esp-metrics-cli 0.3.0

ESP Metrics CLI - Command-line tool for validating and uploading project metrics

ESP Metrics CLI

A command-line tool for standardizing metrics collection, validation, and upload across all Espressif projects. This tool enables consistent metrics processing in CI/CD pipelines by providing a unified interface for generating, validating, and submitting project metrics to the central Espressif metrics service.

• ESP Metrics CLI
  • Why ESP Metrics CLI?
    • The Problem
    • The Solution
  • Architecture Overview
  • Getting Started
    • Installation
      • Option 1: Install from PyPI (Recommended)
      • Option 2: Standalone Binary
      • Option 3: From Source (Development)
  • Developer Integration Guide
    • Step 1: Add esp-metrics-cli to Your Project
    • Step 2: Generate Metrics
    • Step 3: Create Metric Definitions
      • Schema Structure
    • Step 4: Integrate with CI Pipeline
      • File Organization
  • Metrics Validation
    • Local Validation
  • Command Reference
    • esp-metrics-cli init
    • esp-metrics-cli validate
    • esp-metrics-cli upload
    • Global Options
  • Configuration
    • Configuration File (Recommended for Local Development)
    • Environment Variables (Required for CI/CD)
    • Bucket Types
  • Integration Points
    • CI/CD Pipeline Integration
    • MinIO Object Storage
    • Backend Metrics Service
    • Local Development Workflow
  • Development

---

Why ESP Metrics CLI?

The Problem

Espressif has multiple projects (esp-idf, esp-bluetooth, esp-matter, etc.) that need to track various metrics like:

• Build metrics: Binary sizes, compilation times, memory usage
• Test metrics: Pass/fail rates, execution times, coverage
• Custom metrics: Project-specific measurements

Without standardization, each project would:

• Implement metrics collection differently
• Use incompatible data formats
• Duplicate validation and upload logic

The Solution

ESP Metrics CLI provides a standardized client-side architecture that:

• Unifies metrics collection across all Espressif projects
• Enforces consistent data schemas via validation
• Centralizes upload logic to object storage
• Enables cross-project analytics and dashboards
• Integrates seamlessly with existing CI/CD pipelines

Architecture Overview

The ESP Metrics CLI follows a three-stage workflow designed for CI/CD integration:

Key Principles:

• Project-side generation: Projects create custom scripts to transform raw artifacts into meaningful metrics
• Standardized validation: All metrics conform to a contracted JSON schema
• Centralized upload: Single tool handles validation and upload to object storage
• Backend integration: Uploaded metrics trigger webhook processing for dashboard/API consumption

Getting Started

Installation

Choose your preferred installation method:

Option 1: Install from PyPI (Recommended)

# Using pip
pip install esp-metrics-cli

# Using uv (faster)
uv tool install esp-metrics-cli

# Verify installation
esp-metrics-cli --help

Option 2: Standalone Binary

Download pre-built binaries from the GitLab Package Registry on the project page.

Option 3: From Source (Development)

# Clone repository
cd esp-metrics-cli

# Install with uv
uv venv && source .venv/bin/activate
uv sync --all-extras

# Verify installation
python -m esp_metrics_cli.cli --help

Developer Integration Guide

Step 1: Add esp-metrics-cli to Your Project

Install the tool in your project environment:

# Add to requirements.txt
echo "esp-metrics-cli>=0.1.0" >> requirements.txt

# Or add to pyproject.toml
esp-metrics-cli = ">=0.1.0"

# Or install directly
pip install esp-metrics-cli

Step 2: Generate Metrics

Create a script that transforms your raw artifacts into JSON which satisfies the format you described in metrics_definitions.yaml.

The script can be implemented in any language, or you can use tools like jq to produce metrics.json in the required format.

The metrics.json file should contain a JSON object where the top-level keys are metric categories (like "builds", "tests", "coverage", etc.) and the values follow the hierarchical structure you defined in your metrics_definitions.yaml. Read more in Step 3: Create Metric Definitions:

Here is an example script:

#!/usr/bin/env python3
# tools/generate_metrics.py

import json


def generate_metrics():
    """Generate metrics from your project artifacts."""

    return {
        'builds': {
            'apps/my_app/build': {
                'size': {'flash_size': 1048576, 'ram_size': 65536},
                'build': {'time': 45.2, 'warnings': 3},
            }
        },
        'tests': {'unit_tests': {'passed': 42, 'failed': 1, 'duration': 12.5}},
        'coverage': {'line_coverage': 85.3, 'branch_coverage': 78.9},
    }


if __name__ == '__main__':
    metrics = generate_metrics()
    with open('metrics.json', 'w') as f:
        json.dump(metrics, f, indent=2)

Notes about the format of JSON file:

• Top-level keys are the metrics and can be anything meaningful to your project.
• You can mix flat metrics and nested structures at the same level
• Metric names must use snake_case with units/dimensions at the end (e.g., time_seconds, size_bytes)
• Lists of dictionaries are supported in the metrics structure
• The JSON structure must match what you've defined in your metrics_definitions.yaml Step 3: Create Metric Definitions

Step 3: Create Metric Definitions

Definition files (metrics_definitions.yaml) serve multiple purposes:

1. Schema validation: Ensures your metrics.json file has the correct structure and data types
2. Documentation: Describes what each metric represents and its purpose
3. Type safety: Prevents errors by enforcing consistent data types across uploads
4. API contract: Defines the expected format for service consuming your metrics

Schema Structure

The schema follows JSON Schema specification and should contain:

• type: object - Root level must be an object
• properties - Define your top-level metric categories (e.g. builds, tests, coverage, etc.)
• description - Human-readable explanation of what each section contains
• Data types - number, integer, string, boolean, array, object
• patternProperties - For dynamic keys (like build paths or test suite names)

Create a metrics_definitions.yaml file that defines your project's metrics schema. Look at the example below:

type: object
properties:
  builds:
    type: object
    description: "Build-related metrics organized by path"
    patternProperties:
      "^.+$": # Matches any build path like 'apps/my_app/build'
        type: object
        properties:
          size:
            type: object
            properties:
              flash_size:
                type: integer
                unit: "bytes"
                friendly_name: "Flash Memory Size"
                description: "Flash memory usage in bytes"
              ram_size:
                type: integer
                unit: "bytes"
                friendly_name: "RAM Memory Size"
                description: "RAM memory usage in bytes"
          build:
            type: object
            properties:
              time:
                type: number
                unit: "seconds"
                friendly_name: "Build Time"
                description: "Build compilation time in seconds"
              warnings:
                type: integer
                unit: "count"
                friendly_name: "Build Warnings"
                description: "Number of compiler warnings"
  tests:
    type: object
    description: "Test-related metrics organized by test suite"
    patternProperties:
      "^.+$": # Matches any test suite name like 'unit_tests'
        type: object
        properties:
          passed:
            type: integer
            properties:
              unit:
                type: string
                value: "count"
              friendly_name:
                type: string
                value: "Passed Tests"
              description:
                type: string
                value: "Number of passed tests"
          failed:
            type: integer
            properties:
              unit:
                type: string
                value: "count"
              friendly_name:
                type: string
                value: "Failed Tests"
              description:
                type: string
                value: "Number of failed tests"
          duration:
            type: number
            properties:
              unit:
                type: string
                value: "seconds"
              friendly_name:
                type: string
                value: "Test Duration"
              description:
                type: string
                value: "Total test execution time in seconds"
  coverage:
    type: object
    description: "Code coverage metrics"
    properties:
      line_coverage:
        type: number
        properties:
          unit:
            type: string
            value: "percent"
          friendly_name:
            type: string
            value: "Line Coverage"
          description:
            type: string
            value: "Line coverage percentage"
      branch_coverage:
        type: number
        properties:
          unit:
            type: string
            value: "percent"
          friendly_name:
            type: string
            value: "Branch Coverage"
          description:
            type: string
            value: "Branch coverage percentage"

Step 4: Integrate with CI Pipeline

File Organization

To keep your metrics infrastructure organized and maintainable, create a centralized structure for all metrics-related files.

As an example:

project-root/
├── tools/
│   └── ci/
│       └── metrics/
│           ├── build/
│           │   ├── metrics_definitions.yaml
│           │   └── generate_metrics.py
│           ├── tests/
│           │   ├── metrics_definitions.yaml
│           │   └── generate_metrics.py
│           └── coverage/
│               ├── metrics_definitions.yaml
│               └── generate_metrics.py
└── .gitlab-ci.yml

Key principles:

• Centralized location: All metrics files under one location, e.g. tools/ci/metrics/
• Category-based organization: Separate directories for build/, tests/, coverage/, etc.
• Consistent naming: Each category has its own metrics_definitions.yaml and generate_metrics.py
• Version controlled: All files committed to your repository for reproducibility

Update your existing jobs to generate and upload metrics. Make sure the following variables are defined in your GitLab CI settings:

• ESP_METRICS_S3_SERVER
• ESP_METRICS_S3_ACCESS_KEY
• ESP_METRICS_S3_SECRET_KEY
• ESP_METRICS_S3_LONG_TERM_BUCKET
• ESP_METRICS_S3_SHORT_TERM_BUCKET

The CLI automatically uses the following GitLab CI built-in variables, make sure to have them available in the variables section.

• CI_PROJECT_URL - for project identification
• CI_PROJECT_ID - for project identification
• CI_COMMIT_SHA - for commit identification
• CI_COMMIT_REF_NAME - for branch identification (optional, auto-detected if not set via ESP_METRICS_BRANCH_NAME)

Update your existing jobs to generate and upload metrics in gitlab ci config for the build job that produces build-related metrics:

# .gitlab-ci.yml
variables:
  ESP_METRICS_PROJECT_URL: "$CI_PROJECT_URL"
  ESP_METRICS_PROJECT_ID: "$CI_PROJECT_ID"
  ESP_METRICS_COMMIT_SHA: "$CI_COMMIT_SHA"
  ESP_METRICS_BRANCH_NAME: "$CI_COMMIT_REF_NAME"
build_job:
  stage: build
  script:
    - make build
    - pip install esp-metrics-cli
    - python tools/ci/metrics/build/generate_metrics.py > metrics.json
    - esp-metrics-cli upload -d tools/ci/metrics/build/metrics_definitions.yaml -i metrics.json --verbose

Metrics Validation

Local Validation

Validate your metrics before CI upload:

# Basic validation
esp-metrics-cli validate -d definitions.yaml -i metrics.json

# Verbose validation with detailed output
esp-metrics-cli validate -d definitions.yaml -i metrics.json --verbose

Validation checks:

• Required fields present
• Data types match schema
• Custom validation rules

Command Reference

esp-metrics-cli init

Create a configuration file with default settings. The config file (git ignored by default) is used for the local development. For the production environment it is recommended to use environment variables.

esp-metrics-cli init [OPTIONS]

Options:

• --path TEXT: Configuration file path (default: .esp_metrics.toml)
• --force: Overwrite existing configuration file

Examples:

# Create default config
esp-metrics-cli init

# Create config with custom path
esp-metrics-cli init --path config/metrics.toml

# Overwrite existing config
esp-metrics-cli init --force

esp-metrics-cli validate

Validate metrics against schema definitions.

esp-metrics-cli validate [OPTIONS]

Required Options:

• -d, --definitions PATH: Path to definitions.yaml file
• -i, --input PATH: Path to metrics.json file

Optional Options:

• -v, --verbose: Enable detailed output
• -o, --output PATH: Path to save validated metrics data as JSON file

Examples:

# Basic validation
esp-metrics-cli validate -d definitions.yaml -i metrics.json

# Verbose validation
esp-metrics-cli validate -d definitions.yaml -i metrics.json --verbose

# Validation with output file to save validated metrics
esp-metrics-cli validate -d definitions.yaml -i metrics.json --output validated_metrics.json

esp-metrics-cli upload

Upload metrics to the service. Note: This command automatically validates metrics before upload, so a separate validation step is not required.

esp-metrics-cli upload [OPTIONS]

Required Options:

• -d, --definitions PATH: Path to definitions.yaml file
• -i, --input PATH: Path to metrics.json file

Optional Options:

• -v, --verbose: Enable detailed output
• --bucket-type CHOICE: Override bucket selection (auto|long_term|short_term|custom)
• --custom-bucket TEXT: Custom bucket name (required with --bucket-type=custom)

Examples:

# Upload with auto bucket detection
esp-metrics-cli upload -d definitions.yaml -i metrics.json

# Force upload to long-term bucket
esp-metrics-cli upload -d definitions.yaml -i metrics.json --bucket-type long_term

# Upload to custom bucket
esp-metrics-cli upload -d definitions.yaml -i metrics.json --bucket-type custom --custom-bucket my-bucket

# Verbose upload with detailed validation output
esp-metrics-cli upload -d definitions.yaml -i metrics.json --verbose

Global Options

Available for all commands:

• -c, --config-file PATH: Path to configuration file (for local dev)
• -h, --help: Show help message

Examples:

# Use custom config file
esp-metrics-cli -c /path/to/config.toml validate -d definitions.yaml -i metrics.json

esp-metrics-cli upload -d definitions.yaml -i metrics.json

Configuration

ESP Metrics CLI supports two configuration methods depending on your use case:

Configuration File (Recommended for Local Development)

For local development and testing, use the .esp_metrics.toml configuration file. This method provides:

• Easy editing
• Clear documentation with inline comments
• Persistent settings across development sessions
• Template generation via esp-metrics-cli init

Create and customize your configuration file:

# Generate default configuration file
esp-metrics-cli init

The .esp_metrics.toml file contains all configuration settings with helpful comments. Look at the example.

Environment Variables (Required for CI/CD)

For CI/CD pipelines, use environment variables instead of configuration files. This approach provides:

• Secure credential handling via CI/CD secrets
• No file management
• Dynamic configuration per pipeline run

All configuration can be set via environment variables with the ESP_METRICS_ prefix.

You can take a look at the example.

Bucket Types

The tool supports different bucket types for upload destinations:

• auto (default): Automatically selects bucket based on branch name
  • Release branches → long-term bucket
  • Other branches → short-term bucket
• long_term: Always use long-term bucket (for releases)
• short_term: Always use short-term bucket (for development)
• custom: Use custom bucket name (requires --custom-bucket)

Integration Points

CI/CD Pipeline Integration

ESP Metrics CLI provides a reusable template for uploading metrics generated by your existing jobs. The example is shown in Step 4: Integrate with CI Pipeline.

MinIO Object Storage

The tool uploads metrics to MinIO-compatible S3 storage:

• Authentication: S3 access/secret keys
• Bucket selection: Automatic or manual bucket routing
• File naming: Structured naming convention for easy retrieval
• Retry logic: Automatic retry on upload failures

Upload path structure:

bucket/
├── project_host/
│   ├── projectid_project_path/
│   │   ├── commit_sha/
│   │   │   └── {timestamp}.json
│   │   ├── commit_sha/
│   │   │   └── {timestamp}.json
│   │   └── ...
│   ├── projectid_project_path/
│   │   └── ...
│   └── ...
├── project_host/
│   └── ...
└── ...

Example:

esp-metrics-bucket/
├── https_gitlab_example_com_8080/
│   ├── 1111_myorg_esp-project/
│   │   ├── abc123def456/
│   │   │   ├── 2024-01-15_14-30-00.507.json
│   │   │   └── 2024-01-15_15-45-30.507.json
│   │   └── def789ghi012/
│   │       └── 2024-01-16_09-15-22.507.json
│   └── 2222_myorg_esp-components/
│       └── xyz987uvw654/
│           └── 2024-01-17_11-20-45.507.json
├── https_github_example_com/
│   └── 3333_myteam_my-project/
│       └── 1a2b3c4d5e6f/
│           └── 2024-01-18_16-55-10.507.json
└── https_git_example_com/
    └── 4444_devteam_repo-name/
        └── 9z8y7x6w5v4u/
            └── 2024-01-19_12-00-00.507.json

Backend Metrics Service

Uploaded metrics trigger backend processing:

1. Webhook trigger: MinIO notifies backend service of new uploads
2. Data validation: Backend re-validates metrics schema
3. Database storage: Metrics stored in MongoDB for analytics
4. API exposure: Metrics available via REST API
5. Dashboard updates: Real-time dashboard updates

Integration flow:

CLI Upload → MinIO → Webhook → Backend Service → MongoDB → Dashboard/API

Local Development Workflow

Developer workflow for testing metrics locally using configuration files:

# 1. Set up development environment with configuration file
esp-metrics-cli init
# Edit .esp_metrics.toml with your S3 settings and preferences

# 2. Generate test metrics
python tools/generate_metrics.py

# 3. Upload and validate in one step (if S3 credentials are configured)
esp-metrics-cli upload -d definitions.yaml -i metrics.json --verbose

# Optional: Validate only (for testing schema without uploading)
esp-metrics-cli validate -d definitions.yaml -i metrics.json --verbose

---

Development

This project uses modern Python development tools for maintainability and code quality.

Development setup:

# Clone repository
cd esp-metrics-cli

# Install with development dependencies
uv venv && source .venv/bin/activate
uv sync --all-extras

# Install pre-commit hooks
pre-commit install && pre-commit autoupdate

# Run tests
pytest

# Run quality checks
pre-commit run --all-files

Tools used:

• UV: Fast package manager
• Hatchling: Modern build system
• Ruff: Fast linter and formatter
• Pytest: Testing framework
• Pre-commit: Code quality hooks
• Mypy: Type checking