoding/BOM issues, and so on — each with a stable code and a hint.

Important, up front: markinp is an independent, unofficial utility. It is not affiliated with or endorsed by the authors of Program MARK or RMark — it only reads/writes the .inp file format. It does file I/O and validation only: it never fits models or computes anything statistical. Where I checked the format rules, I leaned on Chapter 2 of Cooch & White's Gentle Introduction.

Scope (being honest): the standard 0/1 encounter-history format (CJS / Jolly-Seber / closed captures) is fully validated. Known-fate, dead-recovery (Brownie), multistrata, and occupancy-style formats are currently only detected and structurally checked — the tool says so rather than pretending. Full support for those is on the roadmap, and I'd genuinely value pointers on the trickier format rules.

Install / links:

pip install markinp

Source, docs, and issue tracker: https://github.com/leonbzt/markinp

If you have .inp files that trip it up — especially edge cases or specialised data types — I'd love a bug report or a (non-sensitive) sample. Thanks for taking a look.