tomp736 / pyfi

python/html app for transaction review

PYFI

Projects

pyfi_core

• Common classes for building transactionviews using transactions, filter rules, and transform rules.

pyfi_server

• Sample app using flask and charts.js.
• Requires datasource.json and views.json config files under /srv/pyfi/etc/.

	datasource.json	views.json
Description	Configuration for transaction data sources.	Configuration for transaction views.
Example	LINK	LINK

Usage:

## Setup
transaction = <[transaction]>
start_date = <datetime>
end_date = <datetime>
time_delta_d = <int>
transaction_view_filters = [
    RegexFieldFilterStrategy('counterpart_data', ".*UBER.*")
]

# exchange rate provider to normalize amounts to a common currency
# linear fit works best with free api access
exchange_rate_provider = LinearFitOpenExchangeRatesProvider(
    date(2021, 1, 1),
    date(2023, 2, 1),
    timedelta(days=60)
)
transaction_transform_currency = CurrencyTranformStrategy("PLN", exchange_rate_provider)

## Extract View
transaction_view_builder = TransactionViewBuilder(transactions)
transaction_view_builder.set_duration(start_date, end_date, timedelta(days=time_delta_d))
transaction_view_builder.add_transform(transaction_transform_currency)
transaction_view_builder.add_filters(transaction_view_filters)
transaction_view_builder_views = transaction_view_builder.get_views()

Web Interface

Chart and controls.

Dataview Editor.

Progress

pyfi_core - Exchange Rates

• Hardcoded implementation
• OpenExchangeRatesProvider (high resulution, api calls grow with data, duplicate calls if same data requested)
• CachedOpenExchangeRatesProvider (high resulution, api calls grow with data up to data resolution)
• LinearFitOpenExchangeRatesProvider (low resolution, api calls fixed to resolution requested using start_date, end_date, timedelta(days))
• UnitTests

pyfi_core - DataSource

• ING CSV
• UnitTests
• Import/Export to pyfi native datastore
• ??

Frontend

• Chart.js Bars
• UI Labels for Inputs
• Stacked Bar
• Cleanup Html