HUD.ai Python Client

The HUD.ai Python Client provides an easy to use wrapper to interact with the HUD.ai API in python applications.

You must first acquire a HUD.ai secret key before you can use this module.

Installation

pip install hudai

Usage

from hudai import Client as HudAiClient

hud_ai = HudAiClient(
    client_id='xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx',
    client_secret='xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx'
)

hud_ai.companies.list()

hud_ai.articles.fetch('17787d76-4198-4775-a49a-b3581c37a482')

Parameters

Parameter	Usage	Example
`client_id`*	Registered Client ID	`'46ef9d9b-89a9-4fd2-84cf-af6de31f2618'`
`client_secret`	Registered Client Secret	`'59170c3e-e2c9-4244-92d8-c3595d4af325'`
`base_url`	Specify an alternate server to request resources from	`'https://stage.api.hud.ai/v1'`
`auth_url`	Specify an alternate server to request auth tokens from	`'https://stage.accounts.hud.ai'`
`redirect_uri`	Path to redirect auth requests to (required for `#get_authorize_uri`)	`'https://app.example.com/oauth/callbacks/hud-ai'`

Client Auth Flow

In order to access HUD.ai data on a user's behalf, they must authorize you to do so. This requires 3 steps.

First, send them to the authorization URL

from flask import Flask, redirect, url_for
from hudai import Client as HudAiClient

app = Flask(__name__)

hud_ai = HudAiClient(
    client_id='xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx',
    client_secret='xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx',
    redirect_uri='https://app.example.com/oauth/callbacks/hud-ai'
)

@app.route('/oauth/authorize/hud-ai')
def hud_ai_authorization():
    return redirect(hud_ai.get_authorize_uri(), code=302)

Now the user will be presented with a dialog screen where they can accept or deny your request. The auth server will redirect back to your app in either case.

Finally, parse the response and get the code and attach it to the client. The code will be exchanged before the next request is made.

@app.route('/oauth/callbacks/hud-ai')
def hud_ai_callback():
    code = request.args.get('code')
    hud_ai.set_auth_code(code)

`client.get_authorize_uri()`

NOTE: This method requires redirect_uri

Returns a URL to direct users to to authorize your application to act on their behalf. You will need to handle the redirect that includes the code

`client.set_auth_code(code)`

Store the code for future exchange for an auth token

`client.refresh_tokens()`

Attempts to ensure that the client has valid auth tokens.

Refreshes known expired tokens
Exchanges auth codes for access tokens
Exchanges client ID/secret for app access tokens

Resources

Notes

* and bolded Type indicates required param
Params indicated with ? are optional keyword params
Date types are automatically converted to/from standard DateTime objects
All list resources are paginated to 50/request, with page being 0-indexed (e.g. page=3 will get you the fourth page)

Entity	Method Base
Article	client.articles
ArticleHighlights	client.article_highlights
ArticleKeyTerm	client.article_key_terms
ArticleTag	client.article_tags
ArticleGeography	client.article_geographies
Company	client.companies
CompanyEvent	client.company_events
CompanyIndustry	client.company_industries
CompanyKeyTerm	client.company_key_terms
Domain	client.domains
Industry	client.industries
KeyTerm	client.key_terms
Person	client.people
PersonKeyTerm	client.person_key_terms
PersonQuote	client.person_quotes
SystemEvent	client.system_events
SystemTask	client.system_tasks
TextCorpus	client.text_corpora
User	client.users
UserCompany	client.user_companies
UserContact	client.user_contacts
UserDigestSubscription	client.user_digest_subscriptions
UserKeyTerm	client.user_key_terms

Deployment

Deploys occur automatically via Travis-CI on tagged commits that build successfully. Should you need to manually build the project, follow these steps:

# Install twine to prevent your password from being set in plaintext
pip install twine
# Build the package
python setup.py sdist
# Upload via twine
twine upload dist/hudai-NEW_VERSION_HERE.tar.gz

FoundryAI/hud-ai-python