[Tool] BTC Purchase Tracker - offline desktop app for DCA buys, no accounts

[darkvector-dev]

Hi everyone,

Usually, everyone who does DCA has the same problem: tracking without mess (dates, prices, amounts of BTC/Sats, Tx/Id etc.). Someone uses Excel, Google Sheets, other online tools, but in this case, we are putting our purchases history on someone else's server (except for Excel, but for this, need to know formulas, and if you ever need to change program for X reason, it's a nightmare).

So i built a tool, an open source desktop app that runs OFFLINE, stores everything local in sqlite, and export .csv and .pdf files. No cloud, no api.


The features:

- Works on Windows (portable, no install) and Linux (AppImage), (MacOS planned if high interest)
- Stores BTC/Sats and fiat amounts as integers (no floating point errors)
- Live search by date, exchange, amount or transaction ID
- Atomic backups and csv and pdf export
- Supports both EUR and USD databases
- Totally open source on GitHub
- Free (GPLv3)



Includes export logs for troubleshooting. NO private data is collected or shared.


I'm sharing it because i hope it saves someone else headache.

Screenshots on the GitHub page


GitHub link: https://github.com/darkvector-dev/BTC-Purchase-Tracker

Release: https://github.com/darkvector-dev/BTC-Purchase-Tracker/releases/tag/v1.1.0




Feedback and bug reports are welcome.

[Cinexa]

Well done. Skimmed out quickly on the GitHub and v1.1.0 release. Offline claim all looks good from the originating source - Qt Widgets + Sql only, local SQLite, sats/cents as whole integers, GPLv3.

That's the way to go for DCA history rather than export everything to a google sheet. (although I can see the point for the latter too) One thing: could you also put the SHA256 hashes as separate files on the release page (they're in the zip for those versions, but not available separately).

Also, the Windows binary is unsigned so Smart Screen will scream at us. I'll try it out with 'dummy' data first and then maybe import an actual CSV. Thanks for open sourcing!

[darkvector-dev]

Thanks for checking the source!

Added the checksums.txt file to the release assets.


For the SmartScreen warning: i chose this way because signing requires a certificate tied to a real identity, and this is not compatible with the privacy nature of this project.
I've updated the readme with instructions:

- first warning
- more info
- run anyway



Let me know how it goes if you try it.

[Cinexa]

Nice, checkums.txt on the release is confirmed - both ZIP hashes match here. Respect on the omit code-signing for privacy - good enough to verify the hash first as a decent proxy.

Tested with dummy data. No Mac build yet so I built v1.1.0 from source (since haven't downloaded Windows binary) - passes official tests, then I dropped in some fake DCA and imported your sample_import.csv - search, total, backup all worked. Looks to be properly offline as well (local SQLite, no network in app itself).

Will play around more; so far so good.

Mac binary would be a bonus when you're ready.

[darkvector-dev]

Thanks a lot for the deep testing, really appreciate it.

About the Mac build: i think i'll work on it soon. My problem is that i don't have a personal Mac for testing, but i'm thinking i can run testing on my friend's iMac (M3 -> ARM version).

[darkvector-dev]

***** UPDATE *****

Just released v1.1.1, which includes support for macOS Apple Silicon (from M1 cpu).

This build was tested successfully on an iMac M3 with macOS Tahoe 26.6.2. During the test all features worked, data entry, search, .csv export and import, database backups and restore. The final build differs from the tested one only for a Windows specific fix, which does not affect macOS.

If anyone has an Apple Silicon Mac and wants to test it, feedback would be appreciated.


Link GitHub v1.1.1: https://github.com/darkvector-dev/BTC-Purchase-Tracker/releases/tag/v1.1.1

[darkvector-dev]

***** UPDATE *****

New tests executed on latest release (v1.1.1):

- Windows 10: Ok
- Windows 11: Ok
- Linux EndeavourOS: Ok
- macOS Tahoe: Ok
- macOS Golden Gate: Ok