Google Cloud Natural Language API Client Library

Warnings

• breaking The Natural Language API v2 introduces breaking changes, including renaming the `language` field to `language_code` in responses and removing fields like `salience` and `wikipedia_url` for entity analysis. Ensure you use the correct client (`language_v1` or `language_v2`) and adapt your code accordingly.
  Fix: For new projects or if you need v2 features, explicitly import `language_v2` (e.g., `from google.cloud import language_v2`). Review the v2 API documentation for specific field changes.
• gotcha Direct API Key authentication is generally not supported for Google Cloud client libraries. Attempting to use an API key might result in authentication errors.
  Fix: Use Application Default Credentials (ADC). For local development, run `gcloud auth application-default login` in your terminal. For deployed applications on Google Cloud, a service account attached to the resource is the preferred method.
• deprecated The direct import `from google.cloud.language import enums` is deprecated. Enums are now nested under the versioned types module (e.g., `language_v1.types.Document.Type`).
  Fix: Update imports to access enums directly via the versioned module, for example, `language_v1.Document.Type.PLAIN_TEXT`.
• gotcha Naming your Python script `google.py`, `language.py`, or similar names can lead to import conflicts (shadowing) with the installed `google.cloud.language` library, causing `ImportError`.
  Fix: Always use distinct names for your local Python files that do not conflict with installed package names.

Install

• pip install google-cloud-language Install latest version

Imports

• LanguageServiceClient

  from google.cloud import language_v1
  client = language_v1.LanguageServiceClient()

  Client classes are typically versioned under `language_v1` or `language_v2` for stability and to distinguish API versions.
• Document.Type

  from google.cloud import language_v1
  document = language_v1.types.Document(content=text, type_=language_v1.Document.Type.PLAIN_TEXT)

  Enums are now accessed directly via the `language_v1` or `language_v2` module's types, e.g., `language_v1.Document.Type` (note `type_` for the field name to avoid conflict with the Python keyword `type`). The `enums` submodule is deprecated.

Quickstart

This quickstart demonstrates how to perform sentiment analysis using the `google-cloud-language` library. It initializes the client and sends a text document for sentiment detection. Ensure your Google Cloud project has the Natural Language API enabled and that you've set up authentication, preferably using Application Default Credentials, for local development or deployed applications.

import os
from google.cloud import language_v1

# Set up Application Default Credentials if running locally
# This line is usually not needed when running on Google Cloud services (e.g., Compute Engine, Cloud Run)
# For local development, ensure GOOGLE_APPLICATION_CREDENTIALS points to a service account key file
# or run `gcloud auth application-default login`
# os.environ['GOOGLE_APPLICATION_CREDENTIALS'] = os.environ.get('GOOGLE_APPLICATION_CREDENTIALS', '')

def analyze_sentiment(text_content):
    """Analyzes the sentiment of the provided text."""
    client = language_v1.LanguageServiceClient()

    # Available document types: PLAIN_TEXT, HTML
    document = language_v1.types.Document(
        content=text_content, type_=language_v1.Document.Type.PLAIN_TEXT
    )

    # Detects the sentiment of the document
    sentiment = client.analyze_sentiment(request={'document': document}).document_sentiment

    print(f"Text: {text_content}")
    print(f"Sentiment Score: {sentiment.score}")
    print(f"Sentiment Magnitude: {sentiment.magnitude}")

# Example usage
if __name__ == '__main__':
    analyze_sentiment("Hello, world! This is a great day.")
    analyze_sentiment("I am very unhappy with the service.")