purchase-history-parsers

Parse store purchase histories into plain Python records: what you bought, when, for how much, and how you paid, going as far back as each store's own history.

Part of infinparse. Each parser takes one file you exported from your own account and returns a list of Purchase dataclasses. No parser logs in or makes network calls. For stores with no export button, the browser exporters produce the file from your own logged-in browser session.

Supported stores

Any currency works; each row keeps the currency the store shows. Some parsers expect English-language pages; see each runbook.

Install and use

pip install purchase-history-parsers

One dependency: Beautiful Soup, for reading saved HTML pages.

from purchase_history_parsers import parser

rows = parser("steam").parse("steam-purchase-history.PERSONAL.html")   # list[Purchase]

parser("mypkg.mystore") loads your own parser by dotted path, so you don't need a fork.

Exporters

The exporters are small JavaScript files that run in your browser, on the store's own page, while you're logged in. Each loads your full history and downloads the file the matching parser reads. Install one in a userscript manager (Tampermonkey, Violentmonkey) to get an "Export" button on the page, or paste it into the browser's developer console to run it once. Your password never touches this package, and the Python parsers never go online.

What an exporter saves is only what its parser reads: the history rows, with no page scripts, session data, account email or profile. It is still your purchase history, so the file name says PERSONAL and the page carries a banner. Keep it private: never attach it to an issue. To report a parsing problem, describe the layout or share masked structure instead.

The model

Purchase

Licence

MIT. See LICENSE.