You could also use Resolver One[1], which sadly is no longer in development[2]. It is a commercial spreadsheet written in IronPython that uses (almost?) vanilla python files as documents. Needless to say it is extremely scriptable.

I'd love to take a stab at making a lite version in QT or wxPython running in CPython... Sadly, I've gotten too used to Google Spreadsheets to really need a desktop version

Also, of course, LibreOffice has support for Python macros.

[1]: http://www.resolversystems.com/products/resolver-one/

[2]: http://www.resolversystems.com/eol.php

Not sure why estebank's comment was downvoted; Resolver One was truly awesome. DataNitro's stuff is also interesting, but it's a different approach, and Resolver One's model and work shouldn't be forgotten.

> Not sure why estebank's comment was downvoted

I gave an upvote. I suspect others downvoted the comment because it spoke well of Resolver One, which is competing in the same space as DataNitro (a YC company).

Also the PySpread project which is still active: http://manns.github.io/pyspread/

Are there docs on python macros for LibreOffice? I'm having trouble tracking them down if there are...

Thanks, but as far as I can tell that just shows you how to launch scripts/macros in various ways (menu item, etc). However, it doesn't document what you can do with the macros...coding conventions...interacting with data...etc.

http://wiki.openoffice.org/wiki/Python_as_a_macro_language#S... has a sample hello world script that can be run as a macro

But I agree in general that the documentation is paltry. Good opportunity for an interested person to do some more digging, write a few blog posts and possibly contribute to the documentation :)

Great...thank you!

Personal finance software is probably one of the most commonly attacked problems in the development world, and yet there are still a ton of people just building custom spreadsheets.

I've tried Quicken and Mint and PageOnce (now called Check) and a number of other things, and I've still found that the best solution for me is a mix of GnuCash and Spreadsheets.

Why hasn't this problem been solved yet?

I've discussed this at length elsewhere, so here's my abridged opinion.

The main issue is that a "mainstream" personal finance app must have the following two features:

- Automatic download of financial transactions

- Intelligent categorization from "STBCKS3134" to "Starbucks [Dining]"

Any tool that doesn't have those two features is doomed to be "niche"... used only by people who actually enjoy and/or are disciplined enough to enter their own data regularly.

In short, we need a "Twilio for personal finance" so that people can build creative solutions and not have to reinvent the wheel. There are startups out there that are trying to solve this.

It hasn't been solved because it's not one problem. People have different wants and needs and expectations and banks and tax codes and tolerances for entering data, and are lazy when it comes to testing existing options. Why spend hours searching and signing up for trials and learning new UIs when they know how to make a spreadsheet that does what they want?

But isn't the main purpose of personal finance to influence spending decisions? An ideal solution simply just answers the question "Can I afford to X?"

I've not used any software that helps solve that problem in an elegant way, and maybe I'm just minimizing the problem, but it seems like every path to answering that question is simple arithmetic provided you know what other expenses you have in addition to X.

YNAB is the best option I have found.

For me - data entry. Form-based software takes too long to fill out. With a spreadsheet, I can import csv files or copy and paste data easily. Developers often talk about Excel being misused as a database, and I think the root cause is the same - spreadsheets provide a very friendly user interface for adding and editing data.

Because there isn't one problem with one solution? Case in point: I think most people would not find that the best solution for them is GnuCash and Spreadsheets.

Actually, discovered ledger-cli here last week (thanks HN). It's amazing. Abstracting currency to the concept of commodity + price files makes it trivial to track your shares in 401k, and have the current market value print on the balance report. Handles lots. Recurring transactions support period specifications like "every two weeks starting on 2013/10/01". Maps payee's to accounts, and supports payee aliases. Pretty much anywhere you type a string you can also use a regex. State is a text file and the program does not alter state; just reads it and prints out a report. Finally, it provides a CSV convert which makes it easy to convert an export from your bank to a list of double entry transactions (./convert bk_download.csv >> trx.dat).

Love seeing your posts, but would appreciate when I see 'em here, especially when posted by a founder, some mention of your YCombinator Summer 2012 status and backing.

Yup.

Let us know when you're slashvertising...

Every screenshot is blurred. Could fake data not have been used for demonstration purposes?

Yeah, but blurring the screenshots was quicker ;-)

When will people learn that gaussian blur is not a secure redaction method?

I didn't know that, but it was quick, and the security isn't all that important!

There were times in my life when I would have loved to have had this, but nowadays I rarely see spreadsheets (not just excel, but the concept in general) as much of a solution to anything.

spreadsheets power pretty much everything I see when I turn my head around. Dunno why you think the opposite. You're the first person I see to say something like this.

I couldn't get by without the little fellers.

The string concatenation in this make me want to cry...

Oh dear! Sorry to hear that... looking it up I find my method described as 'naive appending', oops! Still, efficiency is of absolutely no importance in this application, and as I said in the post, I'm new to Python...

"Now to get Excel to talk to eBay…"

Should just use python for that too.

I think he intends to =)

You're probably right. I didn't pick up on that =/

I do ;-)

In addition to our regular license, we have a read-only license that lets people run (but not edit) scripts. An Enterprise license lets you give up to 4 people in your organization read-only licenses.