ohlcobject 1.2

Improve performance for handling OHLC against Pandas

为 OHLC 处理去 pandas 化，从而避免大量的 pandas 内存拷贝，达到性能优化的效果（大概 8000%）。

内部存储为 { "low": [], "high": [] }

可将 ohlc_object[num]['key'] 的调用转化为 ohlc_object['key'][num] 的调用。

init

obj = OHLCObject()
obj['open'] = [1, 2, ...]
obj['high'] = [100, 200, ...]
obj['low'] = [1, 2, ...]
obj['close'] = [100, 200, ...]
obj['timestamp'] = [1498924800000, 1498939200000, ...]
return obj

merge (concat OHLC)

Merge obj2 into obj1, automatically sort using 'timestamp' key

obj1.merge(obj2)

read

obj['low'][1]  # best performance
obj[1]['low']
obj[3:][1]['low']