PairCandleCache

API documentation for tradingstrategy.transport.pair_candle_cache.PairCandleCache Python class in Trading Strategy framework.

class PairCandleCache

Bases: object

Context manager for pair candle cache operations.

Handles loading, updating, and saving of cached candle data and metadata. Designed to be used as a context manager to ensure proper file locking.

Example usage:

with PairCandleCache(cache_path) as cache:

partition = cache.metadata.partition_for_fetch(pair_ids, start_time, end_time) # … perform fetches … cache.update([df1, df2]) # Update cache with new data return cache.data[] # filter as needed

__init__(base_path)

Initialize cache with base file path.

Parameters:

base_path (str) – Absolute path without extension (e.g., “/path/to/candles-1h”). Extensions .parquet, .json, .lock will be appended as needed.

Methods

__init__(base_path)	Initialize cache with base file path.
update(new_dataframes, pair_ids, start_time, ...)	Update cache with new candle data.

Attributes

data	Access to the cached candle DataFrame.
metadata	Access to the cache metadata.

__init__(base_path)

Initialize cache with base file path.

Parameters:

base_path (str) – Absolute path without extension (e.g., “/path/to/candles-1h”). Extensions .parquet, .json, .lock will be appended as needed.

property data: DataFrame

Access to the cached candle DataFrame.

property metadata: PairCandleMetadata

Access to the cache metadata.

update(new_dataframes, pair_ids, start_time, end_time)

Update cache with new candle data.

Concatenates new data with existing, removes duplicates, sorts, and saves both the parquet file and metadata.

Parameters:

• new_dataframes (list[pandas.core.frame.DataFrame]) – List of DataFrames containing new candle data to add to cache.

• pair_ids (Collection[int]) – Trading pairs that were included in the fetch operation.

• start_time (datetime) – Start time used for the fetch operation.

• end_time (datetime) – End time used for the fetch operation.

Return type:

None