Release 1.2.1.88ΒΆ

Changing the minor version number from 1 to 2 has taken sometime, but the deprecation of the old DataResampler and DataReplayer have led to it.

The documentation at readthedocs has

The Documentation has been updated to only reference the modern way to do resampling and replaying. It is as easy as:

...
data = backtrader.feeds.BacktraderCSVData(dataname='mydata.csv')  # daily bars
cerebro.resampledata(data, timeframe=backtrader.TimeFrame.Weeks) # to weeks
...

For replaying just change resampledata to replaydata. There are additional ways to do it, but this is the most straightforward interface and probably the only one that will ever get used by anyone.

Following Ticket #60 it was clear that the extension mechanism which allows adding additional lines to data feeds (actually to any lines based object) was not enough to support what was suggested in the ticket.

Hence the implementation of an additional parameter to lines object which allows the complete redefinition of the lines hierarchy (Escape from OHLC Land would be an appropriate film title)

A sample named data-bid-ask has been added to the sources. From the sample:

class BidAskCSV(btfeeds.GenericCSVData):
    linesoverride = True  # discard usual OHLC structure
    # datetime must be present and last
    lines = ('bid', 'ask', 'datetime')
    # datetime (always 1st) and then the desired order for
    params = (
        ('dtformat', '%m/%d/%Y %H:%M:%S'),

        ('datetime', 0),  # field pos 0
        ('bid', 1),  # default field pos 1
        ('ask', 2),  # defult field pos 2
    )

By specifying linesoverride the regular lines inheritance mechanisme is bypassed and the defined lines in the object supersede any previous lines.

The release is available from pypi and can be installed with usual:

pip install backtrader

Or if updating:

pip install backtrader --upgrade

Comments

Fork me on GitHub