A command line tool for transaction tracking and budgeting.
Ideas are taken from beancount and double entry accounting method.
NOTE: Currently only CSVs exported from Commonwealth Bank of Australia are supported.
- Python 3.6+;
poetry;- database back-end (e.g. Postgres)
$ git clone git@github.com:john5f35/account-book-cli.git
$ cd account-book-cli
$ poetry install
First set up a config file:
SQLite:
{
"db": {
"provider": "sqlite",
"filename": "accbook-sqlite.db",
"create_db": true
}
}Postgres:
{
"db": {
"provider": "postgres",
"database": "accbook",
"user": "johnz",
"host": "localhost",
"port": 5432
}
}See PonyORM Doc - Supported databases for supported database and relevant arguments in config file.
SQLite DB file can be created anew; other DBs will need the DB setup and running for abcli to connect to it.
$ poetry shell
$ abcli
Usage: abcli [OPTIONS] COMMAND [ARGS]...
Options:
--log-level [CRITICAL|FATAL|ERROR|WARN|WARNING|INFO|DEBUG|NOTSET]
Set the root logger level
-c, --config PATH Path to config JSON file
--help Show this message and exit.
Commands:
account
balance
budget
csv
transaction
Use ablic csv prep to reformat the CSV for classification.
$ abcli csv prep --help
Usage: abcli csv prep [OPTIONS] CSVPATH
Options:
-a, --this TEXT The operating account name [required]
--help Show this message and exit.
The prepared CSV will have the column format:
| field | description |
|---|---|
date_occurred |
The date on which the transaction occurred (enter into pending state) |
date_resolved |
The date on which the transaction resolved (pending -> resolved) |
amount |
The amount of the transfer |
description |
The description from the bank |
balance |
Balance in the operating account after the transaction |
this |
The account from which the transaction originates; should just be the operating account |
that_auto |
The automatically classified destination (category) account of the transaction |
that_overwrite |
Overriding the automated classification; can be the name of the account, or a split dictionary for transaction split |
ref |
A @ prefixed reference tag to merge relevant transactions in later processing |
Use abcli csv classify to classify prepared transaction CSVs according to a rulebook:
$ abcli csv classify --help
Usage: abcli csv classify [OPTIONS] CSVPATH
Options:
-r, --rulebook FILE Rule book JSON file for assigning accounts/categories;
default can be specified in config.json
--help Show this message and exit.
Both fields be customised into Python dictionary strings, so to split the transaction into finer postings. This can help dealing with aggregated transactions, allowing them to be tracked separately.
E.g.:
date_occurred,date_resolved,amount,description,balance,this,that_auto,that_overwrite,ref
23/06/2019,26/06/2019,-52.50,Group Dine Out,...,Assets:Bank:Commonwealth:Checking,,"{'Expenses:Food&Drink:Restaurant': 15.00, 'Expenses:Misc:Lent': 37.50}",@dinout23jun
The group dine out transaction of $52.50 is split into personal spending of $15.00 in Expenses:Food&Drink:Restaurant category account, and "lent" spending of $37.50.
The reference tag @dinout23jun is attached to this and other pay-back transactions to resolve the $37.50 lent money.
A yaml file that allow keyword or regex matching on transaction description.
Keywords are searched in top-down, first-match fashion.
The matched value will be written to the transaction's that_auto field.
A rule can also override this field in the CSV as well:
keyword:
Dinner: "Expenses:Food&Drink:Restaurants"
food reimb:
this: "Income:Reimbursements"
that_auto: "Expenses:Misc:Lent"
regex:
.*(Comm|Net)Bank.*: "Assets:Bank:Savings"Once abcli csv classify says it's 100% classified, you can run abcli transaction import to import the classified csv file into your database.
Then you can query it using the summary and show command:
$ abcli transaction summary -m 04/2018 -d 2
account amount % of parent
------------------- --------- -------------
Income -$5432.00
├── Work1 -$3521.00 64.82%
└── Work2 -$1911.00 35.18%
Expenses $3681.26
├── Food&Drink $700.39 19.03%
├── Charity $92.00 2.50%
├── Transport $78.85 2.14%
├── Personal $151.86 4.13%
├── Bills&Utilities $128.91 3.50%
├── Misc $83.39 2.27%
├── Housing $640.00 17.39%
├── Government $1595.00 43.33%
├── Healthcare $72.00 1.96%
└── Hobbies $138.86 3.77%
You can write up a budget in YAML format, and run abcli budget progress to get the progress report on the budget:
$ cat testbudget.yaml
date_from: 01/04/2018
date_to: 30/04/2018
items:
'Expenses:Food&Drink': 4000
'Expenses:Bills&Utilities': 130
'Expenses:Transport': 60
'Expenses:Bills&Utilities:Phone': 120
'Expenses:Bills&Utilities:Internet': 20
$ abcli budget progress testbudget.yaml
account_name budgeted % of parent consumed progress
------------------- ---------- ------------- ---------- ----------
Income
Expenses $4330.00 $3681.26 85.02%
├── Food&Drink $4000.00 92.38% $700.39 17.51%
├── Bills&Utilities $270.00 6.24% $128.91 47.74%
│ ├── Phone $120.00 44.44% $120.51 100.43%
│ └── Internet $20.00 7.41% $20.00 100.00%
└── Transport $60.00 1.39% $78.85 131.42%
Assets
Liabilities
As the accounts are presented in trees, sub-categories are summed at parent level.
Pull requests are welcome. For major changes, please open an issue first to discuss what you would like to change.
Please make sure to update tests as appropriate.