analytix#

PyPi version PyPI - Status Downloads GitHub last commit License

CI Read the Docs Maintainability Test Coverage

A simple yet powerful wrapper for the YouTube Analytics API.

CPython versions 3.7 through 3.11-dev and PyPy versions 3.7 through 3.9 are officially supported.

Windows, MacOS, and Linux are all supported.

Features#

  • Pythonic syntax lets you feel right at home

  • Dynamic error handling saves hours of troubleshooting, and makes sure only valid requests count toward your API quota

  • A clever interface allows you to make multiple requests across multiple sessions without reauthorising

  • Extra support allows the native saving of CSV files and conversion to DataFrame objects

  • Easy enough for beginners, but powerful enough for advanced users

What does analytix do?#

The YouTube Studio provides a fantastic interface where creators can view some incredibly detailed analytics for their channel. However, there’s no way to perform programmatical operations on the data to do some proper analysis on it. This is where analytix comes in.

The process of analysing data on the YouTube Studio is comprised of two steps:

  1. Retrieving the data to be analysed and visualised

  2. Presenting that data to the user

analytix aims to handle step one as comprehensively as possible, allowing analysts to use tools such as pandas and Matplotlib to work on the data without having to faff around with Google’s offerings.

Installation#

To install the latest stable version of analytix, use the following command:

pip install analytix

You can also install the latest development version using the following command:

pip install git+https://github.com/parafoxia/analytix

You may need to prefix these commands with a call to the Python interpreter depending on your OS and Python configuration.

Additional support#

You can also install analytix with additional libraries to provide extra functionality:

  • analytix[arrow] — Apache Arrow support (including Feather and Parquet files)

  • analytix[dev] — development dependencies

  • analytix[excel] — support for exporting reports to Excel spreadsheets

  • analytix[modin] — Modin support (note: this installs all engines; if you want to use a specific engine, you will need to do so manually)

  • analytix[pandas] — pandas support (analytix[df] does the same, but is deprecated)

  • analytix[types] — type stubs for type-hinted projects

To install multiple at once, use commas:

pip install "analytix[excel,modin,types]

OAuth authentication#

All requests to the YouTube Analytics API need to be authorised through OAuth 2. In order to do this, you will need a Google Developers project with the YouTube Analytics API enabled. You can find instructions on how to do that in the API setup guide, or on this video.

When analytix boots up for the first time, it will display a link and ask for a code. You’ll need to follow that link, run through all the steps, and enter the code to authorise it. Once that’s done, analytix saves the tokens to the disk (if you plan to run analytix on a server, make sure these are in a safe place). This includes your refresh token, which analytix will automatically use to refresh your access token when needed.

This means you should only have to authorise analytix, at most, every week. More details regarding how and when refresh tokens expire can be found on the Google Identity documentation.

Logging#

If you want to see what analytix is doing, you can enable the packaged logger:

import analytix

analytix.setup_logging()

If anything is going wrong, or analytix appears to be taking a long time to fetch data, try enabling the logger in DEBUG mode.

Usage#

Retrieving reports from the YouTube Analytics API is easy. The below example loads credentials from a secrets file, and gets as much information as possible from the last 28 days:

from analytix import Analytics

client = Analytics.with_secrets("./secrets.json")
report = client.retrieve(dimensions=("day",))
report.to_csv("./analytics.csv")

This can also be done asynchronously:

import datetime as dt

from analytix import AsyncAnalytics

client = await AsyncAnalytics.with_secrets("./secrets.json")
report = await client.retrieve(
    dimensions=("country",),
    start_date=dt.date.today() - dt.timedelta(days=7),
)
await report.to_csv("./async-analytics.csv")

If you want to analyse this data using additional tools such as pandas, you can directly export the report as a DataFrame (note that pandas is an optional dependency – see above):

df = report.to_dataframe()

To read up further, have a look at the documentation, or have a look at some examples.

Contributing#

Contributions are very much welcome! To get started:

License#

The analytix module for Python is licensed under the BSD 3-Clause License.