What solutions did you look at? Why didn’t they fit? E.g. [1,2]

[1] https://www.gnucash.org/index.phtml

[2] http://www.sql-ledger.com/

The best fix product wise I found was [1] but Im afraid its not self hosted and seems a bit abandoned. So basically providing the ledger backend with API calls for transactions & reporting. That would be the perfect fit.

The rest are all linked to paid accountancy software, with liscencing and hosting trickiness.

Gnucash I did check out - it just seems to be a bit too large. What im looking for is only a small subset of the functionality and id worry id have a hard time stripping out the parts we need.

sql-ledger looks interesting though... Have you used it before?

[1] http://subledger.com/

No, I had no need for accounting software so far. I just kept an eye on the oss solutions if need ever arises.

[2] has an oxymoron right on it's homepage:

> Written in easy readable perl code

Perhaps you could elaborate on the use of "oxymoron" here? Just checking, as you apparently do not know your "it's" from your "its".

But seriously, why the bashing? Have you actually looked at the code and determined it is not easily readable? If not, why do you opine that it is unreadable? Prejudice is easily propagated. Please don't.

It's a joke, because perl is famously unreadable. That's all.

I see. So you're just propagating prejudice.

I've been looking for a good invoicing solution along the same lines (OSS and realistically self hostable) and not found much that ticks all the boxes.

Would your project be OSS as well?

To answer the last question, things like this are pretty CRUD-y, so a decent MVC framework/library should be enough to get going.

>I've been looking for a good invoicing solution

Thats what we do :P But its closed source and very localised so not much help. Invoicing is one of those things that looks easy but in practice turns out to be quite tricky and varies by region and vertical - especially when integrated with payments.

> Would your project be OSS as well?

If possible, yes. We have a paid closed-source product also in finance but there is solid rationale to contributing to a localised FOS bookkeeping product. All of our clients are on cracked versions of old accountancy software - we cant integrate with these so the clients end up doing a bunch of unnecessary work.

I'll recommend Odoo. Easy to deploy, configure, and it is fairly simple to make modifications or extensions you need in python.

Demo here: https://demo3.odoo.com/web

Odoo looks pretty impressive