/hledger-scripts

Helpful scripts to do things with your hledger data.

Primary LanguageHaskellMIT LicenseMIT

hledger-scripts

Helpful scripts to do things with your hledger data. I use these to supply data to a grafana dashboard.

Example Dashboard

hledger-to-influxdb

Copies your hledger data to an influxdb database called "finance".

Dependencies: influxdb and stack.

Usage:

$ influx
> drop database finance;
> create database finance;

$ ./hledger-to-influxdb.hs

Emptying, or recreating, the "finance" database is necessary, as this writes all the data, regardless of whether it already exists.

This uses the default journal file, which is ~/.hledger.journal. To use a different one, set the LEDGER_FILE environment variable.

Output:

This gives you a collection of measurements in your database. These are:

  • $value_$aggregate_$type, where
    • $value is the valuation method used:
      • normal, where the literal value is used
      • cost, where the cost value is used
      • market, where the market value is used
    • $aggregate is how the transactions are turned into measurements:
      • txns, means each transaction gets one measurement
      • dailies, means each day gets one measurement
    • $type is what is reported:
      • total, means the total value of accounts is reported
      • delta, means the difference in value of accounts is reported
  • count_total, which is the total number of transactions
  • count_delta, which is 1 for each transaction
  • market, which is the market value of each commodity

Inside these measurements are values for accounts and currencies. It's probably easiest to explore these through a tool like grafana.

I find normal value is good for tracking cash in my wallet; cost value is good for tracking my budget; and market value is good for tracking my net worth. I usually use dailies rather than txns, to avoid weird spikes if there are multiple transactions on the same day. I usually use total for account stats, and delta for transaction stats.

market-prices

Downloads market values of commodities from a few different sources.

Dependencies: python 3.

Usage:

$ cd market-prices

$ ./market-prices.py < commodities.json

Configuration:

The commodities.json file specifies what to download. It is an object where keys are commodities, and values are configuration for how to get the current value.

Here is an example showing off all of the options:

{ "commodities":
  { "BTC":   { "provider": "coinbase", "currency": "USD" }
  , "ETH":   { "provider": "coinbase" }
  , "LTC":   { "provider": "coinbase" }
  , "Euro":  { "provider": "ft_currency", "base": "EUR" }
  , "JPY":   { "provider": "ft_currency" }
  , "VADEA": { "provider": "ft_fund", "isin": "GB00B41XG308" }
  }
, "symbols":
  { "GBP": "£"
  , "USD": "$"
  }
}

There are three providers:

  • coinbase, spot price from coinbase. Arguments are:
    • base, the name of the cryptocoin (defaults to the commodity name)
    • currency, the currency to get the spot price in (defaults to "GBP")
  • ft_currency, exchange rate from Financial Times. Arguments are:
    • base, the name of the foreign currency (defaults to the commodity name)
    • currency, the currency to get the exchange rate in (defaults to "GBP")
  • ft_fund, fund NAV from Financial Times. Arguments are:
    • isin, the ISIN of the fund (defaults to the commodity name)
    • currency, the currency to get the NAV in (defaults to "GBP")

Output:

This gives a sequence of hledger market price directives like so:

P 2018-05-26 BTC $7536.47
P 2018-05-26 ETH £450.84
P 2018-05-26 LTC £90.95
P 2018-05-26 Euro £0.8752
P 2018-05-26 JPY £0.0069
P 2018-05-26 VADEA £212.16

The optional symbols object in the configuration maps currency names to prefix symbols.

Any errors are printed as an hledger comment. For example, if the ISIN for VADEA were incorrect:

P 2018-05-26 BTC $7536.47
P 2018-05-26 ETH £450.84
P 2018-05-26 LTC £90.95
P 2018-05-26 Euro £0.8752
P 2018-05-26 JPY £0.0069
; error processing commodity 'VADEA': could not find price

So there is no danger (in terms of producing an invalid journal) in having a cronjob to run this script and append the results to a file; you'll just need to manually fix any errors.

aggregate-fundingcircle

Aggregates interest payments and fees from a Funding Circle statement into totals. Downloads market values of commodities from a few different sources.

Dependencies: none.

Usage:

$ ./aggregate-fundingcircle.sh path/to/statement.csv

Output:

interest: 832
fees:     78

The aggregates are in pence.