README
¶
PostgreSQL
This directory contains the Spanner PostgreSQL gorm dialect. This dialect is an extension of the standard gorm
PostgreSQL dialect. The dialect uses the Spanner database/sql driver to connect to Spanner. You do not need
PGAdapter to use this dialect.
Connecting
Call the New function in the github.com/googleapis/go-gorm-spanner/postgresql package to create a gorm
connection to a Spanner PostgreSQL database. Specify a Spanner database name as the connection string, e.g.
projects/my-project/instances/my-instance/databases/my-database.
import (
"fmt"
spannerpg "github.com/googleapis/go-gorm-spanner/postgresql"
_ "github.com/googleapis/go-sql-spanner"
"gorm.io/driver/postgres"
"gorm.io/gorm"
)
func helloWorld(projectId, instanceId, databaseId string) error {
db, err := gorm.Open(spannerpg.New(postgres.Config{
DSN: fmt.Sprintf("projects/%s/instances/%s/databases/%s", projectId, instanceId, databaseId),
}), &gorm.Config{})
if err != nil {
return fmt.Errorf("failed to open database connection: %v\n", err)
}
var msg string
if err := db.Raw("SELECT $1::varchar as greeting", "Hello World from Spanner PostgreSQL!").Scan(&msg).Error; err != nil {
return fmt.Errorf("failed to execute query: %v", err)
}
fmt.Println(msg)
return nil
}
Samples
See the samples directory for a list of ready-to-run samples that show how to use Spanner PostgreSQL features with gorm.
Migrations
The Spanner PostgreSQL gorm dialect uses a custom migrator that overrides some of the defaults in the standard PostgreSQL gorm migrator, as Spanner does not support the full DDL dialect of PostgreSQL. This migrator can be used for development purposes.
It is recommended to use a schema management tool for changes to test and production databases. This ensures that you have full control over the schema changes that are made, and makes it easier to use Spanner best-practices for schema management, such as grouping as many DDL statements into one batch as possible.
Supported schema management tools include:
- golang-migrate: https://github.com/golang-migrate/migrate
- Liquibase: https://github.com/cloudspannerecosystem/liquibase-spanner
Documentation
¶
Index ¶
- Constants
- func AutoOrderBy(db *gorm.DB)
- func BeforeUpdate(db *gorm.DB)
- func New(config postgres.Config) gorm.Dialector
- func NewWithSpannerConfig(config postgres.Config, spannerConfig SpannerConfig) gorm.Dialector
- func Open(dsn string) gorm.Dialector
- type Dialector
- func (dialector Dialector) ClauseBuilders() map[string]clause.ClauseBuilder
- func (dialector Dialector) DataTypeOf(field *schema.Field) string
- func (dialector Dialector) Initialize(db *gorm.DB) (err error)
- func (dialector Dialector) Migrator(db *gorm.DB) gorm.Migrator
- func (dialector Dialector) Name() string
- type SpannerConfig
Constants ¶
const ( // ClauseOnConflict for clause.ClauseBuilder ON CONFLICT key ClauseOnConflict = "ON CONFLICT" // ClauseLimit for clause.ClauseBuilder LIMIT key ClauseLimit = "LIMIT" )
Variables ¶
This section is empty.
Functions ¶
func AutoOrderBy ¶
func BeforeUpdate ¶
func NewWithSpannerConfig ¶
func NewWithSpannerConfig(config postgres.Config, spannerConfig SpannerConfig) gorm.Dialector
Types ¶
type Dialector ¶
type Dialector struct {
postgres.Dialector
SpannerConfig SpannerConfig
}
func (Dialector) ClauseBuilders ¶
func (dialector Dialector) ClauseBuilders() map[string]clause.ClauseBuilder
type SpannerConfig ¶
type SpannerConfig struct {
// DisableAutoMigrateBatching turns off DDL batching for AutoMigrate calls.
// Spanner PostgreSQL by default uses DDL batching when AutoMigrate is called, as
// executing multiple DDL statements in a single batch is a lot more efficient
// than executing each statement separately. You should only use this option
// if you are experiencing problems with the automatic batching of DDL
// statements when calling AutoMigrate.
DisableAutoMigrateBatching bool
// AutoOrderByPk automatically adds an ORDER BY <pk> to all queries.
// This flag is primarily intended for testing, as most gorm tests assume that queries will return query results
// in primary key order, even when there is no ORDER BY clause in the query. Spanner does not guarantee this.
AutoOrderByPk bool
// AutoAddPrimaryKey automatically adds a 'generated_id serial' column to tables that are created without a
// primary key. This flag is primarily intended for testing, as some gorm tests assumes that databases support
// tables without a primary key. Spanner does not support this.
AutoAddPrimaryKey bool
}
SpannerConfig contains custom configuration options for Spanner PostgreSQL.
Source Files