Compare commits
1 Commits
main
...
c740d02821
| Author | SHA1 | Date | |
|---|---|---|---|
|
c740d02821
|
@@ -1,24 +0,0 @@
|
|||||||
# Skill Registry — go-migrate
|
|
||||||
|
|
||||||
Generated: 2026-04-21
|
|
||||||
|
|
||||||
## Compact Rules
|
|
||||||
|
|
||||||
### Go conventions
|
|
||||||
- Use existing error wrapping pattern: `fmt.Errorf("context: %w", err)`
|
|
||||||
- Channel-based pipeline — keep goroutine lifecycle clean (close channels in correct order)
|
|
||||||
- No comments unless non-obvious WHY; no docstrings
|
|
||||||
- Prefer named returns only when it aids clarity in short functions
|
|
||||||
- Use `strings.EqualFold` for case-insensitive column name comparison
|
|
||||||
|
|
||||||
### Project conventions
|
|
||||||
- Config structs live in `internal/app/config/`
|
|
||||||
- ETL interfaces live in `internal/app/etl/types.go`
|
|
||||||
- Transformer implementations in `internal/app/etl/transformers/`
|
|
||||||
- Azure operations via `internal/app/azure/main.go`
|
|
||||||
- Per-job transformer creation (not shared) when job has storage config
|
|
||||||
|
|
||||||
## User Skills
|
|
||||||
| Trigger | Skill |
|
|
||||||
|---------|-------|
|
|
||||||
| sdd-* | SDD workflow skills |
|
|
||||||
14
.env.example
14
.env.example
@@ -1,12 +1,2 @@
|
|||||||
SOURCE_DB_URL=sqlserver://sa:password@localhost:1433?database=master&packet+size=32767&loc=UTC
|
PG_FROM_DB_URL=postgresql://postgres:password@localhost:5432/db
|
||||||
TARGET_DB_URL=postgresql://postgres:password@localhost:5432/db
|
PG_TO_DB_URL=postgresql://postgres:password@localhost:5432/db
|
||||||
|
|
||||||
LOG_LEVEL=INFO
|
|
||||||
|
|
||||||
AZ_STORAGE_ENABLED=false
|
|
||||||
AZ_ACCOUNT_NAME=
|
|
||||||
AZ_CONTAINER=
|
|
||||||
AZ_ACCOUNT_KEY=
|
|
||||||
AZ_USE_HTTPS=true
|
|
||||||
AZ_SERVICE_URL=
|
|
||||||
AZ_PREFIX=
|
|
||||||
3
.gitignore
vendored
3
.gitignore
vendored
@@ -4,7 +4,6 @@
|
|||||||
*.dll
|
*.dll
|
||||||
*.so
|
*.so
|
||||||
*.dylib
|
*.dylib
|
||||||
bin/
|
|
||||||
|
|
||||||
# Test binary, built with `go test -c`
|
# Test binary, built with `go test -c`
|
||||||
*.test
|
*.test
|
||||||
@@ -27,5 +26,5 @@ go.work.sum
|
|||||||
|
|
||||||
# Editor/IDE
|
# Editor/IDE
|
||||||
# .idea/
|
# .idea/
|
||||||
.vscode/
|
# .vscode/
|
||||||
.temp
|
.temp
|
||||||
|
|||||||
80
Makefile
80
Makefile
@@ -1,80 +0,0 @@
|
|||||||
.PHONY: build build-linux build-windows build-all clean help
|
|
||||||
|
|
||||||
# Variables
|
|
||||||
BINARY_NAME=go-migrate
|
|
||||||
CMD_PATH=./cmd/go_migrate
|
|
||||||
OUTPUT_DIR=bin
|
|
||||||
VERSION?=$(shell git describe --tags --always --dirty 2>/dev/null || echo "dev")
|
|
||||||
BUILD_TIME=$(shell date -u '+%Y-%m-%d_%H:%M:%S')
|
|
||||||
GIT_COMMIT=$(shell git rev-parse --short HEAD 2>/dev/null || echo "unknown")
|
|
||||||
|
|
||||||
# Flags de compilación
|
|
||||||
LD_FLAGS=-ldflags="-s -w -X main.Version=$(VERSION) -X main.BuildTime=$(BUILD_TIME) -X main.GitCommit=$(GIT_COMMIT)"
|
|
||||||
|
|
||||||
# Default: compilar para el SO actual
|
|
||||||
build: build-$(OS)
|
|
||||||
|
|
||||||
ifeq ($(OS),Windows_NT)
|
|
||||||
build-native: build-windows
|
|
||||||
else
|
|
||||||
build-native: build-linux
|
|
||||||
endif
|
|
||||||
|
|
||||||
# Compilar para Linux (sin CGO para máxima compatibilidad)
|
|
||||||
build-linux:
|
|
||||||
@echo "Compilando para Linux..."
|
|
||||||
@mkdir -p $(OUTPUT_DIR)
|
|
||||||
CGO_ENABLED=0 GOOS=linux GOARCH=amd64 go build \
|
|
||||||
$(LD_FLAGS) \
|
|
||||||
-o $(OUTPUT_DIR)/$(BINARY_NAME)-linux-amd64 \
|
|
||||||
$(CMD_PATH)
|
|
||||||
@echo "Binario creado: $(OUTPUT_DIR)/$(BINARY_NAME)-linux-amd64"
|
|
||||||
|
|
||||||
# Compilar para Windows
|
|
||||||
build-windows:
|
|
||||||
@echo "Compilando para Windows..."
|
|
||||||
@mkdir -p $(OUTPUT_DIR)
|
|
||||||
CGO_ENABLED=0 GOOS=windows GOARCH=amd64 go build \
|
|
||||||
$(LD_FLAGS) \
|
|
||||||
-o $(OUTPUT_DIR)/$(BINARY_NAME)-windows-amd64.exe \
|
|
||||||
$(CMD_PATH)
|
|
||||||
@echo "Binario creado: $(OUTPUT_DIR)/$(BINARY_NAME)-windows-amd64.exe"
|
|
||||||
|
|
||||||
# Compilar para ambas plataformas
|
|
||||||
build-all: build-linux build-windows
|
|
||||||
@echo ""
|
|
||||||
@echo "Binarios compilados:"
|
|
||||||
@ls -lh $(OUTPUT_DIR)/$(BINARY_NAME)*
|
|
||||||
|
|
||||||
# Compilar para Linux arm64 (opcional, para Raspberry Pi, etc.)
|
|
||||||
build-linux-arm64:
|
|
||||||
@echo "Compilando para Linux ARM64..."
|
|
||||||
@mkdir -p $(OUTPUT_DIR)
|
|
||||||
CGO_ENABLED=0 GOOS=linux GOARCH=arm64 go build \
|
|
||||||
$(LD_FLAGS) \
|
|
||||||
-o $(OUTPUT_DIR)/$(BINARY_NAME)-linux-arm64 \
|
|
||||||
$(CMD_PATH)
|
|
||||||
@echo "Binario creado: $(OUTPUT_DIR)/$(BINARY_NAME)-linux-arm64"
|
|
||||||
|
|
||||||
# Limpiar binarios
|
|
||||||
clean:
|
|
||||||
@echo "Limpiando binarios..."
|
|
||||||
@rm -rf $(OUTPUT_DIR)
|
|
||||||
@echo "Limpieza completada"
|
|
||||||
|
|
||||||
# Ayuda
|
|
||||||
help:
|
|
||||||
@echo "Comandos disponibles:"
|
|
||||||
@echo ""
|
|
||||||
@echo " make build - Compilar para el SO actual (Linux/Windows)"
|
|
||||||
@echo " make build-linux - Compilar para Linux x86_64"
|
|
||||||
@echo " make build-windows - Compilar para Windows x86_64"
|
|
||||||
@echo " make build-linux-arm64 - Compilar para Linux ARM64 (opcional)"
|
|
||||||
@echo " make build-all - Compilar para Linux y Windows"
|
|
||||||
@echo " make clean - Eliminar binarios compilados"
|
|
||||||
@echo " make help - Mostrar esta ayuda"
|
|
||||||
@echo ""
|
|
||||||
@echo "Ejemplos de uso:"
|
|
||||||
@echo " make build-all # Crear binarios para ambas plataformas"
|
|
||||||
@echo " make build-linux OS= # Crear solo para Linux"
|
|
||||||
@echo ""
|
|
||||||
77
cmd/go_migrate/connect.go
Normal file
77
cmd/go_migrate/connect.go
Normal file
@@ -0,0 +1,77 @@
|
|||||||
|
package main
|
||||||
|
|
||||||
|
import (
|
||||||
|
"context"
|
||||||
|
"database/sql"
|
||||||
|
"errors"
|
||||||
|
"fmt"
|
||||||
|
"sync"
|
||||||
|
"time"
|
||||||
|
|
||||||
|
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/config"
|
||||||
|
"github.com/jackc/pgx/v5/pgxpool"
|
||||||
|
_ "github.com/microsoft/go-mssqldb"
|
||||||
|
log "github.com/sirupsen/logrus"
|
||||||
|
)
|
||||||
|
|
||||||
|
func connectToSqlServer() (*sql.DB, error) {
|
||||||
|
db, err := sql.Open("sqlserver", config.App.SourceDbUrl)
|
||||||
|
if err != nil {
|
||||||
|
return nil, fmt.Errorf("Unable to connect to sqlserver: %w", err)
|
||||||
|
}
|
||||||
|
|
||||||
|
ctx, cancel := context.WithTimeout(context.Background(), 20*time.Second)
|
||||||
|
defer cancel()
|
||||||
|
|
||||||
|
if err := db.PingContext(ctx); err != nil {
|
||||||
|
return nil, fmt.Errorf("Unable to ping sqlserver: %w", err)
|
||||||
|
}
|
||||||
|
|
||||||
|
return db, nil
|
||||||
|
}
|
||||||
|
|
||||||
|
func connectToPostgres() (*pgxpool.Pool, error) {
|
||||||
|
ctx, cancel := context.WithTimeout(context.Background(), 20*time.Second)
|
||||||
|
defer cancel()
|
||||||
|
|
||||||
|
pool, err := pgxpool.New(ctx, config.App.TargetDbUrl)
|
||||||
|
if err != nil {
|
||||||
|
return nil, fmt.Errorf("Unable to connect to postgres: %w", err)
|
||||||
|
}
|
||||||
|
|
||||||
|
if err := pool.Ping(ctx); err != nil {
|
||||||
|
pool.Close()
|
||||||
|
return nil, fmt.Errorf("Unable to ping postgres: %w", err)
|
||||||
|
}
|
||||||
|
|
||||||
|
return pool, nil
|
||||||
|
}
|
||||||
|
|
||||||
|
func connectToDatabases() (*sql.DB, *pgxpool.Pool, error) {
|
||||||
|
var sourceDbErr, targetDbErr error
|
||||||
|
var sourceDb *sql.DB
|
||||||
|
var targetDb *pgxpool.Pool
|
||||||
|
var wg sync.WaitGroup
|
||||||
|
|
||||||
|
wg.Go(func() {
|
||||||
|
sourceDb, sourceDbErr = connectToSqlServer()
|
||||||
|
if sourceDbErr != nil {
|
||||||
|
log.Error("Unable to connect to source db: ", sourceDbErr)
|
||||||
|
}
|
||||||
|
})
|
||||||
|
|
||||||
|
wg.Go(func() {
|
||||||
|
targetDb, targetDbErr = connectToPostgres()
|
||||||
|
if targetDbErr != nil {
|
||||||
|
log.Error("Unable to connect to target db: ", targetDbErr)
|
||||||
|
}
|
||||||
|
})
|
||||||
|
|
||||||
|
wg.Wait()
|
||||||
|
|
||||||
|
if sourceDbErr != nil || targetDbErr != nil {
|
||||||
|
return nil, nil, errors.New("Unable to connect to databases")
|
||||||
|
}
|
||||||
|
|
||||||
|
return sourceDb, targetDb, nil
|
||||||
|
}
|
||||||
@@ -3,7 +3,6 @@ package main
|
|||||||
import (
|
import (
|
||||||
"time"
|
"time"
|
||||||
|
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/config"
|
|
||||||
log "github.com/sirupsen/logrus"
|
log "github.com/sirupsen/logrus"
|
||||||
)
|
)
|
||||||
|
|
||||||
@@ -14,13 +13,5 @@ func configureLog() {
|
|||||||
DisableSorting: false,
|
DisableSorting: false,
|
||||||
PadLevelText: true,
|
PadLevelText: true,
|
||||||
})
|
})
|
||||||
|
log.SetLevel(log.DebugLevel)
|
||||||
logLevelEnv := config.App.LogLevel
|
|
||||||
logLevel, err := log.ParseLevel(logLevelEnv)
|
|
||||||
if err != nil {
|
|
||||||
log.Warnf("Nivel de log inválido '%s', usando INFO por defecto", logLevelEnv)
|
|
||||||
logLevel = log.InfoLevel
|
|
||||||
}
|
|
||||||
|
|
||||||
log.SetLevel(logLevel)
|
|
||||||
}
|
}
|
||||||
|
|||||||
@@ -2,18 +2,18 @@ package main
|
|||||||
|
|
||||||
import (
|
import (
|
||||||
"context"
|
"context"
|
||||||
|
"database/sql"
|
||||||
"sync"
|
"sync"
|
||||||
"time"
|
"time"
|
||||||
|
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/azure"
|
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/config"
|
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/config"
|
||||||
dbwrapper "git.ksdemosapps.com/kylesoda/go-migrate/internal/app/db-wrapper"
|
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/db"
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/etl/extractors"
|
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/etl/extractors"
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/etl/loaders"
|
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/etl/loaders"
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/etl/table_analyzers"
|
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/etl/table_analyzers"
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/models"
|
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/etl/transformers"
|
||||||
|
"github.com/jackc/pgx/v5/pgxpool"
|
||||||
log "github.com/sirupsen/logrus"
|
log "github.com/sirupsen/logrus"
|
||||||
"golang.org/x/sync/errgroup"
|
|
||||||
)
|
)
|
||||||
|
|
||||||
func main() {
|
func main() {
|
||||||
@@ -24,7 +24,7 @@ func main() {
|
|||||||
log.Fatalf("error leyendo configuracion: %v", err)
|
log.Fatalf("error leyendo configuracion: %v", err)
|
||||||
}
|
}
|
||||||
|
|
||||||
// log.Debugf("Config: %+v", migrationConfig)
|
log.Debugf("Config: %+v", migrationConfig)
|
||||||
|
|
||||||
startTime := time.Now()
|
startTime := time.Now()
|
||||||
|
|
||||||
@@ -33,33 +33,11 @@ func main() {
|
|||||||
|
|
||||||
log.Info("=== Starting migration ===")
|
log.Info("=== Starting migration ===")
|
||||||
|
|
||||||
var wgConnect errgroup.Group
|
sourceDb, targetDb, connError := connectToDatabases()
|
||||||
var sourceDb, targetDb dbwrapper.DbWrapper
|
if connError != nil {
|
||||||
|
log.Fatal("Connection error: ", connError)
|
||||||
wgConnect.Go(func() error {
|
|
||||||
var err error
|
|
||||||
sourceDb, err = connectWithTimeout(ctx, migrationConfig.SourceDbType, config.App.SourceDbUrl, 20*time.Second)
|
|
||||||
if err != nil {
|
|
||||||
return err
|
|
||||||
}
|
}
|
||||||
|
|
||||||
return nil
|
|
||||||
})
|
|
||||||
|
|
||||||
wgConnect.Go(func() error {
|
|
||||||
var err error
|
|
||||||
targetDb, err = connectWithTimeout(ctx, migrationConfig.TargetDbType, config.App.TargetDbUrl, 20*time.Second)
|
|
||||||
if err != nil {
|
|
||||||
return err
|
|
||||||
}
|
|
||||||
|
|
||||||
return nil
|
|
||||||
})
|
|
||||||
|
|
||||||
if err := wgConnect.Wait(); err != nil {
|
|
||||||
log.Error("Connection error: ", err)
|
|
||||||
return
|
|
||||||
}
|
|
||||||
defer sourceDb.Close()
|
defer sourceDb.Close()
|
||||||
defer targetDb.Close()
|
defer targetDb.Close()
|
||||||
|
|
||||||
@@ -92,14 +70,14 @@ func main() {
|
|||||||
|
|
||||||
func processMigrationJobs(
|
func processMigrationJobs(
|
||||||
ctx context.Context,
|
ctx context.Context,
|
||||||
sourceDb dbwrapper.DbWrapper,
|
sourceDb *sql.DB,
|
||||||
targetDb dbwrapper.DbWrapper,
|
targetDb *pgxpool.Pool,
|
||||||
jobs []config.Job,
|
jobs []config.Job,
|
||||||
maxParallelWorkers int,
|
maxParallelWorkers int,
|
||||||
) []models.JobResult {
|
) []JobResult {
|
||||||
if len(jobs) == 0 {
|
if len(jobs) == 0 {
|
||||||
log.Info("No migration jobs configured")
|
log.Info("No migration jobs configured")
|
||||||
return []models.JobResult{}
|
return []JobResult{}
|
||||||
}
|
}
|
||||||
|
|
||||||
if maxParallelWorkers <= 0 {
|
if maxParallelWorkers <= 0 {
|
||||||
@@ -112,23 +90,16 @@ func processMigrationJobs(
|
|||||||
|
|
||||||
log.Infof("Starting migration with %d parallel worker(s)", maxParallelWorkers)
|
log.Infof("Starting migration with %d parallel worker(s)", maxParallelWorkers)
|
||||||
|
|
||||||
chJobResults := make(chan models.JobResult, len(jobs))
|
chJobResults := make(chan JobResult, len(jobs))
|
||||||
chJobs := make(chan config.Job, len(jobs))
|
chJobs := make(chan config.Job, len(jobs))
|
||||||
var wgJobs sync.WaitGroup
|
var wgJobs sync.WaitGroup
|
||||||
|
|
||||||
|
targetDbWrapper := db.NewPostgresDbWrapper(targetDb)
|
||||||
sourceTableAnalyzer := table_analyzers.NewMssqlTableAnalyzer(sourceDb)
|
sourceTableAnalyzer := table_analyzers.NewMssqlTableAnalyzer(sourceDb)
|
||||||
targetTableAnalyzer := table_analyzers.NewPostgresTableAnalyzer(targetDb)
|
targetTableAnalyzer := table_analyzers.NewPostgresTableAnalyzer(targetDb)
|
||||||
extractor := extractors.NewMssqlExtractor(sourceDb)
|
extractor := extractors.NewMssqlExtractor(sourceDb)
|
||||||
loader := loaders.NewGenericLoader(targetDb)
|
transformer := transformers.NewMssqlTransformer()
|
||||||
|
loader := loaders.NewPostgresLoader(targetDb)
|
||||||
var azureClient *azure.Client
|
|
||||||
if config.App.AzureStorage.Enabled {
|
|
||||||
var err error
|
|
||||||
azureClient, err = azure.NewClient(config.App.AzureStorage)
|
|
||||||
if err != nil {
|
|
||||||
log.Fatalf("Failed to create Azure storage client: %v", err)
|
|
||||||
}
|
|
||||||
}
|
|
||||||
|
|
||||||
for i := range maxParallelWorkers {
|
for i := range maxParallelWorkers {
|
||||||
wgJobs.Go(func() {
|
wgJobs.Go(func() {
|
||||||
@@ -136,14 +107,13 @@ func processMigrationJobs(
|
|||||||
log.Infof("[worker %d] >>> Processing job: %s.%s <<<", i, job.SourceTable.Schema, job.SourceTable.Table)
|
log.Infof("[worker %d] >>> Processing job: %s.%s <<<", i, job.SourceTable.Schema, job.SourceTable.Table)
|
||||||
res := processMigrationJob(
|
res := processMigrationJob(
|
||||||
ctx,
|
ctx,
|
||||||
targetDb,
|
targetDbWrapper,
|
||||||
sourceTableAnalyzer,
|
sourceTableAnalyzer,
|
||||||
targetTableAnalyzer,
|
targetTableAnalyzer,
|
||||||
extractor,
|
extractor,
|
||||||
azureClient,
|
transformer,
|
||||||
loader,
|
loader,
|
||||||
job,
|
job,
|
||||||
targetDb.GetDialect(),
|
|
||||||
)
|
)
|
||||||
|
|
||||||
chJobResults <- res
|
chJobResults <- res
|
||||||
@@ -161,26 +131,10 @@ func processMigrationJobs(
|
|||||||
close(chJobResults)
|
close(chJobResults)
|
||||||
}()
|
}()
|
||||||
|
|
||||||
var finalResults []models.JobResult
|
var finalResults []JobResult
|
||||||
for res := range chJobResults {
|
for res := range chJobResults {
|
||||||
finalResults = append(finalResults, res)
|
finalResults = append(finalResults, res)
|
||||||
}
|
}
|
||||||
|
|
||||||
return finalResults
|
return finalResults
|
||||||
}
|
}
|
||||||
|
|
||||||
func connectWithTimeout(ctx context.Context, dbType string, dbUrl string, timeout time.Duration) (dbwrapper.DbWrapper, error) {
|
|
||||||
localCtx, cancel := context.WithTimeout(ctx, timeout)
|
|
||||||
defer cancel()
|
|
||||||
|
|
||||||
sourceDb, err := dbwrapper.New(dbType)
|
|
||||||
if err != nil {
|
|
||||||
return nil, err
|
|
||||||
}
|
|
||||||
|
|
||||||
if err = sourceDb.Connect(localCtx, dbUrl); err != nil {
|
|
||||||
return nil, err
|
|
||||||
}
|
|
||||||
|
|
||||||
return sourceDb, nil
|
|
||||||
}
|
|
||||||
|
|||||||
13
cmd/go_migrate/metrics.go
Normal file
13
cmd/go_migrate/metrics.go
Normal file
@@ -0,0 +1,13 @@
|
|||||||
|
package main
|
||||||
|
|
||||||
|
import "time"
|
||||||
|
|
||||||
|
type JobResult struct {
|
||||||
|
JobName string
|
||||||
|
StartTime time.Time
|
||||||
|
Duration time.Duration
|
||||||
|
RowsRead int64
|
||||||
|
RowsLoaded int64
|
||||||
|
RowsFailed int64
|
||||||
|
Error error
|
||||||
|
}
|
||||||
@@ -2,54 +2,35 @@ package main
|
|||||||
|
|
||||||
import (
|
import (
|
||||||
"context"
|
"context"
|
||||||
"fmt"
|
|
||||||
"sync"
|
"sync"
|
||||||
"sync/atomic"
|
"sync/atomic"
|
||||||
"time"
|
"time"
|
||||||
|
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/azure"
|
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/config"
|
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/config"
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/custom_errors"
|
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/custom_errors"
|
||||||
dbwrapper "git.ksdemosapps.com/kylesoda/go-migrate/internal/app/db-wrapper"
|
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/db"
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/etl"
|
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/etl"
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/etl/extractors"
|
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/etl/table_analyzers"
|
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/etl/table_analyzers"
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/etl/transformers"
|
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/models"
|
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/models"
|
||||||
log "github.com/sirupsen/logrus"
|
log "github.com/sirupsen/logrus"
|
||||||
"golang.org/x/sync/errgroup"
|
"golang.org/x/sync/errgroup"
|
||||||
)
|
)
|
||||||
|
|
||||||
func buildTruncateQuery(targetDbType, schema, table, truncateMethod string) string {
|
|
||||||
if truncateMethod == "DELETE" {
|
|
||||||
if targetDbType == "postgres" {
|
|
||||||
return fmt.Sprintf(`DELETE FROM "%s"."%s"`, schema, table)
|
|
||||||
}
|
|
||||||
return fmt.Sprintf(`DELETE FROM [%s].[%s]`, schema, table)
|
|
||||||
}
|
|
||||||
|
|
||||||
if targetDbType == "postgres" {
|
|
||||||
return fmt.Sprintf(`TRUNCATE TABLE "%s"."%s"`, schema, table)
|
|
||||||
}
|
|
||||||
return fmt.Sprintf(`TRUNCATE TABLE [%s].[%s]`, schema, table)
|
|
||||||
}
|
|
||||||
|
|
||||||
func processMigrationJob(
|
func processMigrationJob(
|
||||||
ctx context.Context,
|
ctx context.Context,
|
||||||
targetDbWrapper dbwrapper.DbWrapper,
|
// sourceDbWrapper db.DbWrapper,
|
||||||
|
targetDbWrapper db.DbWrapper,
|
||||||
sourceTableAnalyzer etl.TableAnalyzer,
|
sourceTableAnalyzer etl.TableAnalyzer,
|
||||||
targetTableAnalyzer etl.TableAnalyzer,
|
targetTableAnalyzer etl.TableAnalyzer,
|
||||||
extractor etl.Extractor,
|
extractor etl.Extractor,
|
||||||
azureClient *azure.Client,
|
transformer etl.Transformer,
|
||||||
loader etl.Loader,
|
loader etl.Loader,
|
||||||
job config.Job,
|
job config.Job,
|
||||||
targetDbType string,
|
) JobResult {
|
||||||
) models.JobResult {
|
|
||||||
transformer := transformers.NewMssqlTransformer(job.ToStorage, job.SourceTable, azureClient)
|
|
||||||
localCtx, cancel := context.WithCancel(ctx)
|
localCtx, cancel := context.WithCancel(ctx)
|
||||||
defer cancel()
|
defer cancel()
|
||||||
|
|
||||||
result := models.JobResult{
|
result := JobResult{
|
||||||
JobName: job.Name,
|
JobName: job.Name,
|
||||||
StartTime: time.Now(),
|
StartTime: time.Now(),
|
||||||
}
|
}
|
||||||
@@ -85,13 +66,7 @@ func processMigrationJob(
|
|||||||
return result
|
return result
|
||||||
}
|
}
|
||||||
|
|
||||||
preSqlQueries := job.TargetTable.PreSQL
|
for _, query := range job.PreSQL {
|
||||||
if job.TruncateTarget {
|
|
||||||
truncateQuery := buildTruncateQuery(targetDbType, job.TargetTable.Schema, job.TargetTable.Table, job.TruncateMethod)
|
|
||||||
preSqlQueries = append([]string{truncateQuery}, job.TargetTable.PreSQL...)
|
|
||||||
}
|
|
||||||
|
|
||||||
for _, query := range preSqlQueries {
|
|
||||||
if _, err := targetDbWrapper.Exec(localCtx, query); err != nil {
|
if _, err := targetDbWrapper.Exec(localCtx, query); err != nil {
|
||||||
result.Error = err
|
result.Error = err
|
||||||
return result
|
return result
|
||||||
@@ -104,7 +79,6 @@ func processMigrationJob(
|
|||||||
job.SourceTable.TableInfo,
|
job.SourceTable.TableInfo,
|
||||||
job.SourceTable.PrimaryKey,
|
job.SourceTable.PrimaryKey,
|
||||||
job.RowsPerPartition,
|
job.RowsPerPartition,
|
||||||
job.Range,
|
|
||||||
)
|
)
|
||||||
if err != nil {
|
if err != nil {
|
||||||
log.Error("Unexpected error calculating batch ranges: ", err)
|
log.Error("Unexpected error calculating batch ranges: ", err)
|
||||||
@@ -155,9 +129,8 @@ func processMigrationJob(
|
|||||||
|
|
||||||
for range maxExtractors {
|
for range maxExtractors {
|
||||||
wgExtractors.Go(func() {
|
wgExtractors.Go(func() {
|
||||||
extractors.Consume(
|
extractor.Exec(
|
||||||
localCtx,
|
localCtx,
|
||||||
extractor,
|
|
||||||
job.SourceTable,
|
job.SourceTable,
|
||||||
sourceColTypes,
|
sourceColTypes,
|
||||||
job.BatchSize,
|
job.BatchSize,
|
||||||
@@ -241,7 +214,7 @@ func processMigrationJob(
|
|||||||
cancel()
|
cancel()
|
||||||
}()
|
}()
|
||||||
|
|
||||||
for _, query := range job.TargetTable.PostSQL {
|
for _, query := range job.PostSQL {
|
||||||
if _, err := targetDbWrapper.Exec(localCtx, query); err != nil {
|
if _, err := targetDbWrapper.Exec(localCtx, query); err != nil {
|
||||||
result.Error = err
|
result.Error = err
|
||||||
return result
|
return result
|
||||||
@@ -261,9 +234,5 @@ func processMigrationJob(
|
|||||||
result.RowsLoaded = atomic.LoadInt64(&rowsLoaded)
|
result.RowsLoaded = atomic.LoadInt64(&rowsLoaded)
|
||||||
result.RowsFailed = atomic.LoadInt64(&rowsFailed)
|
result.RowsFailed = atomic.LoadInt64(&rowsFailed)
|
||||||
|
|
||||||
if result.RowsRead != result.RowsLoaded {
|
|
||||||
result.Error = fmt.Errorf("Row count mismatch: extracted %d rows but loaded %d rows (failed: %d)", result.RowsRead, result.RowsLoaded, result.RowsFailed)
|
|
||||||
}
|
|
||||||
|
|
||||||
return result
|
return result
|
||||||
}
|
}
|
||||||
|
|||||||
28
config.yaml
28
config.yaml
@@ -30,11 +30,6 @@ jobs:
|
|||||||
table: MANZANA
|
table: MANZANA
|
||||||
pre_sql:
|
pre_sql:
|
||||||
- 'SELECT 1'
|
- 'SELECT 1'
|
||||||
range:
|
|
||||||
min: 1000000
|
|
||||||
max: 2000000
|
|
||||||
is_min_inclusive: false
|
|
||||||
is_max_inclusive: true
|
|
||||||
|
|
||||||
- name: red_puerto
|
- name: red_puerto
|
||||||
enabled: true
|
enabled: true
|
||||||
@@ -49,26 +44,3 @@ jobs:
|
|||||||
- 'SELECT 1'
|
- 'SELECT 1'
|
||||||
post_sql:
|
post_sql:
|
||||||
- "SELECT 1"
|
- "SELECT 1"
|
||||||
|
|
||||||
- name: infraestructura_site_holder__attach
|
|
||||||
source:
|
|
||||||
schema: Infraestructura
|
|
||||||
table: SITE_HOLDER__ATTACH
|
|
||||||
primary_key: GDB_ARCHIVE_OID
|
|
||||||
target:
|
|
||||||
schema: Infraestructura
|
|
||||||
table: SITE_HOLDER__ATTACH
|
|
||||||
to_storage:
|
|
||||||
columns:
|
|
||||||
- source: DATA
|
|
||||||
target: FILE_URL
|
|
||||||
mode: REFERENCE_ONLY
|
|
||||||
max_extractors: 8
|
|
||||||
max_loaders: 4
|
|
||||||
queue_size: 32
|
|
||||||
batch_size: 1
|
|
||||||
retry:
|
|
||||||
attempts: 5
|
|
||||||
base_delay_ms: 1000
|
|
||||||
max_delay_ms: 15000
|
|
||||||
max_jitter_ms: 500
|
|
||||||
|
|||||||
14
go.mod
14
go.mod
@@ -1,13 +1,12 @@
|
|||||||
module git.ksdemosapps.com/kylesoda/go-migrate
|
module git.ksdemosapps.com/kylesoda/go-migrate
|
||||||
|
|
||||||
go 1.26
|
go 1.25.7
|
||||||
|
|
||||||
require (
|
require (
|
||||||
github.com/Azure/azure-sdk-for-go/sdk/storage/azblob v1.6.4
|
|
||||||
github.com/gaspardle/go-mssqlclrgeo v0.0.0-20160129143314-97ceabf987a4
|
github.com/gaspardle/go-mssqlclrgeo v0.0.0-20160129143314-97ceabf987a4
|
||||||
github.com/google/uuid v1.6.0
|
github.com/google/uuid v1.6.0
|
||||||
github.com/ilyakaznacheev/cleanenv v1.5.0
|
|
||||||
github.com/jackc/pgx/v5 v5.9.1
|
github.com/jackc/pgx/v5 v5.9.1
|
||||||
|
github.com/joho/godotenv v1.5.1
|
||||||
github.com/microsoft/go-mssqldb v1.9.8
|
github.com/microsoft/go-mssqldb v1.9.8
|
||||||
github.com/sirupsen/logrus v1.9.4
|
github.com/sirupsen/logrus v1.9.4
|
||||||
github.com/twpayne/go-geom v1.6.1
|
github.com/twpayne/go-geom v1.6.1
|
||||||
@@ -16,20 +15,17 @@ require (
|
|||||||
)
|
)
|
||||||
|
|
||||||
require (
|
require (
|
||||||
github.com/Azure/azure-sdk-for-go/sdk/azcore v1.21.0 // indirect
|
github.com/cenkalti/backoff v2.2.1+incompatible // indirect
|
||||||
github.com/Azure/azure-sdk-for-go/sdk/internal v1.11.2 // indirect
|
github.com/cenkalti/backoff/v5 v5.0.3 // indirect
|
||||||
github.com/BurntSushi/toml v1.6.0 // indirect
|
|
||||||
github.com/golang-sql/civil v0.0.0-20220223132316-b832511892a9 // indirect
|
github.com/golang-sql/civil v0.0.0-20220223132316-b832511892a9 // indirect
|
||||||
github.com/golang-sql/sqlexp v0.1.0 // indirect
|
github.com/golang-sql/sqlexp v0.1.0 // indirect
|
||||||
github.com/jackc/pgpassfile v1.0.0 // indirect
|
github.com/jackc/pgpassfile v1.0.0 // indirect
|
||||||
github.com/jackc/pgservicefile v0.0.0-20240606120523-5a60cdf6a761 // indirect
|
github.com/jackc/pgservicefile v0.0.0-20240606120523-5a60cdf6a761 // indirect
|
||||||
github.com/jackc/puddle/v2 v2.2.2 // indirect
|
github.com/jackc/puddle/v2 v2.2.2 // indirect
|
||||||
github.com/joho/godotenv v1.5.1 // indirect
|
github.com/kr/text v0.2.0 // indirect
|
||||||
github.com/rogpeppe/go-internal v1.14.1 // indirect
|
github.com/rogpeppe/go-internal v1.14.1 // indirect
|
||||||
github.com/shopspring/decimal v1.4.0 // indirect
|
github.com/shopspring/decimal v1.4.0 // indirect
|
||||||
golang.org/x/crypto v0.48.0 // indirect
|
golang.org/x/crypto v0.48.0 // indirect
|
||||||
golang.org/x/net v0.51.0 // indirect
|
|
||||||
golang.org/x/sys v0.41.0 // indirect
|
golang.org/x/sys v0.41.0 // indirect
|
||||||
golang.org/x/text v0.34.0 // indirect
|
golang.org/x/text v0.34.0 // indirect
|
||||||
olympos.io/encoding/edn v0.0.0-20201019073823-d3554ca0b0a3 // indirect
|
|
||||||
)
|
)
|
||||||
|
|||||||
20
go.sum
20
go.sum
@@ -4,25 +4,23 @@ github.com/Azure/azure-sdk-for-go/sdk/azidentity v1.13.1 h1:Hk5QBxZQC1jb2Fwj6mpz
|
|||||||
github.com/Azure/azure-sdk-for-go/sdk/azidentity v1.13.1/go.mod h1:IYus9qsFobWIc2YVwe/WPjcnyCkPKtnHAqUYeebc8z0=
|
github.com/Azure/azure-sdk-for-go/sdk/azidentity v1.13.1/go.mod h1:IYus9qsFobWIc2YVwe/WPjcnyCkPKtnHAqUYeebc8z0=
|
||||||
github.com/Azure/azure-sdk-for-go/sdk/internal v1.11.2 h1:9iefClla7iYpfYWdzPCRDozdmndjTm8DXdpCzPajMgA=
|
github.com/Azure/azure-sdk-for-go/sdk/internal v1.11.2 h1:9iefClla7iYpfYWdzPCRDozdmndjTm8DXdpCzPajMgA=
|
||||||
github.com/Azure/azure-sdk-for-go/sdk/internal v1.11.2/go.mod h1:XtLgD3ZD34DAaVIIAyG3objl5DynM3CQ/vMcbBNJZGI=
|
github.com/Azure/azure-sdk-for-go/sdk/internal v1.11.2/go.mod h1:XtLgD3ZD34DAaVIIAyG3objl5DynM3CQ/vMcbBNJZGI=
|
||||||
github.com/Azure/azure-sdk-for-go/sdk/resourcemanager/storage/armstorage v1.8.1 h1:/Zt+cDPnpC3OVDm/JKLOs7M2DKmLRIIp3XIx9pHHiig=
|
|
||||||
github.com/Azure/azure-sdk-for-go/sdk/resourcemanager/storage/armstorage v1.8.1/go.mod h1:Ng3urmn6dYe8gnbCMoHHVl5APYz2txho3koEkV2o2HA=
|
|
||||||
github.com/Azure/azure-sdk-for-go/sdk/security/keyvault/azkeys v1.4.0 h1:E4MgwLBGeVB5f2MdcIVD3ELVAWpr+WD6MUe1i+tM/PA=
|
github.com/Azure/azure-sdk-for-go/sdk/security/keyvault/azkeys v1.4.0 h1:E4MgwLBGeVB5f2MdcIVD3ELVAWpr+WD6MUe1i+tM/PA=
|
||||||
github.com/Azure/azure-sdk-for-go/sdk/security/keyvault/azkeys v1.4.0/go.mod h1:Y2b/1clN4zsAoUd/pgNAQHjLDnTis/6ROkUfyob6psM=
|
github.com/Azure/azure-sdk-for-go/sdk/security/keyvault/azkeys v1.4.0/go.mod h1:Y2b/1clN4zsAoUd/pgNAQHjLDnTis/6ROkUfyob6psM=
|
||||||
github.com/Azure/azure-sdk-for-go/sdk/security/keyvault/internal v1.2.0 h1:nCYfgcSyHZXJI8J0IWE5MsCGlb2xp9fJiXyxWgmOFg4=
|
github.com/Azure/azure-sdk-for-go/sdk/security/keyvault/internal v1.2.0 h1:nCYfgcSyHZXJI8J0IWE5MsCGlb2xp9fJiXyxWgmOFg4=
|
||||||
github.com/Azure/azure-sdk-for-go/sdk/security/keyvault/internal v1.2.0/go.mod h1:ucUjca2JtSZboY8IoUqyQyuuXvwbMBVwFOm0vdQPNhA=
|
github.com/Azure/azure-sdk-for-go/sdk/security/keyvault/internal v1.2.0/go.mod h1:ucUjca2JtSZboY8IoUqyQyuuXvwbMBVwFOm0vdQPNhA=
|
||||||
github.com/Azure/azure-sdk-for-go/sdk/storage/azblob v1.6.4 h1:jWQK1GI+LeGGUKBADtcH2rRqPxYB1Ljwms5gFA2LqrM=
|
|
||||||
github.com/Azure/azure-sdk-for-go/sdk/storage/azblob v1.6.4/go.mod h1:8mwH4klAm9DUgR2EEHyEEAQlRDvLPyg5fQry3y+cDew=
|
|
||||||
github.com/AzureAD/microsoft-authentication-library-for-go v1.6.0 h1:XRzhVemXdgvJqCH0sFfrBUTnUJSBrBf7++ypk+twtRs=
|
github.com/AzureAD/microsoft-authentication-library-for-go v1.6.0 h1:XRzhVemXdgvJqCH0sFfrBUTnUJSBrBf7++ypk+twtRs=
|
||||||
github.com/AzureAD/microsoft-authentication-library-for-go v1.6.0/go.mod h1:HKpQxkWaGLJ+D/5H8QRpyQXA1eKjxkFlOMwck5+33Jk=
|
github.com/AzureAD/microsoft-authentication-library-for-go v1.6.0/go.mod h1:HKpQxkWaGLJ+D/5H8QRpyQXA1eKjxkFlOMwck5+33Jk=
|
||||||
github.com/BurntSushi/toml v1.2.1/go.mod h1:CxXYINrC8qIiEnFrOxCa7Jy5BFHlXnUU2pbicEuybxQ=
|
|
||||||
github.com/BurntSushi/toml v1.6.0 h1:dRaEfpa2VI55EwlIW72hMRHdWouJeRF7TPYhI+AUQjk=
|
|
||||||
github.com/BurntSushi/toml v1.6.0/go.mod h1:ukJfTF/6rtPPRCnwkur4qwRxa8vTRFBF0uk2lLoLwho=
|
|
||||||
github.com/DATA-DOG/go-sqlmock v1.5.2 h1:OcvFkGmslmlZibjAjaHm3L//6LiuBgolP7OputlJIzU=
|
github.com/DATA-DOG/go-sqlmock v1.5.2 h1:OcvFkGmslmlZibjAjaHm3L//6LiuBgolP7OputlJIzU=
|
||||||
github.com/DATA-DOG/go-sqlmock v1.5.2/go.mod h1:88MAG/4G7SMwSE3CeA0ZKzrT5CiOU3OJ+JlNzwDqpNU=
|
github.com/DATA-DOG/go-sqlmock v1.5.2/go.mod h1:88MAG/4G7SMwSE3CeA0ZKzrT5CiOU3OJ+JlNzwDqpNU=
|
||||||
github.com/alecthomas/assert/v2 v2.10.0 h1:jjRCHsj6hBJhkmhznrCzoNpbA3zqy0fYiUcYZP/GkPY=
|
github.com/alecthomas/assert/v2 v2.10.0 h1:jjRCHsj6hBJhkmhznrCzoNpbA3zqy0fYiUcYZP/GkPY=
|
||||||
github.com/alecthomas/assert/v2 v2.10.0/go.mod h1:Bze95FyfUr7x34QZrjL+XP+0qgp/zg8yS+TtBj1WA3k=
|
github.com/alecthomas/assert/v2 v2.10.0/go.mod h1:Bze95FyfUr7x34QZrjL+XP+0qgp/zg8yS+TtBj1WA3k=
|
||||||
github.com/alecthomas/repr v0.4.0 h1:GhI2A8MACjfegCPVq9f1FLvIBS+DrQ2KQBFZP1iFzXc=
|
github.com/alecthomas/repr v0.4.0 h1:GhI2A8MACjfegCPVq9f1FLvIBS+DrQ2KQBFZP1iFzXc=
|
||||||
github.com/alecthomas/repr v0.4.0/go.mod h1:Fr0507jx4eOXV7AlPV6AVZLYrLIuIeSOWtW57eE/O/4=
|
github.com/alecthomas/repr v0.4.0/go.mod h1:Fr0507jx4eOXV7AlPV6AVZLYrLIuIeSOWtW57eE/O/4=
|
||||||
|
github.com/cenkalti/backoff v2.2.1+incompatible h1:tNowT99t7UNflLxfYYSlKYsBpXdEet03Pg2g16Swow4=
|
||||||
|
github.com/cenkalti/backoff v2.2.1+incompatible/go.mod h1:90ReRw6GdpyfrHakVjL/QHaoyV4aDUVVkXQJJJ3NXXM=
|
||||||
|
github.com/cenkalti/backoff/v5 v5.0.3 h1:ZN+IMa753KfX5hd8vVaMixjnqRZ3y8CuJKRKj1xcsSM=
|
||||||
|
github.com/cenkalti/backoff/v5 v5.0.3/go.mod h1:rkhZdG3JZukswDf7f0cwqPNk4K0sa+F97BxZthm/crw=
|
||||||
|
github.com/creack/pty v1.1.9/go.mod h1:oKZEueFk5CKHvIhNR5MUki03XCEU+Q6VDXinZuGJ33E=
|
||||||
github.com/davecgh/go-spew v1.1.0/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
|
github.com/davecgh/go-spew v1.1.0/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
|
||||||
github.com/davecgh/go-spew v1.1.1 h1:vj9j/u1bqnvCEfJOwUhtlOARqs3+rkHYY13jYWTU97c=
|
github.com/davecgh/go-spew v1.1.1 h1:vj9j/u1bqnvCEfJOwUhtlOARqs3+rkHYY13jYWTU97c=
|
||||||
github.com/davecgh/go-spew v1.1.1/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
|
github.com/davecgh/go-spew v1.1.1/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
|
||||||
@@ -38,8 +36,6 @@ github.com/google/uuid v1.6.0 h1:NIvaJDMOsjHA8n1jAhLSgzrAzy1Hgr+hNrb57e+94F0=
|
|||||||
github.com/google/uuid v1.6.0/go.mod h1:TIyPZe4MgqvfeYDBFedMoGGpEw/LqOeaOT+nhxU+yHo=
|
github.com/google/uuid v1.6.0/go.mod h1:TIyPZe4MgqvfeYDBFedMoGGpEw/LqOeaOT+nhxU+yHo=
|
||||||
github.com/hexops/gotextdiff v1.0.3 h1:gitA9+qJrrTCsiCl7+kh75nPqQt1cx4ZkudSTLoUqJM=
|
github.com/hexops/gotextdiff v1.0.3 h1:gitA9+qJrrTCsiCl7+kh75nPqQt1cx4ZkudSTLoUqJM=
|
||||||
github.com/hexops/gotextdiff v1.0.3/go.mod h1:pSWU5MAI3yDq+fZBTazCSJysOMbxWL1BSow5/V2vxeg=
|
github.com/hexops/gotextdiff v1.0.3/go.mod h1:pSWU5MAI3yDq+fZBTazCSJysOMbxWL1BSow5/V2vxeg=
|
||||||
github.com/ilyakaznacheev/cleanenv v1.5.0 h1:0VNZXggJE2OYdXE87bfSSwGxeiGt9moSR2lOrsHHvr4=
|
|
||||||
github.com/ilyakaznacheev/cleanenv v1.5.0/go.mod h1:a5aDzaJrLCQZsazHol1w8InnDcOX0OColm64SlIi6gk=
|
|
||||||
github.com/jackc/pgpassfile v1.0.0 h1:/6Hmqy13Ss2zCq62VdNG8tM1wchn8zjSGOBJ6icpsIM=
|
github.com/jackc/pgpassfile v1.0.0 h1:/6Hmqy13Ss2zCq62VdNG8tM1wchn8zjSGOBJ6icpsIM=
|
||||||
github.com/jackc/pgpassfile v1.0.0/go.mod h1:CEx0iS5ambNFdcRtxPj5JhEz+xB6uRky5eyVu/W2HEg=
|
github.com/jackc/pgpassfile v1.0.0/go.mod h1:CEx0iS5ambNFdcRtxPj5JhEz+xB6uRky5eyVu/W2HEg=
|
||||||
github.com/jackc/pgservicefile v0.0.0-20240606120523-5a60cdf6a761 h1:iCEnooe7UlwOQYpKFhBabPMi4aNAfoODPEFNiAnClxo=
|
github.com/jackc/pgservicefile v0.0.0-20240606120523-5a60cdf6a761 h1:iCEnooe7UlwOQYpKFhBabPMi4aNAfoODPEFNiAnClxo=
|
||||||
@@ -50,8 +46,8 @@ github.com/jackc/puddle/v2 v2.2.2 h1:PR8nw+E/1w0GLuRFSmiioY6UooMp6KJv0/61nB7icHo
|
|||||||
github.com/jackc/puddle/v2 v2.2.2/go.mod h1:vriiEXHvEE654aYKXXjOvZM39qJ0q+azkZFrfEOc3H4=
|
github.com/jackc/puddle/v2 v2.2.2/go.mod h1:vriiEXHvEE654aYKXXjOvZM39qJ0q+azkZFrfEOc3H4=
|
||||||
github.com/joho/godotenv v1.5.1 h1:7eLL/+HRGLY0ldzfGMeQkb7vMd0as4CfYvUVzLqw0N0=
|
github.com/joho/godotenv v1.5.1 h1:7eLL/+HRGLY0ldzfGMeQkb7vMd0as4CfYvUVzLqw0N0=
|
||||||
github.com/joho/godotenv v1.5.1/go.mod h1:f4LDr5Voq0i2e/R5DDNOoa2zzDfwtkZa6DnEwAbqwq4=
|
github.com/joho/godotenv v1.5.1/go.mod h1:f4LDr5Voq0i2e/R5DDNOoa2zzDfwtkZa6DnEwAbqwq4=
|
||||||
github.com/kr/pretty v0.3.1 h1:flRD4NNwYAUpkphVc1HcthR4KEIFJ65n8Mw5qdRn3LE=
|
github.com/kr/pretty v0.3.0 h1:WgNl7dwNpEZ6jJ9k1snq4pZsg7DOEN8hP9Xw0Tsjwk0=
|
||||||
github.com/kr/pretty v0.3.1/go.mod h1:hoEshYVHaxMs3cyo3Yncou5ZscifuDolrwPKZanG3xk=
|
github.com/kr/pretty v0.3.0/go.mod h1:640gp4NfQd8pI5XOwp5fnNeVWj67G7CFk/SaSQn7NBk=
|
||||||
github.com/kr/text v0.2.0 h1:5Nx0Ya0ZqY2ygV366QzturHI13Jq95ApcVaJBhpS+AY=
|
github.com/kr/text v0.2.0 h1:5Nx0Ya0ZqY2ygV366QzturHI13Jq95ApcVaJBhpS+AY=
|
||||||
github.com/kr/text v0.2.0/go.mod h1:eLer722TekiGuMkidMxC/pM04lWEeraHUUmBw8l2grE=
|
github.com/kr/text v0.2.0/go.mod h1:eLer722TekiGuMkidMxC/pM04lWEeraHUUmBw8l2grE=
|
||||||
github.com/kylelemons/godebug v1.1.0 h1:RPNrshWIDI6G2gRW9EHilWtl7Z6Sb1BR0xunSBf0SNc=
|
github.com/kylelemons/godebug v1.1.0 h1:RPNrshWIDI6G2gRW9EHilWtl7Z6Sb1BR0xunSBf0SNc=
|
||||||
@@ -91,5 +87,3 @@ gopkg.in/check.v1 v1.0.0-20201130134442-10cb98267c6c/go.mod h1:JHkPIbrfpd72SG/EV
|
|||||||
gopkg.in/yaml.v3 v3.0.0-20200313102051-9f266ea9e77c/go.mod h1:K4uyk7z7BCEPqu6E+C64Yfv1cQ7kz7rIZviUmN+EgEM=
|
gopkg.in/yaml.v3 v3.0.0-20200313102051-9f266ea9e77c/go.mod h1:K4uyk7z7BCEPqu6E+C64Yfv1cQ7kz7rIZviUmN+EgEM=
|
||||||
gopkg.in/yaml.v3 v3.0.1 h1:fxVm/GzAzEWqLHuvctI91KS9hhNmmWOoWu0XTYJS7CA=
|
gopkg.in/yaml.v3 v3.0.1 h1:fxVm/GzAzEWqLHuvctI91KS9hhNmmWOoWu0XTYJS7CA=
|
||||||
gopkg.in/yaml.v3 v3.0.1/go.mod h1:K4uyk7z7BCEPqu6E+C64Yfv1cQ7kz7rIZviUmN+EgEM=
|
gopkg.in/yaml.v3 v3.0.1/go.mod h1:K4uyk7z7BCEPqu6E+C64Yfv1cQ7kz7rIZviUmN+EgEM=
|
||||||
olympos.io/encoding/edn v0.0.0-20201019073823-d3554ca0b0a3 h1:slmdOY3vp8a7KQbHkL+FLbvbkgMqmXojpFUO/jENuqQ=
|
|
||||||
olympos.io/encoding/edn v0.0.0-20201019073823-d3554ca0b0a3/go.mod h1:oVgVk4OWVDi43qWBEyGhXgYxt7+ED4iYNpTngSLX2Iw=
|
|
||||||
|
|||||||
@@ -1,87 +0,0 @@
|
|||||||
package azure
|
|
||||||
|
|
||||||
import (
|
|
||||||
"context"
|
|
||||||
"errors"
|
|
||||||
"fmt"
|
|
||||||
"net/url"
|
|
||||||
|
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/config"
|
|
||||||
"github.com/Azure/azure-sdk-for-go/sdk/storage/azblob"
|
|
||||||
)
|
|
||||||
|
|
||||||
var (
|
|
||||||
ErrInvalidConnectionString = errors.New("invalid connection string")
|
|
||||||
ErrContainerNotFound = errors.New("container not found")
|
|
||||||
ErrBlobNotFound = errors.New("blob not found")
|
|
||||||
ErrInvalidInput = errors.New("invalid input parameters")
|
|
||||||
)
|
|
||||||
|
|
||||||
type Client struct {
|
|
||||||
client *azblob.Client
|
|
||||||
azureStorageConfig config.AzureStorageConfig
|
|
||||||
}
|
|
||||||
|
|
||||||
func NewClient(azureStorageConfig config.AzureStorageConfig) (*Client, error) {
|
|
||||||
protocol := "https"
|
|
||||||
if !azureStorageConfig.UseHTTPS {
|
|
||||||
protocol = "http"
|
|
||||||
}
|
|
||||||
|
|
||||||
blobEndpoint, _ := url.JoinPath(azureStorageConfig.ServiceURL, azureStorageConfig.AccountName)
|
|
||||||
connStr := fmt.Sprintf("DefaultEndpointsProtocol=%s;AccountName=%s;AccountKey=%s;BlobEndpoint=%s;",
|
|
||||||
protocol, azureStorageConfig.AccountName, azureStorageConfig.AccountKey, blobEndpoint)
|
|
||||||
|
|
||||||
client, err := azblob.NewClientFromConnectionString(connStr, nil)
|
|
||||||
if err != nil {
|
|
||||||
return nil, fmt.Errorf("creating azure storage client: %w", err)
|
|
||||||
}
|
|
||||||
|
|
||||||
return &Client{
|
|
||||||
client: client,
|
|
||||||
azureStorageConfig: azureStorageConfig,
|
|
||||||
}, nil
|
|
||||||
}
|
|
||||||
|
|
||||||
func (c *Client) CreateContainer(ctx context.Context, containerName string) error {
|
|
||||||
if containerName == "" {
|
|
||||||
return ErrInvalidInput
|
|
||||||
}
|
|
||||||
|
|
||||||
_, err := c.client.CreateContainer(ctx, containerName, nil)
|
|
||||||
if err != nil {
|
|
||||||
return fmt.Errorf("creating container %s: %w", containerName, err)
|
|
||||||
}
|
|
||||||
return nil
|
|
||||||
}
|
|
||||||
|
|
||||||
func (c *Client) UploadBuffer(ctx context.Context, containerName, blobPath string, buffer []byte) error {
|
|
||||||
if containerName == "" || blobPath == "" || buffer == nil {
|
|
||||||
return ErrInvalidInput
|
|
||||||
}
|
|
||||||
|
|
||||||
_, err := c.client.UploadBuffer(ctx, containerName, blobPath, buffer, nil)
|
|
||||||
if err != nil {
|
|
||||||
return fmt.Errorf("uploading blob %s: %w", blobPath, err)
|
|
||||||
}
|
|
||||||
return nil
|
|
||||||
}
|
|
||||||
|
|
||||||
func (c *Client) UploadAndGetURL(ctx context.Context, blobPath string, buffer []byte) (string, error) {
|
|
||||||
if blobPath == "" || buffer == nil {
|
|
||||||
return "", ErrInvalidInput
|
|
||||||
}
|
|
||||||
|
|
||||||
fullPath := blobPath
|
|
||||||
if c.azureStorageConfig.Prefix != "" {
|
|
||||||
fullPath, _ = url.JoinPath(c.azureStorageConfig.Prefix, blobPath)
|
|
||||||
}
|
|
||||||
|
|
||||||
if err := c.UploadBuffer(ctx, c.azureStorageConfig.Container, fullPath, buffer); err != nil {
|
|
||||||
return "", err
|
|
||||||
}
|
|
||||||
|
|
||||||
blobEndpoint, _ := url.JoinPath(c.azureStorageConfig.ServiceURL, c.azureStorageConfig.AccountName)
|
|
||||||
blobURL, _ := url.JoinPath(blobEndpoint, c.azureStorageConfig.Container, fullPath)
|
|
||||||
return blobURL, nil
|
|
||||||
}
|
|
||||||
@@ -1,41 +1,41 @@
|
|||||||
package config
|
package config
|
||||||
|
|
||||||
import (
|
import (
|
||||||
"github.com/ilyakaznacheev/cleanenv"
|
"os"
|
||||||
|
|
||||||
|
"github.com/joho/godotenv"
|
||||||
log "github.com/sirupsen/logrus"
|
log "github.com/sirupsen/logrus"
|
||||||
)
|
)
|
||||||
|
|
||||||
type AzureStorageConfig struct {
|
type appConfig struct {
|
||||||
AccountName string `env:"AZ_ACCOUNT_NAME"`
|
SourceDbUrl string
|
||||||
Container string `env:"AZ_CONTAINER"`
|
TargetDbUrl string
|
||||||
AccountKey string `env:"AZ_ACCOUNT_KEY"`
|
|
||||||
UseHTTPS bool `env:"AZ_USE_HTTPS" env-default:"true"`
|
|
||||||
ServiceURL string `env:"AZ_SERVICE_URL"`
|
|
||||||
Prefix string `env:"AZ_PREFIX"`
|
|
||||||
Enabled bool `env:"AZ_STORAGE_ENABLED"`
|
|
||||||
}
|
}
|
||||||
|
|
||||||
type appConfig struct {
|
func loadEnv() {
|
||||||
SourceDbUrl string `env:"SOURCE_DB_URL" env-required:"true"`
|
err := godotenv.Load()
|
||||||
TargetDbUrl string `env:"TARGET_DB_URL" env-required:"true"`
|
if err != nil {
|
||||||
LogLevel string `env:"LOG_LEVEL" env-default:"INFO"`
|
log.Warn("Warning: could not load .env file")
|
||||||
AzureStorage AzureStorageConfig
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
func getAppConfig() appConfig {
|
func getAppConfig() appConfig {
|
||||||
var cfg appConfig
|
loadEnv()
|
||||||
|
|
||||||
err := cleanenv.ReadConfig(".env", &cfg)
|
sourceDbUrl := os.Getenv("SOURCE_DB_URL")
|
||||||
if err != nil {
|
if sourceDbUrl == "" {
|
||||||
log.Warn("Could not load .env file")
|
log.Fatal("SOURCE_DB_URL environment variable not set")
|
||||||
}
|
}
|
||||||
|
|
||||||
err = cleanenv.ReadEnv(&cfg)
|
targetDbUrl := os.Getenv("TARGET_DB_URL")
|
||||||
if err != nil {
|
if targetDbUrl == "" {
|
||||||
log.Fatalf("Error al cargar variables: %v", err)
|
log.Fatal("TARGET_DB_URL environment variable not set")
|
||||||
}
|
}
|
||||||
|
|
||||||
return cfg
|
return appConfig{
|
||||||
|
SourceDbUrl: sourceDbUrl,
|
||||||
|
TargetDbUrl: targetDbUrl,
|
||||||
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
var App appConfig = getAppConfig()
|
var App appConfig = getAppConfig()
|
||||||
|
|||||||
@@ -14,16 +14,6 @@ type RetryConfig struct {
|
|||||||
MaxJitterMs int `yaml:"max_jitter_ms"`
|
MaxJitterMs int `yaml:"max_jitter_ms"`
|
||||||
}
|
}
|
||||||
|
|
||||||
type ToStorageColumnConfig struct {
|
|
||||||
Source string `yaml:"source"`
|
|
||||||
Target string `yaml:"target"`
|
|
||||||
Mode string `yaml:"mode"`
|
|
||||||
}
|
|
||||||
|
|
||||||
type ToStorageConfig struct {
|
|
||||||
Columns []ToStorageColumnConfig `yaml:"columns"`
|
|
||||||
}
|
|
||||||
|
|
||||||
type JobConfig struct {
|
type JobConfig struct {
|
||||||
MaxExtractors int `yaml:"max_extractors"`
|
MaxExtractors int `yaml:"max_extractors"`
|
||||||
MaxLoaders int `yaml:"max_loaders"`
|
MaxLoaders int `yaml:"max_loaders"`
|
||||||
@@ -36,7 +26,6 @@ type JobConfig struct {
|
|||||||
MaxChunkErrors int `yaml:"max_chunk_errors"`
|
MaxChunkErrors int `yaml:"max_chunk_errors"`
|
||||||
Retry RetryConfig `yaml:"retry"`
|
Retry RetryConfig `yaml:"retry"`
|
||||||
RowsPerPartition int64
|
RowsPerPartition int64
|
||||||
ToStorage ToStorageConfig `yaml:"to_storage"`
|
|
||||||
}
|
}
|
||||||
|
|
||||||
type TableInfo struct {
|
type TableInfo struct {
|
||||||
@@ -44,31 +33,23 @@ type TableInfo struct {
|
|||||||
Table string `yaml:"table"`
|
Table string `yaml:"table"`
|
||||||
}
|
}
|
||||||
|
|
||||||
|
type TargetTableInfo struct {
|
||||||
|
TableInfo `yaml:",inline"`
|
||||||
|
}
|
||||||
|
|
||||||
type SourceTableInfo struct {
|
type SourceTableInfo struct {
|
||||||
TableInfo `yaml:",inline"`
|
TableInfo `yaml:",inline"`
|
||||||
PrimaryKey string `yaml:"primary_key"`
|
PrimaryKey string `yaml:"primary_key"`
|
||||||
}
|
}
|
||||||
|
|
||||||
type TargetTableInfo struct {
|
|
||||||
TableInfo `yaml:",inline"`
|
|
||||||
PreSQL []string `yaml:"pre_sql"`
|
|
||||||
PostSQL []string `yaml:"post_sql"`
|
|
||||||
}
|
|
||||||
|
|
||||||
type RangeConfig struct {
|
|
||||||
Min int64 `yaml:"min"`
|
|
||||||
Max int64 `yaml:"max"`
|
|
||||||
IsMinInclusive bool `yaml:"is_min_inclusive"`
|
|
||||||
IsMaxInclusive bool `yaml:"is_max_inclusive"`
|
|
||||||
}
|
|
||||||
|
|
||||||
type Job struct {
|
type Job struct {
|
||||||
Name string `yaml:"name"`
|
Name string `yaml:"name"`
|
||||||
Enabled bool `yaml:"enabled"`
|
Enabled bool `yaml:"enabled"`
|
||||||
SourceTable SourceTableInfo `yaml:"source"`
|
SourceTable SourceTableInfo `yaml:"source"`
|
||||||
TargetTable TargetTableInfo `yaml:"target"`
|
TargetTable TargetTableInfo `yaml:"target"`
|
||||||
|
PreSQL []string `yaml:"pre_sql"`
|
||||||
|
PostSQL []string `yaml:"post_sql"`
|
||||||
JobConfig `yaml:",inline"`
|
JobConfig `yaml:",inline"`
|
||||||
Range RangeConfig `yaml:"range"`
|
|
||||||
}
|
}
|
||||||
|
|
||||||
type MigrationConfig struct {
|
type MigrationConfig struct {
|
||||||
@@ -95,8 +76,6 @@ func (c *MigrationConfig) UnmarshalYAML(value *yaml.Node) error {
|
|||||||
|
|
||||||
c.MaxParallelWorkers = raw.MaxParallelWorkers
|
c.MaxParallelWorkers = raw.MaxParallelWorkers
|
||||||
c.Defaults = raw.Defaults
|
c.Defaults = raw.Defaults
|
||||||
c.SourceDbType = raw.SourceDbType
|
|
||||||
c.TargetDbType = raw.TargetDbType
|
|
||||||
c.Defaults.RowsPerPartition = int64(raw.Defaults.BatchSize * raw.Defaults.BatchesPerPartition)
|
c.Defaults.RowsPerPartition = int64(raw.Defaults.BatchSize * raw.Defaults.BatchesPerPartition)
|
||||||
|
|
||||||
for _, node := range raw.Jobs {
|
for _, node := range raw.Jobs {
|
||||||
|
|||||||
@@ -103,8 +103,8 @@ func ExtractorErrorHandler(
|
|||||||
if err.HasLastId {
|
if err.HasLastId {
|
||||||
newPartition.ParentId = err.Partition.Id
|
newPartition.ParentId = err.Partition.Id
|
||||||
newPartition.Id = uuid.New()
|
newPartition.Id = uuid.New()
|
||||||
newPartition.Range.Min = err.LastId
|
newPartition.LowerLimit = err.LastId
|
||||||
newPartition.Range.IsMinInclusive = false
|
newPartition.IsLowerLimitInclusive = false
|
||||||
}
|
}
|
||||||
|
|
||||||
requeueWithBackoff(ctx, delay, func() {
|
requeueWithBackoff(ctx, delay, func() {
|
||||||
|
|||||||
@@ -1,19 +0,0 @@
|
|||||||
package dbwrapper
|
|
||||||
|
|
||||||
import "fmt"
|
|
||||||
|
|
||||||
type Factory func() DbWrapper
|
|
||||||
|
|
||||||
var drivers = make(map[string]Factory)
|
|
||||||
|
|
||||||
func Register(name string, factory Factory) {
|
|
||||||
drivers[name] = factory
|
|
||||||
}
|
|
||||||
|
|
||||||
func New(driverType string) (DbWrapper, error) {
|
|
||||||
factory, ok := drivers[driverType]
|
|
||||||
if !ok {
|
|
||||||
return nil, fmt.Errorf("driver not yet supported: %s", driverType)
|
|
||||||
}
|
|
||||||
return factory(), nil
|
|
||||||
}
|
|
||||||
@@ -1,176 +0,0 @@
|
|||||||
package dbwrapper
|
|
||||||
|
|
||||||
import (
|
|
||||||
"context"
|
|
||||||
"database/sql"
|
|
||||||
"fmt"
|
|
||||||
|
|
||||||
mssql "github.com/microsoft/go-mssqldb"
|
|
||||||
)
|
|
||||||
|
|
||||||
func init() {
|
|
||||||
Register("sqlserver", func() DbWrapper {
|
|
||||||
return &mssqlDbWrapper{dialect: "sqlserver"}
|
|
||||||
})
|
|
||||||
}
|
|
||||||
|
|
||||||
type mssqlRowResult struct {
|
|
||||||
row *sql.Row
|
|
||||||
}
|
|
||||||
|
|
||||||
func (mr *mssqlRowResult) Scan(dest ...any) error {
|
|
||||||
return mr.row.Scan(dest...)
|
|
||||||
}
|
|
||||||
|
|
||||||
type mssqlRowsResult struct {
|
|
||||||
columns []string
|
|
||||||
rows *sql.Rows
|
|
||||||
}
|
|
||||||
|
|
||||||
func (mr *mssqlRowsResult) Close() error {
|
|
||||||
return mr.rows.Close()
|
|
||||||
}
|
|
||||||
|
|
||||||
func (mr *mssqlRowsResult) Columns() ([]string, error) {
|
|
||||||
if mr.columns != nil {
|
|
||||||
return mr.columns, nil
|
|
||||||
}
|
|
||||||
|
|
||||||
return mr.rows.Columns()
|
|
||||||
}
|
|
||||||
|
|
||||||
func (mr *mssqlRowsResult) Err() error {
|
|
||||||
return mr.rows.Err()
|
|
||||||
}
|
|
||||||
|
|
||||||
func (mr *mssqlRowsResult) Next() bool {
|
|
||||||
return mr.rows.Next()
|
|
||||||
}
|
|
||||||
|
|
||||||
func (mr *mssqlRowsResult) Scan(dest ...any) error {
|
|
||||||
return mr.rows.Scan(dest...)
|
|
||||||
}
|
|
||||||
|
|
||||||
func (mr *mssqlRowsResult) Values() ([]any, error) {
|
|
||||||
columns, err := mr.Columns()
|
|
||||||
if err != nil {
|
|
||||||
return nil, err
|
|
||||||
}
|
|
||||||
|
|
||||||
rowValues := make([]any, len(columns))
|
|
||||||
scanArgs := make([]any, len(columns))
|
|
||||||
for i := range rowValues {
|
|
||||||
scanArgs[i] = &rowValues[i]
|
|
||||||
}
|
|
||||||
|
|
||||||
if err := mr.rows.Scan(scanArgs...); err != nil {
|
|
||||||
return nil, err
|
|
||||||
}
|
|
||||||
|
|
||||||
return rowValues, nil
|
|
||||||
}
|
|
||||||
|
|
||||||
type mssqlDbWrapper struct {
|
|
||||||
db *sql.DB
|
|
||||||
dialect string
|
|
||||||
}
|
|
||||||
|
|
||||||
func (mw *mssqlDbWrapper) Connect(ctx context.Context, dbUrl string) error {
|
|
||||||
db, err := sql.Open("sqlserver", dbUrl)
|
|
||||||
if err != nil {
|
|
||||||
return err
|
|
||||||
}
|
|
||||||
|
|
||||||
if err := db.PingContext(ctx); err != nil {
|
|
||||||
if err := db.Close(); err != nil {
|
|
||||||
return err
|
|
||||||
}
|
|
||||||
return err
|
|
||||||
}
|
|
||||||
|
|
||||||
mw.db = db
|
|
||||||
return nil
|
|
||||||
}
|
|
||||||
|
|
||||||
func (mw *mssqlDbWrapper) Close() error {
|
|
||||||
return mw.db.Close()
|
|
||||||
}
|
|
||||||
|
|
||||||
func (mw *mssqlDbWrapper) Exec(ctx context.Context, query string, args ...any) (ExecResult, error) {
|
|
||||||
result, execErr := mw.db.ExecContext(ctx, query, args...)
|
|
||||||
if execErr != nil {
|
|
||||||
return ExecResult{}, execErr
|
|
||||||
}
|
|
||||||
|
|
||||||
affectedRows, err := result.RowsAffected()
|
|
||||||
if err != nil {
|
|
||||||
return ExecResult{}, err
|
|
||||||
}
|
|
||||||
|
|
||||||
return ExecResult{AffectedRows: affectedRows}, nil
|
|
||||||
}
|
|
||||||
|
|
||||||
func (mw *mssqlDbWrapper) GetDialect() string {
|
|
||||||
return mw.dialect
|
|
||||||
}
|
|
||||||
|
|
||||||
func (mw *mssqlDbWrapper) Query(ctx context.Context, query string, args ...any) (RowsResult, error) {
|
|
||||||
rows, err := mw.db.QueryContext(ctx, query, args...)
|
|
||||||
if err != nil {
|
|
||||||
return nil, err
|
|
||||||
}
|
|
||||||
|
|
||||||
return &mssqlRowsResult{columns: nil, rows: rows}, nil
|
|
||||||
}
|
|
||||||
|
|
||||||
func (mw *mssqlDbWrapper) QueryRow(ctx context.Context, query string, args ...any) RowResult {
|
|
||||||
row := mw.db.QueryRowContext(ctx, query, args...)
|
|
||||||
return &mssqlRowResult{row: row}
|
|
||||||
}
|
|
||||||
|
|
||||||
func (mw *mssqlDbWrapper) SaveMassive(ctx context.Context, schema string, table string, columnNames []string, rows [][]any) (int64, error) {
|
|
||||||
tx, err := mw.db.BeginTx(ctx, nil)
|
|
||||||
if err != nil {
|
|
||||||
return 0, err
|
|
||||||
}
|
|
||||||
|
|
||||||
fullTableName := fmt.Sprintf("[%s].[%s]", schema, table)
|
|
||||||
|
|
||||||
stmt, err := tx.PrepareContext(ctx, mssql.CopyIn(fullTableName, mssql.BulkOptions{}, columnNames...))
|
|
||||||
if err != nil {
|
|
||||||
tx.Rollback()
|
|
||||||
return 0, err
|
|
||||||
}
|
|
||||||
|
|
||||||
for _, row := range rows {
|
|
||||||
_, err = stmt.ExecContext(ctx, row...)
|
|
||||||
if err != nil {
|
|
||||||
stmt.Close()
|
|
||||||
tx.Rollback()
|
|
||||||
return 0, err
|
|
||||||
}
|
|
||||||
}
|
|
||||||
|
|
||||||
result, err := stmt.ExecContext(ctx)
|
|
||||||
if err != nil {
|
|
||||||
stmt.Close()
|
|
||||||
tx.Rollback()
|
|
||||||
return 0, err
|
|
||||||
}
|
|
||||||
|
|
||||||
if err := stmt.Close(); err != nil {
|
|
||||||
tx.Rollback()
|
|
||||||
return 0, err
|
|
||||||
}
|
|
||||||
|
|
||||||
if err := tx.Commit(); err != nil {
|
|
||||||
return 0, err
|
|
||||||
}
|
|
||||||
|
|
||||||
rowsAffected, raErr := result.RowsAffected()
|
|
||||||
if raErr != nil {
|
|
||||||
return 0, nil
|
|
||||||
}
|
|
||||||
|
|
||||||
return rowsAffected, nil
|
|
||||||
}
|
|
||||||
@@ -1,128 +0,0 @@
|
|||||||
package dbwrapper
|
|
||||||
|
|
||||||
import (
|
|
||||||
"context"
|
|
||||||
"errors"
|
|
||||||
|
|
||||||
"github.com/jackc/pgx/v5"
|
|
||||||
"github.com/jackc/pgx/v5/pgxpool"
|
|
||||||
)
|
|
||||||
|
|
||||||
func init() {
|
|
||||||
Register("postgres", func() DbWrapper {
|
|
||||||
return &postgresDbWrapper{dialect: "postgres"}
|
|
||||||
})
|
|
||||||
}
|
|
||||||
|
|
||||||
type postgresRowResult struct {
|
|
||||||
row pgx.Row
|
|
||||||
}
|
|
||||||
|
|
||||||
func (pr *postgresRowResult) Scan(dest ...any) error {
|
|
||||||
return pr.row.Scan(dest...)
|
|
||||||
}
|
|
||||||
|
|
||||||
type postgresRowsResult struct {
|
|
||||||
columns []string
|
|
||||||
rows pgx.Rows
|
|
||||||
}
|
|
||||||
|
|
||||||
func (pr *postgresRowsResult) Close() error {
|
|
||||||
pr.rows.Close()
|
|
||||||
return nil
|
|
||||||
}
|
|
||||||
|
|
||||||
func (pr *postgresRowsResult) Columns() ([]string, error) {
|
|
||||||
if pr.columns != nil {
|
|
||||||
return pr.columns, nil
|
|
||||||
}
|
|
||||||
|
|
||||||
rawColumns := pr.rows.FieldDescriptions()
|
|
||||||
if rawColumns == nil {
|
|
||||||
return nil, errors.New("error retrieving columns")
|
|
||||||
}
|
|
||||||
|
|
||||||
columns := make([]string, 0, len(rawColumns))
|
|
||||||
for _, rc := range rawColumns {
|
|
||||||
columns = append(columns, rc.Name)
|
|
||||||
}
|
|
||||||
|
|
||||||
return columns, nil
|
|
||||||
}
|
|
||||||
|
|
||||||
func (pr *postgresRowsResult) Err() error {
|
|
||||||
return pr.rows.Err()
|
|
||||||
}
|
|
||||||
|
|
||||||
func (pr *postgresRowsResult) Next() bool {
|
|
||||||
return pr.rows.Next()
|
|
||||||
}
|
|
||||||
|
|
||||||
func (pr *postgresRowsResult) Scan(dest ...any) error {
|
|
||||||
return pr.rows.Scan(dest...)
|
|
||||||
}
|
|
||||||
|
|
||||||
func (pr *postgresRowsResult) Values() ([]any, error) {
|
|
||||||
return pr.rows.Values()
|
|
||||||
}
|
|
||||||
|
|
||||||
type postgresDbWrapper struct {
|
|
||||||
db *pgxpool.Pool
|
|
||||||
dialect string
|
|
||||||
}
|
|
||||||
|
|
||||||
func (pw *postgresDbWrapper) Connect(ctx context.Context, dbUrl string) error {
|
|
||||||
pool, err := pgxpool.New(ctx, dbUrl)
|
|
||||||
if err != nil {
|
|
||||||
return err
|
|
||||||
}
|
|
||||||
|
|
||||||
if err := pool.Ping(ctx); err != nil {
|
|
||||||
pool.Close()
|
|
||||||
return err
|
|
||||||
}
|
|
||||||
|
|
||||||
pw.db = pool
|
|
||||||
return nil
|
|
||||||
}
|
|
||||||
|
|
||||||
func (pw *postgresDbWrapper) Close() error {
|
|
||||||
pw.db.Close()
|
|
||||||
return nil
|
|
||||||
}
|
|
||||||
|
|
||||||
func (pw *postgresDbWrapper) Exec(ctx context.Context, query string, args ...any) (ExecResult, error) {
|
|
||||||
result, err := pw.db.Exec(ctx, query, args...)
|
|
||||||
if err != nil {
|
|
||||||
return ExecResult{}, err
|
|
||||||
}
|
|
||||||
|
|
||||||
return ExecResult{AffectedRows: result.RowsAffected()}, nil
|
|
||||||
}
|
|
||||||
|
|
||||||
func (pw *postgresDbWrapper) GetDialect() string {
|
|
||||||
return pw.dialect
|
|
||||||
}
|
|
||||||
|
|
||||||
func (pw *postgresDbWrapper) Query(ctx context.Context, query string, args ...any) (RowsResult, error) {
|
|
||||||
rows, err := pw.db.Query(ctx, query, args...)
|
|
||||||
if err != nil {
|
|
||||||
return nil, err
|
|
||||||
}
|
|
||||||
|
|
||||||
return &postgresRowsResult{columns: nil, rows: rows}, nil
|
|
||||||
}
|
|
||||||
|
|
||||||
func (pw *postgresDbWrapper) QueryRow(ctx context.Context, query string, args ...any) RowResult {
|
|
||||||
row := pw.db.QueryRow(ctx, query, args...)
|
|
||||||
return &postgresRowResult{row: row}
|
|
||||||
}
|
|
||||||
|
|
||||||
func (pw *postgresDbWrapper) SaveMassive(ctx context.Context, schema string, table string, columnNames []string, rows [][]any) (int64, error) {
|
|
||||||
affectedRows, err := pw.db.CopyFrom(ctx, pgx.Identifier{schema, table}, columnNames, pgx.CopyFromRows(rows))
|
|
||||||
if err != nil {
|
|
||||||
return 0, err
|
|
||||||
}
|
|
||||||
|
|
||||||
return affectedRows, nil
|
|
||||||
}
|
|
||||||
@@ -1,35 +0,0 @@
|
|||||||
package dbwrapper
|
|
||||||
|
|
||||||
import (
|
|
||||||
"context"
|
|
||||||
"errors"
|
|
||||||
)
|
|
||||||
|
|
||||||
var MethodNotSupported error = errors.New("Method not supported by driver... yet :P")
|
|
||||||
|
|
||||||
type ExecResult struct {
|
|
||||||
AffectedRows int64
|
|
||||||
}
|
|
||||||
|
|
||||||
type RowsResult interface {
|
|
||||||
Close() error
|
|
||||||
Columns() ([]string, error)
|
|
||||||
Err() error
|
|
||||||
Next() bool
|
|
||||||
Scan(dest ...any) error
|
|
||||||
Values() ([]any, error)
|
|
||||||
}
|
|
||||||
|
|
||||||
type RowResult interface {
|
|
||||||
Scan(dest ...any) error
|
|
||||||
}
|
|
||||||
|
|
||||||
type DbWrapper interface {
|
|
||||||
Close() error
|
|
||||||
Connect(ctx context.Context, dbUrl string) error
|
|
||||||
Exec(ctx context.Context, query string, args ...any) (ExecResult, error)
|
|
||||||
GetDialect() string
|
|
||||||
Query(ctx context.Context, query string, args ...any) (RowsResult, error)
|
|
||||||
QueryRow(ctx context.Context, query string, args ...any) RowResult
|
|
||||||
SaveMassive(ctx context.Context, schema string, table string, columnNames []string, rows [][]any) (int64, error)
|
|
||||||
}
|
|
||||||
30
internal/app/db/mssql.go
Normal file
30
internal/app/db/mssql.go
Normal file
@@ -0,0 +1,30 @@
|
|||||||
|
package db
|
||||||
|
|
||||||
|
import (
|
||||||
|
"context"
|
||||||
|
"database/sql"
|
||||||
|
)
|
||||||
|
|
||||||
|
type MssqlDbWrapper struct {
|
||||||
|
db *sql.DB
|
||||||
|
}
|
||||||
|
|
||||||
|
func NewMssqlDbWrapper(db *sql.DB) DbWrapper {
|
||||||
|
return &MssqlDbWrapper{db: db}
|
||||||
|
}
|
||||||
|
|
||||||
|
func (wrapper *MssqlDbWrapper) Exec(ctx context.Context, query string, args ...any) (DbWrapperResult, error) {
|
||||||
|
result, execErr := wrapper.db.ExecContext(ctx, query, args...)
|
||||||
|
if execErr != nil {
|
||||||
|
return DbWrapperResult{}, execErr
|
||||||
|
}
|
||||||
|
|
||||||
|
affectedRows, err := result.RowsAffected()
|
||||||
|
if err != nil {
|
||||||
|
return DbWrapperResult{}, err
|
||||||
|
}
|
||||||
|
|
||||||
|
return DbWrapperResult{
|
||||||
|
AffectedRows: affectedRows,
|
||||||
|
}, nil
|
||||||
|
}
|
||||||
47
internal/app/db/postgres.go
Normal file
47
internal/app/db/postgres.go
Normal file
@@ -0,0 +1,47 @@
|
|||||||
|
package db
|
||||||
|
|
||||||
|
import (
|
||||||
|
"context"
|
||||||
|
"fmt"
|
||||||
|
|
||||||
|
"github.com/jackc/pgx/v5/pgxpool"
|
||||||
|
)
|
||||||
|
|
||||||
|
func Connect(ctx context.Context, dbURL string) (*pgxpool.Pool, error) {
|
||||||
|
pool, err := pgxpool.New(ctx, dbURL)
|
||||||
|
if err != nil {
|
||||||
|
return nil, fmt.Errorf("unable to connect to database: %w", err)
|
||||||
|
}
|
||||||
|
|
||||||
|
if err := pool.Ping(ctx); err != nil {
|
||||||
|
pool.Close()
|
||||||
|
return nil, fmt.Errorf("unable to ping database: %w", err)
|
||||||
|
}
|
||||||
|
|
||||||
|
return pool, nil
|
||||||
|
}
|
||||||
|
|
||||||
|
func Close(pool *pgxpool.Pool) {
|
||||||
|
if pool != nil {
|
||||||
|
pool.Close()
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
type PostgresDbWrapper struct {
|
||||||
|
db *pgxpool.Pool
|
||||||
|
}
|
||||||
|
|
||||||
|
func NewPostgresDbWrapper(db *pgxpool.Pool) DbWrapper {
|
||||||
|
return &PostgresDbWrapper{db: db}
|
||||||
|
}
|
||||||
|
|
||||||
|
func (wrapper *PostgresDbWrapper) Exec(ctx context.Context, query string, args ...any) (DbWrapperResult, error) {
|
||||||
|
result, err := wrapper.db.Exec(ctx, query, args...)
|
||||||
|
if err != nil {
|
||||||
|
return DbWrapperResult{}, err
|
||||||
|
}
|
||||||
|
|
||||||
|
return DbWrapperResult{
|
||||||
|
AffectedRows: result.RowsAffected(),
|
||||||
|
}, nil
|
||||||
|
}
|
||||||
11
internal/app/db/types.go
Normal file
11
internal/app/db/types.go
Normal file
@@ -0,0 +1,11 @@
|
|||||||
|
package db
|
||||||
|
|
||||||
|
import "context"
|
||||||
|
|
||||||
|
type DbWrapperResult struct {
|
||||||
|
AffectedRows int64
|
||||||
|
}
|
||||||
|
|
||||||
|
type DbWrapper interface {
|
||||||
|
Exec(ctx context.Context, query string, args ...any) (DbWrapperResult, error)
|
||||||
|
}
|
||||||
@@ -1,101 +0,0 @@
|
|||||||
package extractors
|
|
||||||
|
|
||||||
import (
|
|
||||||
"context"
|
|
||||||
"errors"
|
|
||||||
"slices"
|
|
||||||
"strings"
|
|
||||||
"sync"
|
|
||||||
"sync/atomic"
|
|
||||||
|
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/config"
|
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/custom_errors"
|
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/etl"
|
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/models"
|
|
||||||
)
|
|
||||||
|
|
||||||
func Consume(
|
|
||||||
ctx context.Context,
|
|
||||||
extractor etl.Extractor,
|
|
||||||
tableInfo config.SourceTableInfo,
|
|
||||||
columns []models.ColumnType,
|
|
||||||
batchSize int,
|
|
||||||
chPartitionsIn <-chan models.Partition,
|
|
||||||
chBatchesOut chan<- models.Batch,
|
|
||||||
chErrorsOut chan<- custom_errors.ExtractorError,
|
|
||||||
chJobErrorsOut chan<- custom_errors.JobError,
|
|
||||||
wgActivePartitions *sync.WaitGroup,
|
|
||||||
rowsRead *int64,
|
|
||||||
) {
|
|
||||||
indexPrimaryKey := slices.IndexFunc(columns, func(col models.ColumnType) bool {
|
|
||||||
return strings.EqualFold(col.Name(), tableInfo.PrimaryKey)
|
|
||||||
})
|
|
||||||
|
|
||||||
if indexPrimaryKey == -1 {
|
|
||||||
select {
|
|
||||||
case <-ctx.Done():
|
|
||||||
return
|
|
||||||
case chJobErrorsOut <- custom_errors.JobError{
|
|
||||||
ShouldCancelJob: true,
|
|
||||||
Msg: "Primary key not found in provided columns",
|
|
||||||
}:
|
|
||||||
}
|
|
||||||
|
|
||||||
return
|
|
||||||
}
|
|
||||||
|
|
||||||
for {
|
|
||||||
if ctx.Err() != nil {
|
|
||||||
return
|
|
||||||
}
|
|
||||||
|
|
||||||
select {
|
|
||||||
case <-ctx.Done():
|
|
||||||
return
|
|
||||||
case partition, ok := <-chPartitionsIn:
|
|
||||||
if !ok {
|
|
||||||
return
|
|
||||||
}
|
|
||||||
|
|
||||||
rowsReadResult, err := extractor.Exec(
|
|
||||||
ctx,
|
|
||||||
tableInfo,
|
|
||||||
columns,
|
|
||||||
batchSize,
|
|
||||||
partition,
|
|
||||||
indexPrimaryKey,
|
|
||||||
chBatchesOut,
|
|
||||||
)
|
|
||||||
|
|
||||||
if rowsReadResult > 0 {
|
|
||||||
atomic.AddInt64(rowsRead, int64(rowsReadResult))
|
|
||||||
}
|
|
||||||
|
|
||||||
if err != nil {
|
|
||||||
if exError, ok := errors.AsType[*custom_errors.ExtractorError](err); ok {
|
|
||||||
select {
|
|
||||||
case <-ctx.Done():
|
|
||||||
return
|
|
||||||
case chErrorsOut <- *exError:
|
|
||||||
}
|
|
||||||
} else if jobError, ok := errors.AsType[*custom_errors.JobError](err); ok {
|
|
||||||
select {
|
|
||||||
case <-ctx.Done():
|
|
||||||
return
|
|
||||||
case chJobErrorsOut <- *jobError:
|
|
||||||
}
|
|
||||||
} else {
|
|
||||||
select {
|
|
||||||
case <-ctx.Done():
|
|
||||||
return
|
|
||||||
case chErrorsOut <- custom_errors.ExtractorError{Partition: partition, Msg: err.Error()}:
|
|
||||||
}
|
|
||||||
}
|
|
||||||
|
|
||||||
continue
|
|
||||||
}
|
|
||||||
|
|
||||||
wgActivePartitions.Done()
|
|
||||||
}
|
|
||||||
}
|
|
||||||
}
|
|
||||||
@@ -5,22 +5,24 @@ import (
|
|||||||
"database/sql"
|
"database/sql"
|
||||||
"errors"
|
"errors"
|
||||||
"fmt"
|
"fmt"
|
||||||
|
"slices"
|
||||||
"strings"
|
"strings"
|
||||||
|
"sync"
|
||||||
|
"sync/atomic"
|
||||||
|
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/config"
|
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/config"
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/convert"
|
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/convert"
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/custom_errors"
|
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/custom_errors"
|
||||||
dbwrapper "git.ksdemosapps.com/kylesoda/go-migrate/internal/app/db-wrapper"
|
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/etl"
|
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/etl"
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/models"
|
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/models"
|
||||||
"github.com/google/uuid"
|
"github.com/google/uuid"
|
||||||
)
|
)
|
||||||
|
|
||||||
type MssqlExtractor struct {
|
type MssqlExtractor struct {
|
||||||
db dbwrapper.DbWrapper
|
db *sql.DB
|
||||||
}
|
}
|
||||||
|
|
||||||
func NewMssqlExtractor(db dbwrapper.DbWrapper) etl.Extractor {
|
func NewMssqlExtractor(db *sql.DB) etl.Extractor {
|
||||||
return &MssqlExtractor{db: db}
|
return &MssqlExtractor{db: db}
|
||||||
}
|
}
|
||||||
|
|
||||||
@@ -29,9 +31,6 @@ func buildExtractQueryMssql(
|
|||||||
columns []models.ColumnType,
|
columns []models.ColumnType,
|
||||||
includeRange bool,
|
includeRange bool,
|
||||||
isMinInclusive bool,
|
isMinInclusive bool,
|
||||||
isMaxInclusive bool,
|
|
||||||
hasMin bool,
|
|
||||||
hasMax bool,
|
|
||||||
) string {
|
) string {
|
||||||
var sbQuery strings.Builder
|
var sbQuery strings.Builder
|
||||||
|
|
||||||
@@ -55,32 +54,15 @@ func buildExtractQueryMssql(
|
|||||||
|
|
||||||
fmt.Fprintf(&sbQuery, " FROM [%s].[%s]", tableInfo.Schema, tableInfo.Table)
|
fmt.Fprintf(&sbQuery, " FROM [%s].[%s]", tableInfo.Schema, tableInfo.Table)
|
||||||
|
|
||||||
if includeRange && (hasMin || hasMax) {
|
if includeRange {
|
||||||
sbQuery.WriteString(" WHERE ")
|
fmt.Fprintf(&sbQuery, " WHERE [%s]", tableInfo.PrimaryKey)
|
||||||
|
|
||||||
if hasMin {
|
|
||||||
fmt.Fprintf(&sbQuery, "[%s]", tableInfo.PrimaryKey)
|
|
||||||
if isMinInclusive {
|
if isMinInclusive {
|
||||||
sbQuery.WriteString(" >=")
|
sbQuery.WriteString(" >=")
|
||||||
} else {
|
} else {
|
||||||
sbQuery.WriteString(" >")
|
sbQuery.WriteString(" >")
|
||||||
}
|
}
|
||||||
sbQuery.WriteString(" @min")
|
|
||||||
}
|
|
||||||
|
|
||||||
if hasMin && hasMax {
|
fmt.Fprintf(&sbQuery, " @min AND [%s] <= @max", tableInfo.PrimaryKey)
|
||||||
sbQuery.WriteString(" AND ")
|
|
||||||
}
|
|
||||||
|
|
||||||
if hasMax {
|
|
||||||
fmt.Fprintf(&sbQuery, "[%s]", tableInfo.PrimaryKey)
|
|
||||||
if isMaxInclusive {
|
|
||||||
sbQuery.WriteString(" <=")
|
|
||||||
} else {
|
|
||||||
sbQuery.WriteString(" <")
|
|
||||||
}
|
|
||||||
sbQuery.WriteString(" @max")
|
|
||||||
}
|
|
||||||
}
|
}
|
||||||
|
|
||||||
fmt.Fprintf(&sbQuery, " ORDER BY [%s] ASC", tableInfo.PrimaryKey)
|
fmt.Fprintf(&sbQuery, " ORDER BY [%s] ASC", tableInfo.PrimaryKey)
|
||||||
@@ -116,7 +98,7 @@ func errorFromLastRow(
|
|||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
func (mssqlEx *MssqlExtractor) Exec(
|
func (mssqlEx *MssqlExtractor) ProcessPartition(
|
||||||
ctx context.Context,
|
ctx context.Context,
|
||||||
tableInfo config.SourceTableInfo,
|
tableInfo config.SourceTableInfo,
|
||||||
columns []models.ColumnType,
|
columns []models.ColumnType,
|
||||||
@@ -125,20 +107,18 @@ func (mssqlEx *MssqlExtractor) Exec(
|
|||||||
indexPrimaryKey int,
|
indexPrimaryKey int,
|
||||||
chBatchesOut chan<- models.Batch,
|
chBatchesOut chan<- models.Batch,
|
||||||
) (int, error) {
|
) (int, error) {
|
||||||
hasMin := partition.HasRange && partition.Range.Min > 0
|
query := buildExtractQueryMssql(tableInfo, columns, partition.ShouldUseRange, partition.IsLowerLimitInclusive)
|
||||||
hasMax := partition.HasRange && partition.Range.Max > 0
|
|
||||||
query := buildExtractQueryMssql(tableInfo, columns, partition.HasRange, partition.Range.IsMinInclusive, partition.Range.IsMaxInclusive, hasMin, hasMax)
|
|
||||||
|
|
||||||
var queryArgs []any
|
var queryArgs []any
|
||||||
if hasMin {
|
if partition.ShouldUseRange {
|
||||||
queryArgs = append(queryArgs, sql.Named("min", partition.Range.Min))
|
queryArgs = append(queryArgs,
|
||||||
}
|
sql.Named("min", partition.LowerLimit),
|
||||||
if hasMax {
|
sql.Named("max", partition.UpperLimit),
|
||||||
queryArgs = append(queryArgs, sql.Named("max", partition.Range.Max))
|
)
|
||||||
}
|
}
|
||||||
|
|
||||||
rowsRead := 0
|
rowsRead := 0
|
||||||
rows, err := mssqlEx.db.Query(ctx, query, queryArgs...)
|
rows, err := mssqlEx.db.QueryContext(ctx, query, queryArgs...)
|
||||||
if err != nil {
|
if err != nil {
|
||||||
return rowsRead, &custom_errors.ExtractorError{Partition: partition, HasLastId: false, Msg: err.Error()}
|
return rowsRead, &custom_errors.ExtractorError{Partition: partition, HasLastId: false, Msg: err.Error()}
|
||||||
}
|
}
|
||||||
@@ -207,3 +187,90 @@ func (mssqlEx *MssqlExtractor) Exec(
|
|||||||
|
|
||||||
return rowsRead, nil
|
return rowsRead, nil
|
||||||
}
|
}
|
||||||
|
|
||||||
|
func (mssqlEx *MssqlExtractor) Exec(
|
||||||
|
ctx context.Context,
|
||||||
|
tableInfo config.SourceTableInfo,
|
||||||
|
columns []models.ColumnType,
|
||||||
|
batchSize int,
|
||||||
|
chPartitionsIn <-chan models.Partition,
|
||||||
|
chBatchesOut chan<- models.Batch,
|
||||||
|
chErrorsOut chan<- custom_errors.ExtractorError,
|
||||||
|
chJobErrorsOut chan<- custom_errors.JobError,
|
||||||
|
wgActivePartitions *sync.WaitGroup,
|
||||||
|
rowsRead *int64,
|
||||||
|
) {
|
||||||
|
indexPrimaryKey := slices.IndexFunc(columns, func(col models.ColumnType) bool {
|
||||||
|
return strings.EqualFold(col.Name(), tableInfo.PrimaryKey)
|
||||||
|
})
|
||||||
|
|
||||||
|
if indexPrimaryKey == -1 {
|
||||||
|
select {
|
||||||
|
case <-ctx.Done():
|
||||||
|
return
|
||||||
|
case chJobErrorsOut <- custom_errors.JobError{
|
||||||
|
ShouldCancelJob: true,
|
||||||
|
Msg: "Primary key not found in provided columns",
|
||||||
|
}:
|
||||||
|
}
|
||||||
|
|
||||||
|
return
|
||||||
|
}
|
||||||
|
|
||||||
|
for {
|
||||||
|
if ctx.Err() != nil {
|
||||||
|
return
|
||||||
|
}
|
||||||
|
|
||||||
|
select {
|
||||||
|
case <-ctx.Done():
|
||||||
|
return
|
||||||
|
case partition, ok := <-chPartitionsIn:
|
||||||
|
if !ok {
|
||||||
|
return
|
||||||
|
}
|
||||||
|
|
||||||
|
rowsReadResult, err := mssqlEx.ProcessPartition(
|
||||||
|
ctx,
|
||||||
|
tableInfo,
|
||||||
|
columns,
|
||||||
|
batchSize,
|
||||||
|
partition,
|
||||||
|
indexPrimaryKey,
|
||||||
|
chBatchesOut,
|
||||||
|
)
|
||||||
|
|
||||||
|
if rowsReadResult > 0 {
|
||||||
|
atomic.AddInt64(rowsRead, int64(rowsReadResult))
|
||||||
|
}
|
||||||
|
|
||||||
|
if err != nil {
|
||||||
|
var exError *custom_errors.ExtractorError
|
||||||
|
var jobError *custom_errors.JobError
|
||||||
|
if errors.As(err, &exError) {
|
||||||
|
select {
|
||||||
|
case <-ctx.Done():
|
||||||
|
return
|
||||||
|
case chErrorsOut <- *exError:
|
||||||
|
}
|
||||||
|
} else if errors.As(err, &jobError) {
|
||||||
|
select {
|
||||||
|
case <-ctx.Done():
|
||||||
|
return
|
||||||
|
case chJobErrorsOut <- *jobError:
|
||||||
|
}
|
||||||
|
} else {
|
||||||
|
select {
|
||||||
|
case <-ctx.Done():
|
||||||
|
return
|
||||||
|
case chErrorsOut <- custom_errors.ExtractorError{Partition: partition, Msg: err.Error()}:
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
continue
|
||||||
|
}
|
||||||
|
|
||||||
|
wgActivePartitions.Done()
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|||||||
@@ -2,34 +2,28 @@ package extractors
|
|||||||
|
|
||||||
import (
|
import (
|
||||||
"context"
|
"context"
|
||||||
|
"errors"
|
||||||
"fmt"
|
"fmt"
|
||||||
"strings"
|
"strings"
|
||||||
|
"sync"
|
||||||
|
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/config"
|
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/config"
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/custom_errors"
|
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/custom_errors"
|
||||||
dbwrapper "git.ksdemosapps.com/kylesoda/go-migrate/internal/app/db-wrapper"
|
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/etl"
|
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/etl"
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/models"
|
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/models"
|
||||||
"github.com/google/uuid"
|
"github.com/google/uuid"
|
||||||
|
"github.com/jackc/pgx/v5/pgxpool"
|
||||||
)
|
)
|
||||||
|
|
||||||
type PostgresExtractor struct {
|
type PostgresExtractor struct {
|
||||||
db dbwrapper.DbWrapper
|
db *pgxpool.Pool
|
||||||
}
|
}
|
||||||
|
|
||||||
func NewPostgresExtractor(db dbwrapper.DbWrapper) etl.Extractor {
|
func NewPostgresExtractor(pool *pgxpool.Pool) etl.Extractor {
|
||||||
return &PostgresExtractor{db: db}
|
return &PostgresExtractor{db: pool}
|
||||||
}
|
}
|
||||||
|
|
||||||
func buildExtractQueryPostgres(
|
func buildExtractQueryPostgres(sourceDbInfo config.SourceTableInfo, columns []models.ColumnType) string {
|
||||||
sourceDbInfo config.SourceTableInfo,
|
|
||||||
columns []models.ColumnType,
|
|
||||||
includeRange bool,
|
|
||||||
isMinInclusive bool,
|
|
||||||
isMaxInclusive bool,
|
|
||||||
hasMin bool,
|
|
||||||
hasMax bool,
|
|
||||||
) string {
|
|
||||||
var sbColumns strings.Builder
|
var sbColumns strings.Builder
|
||||||
|
|
||||||
if len(columns) == 0 {
|
if len(columns) == 0 {
|
||||||
@@ -54,44 +48,10 @@ func buildExtractQueryPostgres(
|
|||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
query := fmt.Sprintf(`SELECT %s FROM "%s"."%s"`, sbColumns.String(), sourceDbInfo.Schema, sourceDbInfo.Table)
|
return fmt.Sprintf(`SELECT %s FROM "%s"."%s" ORDER BY "%s" ASC`, sbColumns.String(), sourceDbInfo.Schema, sourceDbInfo.Table, sourceDbInfo.PrimaryKey)
|
||||||
|
|
||||||
if includeRange && (hasMin || hasMax) {
|
|
||||||
query += " WHERE "
|
|
||||||
paramIdx := 1
|
|
||||||
|
|
||||||
if hasMin {
|
|
||||||
query += fmt.Sprintf(`"%s"`, sourceDbInfo.PrimaryKey)
|
|
||||||
if isMinInclusive {
|
|
||||||
query += " >="
|
|
||||||
} else {
|
|
||||||
query += " >"
|
|
||||||
}
|
|
||||||
query += fmt.Sprintf(" $%d", paramIdx)
|
|
||||||
paramIdx++
|
|
||||||
}
|
|
||||||
|
|
||||||
if hasMin && hasMax {
|
|
||||||
query += " AND "
|
|
||||||
}
|
|
||||||
|
|
||||||
if hasMax {
|
|
||||||
query += fmt.Sprintf(`"%s"`, sourceDbInfo.PrimaryKey)
|
|
||||||
if isMaxInclusive {
|
|
||||||
query += " <="
|
|
||||||
} else {
|
|
||||||
query += " <"
|
|
||||||
}
|
|
||||||
query += fmt.Sprintf(" $%d", paramIdx)
|
|
||||||
}
|
|
||||||
}
|
|
||||||
|
|
||||||
query += fmt.Sprintf(` ORDER BY "%s" ASC`, sourceDbInfo.PrimaryKey)
|
|
||||||
|
|
||||||
return query
|
|
||||||
}
|
}
|
||||||
|
|
||||||
func (postgresEx *PostgresExtractor) Exec(
|
func (postgresEx *PostgresExtractor) ProcessPartition(
|
||||||
ctx context.Context,
|
ctx context.Context,
|
||||||
tableInfo config.SourceTableInfo,
|
tableInfo config.SourceTableInfo,
|
||||||
columns []models.ColumnType,
|
columns []models.ColumnType,
|
||||||
@@ -100,20 +60,14 @@ func (postgresEx *PostgresExtractor) Exec(
|
|||||||
indexPrimaryKey int,
|
indexPrimaryKey int,
|
||||||
chBatchesOut chan<- models.Batch,
|
chBatchesOut chan<- models.Batch,
|
||||||
) (int, error) {
|
) (int, error) {
|
||||||
hasMin := partition.HasRange && partition.Range.Min > 0
|
query := buildExtractQueryPostgres(tableInfo, columns)
|
||||||
hasMax := partition.HasRange && partition.Range.Max > 0
|
|
||||||
query := buildExtractQueryPostgres(tableInfo, columns, partition.HasRange, partition.Range.IsMinInclusive, partition.Range.IsMaxInclusive, hasMin, hasMax)
|
|
||||||
|
|
||||||
var queryArgs []any
|
if partition.ShouldUseRange {
|
||||||
if hasMin {
|
return 0, errors.New("Batch config not yet supported")
|
||||||
queryArgs = append(queryArgs, partition.Range.Min)
|
|
||||||
}
|
|
||||||
if hasMax {
|
|
||||||
queryArgs = append(queryArgs, partition.Range.Max)
|
|
||||||
}
|
}
|
||||||
|
|
||||||
rowsRead := 0
|
rowsRead := 0
|
||||||
rows, err := postgresEx.db.Query(ctx, query, queryArgs...)
|
rows, err := postgresEx.db.Query(ctx, query)
|
||||||
if err != nil {
|
if err != nil {
|
||||||
return rowsRead, &custom_errors.ExtractorError{Partition: partition, HasLastId: false, Msg: err.Error()}
|
return rowsRead, &custom_errors.ExtractorError{Partition: partition, HasLastId: false, Msg: err.Error()}
|
||||||
}
|
}
|
||||||
@@ -124,7 +78,7 @@ func (postgresEx *PostgresExtractor) Exec(
|
|||||||
for rows.Next() {
|
for rows.Next() {
|
||||||
values, err := rows.Values()
|
values, err := rows.Values()
|
||||||
if err != nil {
|
if err != nil {
|
||||||
return rowsRead, &custom_errors.ExtractorError{Partition: partition, HasLastId: false, Msg: err.Error()}
|
return rowsRead, errors.New("Unexpected error reading rows from source")
|
||||||
}
|
}
|
||||||
rowsRead++
|
rowsRead++
|
||||||
|
|
||||||
@@ -142,7 +96,7 @@ func (postgresEx *PostgresExtractor) Exec(
|
|||||||
}
|
}
|
||||||
|
|
||||||
if err := rows.Err(); err != nil {
|
if err := rows.Err(); err != nil {
|
||||||
return rowsRead, &custom_errors.ExtractorError{Partition: partition, HasLastId: false, Msg: err.Error()}
|
return rowsRead, errors.New("Unexpected error reading rows from source")
|
||||||
}
|
}
|
||||||
|
|
||||||
if len(batchRows) > 0 {
|
if len(batchRows) > 0 {
|
||||||
@@ -155,3 +109,17 @@ func (postgresEx *PostgresExtractor) Exec(
|
|||||||
|
|
||||||
return rowsRead, nil
|
return rowsRead, nil
|
||||||
}
|
}
|
||||||
|
|
||||||
|
func (postgresEx *PostgresExtractor) Exec(
|
||||||
|
ctx context.Context,
|
||||||
|
tableInfo config.SourceTableInfo,
|
||||||
|
columns []models.ColumnType,
|
||||||
|
batchSize int,
|
||||||
|
chPartitionsIn <-chan models.Partition,
|
||||||
|
chBatchesOut chan<- models.Batch,
|
||||||
|
chErrorsOut chan<- custom_errors.ExtractorError,
|
||||||
|
chJobErrorsOut chan<- custom_errors.JobError,
|
||||||
|
wgActivePartitions *sync.WaitGroup,
|
||||||
|
rowsRead *int64,
|
||||||
|
) {
|
||||||
|
}
|
||||||
|
|||||||
1
internal/app/etl/extractors/types.go
Normal file
1
internal/app/etl/extractors/types.go
Normal file
@@ -0,0 +1 @@
|
|||||||
|
package extractors
|
||||||
@@ -9,32 +9,43 @@ import (
|
|||||||
|
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/config"
|
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/config"
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/custom_errors"
|
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/custom_errors"
|
||||||
dbwrapper "git.ksdemosapps.com/kylesoda/go-migrate/internal/app/db-wrapper"
|
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/etl"
|
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/etl"
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/models"
|
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/models"
|
||||||
|
"github.com/jackc/pgx/v5"
|
||||||
"github.com/jackc/pgx/v5/pgconn"
|
"github.com/jackc/pgx/v5/pgconn"
|
||||||
|
"github.com/jackc/pgx/v5/pgxpool"
|
||||||
)
|
)
|
||||||
|
|
||||||
type GenericLoader struct {
|
type PostgresLoader struct {
|
||||||
db dbwrapper.DbWrapper
|
db *pgxpool.Pool
|
||||||
}
|
}
|
||||||
|
|
||||||
func NewGenericLoader(db dbwrapper.DbWrapper) etl.Loader {
|
func NewPostgresLoader(pool *pgxpool.Pool) etl.Loader {
|
||||||
return &GenericLoader{db: db}
|
return &PostgresLoader{db: pool}
|
||||||
}
|
}
|
||||||
|
|
||||||
func (gl *GenericLoader) ProcessBatch(
|
func mapSlice[T any, V any](input []T, mapper func(T) V) []V {
|
||||||
|
result := make([]V, len(input))
|
||||||
|
|
||||||
|
for i, v := range input {
|
||||||
|
result[i] = mapper(v)
|
||||||
|
}
|
||||||
|
|
||||||
|
return result
|
||||||
|
}
|
||||||
|
|
||||||
|
func (postgresLd *PostgresLoader) ProcessBatch(
|
||||||
ctx context.Context,
|
ctx context.Context,
|
||||||
tableInfo config.TargetTableInfo,
|
tableInfo config.TargetTableInfo,
|
||||||
colNames []string,
|
colNames []string,
|
||||||
batch models.Batch,
|
batch models.Batch,
|
||||||
) (int, error) {
|
) (int, error) {
|
||||||
_, err := gl.db.SaveMassive(
|
tableId := pgx.Identifier{tableInfo.Schema, tableInfo.Table}
|
||||||
|
_, err := postgresLd.db.CopyFrom(
|
||||||
ctx,
|
ctx,
|
||||||
tableInfo.Schema,
|
tableId,
|
||||||
tableInfo.Table,
|
|
||||||
colNames,
|
colNames,
|
||||||
batch.Rows,
|
pgx.CopyFromRows(batch.Rows),
|
||||||
)
|
)
|
||||||
|
|
||||||
if err != nil {
|
if err != nil {
|
||||||
@@ -43,7 +54,7 @@ func (gl *GenericLoader) ProcessBatch(
|
|||||||
if pgErr.Code == "23505" {
|
if pgErr.Code == "23505" {
|
||||||
return 0, &custom_errors.JobError{
|
return 0, &custom_errors.JobError{
|
||||||
ShouldCancelJob: true,
|
ShouldCancelJob: true,
|
||||||
Msg: fmt.Sprintf("Fatal error in table %s.%s", tableInfo.Schema, tableInfo.Table),
|
Msg: fmt.Sprintf("Fatal error in table %s", tableId.Sanitize()),
|
||||||
Prev: err,
|
Prev: err,
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
@@ -55,7 +66,7 @@ func (gl *GenericLoader) ProcessBatch(
|
|||||||
return len(batch.Rows), nil
|
return len(batch.Rows), nil
|
||||||
}
|
}
|
||||||
|
|
||||||
func (gl *GenericLoader) Exec(
|
func (postgresLd *PostgresLoader) Exec(
|
||||||
ctx context.Context,
|
ctx context.Context,
|
||||||
tableInfo config.TargetTableInfo,
|
tableInfo config.TargetTableInfo,
|
||||||
columns []models.ColumnType,
|
columns []models.ColumnType,
|
||||||
@@ -82,7 +93,7 @@ func (gl *GenericLoader) Exec(
|
|||||||
return
|
return
|
||||||
}
|
}
|
||||||
|
|
||||||
processedRows, err := gl.ProcessBatch(ctx, tableInfo, colNames, batch)
|
processedRows, err := postgresLd.ProcessBatch(ctx, tableInfo, colNames, batch)
|
||||||
|
|
||||||
if err != nil {
|
if err != nil {
|
||||||
var ldError *custom_errors.LoaderError
|
var ldError *custom_errors.LoaderError
|
||||||
1
internal/app/etl/loaders/types.go
Normal file
1
internal/app/etl/loaders/types.go
Normal file
@@ -0,0 +1 @@
|
|||||||
|
package loaders
|
||||||
@@ -1,11 +0,0 @@
|
|||||||
package loaders
|
|
||||||
|
|
||||||
func mapSlice[T any, V any](input []T, mapper func(T) V) []V {
|
|
||||||
result := make([]V, len(input))
|
|
||||||
|
|
||||||
for i, v := range input {
|
|
||||||
result[i] = mapper(v)
|
|
||||||
}
|
|
||||||
|
|
||||||
return result
|
|
||||||
}
|
|
||||||
@@ -15,22 +15,7 @@ func PartitionRangeGenerator(
|
|||||||
tableInfo config.TableInfo,
|
tableInfo config.TableInfo,
|
||||||
partitionColumn string,
|
partitionColumn string,
|
||||||
rowsPerPartition int64,
|
rowsPerPartition int64,
|
||||||
jobRange config.RangeConfig,
|
|
||||||
) ([]models.Partition, error) {
|
) ([]models.Partition, error) {
|
||||||
if jobRange.Min > 0 {
|
|
||||||
return []models.Partition{{
|
|
||||||
Id: uuid.New(),
|
|
||||||
HasRange: true,
|
|
||||||
RetryCounter: 0,
|
|
||||||
Range: models.PartitionRange{
|
|
||||||
Min: jobRange.Min,
|
|
||||||
Max: jobRange.Max,
|
|
||||||
IsMinInclusive: jobRange.IsMinInclusive,
|
|
||||||
IsMaxInclusive: jobRange.IsMaxInclusive,
|
|
||||||
},
|
|
||||||
}}, nil
|
|
||||||
}
|
|
||||||
|
|
||||||
rowsCount, err := tableAnalyzer.EstimateTotalRows(ctx, tableInfo)
|
rowsCount, err := tableAnalyzer.EstimateTotalRows(ctx, tableInfo)
|
||||||
if err != nil {
|
if err != nil {
|
||||||
return nil, err
|
return nil, err
|
||||||
@@ -39,7 +24,7 @@ func PartitionRangeGenerator(
|
|||||||
if rowsCount <= rowsPerPartition {
|
if rowsCount <= rowsPerPartition {
|
||||||
return []models.Partition{{
|
return []models.Partition{{
|
||||||
Id: uuid.New(),
|
Id: uuid.New(),
|
||||||
HasRange: false,
|
ShouldUseRange: false,
|
||||||
RetryCounter: 0,
|
RetryCounter: 0,
|
||||||
}}, nil
|
}}, nil
|
||||||
|
|
||||||
|
|||||||
@@ -8,17 +8,16 @@ import (
|
|||||||
"time"
|
"time"
|
||||||
|
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/config"
|
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/config"
|
||||||
dbwrapper "git.ksdemosapps.com/kylesoda/go-migrate/internal/app/db-wrapper"
|
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/etl"
|
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/etl"
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/models"
|
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/models"
|
||||||
"github.com/google/uuid"
|
"github.com/google/uuid"
|
||||||
)
|
)
|
||||||
|
|
||||||
type MssqlTableAnalyzer struct {
|
type MssqlTableAnalyzer struct {
|
||||||
db dbwrapper.DbWrapper
|
db *sql.DB
|
||||||
}
|
}
|
||||||
|
|
||||||
func NewMssqlTableAnalyzer(db dbwrapper.DbWrapper) etl.TableAnalyzer {
|
func NewMssqlTableAnalyzer(db *sql.DB) etl.TableAnalyzer {
|
||||||
return &MssqlTableAnalyzer{db: db}
|
return &MssqlTableAnalyzer{db: db}
|
||||||
}
|
}
|
||||||
|
|
||||||
@@ -36,7 +35,7 @@ JOIN sys.types t ON c.user_type_id = t.user_type_id
|
|||||||
LEFT JOIN sys.types bt ON t.is_user_defined = 1 AND bt.user_type_id = t.system_type_id
|
LEFT JOIN sys.types bt ON t.is_user_defined = 1 AND bt.user_type_id = t.system_type_id
|
||||||
JOIN sys.tables st ON c.object_id = st.object_id
|
JOIN sys.tables st ON c.object_id = st.object_id
|
||||||
JOIN sys.schemas s ON st.schema_id = s.schema_id
|
JOIN sys.schemas s ON st.schema_id = s.schema_id
|
||||||
WHERE s.name = @schema AND st.name = @table AND (c.is_hidden = 0 OR (c.graph_type IS NOT NULL AND c.name LIKE '$%'))
|
WHERE s.name = @schema AND st.name = @table AND c.name NOT LIKE 'graph_id%'
|
||||||
ORDER BY c.column_id;`
|
ORDER BY c.column_id;`
|
||||||
|
|
||||||
type rawColumnMssql struct {
|
type rawColumnMssql struct {
|
||||||
@@ -143,7 +142,7 @@ func (ta *MssqlTableAnalyzer) QueryColumnTypes(
|
|||||||
localCtx, cancel := context.WithTimeout(ctx, 20*time.Second)
|
localCtx, cancel := context.WithTimeout(ctx, 20*time.Second)
|
||||||
defer cancel()
|
defer cancel()
|
||||||
|
|
||||||
rows, err := ta.db.Query(localCtx, mssqlColumnMetadataQuery, sql.Named("schema", tableInfo.Schema), sql.Named("table", tableInfo.Table))
|
rows, err := ta.db.QueryContext(localCtx, mssqlColumnMetadataQuery, sql.Named("schema", tableInfo.Schema), sql.Named("table", tableInfo.Table))
|
||||||
if err != nil {
|
if err != nil {
|
||||||
return nil, err
|
return nil, err
|
||||||
}
|
}
|
||||||
@@ -184,11 +183,11 @@ JOIN sys.partitions p ON t.object_id = p.object_id
|
|||||||
WHERE s.name = @schema AND t.name = @table AND p.index_id IN (0, 1)
|
WHERE s.name = @schema AND t.name = @table AND p.index_id IN (0, 1)
|
||||||
GROUP BY t.name`
|
GROUP BY t.name`
|
||||||
|
|
||||||
ctxTimeout, cancel := context.WithTimeout(ctx, 1*time.Minute)
|
ctxTimeout, cancel := context.WithTimeout(ctx, time.Second*20)
|
||||||
defer cancel()
|
defer cancel()
|
||||||
|
|
||||||
var rowsCount int64
|
var rowsCount int64
|
||||||
err := ta.db.QueryRow(ctxTimeout, query, sql.Named("schema", tableInfo.Schema), sql.Named("table", tableInfo.Table)).Scan(&rowsCount)
|
err := ta.db.QueryRowContext(ctxTimeout, query, sql.Named("schema", tableInfo.Schema), sql.Named("table", tableInfo.Table)).Scan(&rowsCount)
|
||||||
if err != nil {
|
if err != nil {
|
||||||
return 0, err
|
return 0, err
|
||||||
}
|
}
|
||||||
@@ -216,10 +215,10 @@ ORDER BY batch_id`,
|
|||||||
tableInfo.Schema,
|
tableInfo.Schema,
|
||||||
tableInfo.Table)
|
tableInfo.Table)
|
||||||
|
|
||||||
ctxTimeout, cancel := context.WithTimeout(ctx, 1*time.Minute)
|
ctxTimeout, cancel := context.WithTimeout(ctx, time.Second*20)
|
||||||
defer cancel()
|
defer cancel()
|
||||||
|
|
||||||
rows, err := ta.db.Query(ctxTimeout, query, sql.Named("maxPartitions", maxPartitions))
|
rows, err := ta.db.QueryContext(ctxTimeout, query, sql.Named("maxPartitions", maxPartitions))
|
||||||
if err != nil {
|
if err != nil {
|
||||||
return nil, err
|
return nil, err
|
||||||
}
|
}
|
||||||
@@ -230,14 +229,12 @@ ORDER BY batch_id`,
|
|||||||
for rows.Next() {
|
for rows.Next() {
|
||||||
partition := models.Partition{
|
partition := models.Partition{
|
||||||
Id: uuid.New(),
|
Id: uuid.New(),
|
||||||
HasRange: true,
|
ShouldUseRange: true,
|
||||||
RetryCounter: 0,
|
RetryCounter: 0,
|
||||||
Range: models.PartitionRange{
|
IsLowerLimitInclusive: true,
|
||||||
IsMinInclusive: true,
|
|
||||||
},
|
|
||||||
}
|
}
|
||||||
|
|
||||||
if err := rows.Scan(&partition.Range.Min, &partition.Range.Max); err != nil {
|
if err := rows.Scan(&partition.LowerLimit, &partition.UpperLimit); err != nil {
|
||||||
return nil, err
|
return nil, err
|
||||||
}
|
}
|
||||||
|
|
||||||
|
|||||||
@@ -6,16 +6,16 @@ import (
|
|||||||
"time"
|
"time"
|
||||||
|
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/config"
|
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/config"
|
||||||
dbwrapper "git.ksdemosapps.com/kylesoda/go-migrate/internal/app/db-wrapper"
|
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/etl"
|
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/etl"
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/models"
|
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/models"
|
||||||
|
"github.com/jackc/pgx/v5/pgxpool"
|
||||||
)
|
)
|
||||||
|
|
||||||
type PostgresTableAnalyzer struct {
|
type PostgresTableAnalyzer struct {
|
||||||
db dbwrapper.DbWrapper
|
db *pgxpool.Pool
|
||||||
}
|
}
|
||||||
|
|
||||||
func NewPostgresTableAnalyzer(db dbwrapper.DbWrapper) etl.TableAnalyzer {
|
func NewPostgresTableAnalyzer(db *pgxpool.Pool) etl.TableAnalyzer {
|
||||||
return &PostgresTableAnalyzer{db: db}
|
return &PostgresTableAnalyzer{db: db}
|
||||||
}
|
}
|
||||||
|
|
||||||
@@ -125,7 +125,7 @@ func (ta *PostgresTableAnalyzer) QueryColumnTypes(
|
|||||||
ctx context.Context,
|
ctx context.Context,
|
||||||
tableInfo config.TableInfo,
|
tableInfo config.TableInfo,
|
||||||
) ([]models.ColumnType, error) {
|
) ([]models.ColumnType, error) {
|
||||||
localCtx, cancel := context.WithTimeout(ctx, 1*time.Minute)
|
localCtx, cancel := context.WithTimeout(ctx, 20*time.Second)
|
||||||
defer cancel()
|
defer cancel()
|
||||||
|
|
||||||
rows, err := ta.db.Query(localCtx, postgresColumnMetadataQuery, tableInfo.Schema, tableInfo.Table)
|
rows, err := ta.db.Query(localCtx, postgresColumnMetadataQuery, tableInfo.Schema, tableInfo.Table)
|
||||||
|
|||||||
@@ -3,32 +3,18 @@ package transformers
|
|||||||
import (
|
import (
|
||||||
"context"
|
"context"
|
||||||
"errors"
|
"errors"
|
||||||
"fmt"
|
|
||||||
"strings"
|
|
||||||
"sync"
|
"sync"
|
||||||
"time"
|
"time"
|
||||||
|
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/azure"
|
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/config"
|
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/custom_errors"
|
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/custom_errors"
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/etl"
|
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/etl"
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/models"
|
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/models"
|
||||||
"github.com/google/uuid"
|
|
||||||
log "github.com/sirupsen/logrus"
|
|
||||||
)
|
)
|
||||||
|
|
||||||
type MssqlTransformer struct {
|
type MssqlTransformer struct{}
|
||||||
toStorage config.ToStorageConfig
|
|
||||||
sourceTable config.SourceTableInfo
|
|
||||||
azureClient *azure.Client
|
|
||||||
}
|
|
||||||
|
|
||||||
func NewMssqlTransformer(toStorage config.ToStorageConfig, sourceTable config.SourceTableInfo, azureClient *azure.Client) etl.Transformer {
|
func NewMssqlTransformer() etl.Transformer {
|
||||||
return &MssqlTransformer{
|
return &MssqlTransformer{}
|
||||||
toStorage: toStorage,
|
|
||||||
sourceTable: sourceTable,
|
|
||||||
azureClient: azureClient,
|
|
||||||
}
|
|
||||||
}
|
}
|
||||||
|
|
||||||
func computeTransformationPlan(columns []models.ColumnType) []etl.ColumnTransformPlan {
|
func computeTransformationPlan(columns []models.ColumnType) []etl.ColumnTransformPlan {
|
||||||
@@ -74,65 +60,6 @@ func computeTransformationPlan(columns []models.ColumnType) []etl.ColumnTransfor
|
|||||||
return plan
|
return plan
|
||||||
}
|
}
|
||||||
|
|
||||||
func computeStorageTransformationPlan(
|
|
||||||
ctx context.Context,
|
|
||||||
azureClient *azure.Client,
|
|
||||||
toStorage config.ToStorageConfig,
|
|
||||||
sourceColumns []models.ColumnType,
|
|
||||||
sourceTable config.SourceTableInfo,
|
|
||||||
) []etl.ColumnTransformPlan {
|
|
||||||
if azureClient == nil || len(toStorage.Columns) == 0 {
|
|
||||||
return nil
|
|
||||||
}
|
|
||||||
|
|
||||||
colIndex := make(map[string]int, len(sourceColumns))
|
|
||||||
for i, col := range sourceColumns {
|
|
||||||
colIndex[strings.ToUpper(col.Name())] = i
|
|
||||||
}
|
|
||||||
|
|
||||||
var plan []etl.ColumnTransformPlan
|
|
||||||
for _, storageCol := range toStorage.Columns {
|
|
||||||
if storageCol.Mode != "REFERENCE_ONLY" {
|
|
||||||
log.Warnf("to_storage: unsupported mode %q for column %s — skipping", storageCol.Mode, storageCol.Source)
|
|
||||||
continue
|
|
||||||
}
|
|
||||||
|
|
||||||
idx, ok := colIndex[strings.ToUpper(storageCol.Source)]
|
|
||||||
if !ok {
|
|
||||||
log.Warnf("to_storage: source column %q not found in source schema — skipping", storageCol.Source)
|
|
||||||
continue
|
|
||||||
}
|
|
||||||
|
|
||||||
sourceColName := storageCol.Source
|
|
||||||
schema := sourceTable.Schema
|
|
||||||
table := sourceTable.Table
|
|
||||||
|
|
||||||
plan = append(plan, etl.ColumnTransformPlan{
|
|
||||||
Index: idx,
|
|
||||||
Fn: func(v any) (any, error) {
|
|
||||||
if v == nil {
|
|
||||||
return nil, nil
|
|
||||||
}
|
|
||||||
b, ok := v.([]byte)
|
|
||||||
if !ok {
|
|
||||||
log.Warnf("to_storage: expected []byte for %s.%s.%s, got %T — passing through",
|
|
||||||
schema, table, sourceColName, v)
|
|
||||||
return v, nil
|
|
||||||
}
|
|
||||||
start := time.Now()
|
|
||||||
blobPath := fmt.Sprintf("%s/%s/%s", schema, table, uuid.New().String())
|
|
||||||
blobURL, err := azureClient.UploadAndGetURL(ctx, blobPath, b)
|
|
||||||
if err != nil {
|
|
||||||
return nil, fmt.Errorf("uploading %s.%s.%s: %w", schema, table, sourceColName, err)
|
|
||||||
}
|
|
||||||
log.Debugf(`Succesfully uploaded "%s", (%vms)`, blobURL, time.Since(start).Milliseconds())
|
|
||||||
return blobURL, nil
|
|
||||||
},
|
|
||||||
})
|
|
||||||
}
|
|
||||||
return plan
|
|
||||||
}
|
|
||||||
|
|
||||||
const processBatchCtxCheck = 4096
|
const processBatchCtxCheck = 4096
|
||||||
|
|
||||||
func (mssqlTr *MssqlTransformer) ProcessBatch(
|
func (mssqlTr *MssqlTransformer) ProcessBatch(
|
||||||
@@ -173,8 +100,6 @@ func (mssqlTr *MssqlTransformer) Exec(
|
|||||||
wgActiveBatches *sync.WaitGroup,
|
wgActiveBatches *sync.WaitGroup,
|
||||||
) {
|
) {
|
||||||
transformationPlan := computeTransformationPlan(columns)
|
transformationPlan := computeTransformationPlan(columns)
|
||||||
storagePlan := computeStorageTransformationPlan(ctx, mssqlTr.azureClient, mssqlTr.toStorage, columns, mssqlTr.sourceTable)
|
|
||||||
transformationPlan = append(transformationPlan, storagePlan...)
|
|
||||||
|
|
||||||
for {
|
for {
|
||||||
if ctx.Err() != nil {
|
if ctx.Err() != nil {
|
||||||
|
|||||||
@@ -10,7 +10,7 @@ import (
|
|||||||
)
|
)
|
||||||
|
|
||||||
type Extractor interface {
|
type Extractor interface {
|
||||||
Exec(
|
ProcessPartition(
|
||||||
ctx context.Context,
|
ctx context.Context,
|
||||||
tableInfo config.SourceTableInfo,
|
tableInfo config.SourceTableInfo,
|
||||||
columns []models.ColumnType,
|
columns []models.ColumnType,
|
||||||
@@ -19,6 +19,19 @@ type Extractor interface {
|
|||||||
indexPrimaryKey int,
|
indexPrimaryKey int,
|
||||||
chBatchesOut chan<- models.Batch,
|
chBatchesOut chan<- models.Batch,
|
||||||
) (int, error)
|
) (int, error)
|
||||||
|
|
||||||
|
Exec(
|
||||||
|
ctx context.Context,
|
||||||
|
tableInfo config.SourceTableInfo,
|
||||||
|
columns []models.ColumnType,
|
||||||
|
batchSize int,
|
||||||
|
chPartitionsIn <-chan models.Partition,
|
||||||
|
chBatchesOut chan<- models.Batch,
|
||||||
|
chErrorsOut chan<- custom_errors.ExtractorError,
|
||||||
|
chJobErrorsOut chan<- custom_errors.JobError,
|
||||||
|
wgActivePartitions *sync.WaitGroup,
|
||||||
|
rowsRead *int64,
|
||||||
|
)
|
||||||
}
|
}
|
||||||
|
|
||||||
type TransformerFunc func(any) (any, error)
|
type TransformerFunc func(any) (any, error)
|
||||||
|
|||||||
@@ -1,10 +1,6 @@
|
|||||||
package models
|
package models
|
||||||
|
|
||||||
import (
|
import "github.com/google/uuid"
|
||||||
"time"
|
|
||||||
|
|
||||||
"github.com/google/uuid"
|
|
||||||
)
|
|
||||||
|
|
||||||
type UnknownRowValues = []any
|
type UnknownRowValues = []any
|
||||||
|
|
||||||
@@ -15,27 +11,12 @@ type Batch struct {
|
|||||||
RetryCounter int
|
RetryCounter int
|
||||||
}
|
}
|
||||||
|
|
||||||
type PartitionRange struct {
|
|
||||||
Min int64
|
|
||||||
Max int64
|
|
||||||
IsMinInclusive bool
|
|
||||||
IsMaxInclusive bool
|
|
||||||
}
|
|
||||||
|
|
||||||
type Partition struct {
|
type Partition struct {
|
||||||
Id uuid.UUID
|
Id uuid.UUID
|
||||||
ParentId uuid.UUID
|
ParentId uuid.UUID
|
||||||
Range PartitionRange
|
LowerLimit int64
|
||||||
HasRange bool
|
UpperLimit int64
|
||||||
|
IsLowerLimitInclusive bool
|
||||||
|
ShouldUseRange bool
|
||||||
RetryCounter int
|
RetryCounter int
|
||||||
}
|
}
|
||||||
|
|
||||||
type JobResult struct {
|
|
||||||
JobName string
|
|
||||||
StartTime time.Time
|
|
||||||
Duration time.Duration
|
|
||||||
RowsRead int64
|
|
||||||
RowsLoaded int64
|
|
||||||
RowsFailed int64
|
|
||||||
Error error
|
|
||||||
}
|
|
||||||
|
|||||||
@@ -1,44 +0,0 @@
|
|||||||
package main
|
|
||||||
|
|
||||||
import (
|
|
||||||
"context"
|
|
||||||
"fmt"
|
|
||||||
"log"
|
|
||||||
"math/rand"
|
|
||||||
"sync"
|
|
||||||
|
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/azure"
|
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/config"
|
|
||||||
)
|
|
||||||
|
|
||||||
func main() {
|
|
||||||
cfg := config.App.AzureStorage
|
|
||||||
containerName := cfg.Container
|
|
||||||
|
|
||||||
client, err := azure.NewClient(cfg)
|
|
||||||
if err != nil {
|
|
||||||
log.Fatalf("Error creando cliente: %v", err)
|
|
||||||
}
|
|
||||||
ctx := context.Background()
|
|
||||||
|
|
||||||
var wg sync.WaitGroup
|
|
||||||
|
|
||||||
for i := 1; i <= 10; i++ {
|
|
||||||
wg.Add(1)
|
|
||||||
go func(id int) {
|
|
||||||
defer wg.Done()
|
|
||||||
|
|
||||||
blobName := fmt.Sprintf("%sarchivo-%d.txt", cfg.Prefix, id)
|
|
||||||
content := fmt.Sprintf("Contenido aleatorio: %d", rand.Intn(100000))
|
|
||||||
|
|
||||||
err := client.UploadBuffer(ctx, containerName, blobName, []byte(content))
|
|
||||||
if err != nil {
|
|
||||||
log.Printf("Fallo al subir %s: %v", blobName, err)
|
|
||||||
} else {
|
|
||||||
fmt.Printf("Subido exitosamente: %s\n", blobName)
|
|
||||||
}
|
|
||||||
}(i)
|
|
||||||
}
|
|
||||||
|
|
||||||
wg.Wait()
|
|
||||||
}
|
|
||||||
58
scripts/backoff-test/main.go
Normal file
58
scripts/backoff-test/main.go
Normal file
@@ -0,0 +1,58 @@
|
|||||||
|
package main
|
||||||
|
|
||||||
|
import (
|
||||||
|
"context"
|
||||||
|
"errors"
|
||||||
|
"fmt"
|
||||||
|
"net/http"
|
||||||
|
"strconv"
|
||||||
|
|
||||||
|
"github.com/cenkalti/backoff/v5"
|
||||||
|
)
|
||||||
|
|
||||||
|
func ExampleRetry() {
|
||||||
|
// Define an operation function that returns a value and an error.
|
||||||
|
// The value can be any type.
|
||||||
|
// We'll pass this operation to Retry function.
|
||||||
|
operation := func() (string, error) {
|
||||||
|
// An example request that may fail.
|
||||||
|
resp, err := http.Get("http://httpbin.org/get")
|
||||||
|
if err != nil {
|
||||||
|
return "", err
|
||||||
|
}
|
||||||
|
defer resp.Body.Close()
|
||||||
|
|
||||||
|
// If we are being rate limited, return a RetryAfter to specify how long to wait.
|
||||||
|
// This will also reset the backoff policy.
|
||||||
|
if resp.StatusCode == 429 {
|
||||||
|
seconds, err := strconv.ParseInt(resp.Header.Get("Retry-After"), 10, 64)
|
||||||
|
if err == nil {
|
||||||
|
return "", backoff.RetryAfter(int(seconds))
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
// In case of non-retriable error, return Permanent error to stop retrying.
|
||||||
|
// For this HTTP example, client errors are non-retriable.
|
||||||
|
if resp.StatusCode >= 400 && resp.StatusCode < 500 {
|
||||||
|
return "", backoff.Permanent(errors.New("bad request"))
|
||||||
|
}
|
||||||
|
|
||||||
|
// Return successful response.
|
||||||
|
return "hello", nil
|
||||||
|
}
|
||||||
|
|
||||||
|
result, err := backoff.Retry(context.TODO(), operation, backoff.WithBackOff(backoff.NewExponentialBackOff()))
|
||||||
|
if err != nil {
|
||||||
|
fmt.Println("Error:", err)
|
||||||
|
return
|
||||||
|
}
|
||||||
|
|
||||||
|
// Operation is successful after retries.
|
||||||
|
|
||||||
|
fmt.Println(result)
|
||||||
|
// Output: hello
|
||||||
|
}
|
||||||
|
|
||||||
|
func main() {
|
||||||
|
ExampleRetry()
|
||||||
|
}
|
||||||
@@ -8,32 +8,13 @@ import (
|
|||||||
"time"
|
"time"
|
||||||
|
|
||||||
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/config"
|
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/config"
|
||||||
|
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/db"
|
||||||
"github.com/jackc/pgx/v5"
|
"github.com/jackc/pgx/v5"
|
||||||
"github.com/jackc/pgx/v5/pgtype"
|
"github.com/jackc/pgx/v5/pgtype"
|
||||||
"github.com/jackc/pgx/v5/pgxpool"
|
"github.com/jackc/pgx/v5/pgxpool"
|
||||||
log "github.com/sirupsen/logrus"
|
log "github.com/sirupsen/logrus"
|
||||||
)
|
)
|
||||||
|
|
||||||
func Connect(ctx context.Context, dbURL string) (*pgxpool.Pool, error) {
|
|
||||||
pool, err := pgxpool.New(ctx, dbURL)
|
|
||||||
if err != nil {
|
|
||||||
return nil, fmt.Errorf("unable to connect to database: %w", err)
|
|
||||||
}
|
|
||||||
|
|
||||||
if err := pool.Ping(ctx); err != nil {
|
|
||||||
pool.Close()
|
|
||||||
return nil, fmt.Errorf("unable to ping database: %w", err)
|
|
||||||
}
|
|
||||||
|
|
||||||
return pool, nil
|
|
||||||
}
|
|
||||||
|
|
||||||
func Close(pool *pgxpool.Pool) {
|
|
||||||
if pool != nil {
|
|
||||||
pool.Close()
|
|
||||||
}
|
|
||||||
}
|
|
||||||
|
|
||||||
func main() {
|
func main() {
|
||||||
log.SetFormatter(&log.TextFormatter{
|
log.SetFormatter(&log.TextFormatter{
|
||||||
FullTimestamp: true,
|
FullTimestamp: true,
|
||||||
@@ -46,8 +27,8 @@ func main() {
|
|||||||
ctxSource, cancel := context.WithTimeout(context.Background(), 20*time.Second)
|
ctxSource, cancel := context.WithTimeout(context.Background(), 20*time.Second)
|
||||||
defer cancel()
|
defer cancel()
|
||||||
|
|
||||||
sourcePool, err := Connect(ctxSource, config.App.SourceDbUrl)
|
sourcePool, err := db.Connect(ctxSource, config.App.SourceDbUrl)
|
||||||
defer Close(sourcePool)
|
defer db.Close(sourcePool)
|
||||||
if err != nil {
|
if err != nil {
|
||||||
log.Fatal(err)
|
log.Fatal(err)
|
||||||
}
|
}
|
||||||
@@ -56,8 +37,8 @@ func main() {
|
|||||||
ctxTarget, cancel := context.WithTimeout(context.Background(), 20*time.Second)
|
ctxTarget, cancel := context.WithTimeout(context.Background(), 20*time.Second)
|
||||||
defer cancel()
|
defer cancel()
|
||||||
|
|
||||||
targetPool, err := Connect(ctxTarget, config.App.TargetDbUrl)
|
targetPool, err := db.Connect(ctxTarget, config.App.TargetDbUrl)
|
||||||
defer Close(targetPool)
|
defer db.Close(targetPool)
|
||||||
if err != nil {
|
if err != nil {
|
||||||
log.Fatal(err)
|
log.Fatal(err)
|
||||||
}
|
}
|
||||||
|
|||||||
Reference in New Issue
Block a user