forexpy is a python command-line package that can download currency exchange rates from multiple soruces. The package can provide forex data for storage, strategy development, etc.

• Installation
• Data Sources
• Usage
  • Available Currency pairs
  • Available TimeFrames
  • Dukascopy
  • Metatrader5
  • HistData
• Disclaimer
• Contact

The library is currently not available on PyPI. Clone the repo by downloading it as a zip file or run:

git clone https://github.com/zeta-zetra/forexpy.git

Once the folder is on your machine, install the requirements by running:

pip install -r requirements.txt

Change directory into the /forexpy/src/. This is where all commands will be run from.

Here is a list of data sources:

• MetaTrader5

• Dukascopy

• Histdata.com

There is one main function in the library which is fetch.py. The function does all the heavy lifting of calling all the other functions to get the given task accomplished. The fetch.py has options. To get a view of the options, run python fetch.py --help:

Usage: fetch.py [OPTIONS]

Options:
--source TEXT  Source for the forex data
--symbol TEXT  Currency pair to download
--start TEXT   Start date to download the data
--end TEXT     End date to download the data
--tf TEXT      Timeframe to download
--n INTEGER    Number of candles to download
--output TEXT  Type of output for the data
--path TEXT    Full Path to save the data
--keep TEXT    Keep the .bi5 from Dukascopy or the zip files from HistData.
                To keep the files set: --keep T
--help         Show this message and exit.

When you run python info.py, you will get information about available source, pairs, etc. Here is a sample output:

> Here are the available sources:
    1. dukas
    2. metatrader
    3. hist

> Here are the available pairs:
    1. AUDUSD
    2. EURUSD
    3. GBPUSD
        .
        .
        .

> Here are the available timeframes:
    1. tick
    2. 1m
    3. 5m
       .
       .
       . 

> Here are the available output types:
    1. csv

• AUDUSD

• EURUSD

• GBPUSD

• NZDUSD

• USDCAD

• USDCHF

• USDJPY

• Tick

• 1 Min

• 5 Min

• 15 Min

• 30 Min

• 1 Hour

• 4 Hour

• 1 Day

Note: To arrive at some timeframes, 1 minute data is downloaded and resampled to the desired timeframe.

By default if no source is provided, the data will be fetched from Dukascopy.

When you run the code below:

python fetch.py 

you are asking data for EURUSD tick from Dukascopy for today. No data will be downloaded.. Dukascopy does not provide real-time data, try Metatrader5.

For Dukascopy make sure your start and/or end dates are at least a month before the current date. This will download EURUSD tick data for 2022-10-02 only.

python fetch.py --start 20221002  

You can download data between a range of dates. This will download EURUSD tick between 2022-10-02 and 2022-11-10

python fetch.py --start 20221002 --end 20221110  

You can download for different timeframes as well:

1. 1 Min

    python fetch.py --start 20221002 --tf 1m 
   

2. 5 Min

    python fetch.py --start 20221002 --end 20221015 --tf 5m
   

3. 15 Min

    python fetch.py --start 20221102 --end 20221115 --tf 15m
   

4. 30 Min

    python fetch.py --start 20221002 --end 20221015 --tf 30m
   

5. 1 Hour

    python fetch.py --symbol USDJPY --start 20221002 --end 20221015 --tf 1h
   

6. 1 Day

    python fetch.py --symbol GBPUSD  --start 20221002 --end 20221015 --tf 1d
   

When downloading data from Dukascopy, .bi5 files are saved in a folder and then used to create the final csv file. The --keep argument is used to tell forexpy if these files should be deleted or not. By default there are deleted. To keep them, run:

python fetch.py --start 20221002 --end 20221015 --tf 1d --keep T 

If you have a folder of .bi5 files then you can use decompressor.py to decompress them. Simply run:

python decompressor.py --path XXXX 

You can also state where the final output should be saved:

python decompressor.py --path XXXX --save XXXX

To source data from your broker on Metatrader5, make sure Metatrader5 is installed.

python fetch.py --source metatrader 

The above code will fetch the latest 10 EURUSD tick data for today.

To get the latest 100 tick data for a symbol, you run:

python fetch.py --source metatrader --symbol EURUSD --n 100

Change --n according to your needs. Be mindful that there is a limit.

Date ranges and different timeframes are supported for MetaTrader. See the Dukascopy section for examples. Just add the --source metatrader argument.

You can source data from HistData, by running:

python fetch.py --source hist 

The above code will fetch tick EURUSD for the current month.

Date ranges and different timeframes are supported for HistData. See the Dukascopy section for examples. Just add the --source hist argument.

When data is downloaded from HistData, a number of zip files need to be saved. By default forexpy will delete these zip files after the final output. You may want to keep them. Run the following:

python fetch.py --source hist --keep T

I have no affiliation with the data sources used in the package.

If you want to contact me: [email protected]