finflow-sankey 0.3.0

Polars-first financial statement Sankey visualization library

FinFlow Sankey

Polars-first financial statement Sankey visualization library.

import polars as pl
from finflow_sankey import FinancialSankey

df = pl.DataFrame({
    "account": ["Revenue", "Cost of Revenue", "Operating Expenses", "Tax", "Net Income"],
    "value": [100_000_000.0, -40_000_000.0, -30_000_000.0, -10_000_000.0, 20_000_000.0],
    "period": ["FY2025"] * 5,
    "currency": ["USD"] * 5,
    "statement": ["income_statement"] * 5,
    "section": ["revenue", "cost_of_revenue", "operating_expenses", "tax", "profit"],
})

FinancialSankey.income_statement(df, period="FY2025", currency="USD").validate().export_html(
    "income_statement.html", title="FY2025 Income Statement"
)

Installation

pip install finflow-sankey

Requires Python 3.9 or later.

Features

• Polars-first: Native pl.DataFrame / pl.LazyFrame support. No pandas required.
• Accounting-aware validation: Period/currency checks, reconciliation validation.
• Role-based color palette: Revenue, costs, profit, cash flow each have distinct colors.
• Customizable themes: default, monochrome, colorblind_safe, minimal, dark, plus custom YAML palettes.
• Runtime palette override: Change colors via dict or YAML without modifying source.
• Account mapping: Map raw account names to standard sections via dict or YAML.
• Reference layout (opt-in): Detailed account nodes with fixed layout for presentations.
• HTML export: One-line export helper.
• Multi-period comparison: Compare two periods in a single Sankey.

Input Format

Your DataFrame must contain these columns:

Column	Type	Description
account	str	Account name shown on nodes.
value	float	Signed numeric value. Expenses are typically negative.
period	str	Period label (e.g. FY2025, 2024-12-31).
currency	str	Currency label (e.g. USD, KRW tn).
statement	str	One of income_statement, cash_flow_statement, balance_sheet.
section	str	Standard section. See table below.

Supported Sections

Statement	Sections
Income statement	revenue, cost_of_revenue, cost, expense, operating_expenses, tax, non_operating_items, profit
Cash flow	beginning_cash, operating_cash_flow, investing_cash_flow, financing_cash_flow, fx_effect, ending_cash
Balance sheet	asset, current_asset, non_current_asset, liability, current_liability, non_current_liability, equity

Quick Examples

Income Statement

fig = FinancialSankey.income_statement(df, period="FY2025", currency="USD").validate().render(
    title="FY2025 Income Statement"
)
fig.show()

Reference Layout (Detailed View)

fig = FinancialSankey.income_statement(df, layout="reference").validate().render(
    title="Detailed Income Statement"
)

Cash Flow Statement

fig = FinancialSankey.cash_flow_statement(df_cf, period="FY2025", currency="USD").validate().render(
    title="Cash Flow Bridge"
)

Balance Sheet Reconciliation

fig = FinancialSankey.balance_sheet_reconciliation(df_bs, as_of="2025-12-31", currency="USD").validate().render(
    title="Balance Sheet"
)

Multi-Period Comparison

fig = FinancialSankey.multi_period_compare(df_multi, currency="USD").validate().render(
    title="YoY Comparison"
)

Themes & Palettes

# Built-in theme
fig = FinancialSankey.income_statement(df).validate().render(theme="dark")

# Custom YAML palette
fig = FinancialSankey.income_statement(df).validate().render(palette="./my_palette.yaml")

# Runtime dict override
fig = FinancialSankey.income_statement(df).validate().render(
    palette={"revenue": "#0055FF", "profit": "#00AA55"}
)

Account Mapping

Map raw account names to standard sections when your data uses different labels:

mapping = {
    "revenue": ["Net Sales", "Sales Revenue"],
    "cost_of_revenue": ["COGS", "Cost of Goods Sold"],
    "operating_expenses": ["SG&A", "R&D"],
    "tax": ["Income Tax Expense"],
    "profit": ["Net Income"],
}

fig = FinancialSankey.income_statement(df, mapping=mapping).validate().render()

HTML Export

FinancialSankey.income_statement(df).validate().export_html(
    "income_statement.html", title="FY2025 Income Statement"
)

Data Adapters

CSV

from finflow_sankey.adapters import load_income_statement_csv

df = load_income_statement_csv("data.csv", period="FY2025", currency="USD")
FinancialSankey.income_statement(df).validate().render()

Excel

from finflow_sankey.adapters import load_income_statement_excel

df = load_income_statement_excel("data.xlsx", sheet_name="IS", period="FY2025", currency="USD")
FinancialSankey.income_statement(df).validate().render()

Dartlab CLI

from finflow_sankey.adapters import DartlabAdapter

adapter = DartlabAdapter("005930")
df = adapter.load_income_statement()
FinancialSankey.income_statement(df).validate().render()

DART OpenAPI (dart-fss)

from finflow_sankey.adapters.dart_fss import DartFssAdapter

adapter = DartFssAdapter(corp_code="00126380")
df = adapter.load_income_statement()
FinancialSankey.income_statement(df).validate().render()

Requires DART_API_KEY environment variable and pip install dart-fss.

Real-World Data

See examples/dartlab_integration.py for fetching live Samsung Electronics data via Dartlab CLI.

Documentation

Full documentation is available at https://min9lin9.github.io/finflow-sankey/.

Roadmap

See ROADMAP.md for planned features and the path to v1.0.

Community

• GitHub Discussions for questions and usage help
• Issue Templates for bugs and feature requests
• CONTRIBUTING.md for development guidelines

Development

pip install -e ".[dev]"
python -m pytest tests/ -q
ruff check finflow_sankey tests

Project Structure

finflow_sankey/
  core/           # schema, validation, normalization, graph, palette, mapper
  templates/      # income_statement, cash_flow, balance_sheet, multi_period
  renderers/      # plotly renderer
  palettes/       # YAML color palettes
  mappings/       # YAML account mappings
  examples/       # usage examples
tests/            # pytest suite
docs/             # mkdocs documentation

License

MIT