SierraChartUtilities/DataConvertionUtilities.py at main · john-yan/SierraChartUtilities · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125

import pandas as pd
import numpy as np
import struct
import random

def SCIDReaderGen(scid):
    s_IntradayHeader = '=4sIIHHI36s'
    s_IntradayRecord = '=qffffIIII'
    header = scid.read(struct.calcsize(s_IntradayHeader))
    yield header

    while True:
        buf = scid.read(struct.calcsize(s_IntradayRecord))
        if not buf or len(buf) != struct.calcsize(s_IntradayRecord):
            break
        yield struct.unpack(s_IntradayRecord, buf)

def ConvertSCIDToDF(scid_file, datetime_is_seconds=True):
    reader = SCIDReaderGen(scid_file)

    # skip the header
    next(reader)

    raw = pd.DataFrame(reader, columns=[
        'StartDateTime',
        'OpenPrice',
        'HighPrice',
        'LowPrice',
        'LastPrice',
        'NumTrades',
        'Volume',
        'BidVolume',
        'AskVolume'
    ])

    # convert to unix timestamp
    raw.StartDateTime = raw.StartDateTime - (25569 * 86400 * 1000000)

    if datetime_is_seconds:
        raw.StartDateTime = (raw.StartDateTime / 1000000).astype('int')

    return raw

def SCDepthReaderGen(depth):
    s_MarketDepthFileHeader = '=4I48s'
    s_MarketDepthFileRecord = '=qbbhfII'
    header = depth.read(struct.calcsize(s_MarketDepthFileHeader))

    yield struct.unpack(s_MarketDepthFileHeader, header)

    while True:
        record = depth.read(struct.calcsize(s_MarketDepthFileRecord))
        if not record or len(record) != struct.calcsize(s_MarketDepthFileRecord):
            break
        yield struct.unpack(s_MarketDepthFileRecord, record)

def ConvertSCDepthToDF(depth_file):
    reader = SCDepthReaderGen(depth_file)

    # skip the header
    next(reader)

    raw = pd.DataFrame(reader, columns=[
        'DateTime',
        'Command',
        'Flags',
        'NumOrders',
        'Price',
        'Quantity',
        'Reserved'
    ])

    # convert to unix timestamp
    raw.DateTime = raw.DateTime - (25569 * 86400) * 1000_000
    raw.drop(columns=['Reserved'], inplace=True)

    return raw

def ConvertRaw2Tick(raw_df):
    raw_df['AtBidOrAsk'] = (raw_df.Volume == raw_df.AskVolume).astype(np.int32) + 1
    check = (raw_df.Volume != raw_df.AskVolume) & (raw_df.Volume != raw_df.BidVolume)
    if check.any() == True:
        print('Correcting these errors')
        print(raw_df[check])
        for i in raw_df[check].index:
            if raw_df.iloc[i].HighPrice != raw_df.iloc[i].LowPrice:
                at_ask = abs(raw_df.iloc[i].LastPrice - raw_df.iloc[i].HighPrice) > abs(raw_df.iloc[i].LastPrice - raw_df.iloc[i].LowPrice)
                raw_df.at[i, 'AtBidOrAsk'] = int(at_ask) + 1
            else:
                raw_df.at[i, 'AtBidOrAsk'] = random.choice([1, 2])

        print('After corrections')
        print(raw_df[check])

    return pd.DataFrame({
        'DateTime': raw_df.StartDateTime,
        'Price': raw_df.LastPrice,
        'Volume': raw_df.Volume,
        'AtBidOrAsk': raw_df.AtBidOrAsk
    })

def ConvertTick2OHLC(ticks, period=30*60):
    ticks = pd.DataFrame(ticks.reset_index(drop=True))
    periods = (ticks.DateTime / period).astype(np.int32)
    diff = periods.diff()
    diff[0] = 1
    start_index = np.array(ticks[diff != 0].index)
    end_index = np.roll(start_index, -1)
    end_index[-1] = len(ticks)
    new_index = np.arange(0, len(start_index), 1)
    new_index = np.repeat(new_index, end_index - start_index)
    ticks['new_index'] = new_index
    group = ticks.groupby(['new_index'])
    result = pd.DataFrame({
        'StartDateTime': group.DateTime.first(),
        'EndDateTime': group.DateTime.last(),
        'Open': group.Price.first(),
        'High': group.Price.max(),
        'Low': group.Price.min(),
        'Close': group.Price.last(),
        'Volume': group.Volume.sum()
    })

    return result