sqltractor

sqltractor

SQL schema migration tool for Go. Use it in your existing Go code or run commands via the CLI.

Usage

from Terminal

# install
go get github.com/netw00rk/sqltractor/sqltractor-cli

# create new migration file in path
sqltractor-cli -url driver://url -path ./migrations create migration_file_xyz

# apply all available migrations
sqltractor-cli -url driver://url -path ./migrations up

# roll back all migrations
sqltractor-cli -url driver://url -path ./migrations down

# show the current migration version
sqltractor-cli -url driver://url -path ./migrations version

# apply the next n migrations
sqltractor-cli -url driver://url -path ./migrations migrate +1
sqltractor-cli -url driver://url -path ./migrations migrate +2
sqltractor-cli -url driver://url -path ./migrations migrate +n

# roll back the previous n migrations
sqltractor-cli -url driver://url -path ./migrations migrate -1
sqltractor-cli -url driver://url -path ./migrations migrate -2
sqltractor-cli -url driver://url -path ./migrations migrate -n

# go to specific migration
sqltractor-cli -url driver://url -path ./migrations goto 1
sqltractor-cli -url driver://url -path ./migrations goto 10
sqltractor-cli -url driver://url -path ./migrations goto v

in Go code

See GoDoc here: http://godoc.org/github.com/netw00rk/sqltractor/tractor

import "github.com/netw00rk/sqltractor/tractor"

// Import required driver and reader
import "github.com/netw00rk/sqltractor/driver/postgres"
import "github.com/netw00rk/sqltractor/reader/file"


func main() {
    // create tractor struct
    t := &tractor.SqlTractor{
        Driver: postgres.New("driver://url")
        Reader: file.NewFileReader("./path/to/migration/files")
    }

    // UpAsync returning chan of Result structure
    // type Result struct {
    //    File  // applied file
    //    Error // error if something happened
    //}
    for r := range t.UpAsync() {
        if r.Error != nil {
            if r.File != nil {
                fmt.Printf("Error %s while applying file %s", r.Error, r.File.FileName)
            }
        }
        // do something with applied file
    }

    // usage of synchronous wrapper that
    // return slice of applied migration files and error
    files, err := tractor.Up(t)
    if err != nil {
      // do something with error
    }
}

Available Drivers

• PostgreSQL
• Cassandra
• SQLite
• MySQL (experimental)

Need another driver? Just implement the Driver interface and open a PR.

Avaiable readers

• FileReader
• MemoryReader

Migration files

The format of migration files looks like this:

001_initial_plan_to_do_sth.up.sql     # up migration instructions
001_initial_plan_to_do_sth.down.sql   # down migration instructions
002_xxx.up.sql
002_xxx.down.sql
...

Why two files? This way you could still do sth like psql -f ./db/migrations/001_initial_plan_to_do_sth.up.sql and there is no need for any custom markup language to divide up and down migrations. Please note that the filename extension depends on the driver.

Acknowledgements

Many thanks goes to Matthias Kadenbach, https://github.com/mattes and all contributors to the https://github.com/mattes/migrate for the ideas and code