Currently, the Starling Bank mobile app does not have a easy-to-read overview of spending amounts for each day and each month (as shown below - it requires manually selecting each month in the past), which this Python project attempts to solve.
This script uses the Starling Developer API to get the transactions between 2 specified dates for your account, then filter the outbound transactions and group them by day and month with Pandas.
The format of the API response data is as below:
{
"feedItems": [
{
"feedItemUid": "11221122-1122-1122-1122-112211221122",
"categoryUid": "ccddccdd-ccdd-ccdd-ccdd-ccddccddccdd",
"amount": {
"currency": "GBP",
"minorUnits": 123456
},
"sourceAmount": {
"currency": "GBP",
"minorUnits": 123456
},
"direction": "IN",
"updatedAt": "2020-05-02T11:08:45.089Z",
"transactionTime": "2020-05-02T11:08:45.089Z",
"settlementTime": "2020-05-02T11:08:45.089Z",
"retryAllocationUntilTime": "2020-05-02T11:08:45.089Z",
"source": "MASTER_CARD",
"sourceSubType": "CONTACTLESS",
"status": "PENDING",
"transactingApplicationUserUid": "aaaaaaaa-aaaa-4aaa-aaaa-aaaaaaaaaaaa",
"counterPartyType": "MERCHANT",
"counterPartyUid": "68e16af4-c2c3-413b-bf93-1056b90097fa",
"counterPartyName": "Tesco",
"counterPartySubEntityUid": "35d46207-d90e-483c-a40a-128cc4da4bee",
"counterPartySubEntityName": "Tesco Southampton",
"counterPartySubEntityIdentifier": "608371",
"counterPartySubEntitySubIdentifier": "01234567",
"exchangeRate": 0,
"totalFees": 0,
"reference": "TESCO-STORES-6148 SOUTHAMPTON GBR",
"country": "GB",
"spendingCategory": "GROCERIES",
"userNote": "Tax deductable, submit me to payroll",
"roundUp": {
"goalCategoryUid": "68e16af4-c2c3-413b-bf93-1056b90097fa",
"amount": {
"currency": "GBP",
"minorUnits": 123456
}
},
"hasAttachment": true
}
]
}
The graphs for transactions for each day and each month are then shown with matplotlib (as a line graph and bar chart respectively).
The exact amount is displayed when hovering, and the largest 5 transactions are displayed when clicking the line graph or bar chart.
Add "StarlingPersonalAccessToken" as an environmental variable, after creating the personal access token with read permissions for your account at https://developer.starlingbank.com/personal/token
In the config.ini file, set the transactions_start_date to the date you want to start showing data from.
Assuming you have installed Python 3,
in the root project folder run $pip install -r requirements.txt, then $python spending_graph.py.
If you have made some purchases for multiple people and they paid you back later, and want to divide the amount to be more accurate to what you spent, run $python spending_list.py. This will create an Excel file outbound_transactions.xlsx which lists all outbound transactions made.
After identifying the feedItemUid of the transaction to divide, add it to feed_items_to_divide in the config.ini file, along with the number to divide by. Similarly, to ignore transactions completely, add the id to feed_items_to_ignore.