pyEX 0.5.0

Rest API to IEX

Python interface to IEX Cloud

Referral

Please subscribe to IEX Cloud using my referral code.

Getting Started

Install

Install from pip

pip install pyEX

of from source

python setup.py install

Extensions

• pyEX[async]: asyncio integration for streaming APIs
• pyEX[studies]: Technical indicators and other calculations

Demos + Docs

• Demo Notebook - IEX Cloud
• Streaming Notebook - IEX Cloud
• Read The Docs!

Overview

pyEX supports the IEX Cloud api through 2 interfaces. The first is a simple function call, passing in the api version and token as arguments

In [1]: import pyEX as p

In [2]: p.chart?
Signature: p.chart(symbol, timeframe='1m', date=None, token='', version='', filter='')
Docstring:
Historical price/volume data, daily and intraday

https://iexcloud.io/docs/api/#historical-prices
Data Schedule
1d: -9:30-4pm ET Mon-Fri on regular market trading days
    -9:30-1pm ET on early close trading days
All others:
    -Prior trading day available after 4am ET Tue-Sat

Args:
    symbol (str); Ticker to request
    timeframe (str); Timeframe to request e.g. 1m
    date (datetime): date, if requesting intraday
    token (str); Access token
    version (str); API version
    filter (str); filters: https://iexcloud.io/docs/api/#filter-results

Returns:
    dict: result

For most calls, there is a convenience method that returns a dataframe as well:

In [5]: [_ for _ in dir(p) if _.endswith('DF')]
Out[5]:
['advancedStatsDF',
 'auctionDF',
 'balanceSheetDF',
 'batchDF',
 'bookDF',
 'bulkBatchDF',
 'bulkMinuteBarsDF',
 'calendarDF',
...

Since the token rarely changes, we have a Client object for convenience:

In [6]: p.Client?
Init signature: p.Client(api_token=None, version='v1', api_limit=5)
Docstring:
IEX Cloud Client

Client has access to all methods provided as standalone, but in an authenticated way

Args:
    api_token (str): api token (can pickup from IEX_TOKEN environment variable)
    version (str): api version to use (defaults to v1)
                      set version to 'sandbox' to run against the IEX sandbox
    api_limit (int): cache calls in this interval
File:           ~/Programs/projects/iex/pyEX/pyEX/client.py
Type:           type
Subclasses:

The client will automatically pick up the API key from the environment variable IEX_TOKEN, or it can be passed as an argument. To use the IEX Cloud test environment, simple set version='sandbox'.

In [8]: c = p.Client(version='sandbox')

In [9]: c.chartDF('AAPL').head()
Out[9]:
              open   close    high     low    volume   uOpen  uClose   uHigh    uLow   uVolume  change  changePercent   label  changeOverTime
date
2019-11-27  271.31  274.04  277.09  268.75  16994433  267.69  271.99  271.82  266.32  16811747    0.00         0.0000  Nov 27        0.000000
2019-11-29  271.30  272.19  280.00  279.20  12135259  270.90  275.02  270.00  267.10  11927464   -0.60        -0.2255  Nov 29       -0.002232
2019-12-02  279.96  265.23  276.41  267.93  23831255  279.97  266.80  281.32  269.29  24607845   -3.20        -1.1646   Dec 2       -0.013820
2019-12-03  261.54  271.05  259.96  262.09  30331487  259.87  271.34  269.02  260.71  30518449   -4.93        -1.8450   Dec 3       -0.032745
2019-12-04  272.81  273.56  271.26  267.06  17109161  267.30  262.82  274.99  270.83  17230517    2.39         0.8955   Dec 4       -0.023411

Improvements over native API, other libraries, etc

• pyEX will transparently cache requests according to the refresh interval as defined on the IEX Cloud website (and in the docstrings), to avoid wasting credits. It can also cache to disk, or integrate with your own custom caching scheme.
• pyEX fully implements the streaming APIs

Other enhancements

• pyEX-studies: pyEX integration with TA-Lib and other libraries, for technical analysis and other metrics on top of the IEX data
• pyEX-caching: persistent, queryable caching for pyEX function calls. Minimize your spend and maximize your performance
• pyEX-zipline: Zipline integration for IEX data

Demo

Rules Engine

pyEX implements methods for interacting with the Rules Engine.

rule = {
        'conditions': [['changePercent','>',500],
                       ['latestPrice','>',100000]],
        'outputs': [{'frequency': 60,
                     'method': 'email',
                     'to': 'your_email@domain'
                    }]
        }

c.createRule(rule, 'MyTestRule', 'AAPL', 'all')  # returns {"id": <ruleID>, "weight": 2}

c.rules()  # list all rules
c.ruleInfo("<ruleID>")
c.ruleOutput("<ruleID>")
c.pauseRule("<ruleID>")
c.resumeRule("<ruleID>")
c.deleteRule("<ruleID>")

We also provide helper classes in python for constructing rules such that they abide by the rules schema (dictated in the schema() helper function)

Methods

• schema
• lookup
• create
• pause
• resume
• edit
• rule (get info)
• rules (list all)
• output

Data

pyEX provides wrappers around both static and SSE streaming data. For most static data endpoints, we provide both JSON and DataFrame return functions. For market data endpoints, we provide async wrappers as well using aiohttp (to install the dependencies, pip install pyEX[async]).

DataFrame functions will have the suffix DF, and async functions will have the suffix Async.

SSE streaming data can either be used with callbacks:

newsSSE('AAPL', on_data=my_function_todo_on_data)

or via async generators (after installing pyEX[async]):

async for data in newsSSE('AAPL'):

Full API

Please see the readthedocs for a full API spec. Implemented methods are provided in CATALOG.md.

All methods share a common naming convention. If the API method is called technicals, there will be technicals and technicalsDF methods on the client. Additionally, most methods are provided in a scope, e.g. wti is available as client.wti and client.commodities.wti, analystDays from Wall Street Horizon is available as client.premium.analystDays, etc.

Development

See CONTRIBUTING.md for guidelines.

License

This software is licensed under the Apache 2.0 license. See the LICENSE and AUTHORS files for details.