Reckon automagically converts CSV files for use with the command-line accounting tool Ledger. It also helps you to select the correct accounts associated with the CSV data using Bayesian machine learning.

Assuming you have Ruby and Rubygems installed on your system, simply run

gem install --user reckon

First, login to your bank and export your transaction data as a CSV file.

To see how the CSV parses:

reckon -f bank.csv -p

If your CSV file has a header on the first line, include --contains-header.

To convert to ledger format and label everything, do:

reckon -f bank.csv -o output.dat

To have reckon learn from an existing ledger file, provide it with -l:

reckon -f bank.csv -l 2010.dat -o output.dat

Learn more:

> reckon -h

Usage: Reckon.rb [options]

-f, --file FILE                  The CSV file to parse
-a, --account NAME               The Ledger Account this file is for
-v, --[no-]verbose               Run verbosely
-i, --inverse                    Use the negative of each amount
-p, --print-table                Print out the parsed CSV in table form
-o, --output-file FILE           The ledger file to append to
-l, --learn-from FILE            An existing ledger file to learn accounts from
    --ignore-columns 1,2,5
                                 Columns to ignore, starts from 1
    --money-column 2
                                 Column number of the money column, starts from 1
    --money-columns 2,3
                                 Column number of the money columns, starts from 1 (1 or 2 columns)
    --raw-money
                                 Don't format money column (for stocks)
    --date-column 3
                                 Column number of the date column, starts from 1
    --contains-header [N]
                                 Skip N header rows - default 1
    --csv-separator ','
                                 CSV separator (default ',')
    --comma-separates-cents
                                 Use comma to separate cents ($100,50 vs. $100.50)
    --encoding 'UTF-8'
                                 Specify an encoding for the CSV file
-c, --currency '$'               Currency symbol to use - default $ (ex £, EUR)
    --date-format FORMAT
                                 CSV file date format (see `date` for format)
    --ledger-date-format FORMAT
                                 Ledger date format (see `date` for format)
-u, --unattended                 Don't ask questions and guess all the accounts automatically. Use with --learn-from or --account-tokens options.
-t, --account-tokens FILE        YAML file with manually-assigned tokens for each account (see README)
    --table-output-file FILE
    --default-into-account NAME
                                 Default into account
    --default-outof-account NAME
                                 Default 'out of' account
    --fail-on-unknown-account
                                 Fail on unmatched transactions.
    --suffixed
                                 Append currency symbol as a suffix.
-h, --help                       Show this message
    --version                    Show version

If you find CSV files that it can't parse, send me examples or pull requests!

You can run reckon in a non-interactive mode. To guess the accounts reckon can use an existing ledger file or a token file with keywords.

reckon --unattended -a Checking -l 2010.dat -f bank.csv -o ledger.dat

reckon --unattended -a Checking --account-tokens tokens.yaml -f bank.csv -o ledger.dat

In unattended mode, you can use STDIN to read your csv data, by specifying - as the argument to -f.

csv_file_generator | reckon --unattended -a Checking -l 2010.dat -o ledger.dat -f -

The account tokens file provides a way to teach reckon about what tokens are associated with an account. As an example, this tokens.yaml file:

Expenses:
  Bank:
    - 'ING Direct Deposit'

Would tokenize to 'ING', 'Direct' and 'Deposit'. The matcher would then suggest matches to transactions that included those tokens. (ex 'Chase Direct Deposit')

Here's an example of tokens.yaml:

Income:
  Salary:
    - 'LÖN'
    - 'Salary'
Expenses:
  Bank:
    - 'Comission'
    - 'MasterCard'
  Rent:
    - '0011223344' # Landlord bank number
  Hosting:
    - /hosting/i # This regexp will catch descriptions such as WebHosting or filehosting
'[Internal:Transfer]': # Virtual account
  - '4433221100' # Your own account number

Reckon will use Income:Unknown or Expenses:Unknown if it can't match a transaction to an account.

You can override these names with the --default_outof_account and --default_into_account options.

If, in the above example, you'd prefer to match any transaction that contains the string 'ING Direct Deposit' you have to use a regex:

Expenses:
  Bank:
    - /ING Direct Deposit/

We encourage you to contribute to Reckon! Here is some information to help you.

1. Fork the project.
2. Make your feature addition or bug fix.
3. Add tests for it. This is important so I don't break it in a future version unintentionally.
4. Commit, do not mess with rakefile, version, or history.
   • (if you want to have your own version, that is fine but bump version in a commit by itself I can ignore when I pull)
5. Send me a pull request. Bonus points for topic branches.

Reckon has integration test located in spec/integration. These are integration and regression tests for reckon.

Run all the tests:

./spec/integration/test.sh

Run a single test

./spec/integration/test.sh chase/account_tokens_and_regex

Each test has it's own directory, which you can add any files you want, but the following files are required:

• test_args - arguments to add to the reckon command to test against, can specify --unattended, -f input.csv, etc
• output.ledger - the expected ledger file output

If the result of running reckon with test_args does not match output.ledger, then the test fails.

Most tests will specify --unattended, otherwise reckon prompts for keyboard input.

The convention is to use input.csv as the input file, and tokens.yml as the tokens file, but it is not required.

Copyright (c) 2013 Andrew Cantino (@cantino). See LICENSE for details.

Thanks to @BlackEdder for many contributions!

Currently maintained by @benprew. Thank you!