Database

Classes

QuantmodDB

Python SDK for registering instruments, ingesting OHLCV price data, and retrieving historical market data from Supabase.

Assumptions: - Supabase tables instruments and prices already exist - UNIQUE constraint on instruments.symbol - UNIQUE constraint on prices (instrument_id, date) - RLS disabled OR service-role key is used - getData() returns DataFrame indexed by datetime with Open, High, Low, Close, Volume columns

Functions

get_asset_prices

get_asset_prices(symbols: Optional[Union[str, List[str]]] = None, start_date: Optional[str] = None, end_date: Optional[str] = None, column: str = 'close') -> pd.DataFrame

Retrieve OHLCV data in wide format (symbols as columns).

Parameters:

Name	Type	Description	Default
`symbols`	`str \| list[str]`	Filter by symbol(s)	`None`
`start_date`	`str`	Start date filter (YYYY-MM-DD)	`None`
`end_date`	`str`	End date filter (YYYY-MM-DD)	`None`
`column`	`str`	Price column to pivot ('open', 'high', 'low', 'close', 'volume')	`'close'`

Returns:

Type	Description
`DataFrame`	Index: date, Columns: symbols

get_instrument_id

get_instrument_id(symbol: str) -> Optional[int]

Get instrument ID for a symbol.

Parameters:

Name	Type	Description	Default
`symbol`	`str`	Instrument symbol	required

Returns:

Type	Description
`int \| None`	Instrument ID if found, None otherwise

get_latest_prices

get_latest_prices(symbols: Optional[Union[str, List[str]]] = None) -> pd.DataFrame

Get the most recent price for each symbol.

Parameters:

Name	Type	Description	Default
`symbols`	`str \| list[str]`	Filter by symbol(s)	`None`

Returns:

Type	Description
`DataFrame`	Latest price record for each symbol

get_prices

get_prices(symbols: Optional[Union[str, List[str]]] = None, start_date: Optional[str] = None, end_date: Optional[str] = None, limit: Optional[int] = None) -> pd.DataFrame

Retrieve OHLCV data in long format.

Parameters:

Name	Type	Description	Default
`symbols`	`str \| list[str]`	Filter by symbol(s). If None, returns all.	`None`
`start_date`	`str`	Start date filter (YYYY-MM-DD)	`None`
`end_date`	`str`	End date filter (YYYY-MM-DD)	`None`
`limit`	`int`	Maximum number of records to return	`None`

Returns:

Type	Description
`DataFrame`	Columns: date, open, high, low, close, volume, symbol Sorted by date ascending

list_instruments

list_instruments(symbols: Optional[Union[str, List[str]]] = None, exchange: Optional[str] = None, asset_class: Optional[str] = None) -> pd.DataFrame

List registered instruments with optional filters.

Parameters:

Name	Type	Description	Default
`symbols`	`str \| list[str]`	Filter by specific symbol(s)	`None`
`exchange`	`str`	Filter by exchange	`None`
`asset_class`	`str`	Filter by asset class	`None`

Returns:

Type	Description
`DataFrame`	Registered instruments with metadata

load_history

load_history(symbols: Union[str, List[str]], start_date: str, end_date: Optional[str] = None) -> Dict[str, int]

Load historical OHLCV data for one or multiple symbols.

All symbols MUST be registered before loading data. Use register() first to add instruments with metadata.

Parameters:

Name	Type	Description	Default
`symbols`	`str \| list[str]`	Symbol(s) to load data for (must be pre-registered)	required
`start_date`	`str`	Start date in YYYY-MM-DD format	required
`end_date`	`str`	End date in YYYY-MM-DD format (defaults to today)	`None`

Returns:

Type	Description
`dict`	{symbol: num_records_loaded}

Raises:

Type	Description
`ValueError`	If any symbol is not registered

register

register(instruments: Union[dict, List[dict]]) -> List[int]

Register one or multiple instruments with metadata. Idempotent: existing symbols are updated with new metadata.

Parameters:

Name	Type	Description	Default
`instruments`	`dict \| list[dict]`	Instrument(s) to register. Each dict must contain 'symbol'. Optional keys: name, exchange, asset_class, instrument_type	required

Returns:

Type	Description
`list[int]`	Instrument IDs (existing or newly created)

Raises:

Type	Description
`ValueError`	If instrument dict is missing 'symbol' key