feat: integrate azure storage handling in migration pipeline

Stream transformer output to an Azure blob container when a job declares ToStorage, with per-job prefix and credential configuration.

internal/app/etl/extractors/postgres.go

@@ -2,10 +2,8 @@ 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"
@@ -23,7 +21,15 @@ func NewPostgresExtractor(db dbwrapper.DbWrapper) etl.Extractor {
 	return &PostgresExtractor{db: db}
 }
 
-func buildExtractQueryPostgres(sourceDbInfo config.SourceTableInfo, columns []models.ColumnType) string {
+func buildExtractQueryPostgres(
+	sourceDbInfo config.SourceTableInfo,
+	columns []models.ColumnType,
+	includeRange bool,
+	isMinInclusive bool,
+	isMaxInclusive bool,
+	hasMin bool,
+	hasMax bool,
+) string {
 	var sbColumns strings.Builder
 
 	if len(columns) == 0 {
@@ -48,10 +54,44 @@ func buildExtractQueryPostgres(sourceDbInfo config.SourceTableInfo, columns []mo
 		}
 	}
 
-	return fmt.Sprintf(`SELECT %s FROM "%s"."%s"  ORDER BY "%s" ASC`, sbColumns.String(), sourceDbInfo.Schema, sourceDbInfo.Table, sourceDbInfo.PrimaryKey)
+	query := fmt.Sprintf(`SELECT %s FROM "%s"."%s"`, sbColumns.String(), sourceDbInfo.Schema, sourceDbInfo.Table)
+
+	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) ProcessPartition(
+func (postgresEx *PostgresExtractor) Exec(
 	ctx context.Context,
 	tableInfo config.SourceTableInfo,
 	columns []models.ColumnType,
@@ -60,14 +100,20 @@ func (postgresEx *PostgresExtractor) ProcessPartition(
 	indexPrimaryKey int,
 	chBatchesOut chan<- models.Batch,
 ) (int, error) {
-	query := buildExtractQueryPostgres(tableInfo, columns)
+	hasMin := partition.HasRange && partition.Range.Min > 0
+	hasMax := partition.HasRange && partition.Range.Max > 0
+	query := buildExtractQueryPostgres(tableInfo, columns, partition.HasRange, partition.Range.IsMinInclusive, partition.Range.IsMaxInclusive, hasMin, hasMax)
 
-	if partition.HasRange {
-		return 0, errors.New("Batch config not yet supported")
+	var queryArgs []any
+	if hasMin {
+		queryArgs = append(queryArgs, partition.Range.Min)
+	}
+	if hasMax {
+		queryArgs = append(queryArgs, partition.Range.Max)
 	}
 
 	rowsRead := 0
-	rows, err := postgresEx.db.Query(ctx, query)
+	rows, err := postgresEx.db.Query(ctx, query, queryArgs...)
 	if err != nil {
 		return rowsRead, &custom_errors.ExtractorError{Partition: partition, HasLastId: false, Msg: err.Error()}
 	}
@@ -78,7 +124,7 @@ func (postgresEx *PostgresExtractor) ProcessPartition(
 	for rows.Next() {
 		values, err := rows.Values()
 		if err != nil {
-			return rowsRead, errors.New("Unexpected error reading rows from source")
+			return rowsRead, &custom_errors.ExtractorError{Partition: partition, HasLastId: false, Msg: err.Error()}
 		}
 		rowsRead++
 
@@ -96,7 +142,7 @@ func (postgresEx *PostgresExtractor) ProcessPartition(
 	}
 
 	if err := rows.Err(); err != nil {
-		return rowsRead, errors.New("Unexpected error reading rows from source")
+		return rowsRead, &custom_errors.ExtractorError{Partition: partition, HasLastId: false, Msg: err.Error()}
 	}
 
 	if len(batchRows) > 0 {
@@ -109,17 +155,3 @@ func (postgresEx *PostgresExtractor) ProcessPartition(
 
 	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,
-) {
-}