google-cloud-smalltalk

Pharo Smalltalk client libraries for Google Cloud Platform services.

Disclaimer

This is an unofficial, community-maintained library and is not affiliated with, endorsed, or supported by Google LLC. Use at your own risk.

Features

Package	Description	Documentation
Google-GenAI	Gemini AI text generation, chat, function calling, and structured output	docs/gemini.md
Google-BigQuery	BigQuery dataset management	docs/bigquery.md
Google-Auth	GCP authentication	docs/getting-started.md

Requirements

Smalltalk	Version
Pharo	12.0, 13.0

Installation

Metacello new
  baseline: 'GoogleCloud';
  repository: 'github://newapplesho/google-cloud-smalltalk:main/src';
  load.

Quick Start

1. Set up authentication

Smalltalk platform environment at: 'GOOGLE_APPLICATION_CREDENTIALS' put:(FileLocator home / 'service_account_key.json') pathString.

See Getting Started for more authentication options.

2. Use Gemini AI

client := GoogleGenAIClient new
    config: (GoogleGenAIConfig new
        projectId: 'your-gcp-project-id';
        location: 'global';  "Gemini 2.5 models are served via the global endpoint on Vertex AI"
        useVertexAI: true;
        yourself).

response := client generateContent: 'Explain Smalltalk in one sentence.'.
Transcript show: response text.

The default model is gemini-2.5-flash. See Gemini Documentation for chat, function calling, code generation, and more.

Documentation

• Getting Started - Authentication setup
• Gemini AI - Text generation, chat, function calling, structured output
• BigQuery - Dataset management
• Development - Building and testing the library in Pharo
• Roadmap - Planned features

Development

This is a Pharo / Tonel project. With a local Pharo image (make setup):

make load    # load/reload the project into the image
make test    # run the offline test suite
make ui      # open the Pharo GUI

See docs/development.md for the full workflow.