Note. Both use JSON format, so they are very compatible.

Environment

From here for the time being. Omitting the environment construction of python

Install MongoDB on Windows

Don't forget to put the bin in your PATH. (I forgot) http://qiita.com/yoh-nak/items/f0c429f10347ae7ec98b http://qiita.com/t-koyama/items/9b8804cbec59b3c93eb0

Install pymongo

pip install pymongo

Install oandapy (python wrapper for oanda REST API)

pip install oandapy

Instructions

MongoDB Primer

http://qiita.com/saba1024/items/f2ad56f2a3ba7aaf8521 http://www.cuspy.org/diary/2012-04-17/the-little-mongodb-book-ja.pdf

How to use Oanda API

http://developer.oanda.com/docs/jp/ By the way, I haven't tried using the sandbox environment, so I recommend using the fxTrade Practice environment from the beginning.

How to use pymongo

http://api.mongodb.com/python/

How to use oandapy

https://github.com/oanda/oandapy

Code example

Hit the API with oandapy

import oandapy

oanda = oandapy.API(environment="practice", access_token="your_token")
response = oanda.get_history(instrument="EUR_USD",granularity="D",count=500)
EUR_USD_D = response.get("candles")

With this alone, the past daily euro-dollar data (500 steps) can be obtained. You can also specify start and end.

Put data in MongoDB

from pymongo import MongoClient

client = MongoClient('localhost',27017)
db = client.ex_rate
collection = db.eur_usd_d
collection.insert_many(d for d in EUR_USD_D)

With this alone, the daily data (500 steps) acquired above in JSON format can be stored in the collection (table) [eur_usd_d] of the MongoDB database [ex_rate]. Even if the contents of MongoDB are empty, they create databases and collections without permission.

Get a list of available stocks

import oandapy
import pandas as pd

oanda = oandapy.API(environment="practice", access_token="your_token")

Enter the account id for #account_id (not the login id) response = oanda.get_instruments(account_id="xxxxxxx") insts = response.get("instruments") #In pandas for easy viewing df = pd.DataFrame(list(insts)) df.head()

↓ Output

displayName	instrument	maxTradeUnits	pip
0	AUD/CAD	AUD_CAD	10000000	0.0001
1	AUD/CHF	AUD_CHF	10000000	0.0001
2	AUD/HKD	AUD_HKD	10000000	0.0001
3	AUD/JPY	AUD_JPY	10000000	0.01
4	AUD/NZD	AUD_NZD	10000000	0.0001

If you want to get more than 5000 steps of data

You can get it continuously even if the number you want to get exceeds the upper limit of 5000 by the following procedure. There may be a smarter way. -Get the latest data group and get the top (oldest) "time" -Reformat the "time" data and issue a request again as "end"

import oandapy
import datetime as dt

oanda = oandapy.API(environment="practice", 
access_token="your_token")

response1 = oanda.get_history(instrument="EUR_USD",granularity="D",count=2)
EUR_USD_H1 = response1.get("candles")

#Get the time of the first data and convert RFC3339 format to python datetime format dtime = dt.datetime.strptime(EUR_USD_H1[0]['time'],'%Y-%m-%dT%H:%M:%S.%fZ') #Convert to RFC3339 format again rfc_endtime = dtime.isoformat('T')

response2 = oanda.get_history(instrument="EUR_USD",granularity="D",end=rfc_endtime,count=2)
EUR_USD_H2 = response2.get("candles")

for d in EUR_USD_H2:
    print(d)
for d in EUR_USD_H1:
    print(d)