1 Commits

39 changed files with 834 additions and 1631 deletions

View File

@@ -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 |

View File

@@ -1,12 +1,2 @@
SOURCE_DB_URL=sqlserver://sa:password@localhost:1433?database=master&packet+size=32767&loc=UTC
TARGET_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=
PG_FROM_DB_URL=postgresql://postgres:password@localhost:5432/db
PG_TO_DB_URL=postgresql://postgres:password@localhost:5432/db

2
.gitignore vendored
View File

@@ -27,5 +27,5 @@ go.work.sum
# Editor/IDE
# .idea/
.vscode/
# .vscode/
.temp

View File

@@ -3,7 +3,6 @@ package main
import (
"time"
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/config"
log "github.com/sirupsen/logrus"
)
@@ -14,13 +13,5 @@ func configureLog() {
DisableSorting: false,
PadLevelText: true,
})
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)
log.SetLevel(log.DebugLevel)
}

View File

@@ -5,12 +5,12 @@ import (
"sync"
"time"
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/azure"
"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-wrapper"
"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/table_analyzers"
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/etl/transformers"
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/models"
log "github.com/sirupsen/logrus"
"golang.org/x/sync/errgroup"
@@ -24,7 +24,7 @@ func main() {
log.Fatalf("error leyendo configuracion: %v", err)
}
// log.Debugf("Config: %+v", migrationConfig)
log.Debugf("Config: %+v", migrationConfig)
startTime := time.Now()
@@ -118,17 +118,9 @@ func processMigrationJobs(
sourceTableAnalyzer := table_analyzers.NewMssqlTableAnalyzer(sourceDb)
targetTableAnalyzer := table_analyzers.NewPostgresTableAnalyzer(targetDb)
extractor := extractors.NewExtractor(sourceDb)
loader := loaders.NewGenericLoader(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)
}
}
extractor := extractors.NewMssqlExtractor(sourceDb)
transformer := transformers.NewMssqlTransformer()
loader := loaders.NewPostgresLoader(targetDb)
for i := range maxParallelWorkers {
wgJobs.Go(func() {
@@ -140,10 +132,9 @@ func processMigrationJobs(
sourceTableAnalyzer,
targetTableAnalyzer,
extractor,
azureClient,
transformer,
loader,
job,
targetDb.GetDialect(),
)
chJobResults <- res

View File

@@ -7,48 +7,26 @@ import (
"sync/atomic"
"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"
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/extractors"
"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/transformers"
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/models"
log "github.com/sirupsen/logrus"
"golang.org/x/sync/errgroup"
)
const jobErrorsChannelSize int = 100
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(
ctx context.Context,
targetDbWrapper dbwrapper.DbWrapper,
sourceTableAnalyzer etl.TableAnalyzer,
targetTableAnalyzer etl.TableAnalyzer,
extractor extractors.GenericExtractor,
azureClient *azure.Client,
loader loaders.GenericLoader,
extractor etl.Extractor,
transformer etl.Transformer,
loader etl.Loader,
job config.Job,
targetDbType string,
) models.JobResult {
transformer := transformers.NewMssqlTransformer(job.ToStorage, job.SourceTable, azureClient)
localCtx, cancel := context.WithCancel(ctx)
defer cancel()
@@ -88,13 +66,7 @@ func processMigrationJob(
return result
}
preSqlQueries := job.TargetTable.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 {
for _, query := range job.PreSQL {
if _, err := targetDbWrapper.Exec(localCtx, query); err != nil {
result.Error = err
return result
@@ -107,17 +79,17 @@ func processMigrationJob(
job.SourceTable.TableInfo,
job.SourceTable.PrimaryKey,
job.RowsPerPartition,
job.Range,
)
if err != nil {
log.Error("Unexpected error calculating batch ranges: ", err)
}
chJobErrors := make(chan custom_errors.JobError, jobErrorsChannelSize)
chLoadersErrors := make(chan custom_errors.LoaderError, job.ExtractorQueueSize)
chPartitions := make(chan models.Partition, job.ExtractorQueueSize)
chBatchesRaw := make(chan models.Batch, job.ExtractorQueueSize)
chBatchesTransformed := make(chan models.Batch, job.TransformerQueueSize)
chJobErrors := make(chan custom_errors.JobError, job.QueueSize)
chExtractorErrors := make(chan custom_errors.ExtractorError, job.QueueSize)
chLoadersErrors := make(chan custom_errors.LoaderError, job.QueueSize)
chPartitions := make(chan models.Partition, job.QueueSize)
chBatchesRaw := make(chan models.Batch, job.QueueSize)
chBatchesTransformed := make(chan models.Batch, job.QueueSize)
var wgActivePartitions sync.WaitGroup
var wgActiveBatches sync.WaitGroup
@@ -133,10 +105,19 @@ func processMigrationJob(
}
}()
go custom_errors.ExtractorErrorHandler(
localCtx,
job.Retry,
job.MaxPartitionErrrors,
chExtractorErrors,
chPartitions,
chJobErrors,
&wgActivePartitions,
)
go custom_errors.LoaderErrorHandler(
localCtx,
job.Retry,
job.MaxExtractorBatchErrors,
job.MaxChunkErrors,
chLoadersErrors,
chBatchesTransformed,
chJobErrors,
@@ -148,18 +129,17 @@ func processMigrationJob(
for range maxExtractors {
wgExtractors.Go(func() {
extractor.Consume(
extractor.Exec(
localCtx,
job.SourceTable,
sourceColTypes,
job.ExtractorBatchSize,
job.Retry,
job.BatchSize,
chPartitions,
chBatchesRaw,
chExtractorErrors,
chJobErrors,
&wgActivePartitions,
&rowsRead,
job.SourceTable.FromJsonColumns,
)
})
}
@@ -190,12 +170,12 @@ func processMigrationJob(
for range job.MaxLoaders {
wgLoaders.Go(func() {
loader.Consume(
loader.Exec(
localCtx,
job.TargetTable,
targetColTypes,
job.Retry,
chBatchesTransformed,
chLoadersErrors,
chJobErrors,
&wgActiveBatches,
&rowsLoaded,
@@ -210,6 +190,8 @@ func processMigrationJob(
log.Debugf("wgActivePartitions is empty (%v)", job.Name)
close(chPartitions)
log.Debugf("chPartitions is closed (%v)", job.Name)
close(chExtractorErrors)
log.Debugf("chExtractorErrors is closed (%v)", job.Name)
wgExtractors.Wait()
log.Debugf("wgExtractors is empty (%v)", job.Name)
@@ -232,7 +214,7 @@ func processMigrationJob(
cancel()
}()
for _, query := range job.TargetTable.PostSQL {
for _, query := range job.PostSQL {
if _, err := targetDbWrapper.Exec(localCtx, query); err != nil {
result.Error = err
return result

View File

@@ -3,19 +3,15 @@ source_db_type: sqlserver
target_db_type: postgres
defaults:
batches_per_partition: 8
max_extractors: 2
extractor_batch_size: 25000
extractor_queue_size: 8
max_transformers: 2
transformer_batch_size: 25000
transformer_queue_size: 8
max_loaders: 4
loader_batch_size: 25000
queue_size: 8
batch_size: 25000
batches_per_partition: 8
truncate_target: true
truncate_method: TRUNCATE # TRUNCATE | DELETE
max_partition_errrors: 5
max_extractor_batch_errors: 5
max_chunk_errors: 5
retry:
attempts: 3
base_delay_ms: 500
@@ -34,6 +30,7 @@ jobs:
table: MANZANA
pre_sql:
- 'SELECT 1'
# - 'TRUNCATE TABLE "Cartografia"."MANZANA"'
range:
min: 1000000
max: 2000000
@@ -46,36 +43,11 @@ jobs:
schema: Red
table: PUERTO
primary_key: ID_PUERTO
from_json:
- column: $node_id*
field: id
target:
schema: Red
table: PUERTO
pre_sql:
- 'SELECT 1'
# - 'TRUNCATE TABLE "Red"."PUERTO"'
post_sql:
- "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
View File

@@ -1,13 +1,12 @@
module git.ksdemosapps.com/kylesoda/go-migrate
go 1.26
go 1.25.7
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/google/uuid v1.6.0
github.com/ilyakaznacheev/cleanenv v1.5.0
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/sirupsen/logrus v1.9.4
github.com/twpayne/go-geom v1.6.1
@@ -16,20 +15,17 @@ require (
)
require (
github.com/Azure/azure-sdk-for-go/sdk/azcore v1.21.0 // indirect
github.com/Azure/azure-sdk-for-go/sdk/internal v1.11.2 // indirect
github.com/BurntSushi/toml v1.6.0 // indirect
github.com/cenkalti/backoff v2.2.1+incompatible // indirect
github.com/cenkalti/backoff/v5 v5.0.3 // indirect
github.com/golang-sql/civil v0.0.0-20220223132316-b832511892a9 // indirect
github.com/golang-sql/sqlexp v0.1.0 // indirect
github.com/jackc/pgpassfile v1.0.0 // indirect
github.com/jackc/pgservicefile v0.0.0-20240606120523-5a60cdf6a761 // 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/shopspring/decimal v1.4.0 // 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/text v0.34.0 // indirect
olympos.io/encoding/edn v0.0.0-20201019073823-d3554ca0b0a3 // indirect
)

20
go.sum
View File

@@ -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/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/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/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/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/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/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/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/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 v1.1.1 h1:vj9j/u1bqnvCEfJOwUhtlOARqs3+rkHYY13jYWTU97c=
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/hexops/gotextdiff v1.0.3 h1:gitA9+qJrrTCsiCl7+kh75nPqQt1cx4ZkudSTLoUqJM=
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/go.mod h1:CEx0iS5ambNFdcRtxPj5JhEz+xB6uRky5eyVu/W2HEg=
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/joho/godotenv v1.5.1 h1:7eLL/+HRGLY0ldzfGMeQkb7vMd0as4CfYvUVzLqw0N0=
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 v0.3.1/go.mod h1:hoEshYVHaxMs3cyo3Yncou5ZscifuDolrwPKZanG3xk=
github.com/kr/pretty v0.3.0 h1:WgNl7dwNpEZ6jJ9k1snq4pZsg7DOEN8hP9Xw0Tsjwk0=
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/go.mod h1:eLer722TekiGuMkidMxC/pM04lWEeraHUUmBw8l2grE=
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 v3.0.1 h1:fxVm/GzAzEWqLHuvctI91KS9hhNmmWOoWu0XTYJS7CA=
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=

View File

@@ -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
}

View File

@@ -1,41 +1,41 @@
package config
import (
"github.com/ilyakaznacheev/cleanenv"
"os"
"github.com/joho/godotenv"
log "github.com/sirupsen/logrus"
)
type AzureStorageConfig struct {
AccountName string `env:"AZ_ACCOUNT_NAME"`
Container string `env:"AZ_CONTAINER"`
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 {
SourceDbUrl string
TargetDbUrl string
}
type appConfig struct {
SourceDbUrl string `env:"SOURCE_DB_URL" env-required:"true"`
TargetDbUrl string `env:"TARGET_DB_URL" env-required:"true"`
LogLevel string `env:"LOG_LEVEL" env-default:"INFO"`
AzureStorage AzureStorageConfig
func loadEnv() {
err := godotenv.Load()
if err != nil {
log.Warn("Warning: could not load .env file")
}
}
func getAppConfig() appConfig {
var cfg appConfig
loadEnv()
err := cleanenv.ReadConfig(".env", &cfg)
if err != nil {
log.Warn("Could not load .env file")
sourceDbUrl := os.Getenv("SOURCE_DB_URL")
if sourceDbUrl == "" {
log.Fatal("SOURCE_DB_URL environment variable not set")
}
err = cleanenv.ReadEnv(&cfg)
if err != nil {
log.Fatalf("Error al cargar variables: %v", err)
targetDbUrl := os.Getenv("TARGET_DB_URL")
if targetDbUrl == "" {
log.Fatal("TARGET_DB_URL environment variable not set")
}
return cfg
return appConfig{
SourceDbUrl: sourceDbUrl,
TargetDbUrl: targetDbUrl,
}
}
var App appConfig = getAppConfig()

View File

@@ -14,38 +14,18 @@ type RetryConfig struct {
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 {
BatchesPerPartition int `yaml:"batches_per_partition"`
MaxExtractors int `yaml:"max_extractors"`
ExtractorBatchSize int `yaml:"extractor_batch_size"`
ExtractorQueueSize int `yaml:"extractor_queue_size"`
MaxTransformers int `yaml:"max_transformers"`
TransformerBatchSize int `yaml:"transformer_batch_size"`
TransformerQueueSize int `yaml:"transformer_queue_size"`
MaxLoaders int `yaml:"max_loaders"`
LoaderBatchSize int `yaml:"loader_batch_size"`
QueueSize int `yaml:"queue_size"`
BatchSize int `yaml:"batch_size"`
BatchesPerPartition int `yaml:"batches_per_partition"`
TruncateTarget bool `yaml:"truncate_target"`
TruncateMethod string `yaml:"truncate_method"`
MaxPartitionErrrors int `yaml:"max_partition_errrors"`
MaxExtractorBatchErrors int `yaml:"max_extractor_batch_errors"`
MaxChunkErrors int `yaml:"max_chunk_errors"`
Retry RetryConfig `yaml:"retry"`
RowsPerPartition int64
ToStorage ToStorageConfig `yaml:"to_storage"`
}
type FromJsonItem struct {
Column string `yaml:"column"`
Field string `yaml:"field"`
}
type TableInfo struct {
@@ -53,23 +33,13 @@ type TableInfo struct {
Table string `yaml:"table"`
}
type TargetTableInfo struct {
TableInfo `yaml:",inline"`
}
type SourceTableInfo struct {
TableInfo `yaml:",inline"`
PrimaryKey string `yaml:"primary_key"`
FromJsonColumns []FromJsonItem `yaml:"from_json"`
}
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 {
@@ -77,8 +47,15 @@ type Job struct {
Enabled bool `yaml:"enabled"`
SourceTable SourceTableInfo `yaml:"source"`
TargetTable TargetTableInfo `yaml:"target"`
PreSQL []string `yaml:"pre_sql"`
PostSQL []string `yaml:"post_sql"`
JobConfig `yaml:",inline"`
Range RangeConfig `yaml:"range"`
Range struct {
Min int64 `yaml:"min"`
Max int64 `yaml:"max"`
IsMinInclusive bool `yaml:"is_min_inclusive"`
IsMaxInclusive bool `yaml:"is_max_inclusive"`
}
}
type MigrationConfig struct {
@@ -107,7 +84,7 @@ func (c *MigrationConfig) UnmarshalYAML(value *yaml.Node) error {
c.Defaults = raw.Defaults
c.SourceDbType = raw.SourceDbType
c.TargetDbType = raw.TargetDbType
c.Defaults.RowsPerPartition = int64(raw.Defaults.ExtractorBatchSize * raw.Defaults.BatchesPerPartition)
c.Defaults.RowsPerPartition = int64(raw.Defaults.BatchSize * raw.Defaults.BatchesPerPartition)
for _, node := range raw.Jobs {
job := Job{
@@ -118,7 +95,7 @@ func (c *MigrationConfig) UnmarshalYAML(value *yaml.Node) error {
return err
}
job.RowsPerPartition = int64(job.ExtractorBatchSize * job.BatchesPerPartition)
job.RowsPerPartition = int64(job.BatchSize * job.BatchesPerPartition)
c.Jobs = append(c.Jobs, job)
}

View File

@@ -6,7 +6,7 @@ import (
"time"
)
func ComputeBackoffDelay(retryCounter int, baseDelayMs int, maxDelayMs int, maxJitterMs int) time.Duration {
func computeBackoffDelay(retryCounter int, baseDelayMs int, maxDelayMs int, maxJitterMs int) time.Duration {
if retryCounter < 0 {
retryCounter = 0
}

View File

@@ -1,7 +1,13 @@
package custom_errors
import (
"context"
"fmt"
"sync"
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/config"
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/models"
"github.com/google/uuid"
)
type ExtractorError struct {
@@ -14,3 +20,100 @@ type ExtractorError struct {
func (e *ExtractorError) Error() string {
return e.Msg
}
func ExtractorErrorHandler(
ctx context.Context,
retryConfig config.RetryConfig,
maxPartitionErrors int,
chErrorsIn <-chan ExtractorError,
chPartitionsOut chan<- models.Partition,
chJobErrorsOut chan<- JobError,
wgActivePartitions *sync.WaitGroup,
) {
definitiveErrors := 0
for {
if ctx.Err() != nil {
return
}
select {
case <-ctx.Done():
return
case err, ok := <-chErrorsIn:
if !ok {
return
}
if err.Partition.RetryCounter >= retryConfig.Attempts {
wgActivePartitions.Done()
definitiveErrors++
jobError := JobError{
ShouldCancelJob: false,
Msg: fmt.Sprintf("Partition %v reached max retries (%d)", err.Partition.Id, retryConfig.Attempts),
Prev: &err,
}
select {
case chJobErrorsOut <- jobError:
case <-ctx.Done():
return
}
if maxPartitionErrors > 0 && definitiveErrors >= maxPartitionErrors {
fatalError := JobError{
ShouldCancelJob: true,
Msg: fmt.Sprintf("Partition error limit reached (%d)", maxPartitionErrors),
Prev: &err,
}
select {
case chJobErrorsOut <- fatalError:
case <-ctx.Done():
return
}
}
continue
} else {
jobError := JobError{
ShouldCancelJob: false,
Msg: fmt.Sprintf("Temporal error in partition %v (retries: %d)", err.Partition.Id, err.Partition.RetryCounter),
Prev: &err,
}
select {
case chJobErrorsOut <- jobError:
case <-ctx.Done():
return
}
}
newPartition := err.Partition
newPartition.RetryCounter++
delay := computeBackoffDelay(
newPartition.RetryCounter,
retryConfig.BaseDelayMs,
retryConfig.MaxDelayMs,
retryConfig.MaxJitterMs,
)
if err.HasLastId {
newPartition.ParentId = err.Partition.Id
newPartition.Id = uuid.New()
newPartition.Range.Min = err.LastId
newPartition.Range.IsMinInclusive = false
}
requeueWithBackoff(ctx, delay, func() {
select {
case chPartitionsOut <- newPartition:
case <-ctx.Done():
return
}
})
}
}
}

View File

@@ -88,7 +88,7 @@ func LoaderErrorHandler(
}
err.Batch.RetryCounter++
delay := ComputeBackoffDelay(
delay := computeBackoffDelay(
err.Batch.RetryCounter,
retryConfig.BaseDelayMs,
retryConfig.MaxDelayMs,

View File

@@ -1,7 +0,0 @@
package db_dialects
const (
SqlServer string = "sqlserver"
Postgres string = "postgres"
Null string = "null"
)

View File

@@ -4,18 +4,13 @@ import (
"context"
"database/sql"
"fmt"
"strings"
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/config"
dbdialects "git.ksdemosapps.com/kylesoda/go-migrate/internal/app/db-wrapper/db_dialects"
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/models"
mssql "github.com/microsoft/go-mssqldb"
"github.com/sirupsen/logrus"
)
func init() {
Register(dbdialects.SqlServer, func() DbWrapper {
return &mssqlDbWrapper{dialect: dbdialects.SqlServer}
Register("sqlserver", func() DbWrapper {
return &mssqlDbWrapper{dialect: "sqlserver"}
})
}
@@ -179,142 +174,3 @@ func (mw *mssqlDbWrapper) SaveMassive(ctx context.Context, schema string, table
return rowsAffected, nil
}
func buildExtractQueryMssql(q ExtractionQuery) (string, error) {
var sbQuery strings.Builder
sbQuery.WriteString("SELECT ")
hasRegularColumns := len(q.Columns) > 0
hasJsonColumns := len(q.FromJsonColumns) > 0
// logrus.Debugf("Extraction query: %+v", q)
resolvedJson := make(map[string][]config.FromJsonItem, len(q.FromJsonColumns))
if hasJsonColumns {
for _, jsonConfig := range q.FromJsonColumns {
actualColumnName, err := findColumnByPattern(q.Columns, jsonConfig.Column)
if err != nil {
return "", err
}
resolvedJson[actualColumnName] = append(resolvedJson[actualColumnName], jsonConfig)
}
}
selectParts := make([]string, 0, len(q.Columns)+len(q.FromJsonColumns))
if hasRegularColumns {
for _, col := range q.Columns {
jsonConfigs, isJsonColumn := resolvedJson[col.Name()]
if isJsonColumn {
for _, jsonConfig := range jsonConfigs {
jsonPath := buildJsonPathMssql(jsonConfig.Field)
jsonExpr := fmt.Sprintf("JSON_VALUE([%s], '%s') AS [%s]", col.Name(), jsonPath, col.Name())
selectParts = append(selectParts, jsonExpr)
}
continue
}
colExpr := fmt.Sprintf("[%s]", col.Name())
switch col.Type() {
case "GEOMETRY":
colExpr = fmt.Sprintf("[%s].STAsBinary() AS [%s]", col.Name(), col.Name())
}
selectParts = append(selectParts, colExpr)
}
} else if !hasJsonColumns {
selectParts = append(selectParts, "*")
}
for i, part := range selectParts {
sbQuery.WriteString(part)
if i < len(selectParts)-1 {
sbQuery.WriteString(", ")
}
}
fmt.Fprintf(&sbQuery, " FROM [%s].[%s]", q.Schema, q.Table)
if q.LowerLimit.IsValid || q.UpperLimit.IsValid {
sbQuery.WriteString(" WHERE ")
if q.LowerLimit.IsValid {
fmt.Fprintf(&sbQuery, "[%s]", q.PrimaryKey)
if q.LowerLimit.IsInclusive {
sbQuery.WriteString(" >=")
} else {
sbQuery.WriteString(" >")
}
sbQuery.WriteString(" @min")
}
if q.LowerLimit.IsValid && q.UpperLimit.IsValid {
sbQuery.WriteString(" AND ")
}
if q.UpperLimit.IsValid {
fmt.Fprintf(&sbQuery, "[%s]", q.PrimaryKey)
if q.UpperLimit.IsInclusive {
sbQuery.WriteString(" <=")
} else {
sbQuery.WriteString(" <")
}
sbQuery.WriteString(" @max")
}
}
fmt.Fprintf(&sbQuery, " ORDER BY [%s] ASC", q.PrimaryKey)
return sbQuery.String(), nil
}
func findColumnByPattern(columns []models.ColumnType, pattern string) (string, error) {
if pattern == "" {
return "", fmt.Errorf("column pattern cannot be empty")
}
if before, ok := strings.CutSuffix(pattern, "*"); ok {
prefix := before
for _, col := range columns {
if strings.HasPrefix(col.Name(), prefix) {
return col.Name(), nil
}
}
return "", fmt.Errorf("no column found matching pattern '%s'", pattern)
}
for _, col := range columns {
if col.Name() == pattern {
return col.Name(), nil
}
}
return "", fmt.Errorf("column '%s' not found in table columns", pattern)
}
func (mw *mssqlDbWrapper) QueryFromObject(ctx context.Context, q ExtractionQuery) (RowsResult, error) {
queryString, err := buildExtractQueryMssql(q)
if err != nil {
return nil, err
}
logrus.Debugf("Query: %s", queryString)
var queryArgs []any
if q.LowerLimit.IsValid {
queryArgs = append(queryArgs, sql.Named("min", q.LowerLimit.Value))
}
if q.UpperLimit.IsValid {
queryArgs = append(queryArgs, sql.Named("max", q.UpperLimit.Value))
}
return mw.Query(ctx, queryString, queryArgs...)
}
func buildJsonPathMssql(field string) string {
if len(field) > 0 && field[0] == '.' {
field = field[1:]
}
return "$." + field
}

View File

@@ -1,396 +0,0 @@
package dbwrapper
import (
"strings"
"testing"
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/config"
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/models"
)
func TestBuildExtractQueryMssql_NoJsonColumns(t *testing.T) {
q := ExtractionQuery{
Schema: "dbo",
Table: "Users",
PrimaryKey: "ID",
Columns: []models.ColumnType{
models.NewColumnType("ID", false, false, "INT", "int", "INT", false, 0, 0, 0),
models.NewColumnType("Name", true, false, "VARCHAR", "varchar", "VARCHAR", true, 255, 0, 0),
},
FromJsonColumns: []config.FromJsonItem{},
LowerLimit: ExtractorQueryLimit{IsValid: false},
UpperLimit: ExtractorQueryLimit{IsValid: false},
}
query, err := buildExtractQueryMssql(q)
if err != nil {
t.Fatalf("Expected no error, got: %v", err)
}
if !strings.Contains(query, "SELECT [ID], [Name]") {
t.Errorf("Expected columns in query, got: %s", query)
}
if !strings.Contains(query, "FROM [dbo].[Users]") {
t.Errorf("Expected FROM clause, got: %s", query)
}
if !strings.Contains(query, "ORDER BY [ID] ASC") {
t.Errorf("Expected ORDER BY clause, got: %s", query)
}
}
func TestBuildExtractQueryMssql_WithJsonColumns_ExactColumnMatch(t *testing.T) {
// Test that the actual column name is used as alias, not a generated one
q := ExtractionQuery{
Schema: "dbo",
Table: "Events",
PrimaryKey: "EventID",
Columns: []models.ColumnType{
models.NewColumnType("EventID", false, false, "INT", "int", "INT", false, 0, 0, 0),
models.NewColumnType("EventData", true, false, "VARCHAR", "varchar", "VARCHAR", true, 500, 0, 0),
},
FromJsonColumns: []config.FromJsonItem{
{Column: "EventData", Field: ".userId"},
{Column: "EventData", Field: ".timestamp"},
},
LowerLimit: ExtractorQueryLimit{IsValid: false},
UpperLimit: ExtractorQueryLimit{IsValid: false},
}
query, err := buildExtractQueryMssql(q)
if err != nil {
t.Fatalf("Expected no error, got: %v", err)
}
if !strings.HasPrefix(query, "SELECT [EventID], JSON_VALUE([EventData], '$.userId') AS [EventData], JSON_VALUE([EventData], '$.timestamp') AS [EventData]") {
t.Errorf("Expected JSON columns to replace EventData in-order, got: %s", query)
}
if strings.Contains(query, "SELECT [EventID], [EventData]") {
t.Errorf("Expected EventData to be replaced by JSON extraction, got: %s", query)
}
// Alias should be exactly "EventData", not "EventData_userId"
if !strings.Contains(query, "JSON_VALUE([EventData], '$.userId') AS [EventData]") {
t.Errorf("Expected JSON alias to be [EventData], got: %s", query)
}
if !strings.Contains(query, "JSON_VALUE([EventData], '$.timestamp') AS [EventData]") {
t.Errorf("Expected JSON alias to be [EventData], got: %s", query)
}
// Should have comma separating them
if !strings.Contains(query, "JSON_VALUE([EventData], '$.userId') AS [EventData], JSON_VALUE([EventData], '$.timestamp') AS [EventData]") {
t.Errorf("Expected comma-separated JSON values, got: %s", query)
}
}
func TestBuildExtractQueryMssql_WithWildcardPattern(t *testing.T) {
// Test that wildcard pattern matching finds the correct column
q := ExtractionQuery{
Schema: "dbo",
Table: "Events",
PrimaryKey: "ID",
Columns: []models.ColumnType{
models.NewColumnType("ID", false, false, "INT", "int", "INT", false, 0, 0, 0),
models.NewColumnType("NodeMetadata", true, false, "VARCHAR", "varchar", "VARCHAR", true, 500, 0, 0),
},
FromJsonColumns: []config.FromJsonItem{
{Column: "NodeMeta*", Field: ".id"},
},
LowerLimit: ExtractorQueryLimit{IsValid: false},
UpperLimit: ExtractorQueryLimit{IsValid: false},
}
query, err := buildExtractQueryMssql(q)
if err != nil {
t.Fatalf("Expected no error, got: %v", err)
}
// Should find "NodeMetadata" from pattern "NodeMeta*" and use it as alias
if !strings.Contains(query, "JSON_VALUE([NodeMetadata], '$.id') AS [NodeMetadata]") {
t.Errorf("Expected to find and use NodeMetadata column by pattern, got: %s", query)
}
if strings.Contains(query, "SELECT [ID], [NodeMetadata]") {
t.Errorf("Expected NodeMetadata to be replaced by JSON extraction, got: %s", query)
}
}
func TestBuildExtractQueryMssql_ColumnNotFound_Error(t *testing.T) {
// Test that an error is returned when column is not found
q := ExtractionQuery{
Schema: "dbo",
Table: "Events",
PrimaryKey: "ID",
Columns: []models.ColumnType{
models.NewColumnType("ID", false, false, "INT", "int", "INT", false, 0, 0, 0),
},
FromJsonColumns: []config.FromJsonItem{
{Column: "NonExistentColumn", Field: ".id"},
},
LowerLimit: ExtractorQueryLimit{IsValid: false},
UpperLimit: ExtractorQueryLimit{IsValid: false},
}
query, err := buildExtractQueryMssql(q)
if err == nil {
t.Fatalf("Expected error for missing column, got no error. Query: %s", query)
}
if !strings.Contains(err.Error(), "NonExistentColumn") {
t.Errorf("Expected error message to contain column name, got: %v", err)
}
}
func TestBuildExtractQueryMssql_WildcardPatternNotMatched_Error(t *testing.T) {
// Test that an error is returned when wildcard pattern doesn't match any column
q := ExtractionQuery{
Schema: "dbo",
Table: "Events",
PrimaryKey: "ID",
Columns: []models.ColumnType{
models.NewColumnType("ID", false, false, "INT", "int", "INT", false, 0, 0, 0),
models.NewColumnType("EventData", true, false, "VARCHAR", "varchar", "VARCHAR", true, 500, 0, 0),
},
FromJsonColumns: []config.FromJsonItem{
{Column: "NonMatching*", Field: ".id"},
},
LowerLimit: ExtractorQueryLimit{IsValid: false},
UpperLimit: ExtractorQueryLimit{IsValid: false},
}
query, err := buildExtractQueryMssql(q)
if err == nil {
t.Fatalf("Expected error for non-matching wildcard pattern, got no error. Query: %s", query)
}
if !strings.Contains(err.Error(), "NonMatching*") {
t.Errorf("Expected error message to contain pattern, got: %v", err)
}
}
func TestBuildExtractQueryMssql_NestedJsonFields(t *testing.T) {
q := ExtractionQuery{
Schema: "dbo",
Table: "Data",
PrimaryKey: "ID",
Columns: []models.ColumnType{
models.NewColumnType("ID", false, false, "INT", "int", "INT", false, 0, 0, 0),
models.NewColumnType("NodeData", true, false, "VARCHAR", "varchar", "VARCHAR", true, 500, 0, 0),
},
FromJsonColumns: []config.FromJsonItem{
{Column: "NodeData", Field: ".user.name"},
{Column: "NodeData", Field: ".user.email"},
},
LowerLimit: ExtractorQueryLimit{IsValid: false},
UpperLimit: ExtractorQueryLimit{IsValid: false},
}
query, err := buildExtractQueryMssql(q)
if err != nil {
t.Fatalf("Expected no error, got: %v", err)
}
if !strings.Contains(query, "JSON_VALUE([NodeData], '$.user.name') AS [NodeData]") {
t.Errorf("Expected nested JSON path for user.name, got: %s", query)
}
if !strings.Contains(query, "JSON_VALUE([NodeData], '$.user.email') AS [NodeData]") {
t.Errorf("Expected nested JSON path for user.email, got: %s", query)
}
if strings.Contains(query, "SELECT [ID], [NodeData]") {
t.Errorf("Expected NodeData to be replaced by JSON extraction, got: %s", query)
}
}
func TestBuildExtractQueryMssql_WithRangeLimits(t *testing.T) {
q := ExtractionQuery{
Schema: "dbo",
Table: "Products",
PrimaryKey: "ProductID",
Columns: []models.ColumnType{
models.NewColumnType("ProductID", false, false, "INT", "int", "INT", false, 0, 0, 0),
models.NewColumnType("Details", true, false, "VARCHAR", "varchar", "VARCHAR", true, 500, 0, 0),
},
FromJsonColumns: []config.FromJsonItem{
{Column: "Details", Field: ".price"},
},
LowerLimit: ExtractorQueryLimit{IsValid: true, IsInclusive: true, Value: 100},
UpperLimit: ExtractorQueryLimit{IsValid: true, IsInclusive: false, Value: 500},
}
query, err := buildExtractQueryMssql(q)
if err != nil {
t.Fatalf("Expected no error, got: %v", err)
}
if !strings.Contains(query, "WHERE [ProductID] >= @min") {
t.Errorf("Expected WHERE clause with >=, got: %s", query)
}
if !strings.Contains(query, "[ProductID] < @max") {
t.Errorf("Expected upper limit with <, got: %s", query)
}
if !strings.Contains(query, "JSON_VALUE([Details], '$.price') AS [Details]") {
t.Errorf("Expected JSON_VALUE for Details, got: %s", query)
}
if strings.Contains(query, "SELECT [ProductID], [Details]") {
t.Errorf("Expected Details to be replaced by JSON extraction, got: %s", query)
}
}
func TestBuildJsonPathMssql(t *testing.T) {
tests := []struct {
input string
expected string
}{
{".id", "$.id"},
{"id", "$.id"},
{".user.name", "$.user.name"},
{"user.name", "$.user.name"},
{".location.coordinates.lat", "$.location.coordinates.lat"},
{"", "$."},
}
for _, tt := range tests {
result := buildJsonPathMssql(tt.input)
if result != tt.expected {
t.Errorf("buildJsonPathMssql(%q) = %q, want %q", tt.input, result, tt.expected)
}
}
}
func TestFindColumnByPattern_ExactMatch(t *testing.T) {
columns := []models.ColumnType{
models.NewColumnType("ID", false, false, "INT", "int", "INT", false, 0, 0, 0),
models.NewColumnType("Metadata", true, false, "VARCHAR", "varchar", "VARCHAR", true, 500, 0, 0),
models.NewColumnType("EventData", true, false, "VARCHAR", "varchar", "VARCHAR", true, 500, 0, 0),
}
result, err := findColumnByPattern(columns, "Metadata")
if err != nil {
t.Fatalf("Expected no error, got: %v", err)
}
if result != "Metadata" {
t.Errorf("Expected 'Metadata', got '%s'", result)
}
}
func TestFindColumnByPattern_WildcardMatch(t *testing.T) {
columns := []models.ColumnType{
models.NewColumnType("ID", false, false, "INT", "int", "INT", false, 0, 0, 0),
models.NewColumnType("NodeMetadata", true, false, "VARCHAR", "varchar", "VARCHAR", true, 500, 0, 0),
models.NewColumnType("EventData", true, false, "VARCHAR", "varchar", "VARCHAR", true, 500, 0, 0),
}
result, err := findColumnByPattern(columns, "NodeMeta*")
if err != nil {
t.Fatalf("Expected no error, got: %v", err)
}
if result != "NodeMetadata" {
t.Errorf("Expected 'NodeMetadata', got '%s'", result)
}
}
func TestFindColumnByPattern_NotFound(t *testing.T) {
columns := []models.ColumnType{
models.NewColumnType("ID", false, false, "INT", "int", "INT", false, 0, 0, 0),
models.NewColumnType("Metadata", true, false, "VARCHAR", "varchar", "VARCHAR", true, 500, 0, 0),
}
result, err := findColumnByPattern(columns, "NonExistent")
if err == nil {
t.Fatalf("Expected error, got no error. Result: %s", result)
}
if !strings.Contains(err.Error(), "NonExistent") {
t.Errorf("Expected error to contain column name, got: %v", err)
}
}
func TestFindColumnByPattern_WildcardNotFound(t *testing.T) {
columns := []models.ColumnType{
models.NewColumnType("ID", false, false, "INT", "int", "INT", false, 0, 0, 0),
models.NewColumnType("Metadata", true, false, "VARCHAR", "varchar", "VARCHAR", true, 500, 0, 0),
}
result, err := findColumnByPattern(columns, "Event*")
if err == nil {
t.Fatalf("Expected error, got no error. Result: %s", result)
}
if !strings.Contains(err.Error(), "Event*") {
t.Errorf("Expected error to contain pattern, got: %v", err)
}
}
func TestBuildExtractQueryMssql_OnlyJsonColumns(t *testing.T) {
// Test when all columns are used via JSON extraction
q := ExtractionQuery{
Schema: "dbo",
Table: "Data",
PrimaryKey: "ID",
Columns: []models.ColumnType{
models.NewColumnType("ID", false, false, "INT", "int", "INT", false, 0, 0, 0),
models.NewColumnType("JsonData", true, false, "VARCHAR", "varchar", "VARCHAR", true, 500, 0, 0),
},
FromJsonColumns: []config.FromJsonItem{
{Column: "JsonData", Field: ".field1"},
},
LowerLimit: ExtractorQueryLimit{IsValid: false},
UpperLimit: ExtractorQueryLimit{IsValid: false},
}
query, err := buildExtractQueryMssql(q)
if err != nil {
t.Fatalf("Expected no error, got: %v", err)
}
if !strings.HasPrefix(query, "SELECT [ID], JSON_VALUE([JsonData], '$.field1') AS [JsonData]") {
t.Errorf("Expected JsonData to be replaced by JSON extraction, got: %s", query)
}
if strings.Contains(query, "SELECT [ID], [JsonData]") {
t.Errorf("Expected JsonData to be excluded from raw selection, got: %s", query)
}
}
func TestBuildExtractQueryMssql_JsonColumnsReplaceInOrder(t *testing.T) {
q := ExtractionQuery{
Schema: "dbo",
Table: "Users",
PrimaryKey: "UserID",
Columns: []models.ColumnType{
models.NewColumnType("UserID", false, false, "INT", "int", "INT", false, 0, 0, 0),
models.NewColumnType("Name", true, false, "VARCHAR", "varchar", "VARCHAR", false, 255, 0, 0),
models.NewColumnType("Email", true, false, "VARCHAR", "varchar", "VARCHAR", false, 255, 0, 0),
models.NewColumnType("Metadata", true, false, "NVARCHAR", "nvarchar", "NVARCHAR", true, 4000, 0, 0),
models.NewColumnType("Profile", true, false, "NVARCHAR", "nvarchar", "NVARCHAR", true, 4000, 0, 0),
models.NewColumnType("Settings", true, false, "NVARCHAR", "nvarchar", "NVARCHAR", true, 4000, 0, 0),
},
FromJsonColumns: []config.FromJsonItem{
{Column: "Metadata", Field: ".id"},
{Column: "Profile", Field: ".id"},
{Column: "Settings", Field: ".id"},
},
LowerLimit: ExtractorQueryLimit{IsValid: false},
UpperLimit: ExtractorQueryLimit{IsValid: false},
}
query, err := buildExtractQueryMssql(q)
if err != nil {
t.Fatalf("Expected no error, got: %v", err)
}
expected := "SELECT [UserID], [Name], [Email], JSON_VALUE([Metadata], '$.id') AS [Metadata], JSON_VALUE([Profile], '$.id') AS [Profile], JSON_VALUE([Settings], '$.id') AS [Settings] FROM [dbo].[Users] ORDER BY [UserID] ASC"
if query != expected {
t.Errorf("Unexpected query.\nExpected: %s\nGot: %s", expected, query)
}
}

View File

@@ -3,17 +3,14 @@ package dbwrapper
import (
"context"
"errors"
"fmt"
"strings"
dbdialects "git.ksdemosapps.com/kylesoda/go-migrate/internal/app/db-wrapper/db_dialects"
"github.com/jackc/pgx/v5"
"github.com/jackc/pgx/v5/pgxpool"
)
func init() {
Register(dbdialects.Postgres, func() DbWrapper {
return &postgresDbWrapper{dialect: dbdialects.Postgres}
Register("postgres", func() DbWrapper {
return &postgresDbWrapper{dialect: "postgres"}
})
}
@@ -129,75 +126,3 @@ func (pw *postgresDbWrapper) SaveMassive(ctx context.Context, schema string, tab
return affectedRows, nil
}
func (pw *postgresDbWrapper) QueryFromObject(ctx context.Context, q ExtractionQuery) (RowsResult, error) {
var sbQuery strings.Builder
sbQuery.WriteString("SELECT ")
if len(q.Columns) == 0 {
sbQuery.WriteString("*")
} else {
for i, col := range q.Columns {
switch col.Type() {
case "GEOMETRY":
fmt.Fprintf(&sbQuery, `ST_AsEWKB("%s") AS "%s"`, col.Name(), col.Name())
default:
fmt.Fprintf(&sbQuery, `"%s"`, col.Name())
}
if i < len(q.Columns)-1 {
sbQuery.WriteString(", ")
}
}
}
fmt.Fprintf(&sbQuery, ` FROM "%s"."%s"`, q.Schema, q.Table)
if q.LowerLimit.IsValid || q.UpperLimit.IsValid {
sbQuery.WriteString(" WHERE ")
paramIdx := 1
if q.LowerLimit.IsValid {
fmt.Fprintf(&sbQuery, `"%s"`, q.PrimaryKey)
if q.LowerLimit.IsInclusive {
sbQuery.WriteString(" >=")
} else {
sbQuery.WriteString(" >")
}
fmt.Fprintf(&sbQuery, " $%d", paramIdx)
paramIdx++
}
if q.LowerLimit.IsValid && q.UpperLimit.IsValid {
sbQuery.WriteString(" AND ")
}
if q.UpperLimit.IsValid {
fmt.Fprintf(&sbQuery, `"%s"`, q.PrimaryKey)
if q.UpperLimit.IsInclusive {
sbQuery.WriteString(" <=")
} else {
sbQuery.WriteString(" <")
}
fmt.Fprintf(&sbQuery, " $%d", paramIdx)
paramIdx++
}
}
fmt.Fprintf(&sbQuery, ` ORDER BY "%s" ASC`, q.PrimaryKey)
queryString := sbQuery.String()
var queryArgs []any
if q.LowerLimit.IsValid {
queryArgs = append(queryArgs, q.LowerLimit.Value)
}
if q.UpperLimit.IsValid {
queryArgs = append(queryArgs, q.UpperLimit.Value)
}
return pw.Query(ctx, queryString, queryArgs...)
}

View File

@@ -3,9 +3,6 @@ package dbwrapper
import (
"context"
"errors"
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/config"
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/models"
)
var MethodNotSupported error = errors.New("Method not supported by driver... yet :P")
@@ -27,22 +24,6 @@ type RowResult interface {
Scan(dest ...any) error
}
type ExtractorQueryLimit struct {
IsValid bool
IsInclusive bool
Value int64
}
type ExtractionQuery struct {
Schema string
Table string
PrimaryKey string
Columns []models.ColumnType
LowerLimit ExtractorQueryLimit
UpperLimit ExtractorQueryLimit
FromJsonColumns []config.FromJsonItem
}
type DbWrapper interface {
Close() error
Connect(ctx context.Context, dbUrl string) error
@@ -51,5 +32,4 @@ type DbWrapper interface {
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)
QueryFromObject(ctx context.Context, query ExtractionQuery) (RowsResult, error)
}

View File

@@ -1,96 +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/models"
"github.com/sirupsen/logrus"
)
func (ex *GenericExtractor) Consume(
ctx context.Context,
tableInfo config.SourceTableInfo,
columns []models.ColumnType,
batchSize int,
retryConfig config.RetryConfig,
chPartitionsIn <-chan models.Partition,
chBatchesOut chan<- models.Batch,
chErrorsOut chan<- custom_errors.JobError,
wgActivePartitions *sync.WaitGroup,
rowsRead *int64,
fromJsonColumns []config.FromJsonItem,
) {
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 chErrorsOut <- 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 := ex.ProcessPartitionWithRetries(
ctx,
tableInfo,
columns,
batchSize,
partition,
indexPrimaryKey,
retryConfig,
chBatchesOut,
fromJsonColumns,
)
wgActivePartitions.Done()
if rowsReadResult > 0 {
current := atomic.LoadInt64(rowsRead)
logrus.Debugf("Rows read: +%v [current=%v] (%s.%s)", rowsReadResult, current, tableInfo.Schema, tableInfo.Table)
atomic.AddInt64(rowsRead, int64(rowsReadResult))
}
if err != nil {
if jobError, ok := errors.AsType[*custom_errors.JobError](err); ok {
select {
case <-ctx.Done():
return
case chErrorsOut <- *jobError:
}
} else {
select {
case <-ctx.Done():
return
case chErrorsOut <- custom_errors.JobError{ShouldCancelJob: false, Msg: err.Error(), Prev: err}:
}
}
}
}
}
}

View File

@@ -1,42 +0,0 @@
package extractors
import (
"context"
dbwrapper "git.ksdemosapps.com/kylesoda/go-migrate/internal/app/db-wrapper"
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/models"
"github.com/google/uuid"
)
type GenericExtractor struct {
db dbwrapper.DbWrapper
}
func NewExtractor(db dbwrapper.DbWrapper) GenericExtractor {
return GenericExtractor{db: db}
}
func sendBatch(ctx context.Context, chBatchesOut chan<- models.Batch, batch models.Batch) error {
select {
case chBatchesOut <- batch:
return nil
case <-ctx.Done():
return ctx.Err()
}
}
func flush(
ctx context.Context,
partition *models.Partition,
batchSize int,
batchRows []models.UnknownRowValues,
chBatchesOut chan<- models.Batch,
) error {
if len(batchRows) == 0 {
return nil
}
batch := models.Batch{Id: uuid.New(), PartitionId: partition.Id, Rows: batchRows}
batchRows = make([]models.UnknownRowValues, 0, batchSize)
return sendBatch(ctx, chBatchesOut, batch)
}

View File

@@ -0,0 +1,277 @@
package extractors
import (
"context"
"database/sql"
"errors"
"fmt"
"slices"
"strings"
"sync"
"sync/atomic"
"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/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/models"
"github.com/google/uuid"
)
type MssqlExtractor struct {
db dbwrapper.DbWrapper
}
func NewMssqlExtractor(db dbwrapper.DbWrapper) etl.Extractor {
return &MssqlExtractor{db: db}
}
func buildExtractQueryMssql(
tableInfo config.SourceTableInfo,
columns []models.ColumnType,
includeRange bool,
isMinInclusive bool,
) string {
var sbQuery strings.Builder
sbQuery.WriteString("SELECT ")
if len(columns) == 0 {
sbQuery.WriteString("*")
} else {
for i, col := range columns {
fmt.Fprintf(&sbQuery, "[%s]", col.Name())
if col.Type() == "GEOMETRY" {
fmt.Fprintf(&sbQuery, ".STAsBinary() AS [%s]", col.Name())
}
if i < len(columns)-1 {
sbQuery.WriteString(", ")
}
}
}
fmt.Fprintf(&sbQuery, " FROM [%s].[%s]", tableInfo.Schema, tableInfo.Table)
if includeRange {
fmt.Fprintf(&sbQuery, " WHERE [%s]", tableInfo.PrimaryKey)
if isMinInclusive {
sbQuery.WriteString(" >=")
} else {
sbQuery.WriteString(" >")
}
fmt.Fprintf(&sbQuery, " @min AND [%s] <= @max", tableInfo.PrimaryKey)
}
fmt.Fprintf(&sbQuery, " ORDER BY [%s] ASC", tableInfo.PrimaryKey)
return sbQuery.String()
}
func errorFromLastRow(
lastRow models.UnknownRowValues,
indexPrimaryKey int,
partition models.Partition,
previousError error,
) *custom_errors.ExtractorError {
lastIdRawValue := lastRow[indexPrimaryKey]
lastId, ok := convert.ToInt64(lastIdRawValue)
if !ok {
currentPartition := partition
currentPartition.RetryCounter = 3
return &custom_errors.ExtractorError{
Partition: currentPartition,
HasLastId: true,
Msg: fmt.Sprintf("Couldn't cast last id value as int: %s", previousError.Error()),
}
}
return &custom_errors.ExtractorError{
Partition: partition,
HasLastId: true,
LastId: lastId,
Msg: previousError.Error(),
}
}
func (mssqlEx *MssqlExtractor) ProcessPartition(
ctx context.Context,
tableInfo config.SourceTableInfo,
columns []models.ColumnType,
batchSize int,
partition models.Partition,
indexPrimaryKey int,
chBatchesOut chan<- models.Batch,
) (int, error) {
query := buildExtractQueryMssql(tableInfo, columns, partition.HasRange, partition.Range.IsMinInclusive)
var queryArgs []any
if partition.HasRange {
queryArgs = append(queryArgs,
sql.Named("min", partition.Range.Min),
sql.Named("max", partition.Range.Max),
)
}
rowsRead := 0
rows, err := mssqlEx.db.Query(ctx, query, queryArgs...)
if err != nil {
return rowsRead, &custom_errors.ExtractorError{Partition: partition, HasLastId: false, Msg: err.Error()}
}
defer rows.Close()
batchRows := make([]models.UnknownRowValues, 0, batchSize)
for rows.Next() {
rowValues := make([]any, len(columns))
scanArgs := make([]any, len(columns))
for i := range rowValues {
scanArgs[i] = &rowValues[i]
}
if err := rows.Scan(scanArgs...); err != nil {
if len(batchRows) == 0 {
return rowsRead, &custom_errors.ExtractorError{Partition: partition, HasLastId: false, Msg: err.Error()}
}
lastRow := batchRows[len(batchRows)-1]
select {
case chBatchesOut <- models.Batch{Id: uuid.New(), PartitionId: partition.Id, Rows: batchRows, RetryCounter: 0}:
case <-ctx.Done():
return rowsRead, ctx.Err()
}
return rowsRead, errorFromLastRow(lastRow, indexPrimaryKey, partition, err)
}
rowsRead++
batchRows = append(batchRows, rowValues)
if len(batchRows) >= batchSize {
select {
case chBatchesOut <- models.Batch{Id: uuid.New(), PartitionId: partition.Id, Rows: batchRows, RetryCounter: 0}:
case <-ctx.Done():
return rowsRead, ctx.Err()
}
batchRows = make([]models.UnknownRowValues, 0, batchSize)
}
}
if err := rows.Err(); err != nil {
if errors.Is(err, ctx.Err()) {
return rowsRead, ctx.Err()
}
if len(batchRows) > 0 {
lastRow := batchRows[len(batchRows)-1]
return rowsRead, errorFromLastRow(lastRow, indexPrimaryKey, partition, err)
}
return rowsRead, &custom_errors.ExtractorError{Partition: partition, HasLastId: false, Msg: err.Error()}
}
if len(batchRows) > 0 {
select {
case chBatchesOut <- models.Batch{Id: uuid.New(), PartitionId: partition.Id, Rows: batchRows, RetryCounter: 0}:
case <-ctx.Done():
return rowsRead, ctx.Err()
}
}
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()
}
}
}

View File

@@ -0,0 +1,125 @@
package extractors
import (
"context"
"errors"
"fmt"
"strings"
"sync"
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/config"
"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/models"
"github.com/google/uuid"
)
type PostgresExtractor struct {
db dbwrapper.DbWrapper
}
func NewPostgresExtractor(db dbwrapper.DbWrapper) etl.Extractor {
return &PostgresExtractor{db: db}
}
func buildExtractQueryPostgres(sourceDbInfo config.SourceTableInfo, columns []models.ColumnType) string {
var sbColumns strings.Builder
if len(columns) == 0 {
sbColumns.WriteString("*")
} else {
for i, col := range columns {
if col.Type() == "GEOMETRY" {
sbColumns.WriteString(`ST_AsEWKB("`)
sbColumns.WriteString(col.Name())
sbColumns.WriteString(`") AS "`)
sbColumns.WriteString(col.Name())
sbColumns.WriteString(`"`)
} else {
sbColumns.WriteString(`"`)
sbColumns.WriteString(col.Name())
sbColumns.WriteString(`"`)
}
if i < len(columns)-1 {
sbColumns.WriteString(", ")
}
}
}
return fmt.Sprintf(`SELECT %s FROM "%s"."%s" ORDER BY "%s" ASC`, sbColumns.String(), sourceDbInfo.Schema, sourceDbInfo.Table, sourceDbInfo.PrimaryKey)
}
func (postgresEx *PostgresExtractor) ProcessPartition(
ctx context.Context,
tableInfo config.SourceTableInfo,
columns []models.ColumnType,
batchSize int,
partition models.Partition,
indexPrimaryKey int,
chBatchesOut chan<- models.Batch,
) (int, error) {
query := buildExtractQueryPostgres(tableInfo, columns)
if partition.HasRange {
return 0, errors.New("Batch config not yet supported")
}
rowsRead := 0
rows, err := postgresEx.db.Query(ctx, query)
if err != nil {
return rowsRead, &custom_errors.ExtractorError{Partition: partition, HasLastId: false, Msg: err.Error()}
}
defer rows.Close()
batchRows := make([]models.UnknownRowValues, 0, batchSize)
for rows.Next() {
values, err := rows.Values()
if err != nil {
return rowsRead, errors.New("Unexpected error reading rows from source")
}
rowsRead++
batchRows = append(batchRows, values)
if len(batchRows) >= batchSize {
select {
case chBatchesOut <- models.Batch{Id: uuid.New(), PartitionId: partition.Id, Rows: batchRows, RetryCounter: 0}:
case <-ctx.Done():
return rowsRead, ctx.Err()
}
batchRows = make([]models.UnknownRowValues, 0, batchSize)
}
}
if err := rows.Err(); err != nil {
return rowsRead, errors.New("Unexpected error reading rows from source")
}
if len(batchRows) > 0 {
select {
case chBatchesOut <- models.Batch{Id: uuid.New(), PartitionId: partition.Id, Rows: batchRows, RetryCounter: 0}:
case <-ctx.Done():
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,
) {
}

View File

@@ -1,77 +0,0 @@
package extractors
import (
"context"
"errors"
"fmt"
"time"
"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/models"
"github.com/google/uuid"
// "github.com/sirupsen/logrus"
)
func (ex *GenericExtractor) ProcessPartitionWithRetries(
ctx context.Context,
tableInfo config.SourceTableInfo,
columns []models.ColumnType,
batchSize int,
partition models.Partition,
indexPrimaryKey int,
retryConfig config.RetryConfig,
chBatchesOut chan<- models.Batch,
fromJsonColumns []config.FromJsonItem,
) (int64, error) {
var totalRowsRead int64
currentParitition := partition
for {
rowsRead, err := ex.ProcessPartition(
ctx,
tableInfo,
columns,
batchSize,
currentParitition,
indexPrimaryKey,
chBatchesOut,
fromJsonColumns,
)
// logrus.Debugf("Partition %v finished processing (%s.%s)", partition.Id, tableInfo.Schema, tableInfo.Table)
totalRowsRead += rowsRead
if err == nil {
return totalRowsRead, nil
}
if exError, ok := errors.AsType[*custom_errors.ExtractorError](err); ok {
currentParitition.RetryCounter++
if currentParitition.RetryCounter >= retryConfig.Attempts {
return totalRowsRead, &custom_errors.JobError{
Msg: fmt.Sprintf("Partition %v reached max retries", exError.Partition.Id),
Prev: err,
}
}
if exError.HasLastId {
currentParitition.ParentId = exError.Partition.Id
currentParitition.Id = uuid.New()
currentParitition.Range.Min = exError.LastId
currentParitition.Range.IsMinInclusive = false
}
delay := custom_errors.ComputeBackoffDelay(
currentParitition.RetryCounter,
retryConfig.BaseDelayMs,
retryConfig.MaxDelayMs,
retryConfig.MaxJitterMs,
)
time.Sleep(delay)
continue
}
return totalRowsRead, err
}
}

View File

@@ -1,118 +0,0 @@
package extractors
import (
"context"
"fmt"
"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/custom_errors"
dbwrapper "git.ksdemosapps.com/kylesoda/go-migrate/internal/app/db-wrapper"
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/models"
// "github.com/sirupsen/logrus"
)
func errorFromLastPartitionRow(
lastRow models.UnknownRowValues,
indexPrimaryKey int,
partition models.Partition,
previousError error,
) error {
lastIdRawValue := lastRow[indexPrimaryKey]
lastId, ok := convert.ToInt64(lastIdRawValue)
if !ok {
currentPartition := partition
currentPartition.RetryCounter = 3
return &custom_errors.ExtractorError{
Partition: currentPartition,
HasLastId: true,
Msg: fmt.Sprintf("Couldn't cast last id value as int: %s", previousError.Error()),
}
}
return &custom_errors.ExtractorError{
Partition: partition,
HasLastId: true,
LastId: lastId,
Msg: previousError.Error(),
}
}
func (ex *GenericExtractor) ProcessPartition(
ctx context.Context,
tableInfo config.SourceTableInfo,
columns []models.ColumnType,
batchSize int,
partition models.Partition,
indexPrimaryKey int,
chBatchesOut chan<- models.Batch,
fromJsonColumns []config.FromJsonItem,
) (int64, error) {
query := dbwrapper.ExtractionQuery{
Schema: tableInfo.Schema,
Table: tableInfo.Table,
PrimaryKey: tableInfo.PrimaryKey,
Columns: columns,
LowerLimit: dbwrapper.ExtractorQueryLimit{
IsValid: partition.HasRange && partition.Range.Min > 0,
IsInclusive: partition.Range.IsMinInclusive,
Value: partition.Range.Min,
},
UpperLimit: dbwrapper.ExtractorQueryLimit{
IsValid: partition.HasRange && partition.Range.Max > 0,
IsInclusive: partition.Range.IsMaxInclusive,
Value: partition.Range.Max,
},
FromJsonColumns: fromJsonColumns,
}
// logrus.Debugf("Processing partition: %+v (%s.%s)", query, tableInfo.Schema, tableInfo.Table)
rows, err := ex.db.QueryFromObject(ctx, query)
if err != nil {
return 0, err
}
defer rows.Close()
batchRows := make([]models.UnknownRowValues, 0, batchSize)
var rowsRead int64 = 0
for rows.Next() {
rowValues := make([]any, len(columns))
scanArgs := make([]any, len(columns))
for i := range rowValues {
scanArgs[i] = &rowValues[i]
}
if err := rows.Scan(scanArgs...); err != nil {
if len(batchRows) == 0 {
return rowsRead, err
}
if err := flush(ctx, &partition, batchSize, batchRows, chBatchesOut); err != nil {
return rowsRead, err
}
lastRow := batchRows[len(batchRows)-1]
return rowsRead, errorFromLastPartitionRow(lastRow, indexPrimaryKey, partition, err)
}
rowsRead++
batchRows = append(batchRows, rowValues)
if len(batchRows) >= batchSize {
// logrus.Debugf("Batch size reached, flushing batch with %v rows (rowsRead=%v)", len(batchRows), rowsRead)
if err := flush(ctx, &partition, batchSize, batchRows, chBatchesOut); err != nil {
// logrus.Warnf("Error flushing rows: %v", err)
return rowsRead, err
}
batchRows = make([]models.UnknownRowValues, 0, batchSize)
}
}
if err := flush(ctx, &partition, batchSize, batchRows, chBatchesOut); err != nil {
return rowsRead, err
}
return rowsRead, rows.Err()
}

View File

@@ -1,63 +0,0 @@
package loaders
import (
"context"
"errors"
"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/models"
)
func (gl *GenericLoader) Consume(
ctx context.Context,
tableInfo config.TargetTableInfo,
columns []models.ColumnType,
retryConfig config.RetryConfig,
chBatchesIn <-chan models.Batch,
chErrorsOut chan<- custom_errors.JobError,
wgActiveBatches *sync.WaitGroup,
rowsLoaded *int64,
) {
colNames := mapSlice(columns, func(col models.ColumnType) string {
return col.Name()
})
for {
if ctx.Err() != nil {
return
}
select {
case <-ctx.Done():
return
case batch, ok := <-chBatchesIn:
if !ok {
return
}
processedRows, err := gl.ProcessBatchWithRetries(ctx, tableInfo, colNames, retryConfig, batch)
wgActiveBatches.Done()
if err != nil {
if jobError, ok := errors.AsType[*custom_errors.JobError](err); ok {
select {
case <-ctx.Done():
return
case chErrorsOut <- *jobError:
}
} else {
select {
case <-ctx.Done():
return
case chErrorsOut <- custom_errors.JobError{ShouldCancelJob: false, Msg: err.Error(), Prev: err}:
}
}
} else {
atomic.AddInt64(rowsLoaded, int64(processedRows))
}
}
}
}

View File

@@ -1,13 +0,0 @@
package loaders
import (
dbwrapper "git.ksdemosapps.com/kylesoda/go-migrate/internal/app/db-wrapper"
)
type GenericLoader struct {
db dbwrapper.DbWrapper
}
func NewGenericLoader(db dbwrapper.DbWrapper) GenericLoader {
return GenericLoader{db: db}
}

View File

@@ -0,0 +1,127 @@
package loaders
import (
"context"
"errors"
"fmt"
"sync"
"sync/atomic"
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/config"
"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/models"
"github.com/jackc/pgx/v5/pgconn"
)
type PostgresLoader struct {
db dbwrapper.DbWrapper
}
func NewPostgresLoader(db dbwrapper.DbWrapper) etl.Loader {
return &PostgresLoader{db: db}
}
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,
tableInfo config.TargetTableInfo,
colNames []string,
batch models.Batch,
) (int, error) {
_, err := postgresLd.db.SaveMassive(
ctx,
tableInfo.Schema,
tableInfo.Table,
colNames,
batch.Rows,
)
if err != nil {
var pgErr *pgconn.PgError
if errors.As(err, &pgErr) {
if pgErr.Code == "23505" {
return 0, &custom_errors.JobError{
ShouldCancelJob: true,
Msg: fmt.Sprintf("Fatal error in table %s.%s", tableInfo.Schema, tableInfo.Table),
Prev: err,
}
}
}
return 0, &custom_errors.LoaderError{Batch: batch, Msg: err.Error()}
}
return len(batch.Rows), nil
}
func (postgresLd *PostgresLoader) Exec(
ctx context.Context,
tableInfo config.TargetTableInfo,
columns []models.ColumnType,
chBatchesIn <-chan models.Batch,
chErrorsOut chan<- custom_errors.LoaderError,
chJobErrorsOut chan<- custom_errors.JobError,
wgActiveBatches *sync.WaitGroup,
rowsLoaded *int64,
) {
colNames := mapSlice(columns, func(col models.ColumnType) string {
return col.Name()
})
for {
if ctx.Err() != nil {
return
}
select {
case <-ctx.Done():
return
case batch, ok := <-chBatchesIn:
if !ok {
return
}
processedRows, err := postgresLd.ProcessBatch(ctx, tableInfo, colNames, batch)
if err != nil {
var ldError *custom_errors.LoaderError
var jobError *custom_errors.JobError
if errors.As(err, &ldError) {
select {
case <-ctx.Done():
return
case chErrorsOut <- *ldError:
}
} else if errors.As(err, &jobError) {
select {
case <-ctx.Done():
return
case chJobErrorsOut <- *jobError:
}
} else {
select {
case <-ctx.Done():
return
case chErrorsOut <- custom_errors.LoaderError{Batch: batch, Msg: err.Error()}:
}
}
continue
}
wgActiveBatches.Done()
atomic.AddInt64(rowsLoaded, int64(processedRows))
}
}
}

View File

@@ -1,50 +0,0 @@
package loaders
import (
"context"
"errors"
"fmt"
"time"
"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/models"
)
func (gl *GenericLoader) ProcessBatchWithRetries(
ctx context.Context,
tableInfo config.TargetTableInfo,
colNames []string,
retryConfig config.RetryConfig,
batch models.Batch,
) (int64, error) {
for {
rowsLoaded, err := gl.ProcessBatch(ctx, tableInfo, colNames, batch)
if err == nil {
return rowsLoaded, nil
}
if btError, ok := errors.AsType[*custom_errors.LoaderError](err); ok {
batch.RetryCounter++
if batch.RetryCounter >= retryConfig.Attempts {
return rowsLoaded, &custom_errors.JobError{
ShouldCancelJob: false,
Msg: fmt.Sprintf("Temporal error in batch %v (retries: %d)", btError.Batch.Id, btError.Batch.RetryCounter),
Prev: btError,
}
}
delay := custom_errors.ComputeBackoffDelay(
batch.RetryCounter,
retryConfig.BaseDelayMs,
retryConfig.MaxDelayMs,
retryConfig.MaxJitterMs,
)
time.Sleep(delay)
continue
}
return rowsLoaded, err
}
}

View File

@@ -1,43 +0,0 @@
package loaders
import (
"context"
"errors"
"fmt"
"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/models"
"github.com/jackc/pgx/v5/pgconn"
)
func (gl *GenericLoader) ProcessBatch(
ctx context.Context,
tableInfo config.TargetTableInfo,
colNames []string,
batch models.Batch,
) (int64, error) {
_, err := gl.db.SaveMassive(
ctx,
tableInfo.Schema,
tableInfo.Table,
colNames,
batch.Rows,
)
if err != nil {
if pgErr, ok := errors.AsType[*pgconn.PgError](err); ok {
if pgErr.Code == "23505" {
return 0, &custom_errors.JobError{
ShouldCancelJob: true,
Msg: fmt.Sprintf("Fatal error in table %s.%s", tableInfo.Schema, tableInfo.Table),
Prev: err,
}
}
}
return 0, &custom_errors.LoaderError{Batch: batch, Msg: err.Error()}
}
return int64(len(batch.Rows)), nil
}

View File

@@ -0,0 +1 @@
package loaders

View File

@@ -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
}

View File

@@ -7,7 +7,6 @@ import (
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/etl"
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/models"
"github.com/google/uuid"
"github.com/sirupsen/logrus"
)
func PartitionRangeGenerator(
@@ -16,24 +15,8 @@ func PartitionRangeGenerator(
tableInfo config.TableInfo,
partitionColumn string,
rowsPerPartition int64,
jobRange config.RangeConfig,
) ([]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)
logrus.Infof("Estimated rows in source: %v (%s.%s)", rowsCount, tableInfo.Schema, tableInfo.Table)
if err != nil {
return nil, err
}
@@ -53,7 +36,5 @@ func PartitionRangeGenerator(
return nil, err
}
// logrus.Debugf("Partitions: %+v (%s.%s)", partitions, tableInfo.Schema, tableInfo.Table)
return partitions, nil
}

View File

@@ -234,7 +234,6 @@ ORDER BY batch_id`,
RetryCounter: 0,
Range: models.PartitionRange{
IsMinInclusive: true,
IsMaxInclusive: true,
},
}

View File

@@ -3,32 +3,18 @@ package transformers
import (
"context"
"errors"
"fmt"
"strings"
"sync"
"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/etl"
"git.ksdemosapps.com/kylesoda/go-migrate/internal/app/models"
"github.com/google/uuid"
log "github.com/sirupsen/logrus"
)
type MssqlTransformer struct {
toStorage config.ToStorageConfig
sourceTable config.SourceTableInfo
azureClient *azure.Client
}
type MssqlTransformer struct{}
func NewMssqlTransformer(toStorage config.ToStorageConfig, sourceTable config.SourceTableInfo, azureClient *azure.Client) etl.Transformer {
return &MssqlTransformer{
toStorage: toStorage,
sourceTable: sourceTable,
azureClient: azureClient,
}
func NewMssqlTransformer() etl.Transformer {
return &MssqlTransformer{}
}
func computeTransformationPlan(columns []models.ColumnType) []etl.ColumnTransformPlan {
@@ -74,65 +60,6 @@ func computeTransformationPlan(columns []models.ColumnType) []etl.ColumnTransfor
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
func (mssqlTr *MssqlTransformer) ProcessBatch(
@@ -173,8 +100,6 @@ func (mssqlTr *MssqlTransformer) Exec(
wgActiveBatches *sync.WaitGroup,
) {
transformationPlan := computeTransformationPlan(columns)
storagePlan := computeStorageTransformationPlan(ctx, mssqlTr.azureClient, mssqlTr.toStorage, columns, mssqlTr.sourceTable)
transformationPlan = append(transformationPlan, storagePlan...)
for {
if ctx.Err() != nil {

View File

@@ -19,6 +19,19 @@ type Extractor interface {
indexPrimaryKey int,
chBatchesOut chan<- models.Batch,
) (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)
@@ -51,7 +64,18 @@ type Loader interface {
tableInfo config.TargetTableInfo,
colNames []string,
batch models.Batch,
) (int64, error)
) (int, error)
Exec(
ctx context.Context,
tableInfo config.TargetTableInfo,
columns []models.ColumnType,
chBatchesIn <-chan models.Batch,
chErrorsOut chan<- custom_errors.LoaderError,
chJobErrorsOut chan<- custom_errors.JobError,
wgActiveBatches *sync.WaitGroup,
rowsLoaded *int64,
)
}
type TableAnalyzer interface {

View File

@@ -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()
}

View 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()
}