dart-enhanced

Async Python client for Korea's DART (FSS) OpenAPI

An async Python client for the DART (Electronic Disclosure System) OpenAPI. Supports financial statement parsing, company overview, disclosure search, and dividend lookup.

Features

• Async-first — Async HTTP client built on httpx
• Zero pandas dependency — Pure Python dataclasses, no numpy required
• Financial statement parser — Converts raw financial data into structured ParsedFinancials
• Auto ratios — Automatically calculates ROE, ROA, debt ratio, operating margin, FCF, and more
• Fully typed — py.typed marker with type hints on all public APIs
• Corp code cache — In-memory caching of corp code mappings to avoid redundant lookups

Installation

pip install dart-enhanced

Quick Start

import asyncio
from dart_enhanced import DartClient, DartApiError, parse_financial_statements, financials_to_dict

async def main():
    async with DartClient(api_key="YOUR_DART_API_KEY") as client:
        # Company overview for 삼성전자
        info = await client.get_company_info_by_stock("005930")
        print(f"{info.corp_name} ({info.ceo})")
        # → 삼성전자 (한종희)

        # Fetch and parse financial statements
        items = await client.get_financial_statements_by_stock(
            "005930", year=2024, report_type="annual"
        )
        parsed = parse_financial_statements(items)
        print(f"Revenue: {parsed.revenue:,} KRW")
        print(f"ROE: {parsed.roe:.2%}")
        print(f"Debt ratio: {parsed.debt_ratio:.2%}")

        # Convert to dict for API responses
        data = financials_to_dict(parsed)
        print(data["ratios"])

asyncio.run(main())

Error Handling

from dart_enhanced import DartClient, DartApiError

async with DartClient() as client:
    try:
        info = await client.get_company_info_by_stock("999999")
    except ValueError as e:
        print(f"Stock code not found: {e}")
    except DartApiError as e:
        print(f"DART API error: {e} (status={e.status_code})")

DartApiError.status_code contains the DART API error code:

• "013" — No data found
• "020" — Request limit exceeded
• "800" — System maintenance in progress

API Reference

DartClient

client = DartClient(api_key="...")  # or use env var DART_API_KEY

Method	Description
load_corp_codes()	Load full corp code mapping (stock code → corp_code)
get_corp_code(stock_code)	Look up DART corp_code (8 digits) by stock code (6 digits)
get_company_info(corp_code)	Get company overview → CompanyInfo
get_company_info_by_stock(stock_code)	Get company overview by stock code
get_financial_statements(corp_code, year, report_type, consolidated)	Get financial statements → list[FinancialItem]
get_financial_statements_by_stock(stock_code, ...)	Get financial statements by stock code
search_disclosures(corp_code, begin_date, end_date, ...)	Search disclosures
get_dividend_info(corp_code, year, report_type)	Get dividend information
close()	Close the HTTP client (called automatically when using context manager)

Stock code vs corp_code: DART uses its own corporate code (corp_code, 8 digits), which differs from the stock exchange ticker (e.g., 005930). The _by_stock methods handle the stock code → corp_code conversion automatically.

CompanyInfo

Return type for get_company_info():

Field	Type	Description
corp_code	str	DART unique corporate code (8 digits)
corp_name	str	Company name
stock_code	str | None	Stock code (None if unlisted)
stock_name	str | None	Stock name
ceo	str	CEO
corp_cls	str	Y=KOSPI, K=KOSDAQ, N=KONEX, E=Other
address	str	Headquarters address
website	str	Website URL
industry_code	str	Industry code
established	str	Date of establishment (YYYYMMDD)
fiscal_month	str	Fiscal year-end month

FinancialItem

Return type for get_financial_statements() (list):

Field	Type	Description
account_id	str	Account ID
account_name	str	Account name (e.g., "매출액", "자산총계")
current_amount	int	Current period amount
previous_amount	int	Previous period amount
before_previous_amount	int	Period before previous amount
statement_type	str	BS=Balance Sheet, IS=Income Statement, CF=Cash Flow Statement, SCE=Statement of Changes in Equity

Financial Statement Parser

from dart_enhanced import parse_financial_statements, financials_to_dict, ParsedFinancials

parsed = parse_financial_statements(items)  # → ParsedFinancials

Balance Sheet (BS)

Property	Type	Description
total_assets	int	Total assets
total_liabilities	int	Total liabilities
total_equity	int	Total equity
current_assets	int	Current assets
current_liabilities	int	Current liabilities

Income Statement (IS)

Property	Type	Description
revenue	int	Revenue
operating_income	int	Operating income
net_income	int	Net income

Cash Flow Statement (CF)

Property	Type	Description
operating_cash_flow	int	Cash flow from operations
investing_cash_flow	int	Cash flow from investing
financing_cash_flow	int	Cash flow from financing
fcf	int	Free cash flow (operating CF + investing CF, auto-calculated)

Auto-calculated Ratios

Property	Type	Description
debt_ratio	float | None	Debt ratio (total liabilities / total equity)
current_ratio	float | None	Current ratio (current assets / current liabilities)
operating_margin	float | None	Operating margin (operating income / revenue)
net_margin	float | None	Net margin (net income / revenue)
roe	float | None	Return on equity (net income / total equity)
roa	float | None	Return on assets (net income / total assets)
revenue_growth	float | None	Revenue growth rate (vs. previous period)
net_income_growth	float | None	Net income growth rate (vs. previous period)

Ratio properties return None when the denominator is zero or negative.

data = financials_to_dict(parsed)  # Convert to dict for API responses
# → {"balance_sheet": {...}, "income_statement": {...}, "cash_flow": {...}, "ratios": {...}, "growth": {...}}

Report Types

Key	Description
"annual"	Annual report
"half"	Semi-annual report
"q1"	Q1 report
"q3"	Q3 report

Consolidated vs Separate

The consolidated parameter of get_financial_statements():

Value	Description
True (default)	Consolidated financial statements (includes subsidiaries)
False	Separate financial statements (parent entity only)

Consolidated statements (True) are recommended for most analyses. Separate statements are used in special cases such as holding company analysis.

Important Notes

Financial Statement Parser Limitations

• The parser works by matching Korean account names from DART (e.g., "매출액", "자산총계").
• It works correctly for most listed companies, but some companies using non-standard account names may have missing fields.
• Financial sector companies (banks, insurance, securities) have different financial statement structures, and the parser may not recognize some items.

Async Only

This library is async/await only. To use it from synchronous code:

import asyncio
result = asyncio.run(main())

Environment Variables

Variable	Description
DART_API_KEY	DART OpenAPI authentication key (Get one here)

Comparison with Alternatives

Feature	dart-enhanced	opendartreader	dart-fss
Async	O	X	X
pandas-free	O	X	X
Type hints	O	Partial	Partial
Financial parser	O	X	O
Auto ratios (ROE, etc.)	O	X	X
Lightweight	O	X	X

DART API Key

1. Go to DART OpenAPI
2. Sign up and apply for an authentication key
3. Set the issued key as the DART_API_KEY environment variable or pass it to DartClient(api_key="...")

Disclaimer

This library is a technical tool for querying the DART OpenAPI and does not provide investment advice or financial services. Users are responsible for complying with the DART OpenAPI Terms of Use and applicable laws.

License

MIT License. See LICENSE.