framego

README

Framego Manual

framego is a versatile and easy-to-use Go library designed for modern backend development. It simplifies the development process by providing a framework and essential libraries for building robust and scalable backend systems, from gRPC APIs, HTTP APIs to job processing.

Key Features

• Dual Protocol Support: Seamlessly implement gRPC and HTTP APIs. You can easily implement gRPC services and expose corresponding HTTP APIs automatically. Provide a variety of middlewares for gRPC and HTTP services.
• Job Processing Capabilities: Support background tasks and job processing.
• Rich Libraries: Come with essential libraries, including configuration management, database/cache/queue clients, in-memory caching solutions, advanced error handling and logging utilities.
• Ease of Use: Intuitive setup and use, allowing for quick starting your projects.

Repositories

• Framego library: https://github.com/frame-go/framego
• Cookiecutter project template: https://github.com/frame-go/cookiecutter-framego
• Protoc plugin: https://github.com/frame-go/protoc-gen-framego

Getting Started

Prerequisites

Ensure you have installed and properly configured these tools in Shell before initializing project:

• curl
• git
• go
• cookiecutter
• buf
• air (optional)

Git

Ensure you have installed and properly setup Git configuration.

Check whether you can directly clone repository from Git remote without input authentication information.

Go

Ensure you have installed and properly setup Golang development environment.

Check whether you can go get modules from private repository in shell.

Cookiecutter

Cookiecutter is a command-line utility that creates projects from project templates.

Cookiecutter is built by Python. You can use pip to install.

pip install cookiecutter

If you haven't installed pip, follow this guide to install.

Buf

Buf is a building tool to make Protobuf reliable and easy to use.

You can follow this guide to install.

Air

Air is a live-reloading command line utility for Go applications in development. It is optional for this framework.

You can follow this guide to install.

curl -sSfL https://raw.githubusercontent.com/cosmtrek/air/master/install.sh | sh

Create Project

Run this command to create a project folder in the current working directory:

cookiecutter https://github.com/frame-go/cookiecutter-framego

You will need to provide the following template variables:

Variable Name	Description	Example
project_title	Project name, in CapitalizedCase format. This name will be used in project description and documents. Suggest to be full name of project.	Identity and Access Management System
project_name	Project name, in kebab-case format. This name will be used as project root folder name.	iam-server
app_name	App name, in snake_case format. This name will be used as an app name.	security_core
app_package_name	App package name, in flatcase. This name will be used as project root cmd package name and binary name. Suggest aligning with app name, using a single word or abbreviation.	securitycore
service_name	Service name, in snake_case format. This name will be used as service name, DB name.	security_core
service_package_name	Sedervice package name, in flatcase. This name will be used as the api package name. Suggest aligning with service name, using a single word or abbreviation.	securitycore
go_module	Full go module path. Suggest to end with project name.	github.com/example/demo-server
go_version	Golang version, will be used in makefile and CI.	1.16

Initialize Project

Run this command inside the project folder to initialize and install Golang tools:

make init

Build Project

Generate codes and compile the project:

make

Setup DB

Ensure you have MySQL/MariaDB installed and running locally.

Run this command inside the project folder to create a sample DB and tables and grant permission to the test user:

mysql < db/*.sql
mysql -v -e "CREATE USER 'test'@'127.0.0.1' IDENTIFIED BY 'password'; GRANT ALL ON *.* TO 'test'@'127.0.0.1';"

You can change the DB connection config inside configs/debug/config.yaml.

Run Project

You can run the air command inside the project folder to run your project in live reload mode:

air

Or you can simply run the compiled executable inside the bin folder.

Make Commands

The template provides many useful commands by makefile. You can run make commands in the project folder.

Command	Description
make init	Initialize project and install tools.
make / make all	Generate code and compile project. Equal to make fmt && make generate && make build.
make build	Compile project.
make clean	Clear compiled files.
make go-generate	Run go generate to generate go code.
make buf-generate	Run buf generate to generate code from protobuf files and export dependencies protobuf files to buf_vendor folder.
make generate	Equal to make buf-generate && make go-generate .
make fmt	Run go fmt and go vet format and check go code.
make go-lint	Run golangci-lint analysis go code.
make buf-lint	Run buf lint to analysis protobuf code.
make lint	Run all static code analysis. Equal to make fmt && make go-lint && make buf-lint .
make check-go-version	Check whether golang version matches requirement.

Project Structure

Root folder name: {{project_name}}

Path	Description
cmd	Main applications. Each executable should be in a separated package directory. It's common to have a small main function that imports and invokes the code from other packages and nothing else.
cmd/{{app_package_name}}/main.go	Application default executable.
api	API and protocol definitions and generated codes. The data definition can be used by domain and handlers package.
api/{{service_package_name}}/{{service_package_name}}.proto	Service protobuf definition.
buf_vendor	Protobuf dependencies.
configs	Local configuration files
configs/debug/config.yaml	Default configuration files for debug
db	Database creation and migrations sql files.
docs	Documents for the project.
internal	Private application and library code. This is the code you don't want others importing in their applications or libraries.
internal/models	Data models definitions for Data Access layer. The models can be used by db, cache and domain package.
internal/db	Database data access layer. CRUD for database.
internal/cache	Cache data access layer. CRUD for in memory cache and distributed cache.
internal/domain	Business logic layer. This package contains major business logic, invokes code from db and cache packages, and provides functions for handlers.
internal/handlers	Presentation layer. API handlers for HTTP and GRPC interfaces. This package should only contains data validation and conversion, and invokes code from domain package. This package should not invokes code from models, db or cache packages.
tools	External tools used by project.
tools.go	Dummy go file to include external tool dependencies.
go.mod	Go module dependencies configuration.
Makefile	Make command configuration.
.gitignore	Git ignore file configurations.
.golangci.yaml	Golangci-lint configurations.
.dockerignore	Docker ignore file configurations for docker build
Dockerfile	Docker build configurations.
buf.yaml	Buf configurations, includes lint rules and dependencies.
buf.gen.yaml	Buf generate configurations.
.air.toml	Air configuration

Configuration

Applications built from the framework accept configurations from command line arguments, environment variables, configuration file, and Apollo.

The priority of them is: command line arguments > environment variables > configuration file > Apollo.

Basic Configurations

Below are basic configuration items supported:

Configuration Item	Command Line Argument	Environment Variable	Configuration Key	Default Value
Configuration file path	-c, --config-path	CONFIG_PATH	N/A	None
Whether to enable debug mode	-d, --debug	DEBUG	debug	false
Whether to enable human-friendly, colorized log	-b, --beautify-log	BEAUTIFY_LOG	beautify_log	FALSE
Minimal log level: trace, debug, info, warn, error	-l, --log-level	LOG_LEVEL	log_level	debug (if debug==true), info (if debug==false)
Run job	-j, --job	JOB	job	task_executor
Apollo server endpoint	--apollo-server	APOLLO_SERVER _APOLLO_SERVER_	N/A	None
Apollo app ID	--apollo-app-id	APOLLO_APP_ID _APOLLO_APP_ID_	N/A	None
Apollo app access key secret	--apollo-access-key	APOLLO_ACCESS_KEY _APOLLO_ACCESS_KEY_	N/A	None
Apollo cluster	--apollo-cluster	APOLLO_CLUSTER _APOLLO_CLUSTER_	N/A	default
Apollo namespace	--apollo-namespace	APOLLO_NAMESPACE _APOLLO_NAMESPACE_	N/A	config.yaml

App Configuration

To run the application, app configuration must be provided. It can come from either a local config file or remote config (Apollo).

• For local config file, the file path must be provided by command line argument (--config-path) or environment variable (CONFIG_PATH). The framework will try to load the local config file if the config path exists.
• For remote config, the Apollo loading arguments must be provided by command line arguments (--apollo-*) or environment variable (_APOLLO_*_). The framework will try to load Apollo config if Apollo server endpoint and app ID exist.

For the same config key, values in the local config file have higher precedence over remote config.

Besides basic configuration items, you can configure items under app for services and clients configurations.

Below is a sample app configuration config.yaml:

debug: true
beautify_log: true
app:
  name: sample
  observable:
    endpoints:
      http: ":8080"
    modules:
      - pprof
      - metrics
      - swagger
      - channelz
      - grpcui
  services:
    - name: sample
      endpoints:
        grpc: ":9000"
        http: ":8000"
      security:
        grpc:
          key: ./keys/service.pem
          cert: ./keys/service.crt
          ca: ./keys/ca.crt
      middlewares:
        - recovery
        - open_tracing
        - metrics
        - context_logger
        - log_request
        - name: access_control
          policy: ./acl.csv
        - request_validation
  jobs:
    - sample_job
  clients:
    grpc:
      middlewares:
        - open_tracing
        - metrics
      servers:
        - name: auth
          endpoint: "127.0.0.1:9000"
          security:
            key: ./keys/service.pem
            cert: ./keys/service.crt
            ca: ./keys/ca.crt
    databases:
      - name: sample
        database: "sample_db"
        user: "root"
        password: ""
        masters: ["127.0.0.1:3306"]
        slaves: []
    caches:
      - name: sample
        type: redis
        address: "127.0.0.1:6379"
    pulsars:
      - name: sample
        url: "pulsar://127.0.0.1:6650"
        token: "zFbeuKF3jqjfxkQFfOoMeQ"
  id_generator:
    service_id: 1
    key: "c2b4706d47bbddfd6729cb72960c1a3d"

A project is an app, it can has many GRPC/HTTP services under an app, and an optionalobservable service for debug, monitoring, etc. Services are listen on different ports.

Below are configuration items under app.

Path	Description	Example
name	Name of app	iam
observable	Built-in observable services for debug, monitoring, etc.
observable.endpoint.https	HTTP endpoint for observable service.	:8080
observable.modules	Enable built-in observable modules. Details of available modules refer to below.	- pprof
services	List of application services.
services[].name	Name of service	iam
services[].endpoint.grpc	gRPC endpoint for this service.	:9000
services[].endpoint.http	HTTP endpoint for this service.	:8000
services[].security.grpc	gRPC server TLS configuration. Optional, accept insecure connection if not configured.
services[].security.grpc.key	TLS server key.	./keys/service.pem
services[].security.grpc.cert	TLS server certificate chain.	./keys/service.crt
services[].security.grpc.ca	TLS CA for verifying clients certificates.	./key/ca.crt
services[].middlewares	Enable built-in middlewares/interceptors for HTTP/gRPC service. Details of available middlewares refer to below.	- recovery
jobs[]	Jobs enabled in the app. Jobs should be registered by App.AddJob(). Only enabled jobs will be run.	- txn_executor
clients	Clients of dependent service.
clients.gprc	gRPC clients of dependent service.
clients.gprc.middlewares	Enable built-in middlewares/interceptors for all gRPC clients. Details of available middlewares refer to below.	- metrics
clients.grpc.servers	Server of gRPC clients.
clients.grpc.servers[].name	Name of gRPC server to fetch the client interface. This name will also be verified for connection TLS CA is configured.	iam
clients.grpc.servers[].endpoint	Endpoint of gRPC server.	127.0.0.1:9000
clients.grpc.servers[].security	gRPC client TLS configuration. Optional, use insecure connection if not configured.
clients.grpc.servers[].security.key	TLS client key.	./keys/service.pem
clients.grpc.servers[].security.cert	TLS client certificate chain.	./keys/service.crt
clients.grpc.servers[].security.ca	TLS CA for verifying server certificates.	./key/ca.crt
databases	Databases used by app.
databases[].name	Name of database to fetch the client interface.	iam
databases[].database	Database schema name.	iam_db
databases[].user	Database user name.	test_user
databases[].password	Database passwrod.	testpass
databases[].masters	Database master endpoints for read-write query.	["127.0.0.1"]
databases[].slaves	Database slave endpoints for readonly query.	["10.0.0.1:6606", "10.0.0.2:6606"]
caches	Caches used by app.
caches[].name	Name of cache to fetch the client interface.	default
caches[].type	Cache client type. Choices: redis	redis
caches[].address	Cache server address in <host>:<port> format.	127.0.0.1:6379
caches[].username	Optional. Username for authentication.	test_user
caches[].password	Optional. Username for authentication.	testpass
caches[].db	Database to be selected after connecting to the server.	0
pulsars	Pulsar clients
pulsars[].name	Name of pulsar server to fetch the client interface	iam
pulsars[].url	URL of pulsar server	pulsar://10.0.0.1:6650
pulsars[].token	Token for pulsar auth, optional	eyJhbGciOiJU...
id_generator	ID generatior configuration, optional.
id_generator.service_id	Service ID for unique ID generator.	1
id_generator.key	Encrypt key for unique ID generator, 16 bytes, hex encoded.	c2b4706d47bbddfd6729cb72960c1a3d

Observable Service Modules

Below are built-in observable service modules:

Name	Endpoint	Description
pprof	/pprof	Golang standard profiling tool.
channelz	/channelz	gRPC Channelz UI.
swagger	/swagger	Swagger docs UI.
metrics	/metrics	Prometheus metrics endpoint for gRPC and HTTP services.
grpcui	/grpcui	gRPC service interactive UI.

Service Middlewares

Below are built-in middlewares for gRPC/HTTP client/service:

Name	Description	gRPC Service	HTTP Service	gRPC Client
recovery	Auto recover from panic in handler and record details in log.	Y	Y	N
open_tracing	(Not implemented yet) Fetch or generate trace id, put in context; generate and pass request ID for sub requests.	Y	Y	Y
metrics	Record request metrics.	Y	Y	Y
context_logger	Add request metadata into logger and put logger in context. Context logger can be fetched by log.FromContext in handlers.	Y	Y	N
log_request	Record log for each request, includes metadata, error code, latency, etc.	Y	Y	N
request_validation	Validate request data in gRPC client/service by protoc-gen-validate.	Y	N	Y
cors	HTTP CORS handling	N	Y	N
compress	HTTP response compression	N	Y	N
access_control	Request access control based on gRPC TLS certificate and Casbin configuration	Y	N	N

Libraries

errors

fromego/errors module provides useful utilities for error handling, and is compatible with errors module in standard library. Refer to errors document for details.

Documentation

Index

• type App
  • func NewApp(info *AppInfo) App
• type AppInfo
• type DatabaseManager
• type Service

Types

type App

type App = appmgr.App

func NewApp

func NewApp(info *AppInfo) App

type AppInfo

type AppInfo = appmgr.AppInfo

type DatabaseManager

type DatabaseManager = appmgr.DatabaseManager

type Service

type Service = appmgr.Service