Merge branch 'main' into 297-ambiguous-reference

.github/workflows/bats-tests.yml

@@ -34,7 +34,7 @@ jobs:
 
           pip3 install "sqlglot[rs]" pyarrow pandas
 
-          curl -LJO https://github.com/duckdb/duckdb/releases/download/v1.1.3/duckdb_cli-linux-amd64.zip
+          curl -LJO https://github.com/duckdb/duckdb/releases/latest/download/duckdb_cli-linux-amd64.zip
           unzip duckdb_cli-linux-amd64.zip
           chmod +x duckdb
           sudo mv duckdb /usr/local/bin

.github/workflows/mysql-copy-tests.yml

@@ -0,0 +1,99 @@
+name: MySQL Copy Instance Test
+
+on:
+  push:
+    branches: [ "main" ]
+  pull_request:
+    branches: [ "main" ]
+
+jobs:
+  copy-instance-test:
+    runs-on: ubuntu-latest
+    services:
+      source:
+        image: mysql:lts
+        env:
+          MYSQL_ROOT_PASSWORD: root
+        ports:
+          - 13306:3306
+        options: >-
+          --health-cmd="mysqladmin ping"
+          --health-interval=10s
+          --health-timeout=5s
+          --health-retries=3
+
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Set up Go
+        uses: actions/setup-go@v5
+        with:
+          go-version: '1.23'
+
+      - name: Set up Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: '3.13'
+
+      - name: Install dependencies
+        run: |
+          go get .
+
+          pip3 install "sqlglot[rs]"
+
+          curl -LJO https://dev.mysql.com/get/Downloads/MySQL-Shell/mysql-shell_9.1.0-1debian12_amd64.deb
+          sudo apt-get install -y ./mysql-shell_9.1.0-1debian12_amd64.deb
+
+      - name: Setup test data in source MySQL
+        run: |
+          mysqlsh -hlocalhost -P13306 -uroot -proot --sql -e "
+            CREATE DATABASE testdb;
+            USE testdb;
+            CREATE TABLE users (
+              id INT AUTO_INCREMENT PRIMARY KEY,
+              name VARCHAR(100)
+            );
+            INSERT INTO users (name) VALUES ('test1'), ('test2'), ('test3');
+            -- Make a gap in the id sequence
+            INSERT INTO users VALUES (100, 'test100');
+            INSERT INTO users (name) VALUES ('test101');
+
+            -- A table with non-default starting auto_increment value
+            CREATE TABLE items (
+              id INT AUTO_INCREMENT PRIMARY KEY,
+              v BIGINT check (v > 0),
+              name VARCHAR(100)
+            ) AUTO_INCREMENT=1000;
+
+            INSERT INTO items (v, name) VALUES (1, 'item1'), (2, 'item2'), (3, 'item3');
+          "
+
+      - name: Build and start MyDuck Server
+        run: |
+          go build -v
+          ./myduckserver &
+          sleep 5
+
+      - name: Run copy-instance test
+        run: |
+          # Set local_infile to true to allow loading data from files
+          mysqlsh -uroot --no-password --sql -e "SET GLOBAL local_infile = 1;"
+
+          # Copy the data from source MySQL to MyDuck
+          mysqlsh -hlocalhost -P13306 -uroot -proot \
+            -- util copy-instance "mysql://root:@127.0.0.1:3306" \
+            --users false --ignore-version true
+
+          # Verify the data was copied
+          for table in users items; do
+            mysqlsh -hlocalhost -P13306 -uroot -proot --sql -e "
+              SELECT * FROM testdb.$table ORDER BY id;
+            " | tee source_data_$table.tsv
+            mysqlsh -uroot --no-password --sql -e "
+              SELECT * FROM testdb.$table ORDER BY id;
+            " | tee copied_data_$table.tsv
+
+            diff source_data_$table.tsv copied_data_$table.tsv
+          done
+
+          

backend/executor.go

@@ -80,13 +80,12 @@ func (b *DuckBuilder) Build(ctx *sql.Context, root sql.Node, r sql.Row) (sql.Row
 	case *plan.InsertInto:
 		insert := n.(*plan.InsertInto)
 
-		// For AUTO_INCREMENT column, we fallback to the framework if the column is specified.
-		if dst, err := plan.GetInsertable(insert.Destination); err == nil && dst.Schema().HasAutoIncrement() {
-			if len(insert.ColumnNames) == 0 || len(insert.ColumnNames) == len(dst.Schema()) {
-				return b.base.Build(ctx, root, r)
-			}
-		}
-
+		// The handling of auto_increment reset and check constraints is not supported by DuckDB.
+		// We need to fallback to the framework for these cases.
+		// But we want to rewrite LOAD DATA to be handled by DuckDB,
+		// as it is a common way to import data into the database.
+		// Therefore, we ignoring auto_increment and check constraints for LOAD DATA.
+		// So rewriting LOAD DATA is done eagerly here.
 		src := insert.Source
 		if proj, ok := src.(*plan.Project); ok {
 			src = proj.Child
@@ -97,6 +96,20 @@ func (b *DuckBuilder) Build(ctx *sql.Context, root sql.Node, r sql.Row) (sql.Row
 			}
 			return b.base.Build(ctx, root, r)
 		}
+
+		if dst, err := plan.GetInsertable(insert.Destination); err == nil {
+			// For AUTO_INCREMENT column, we fallback to the framework if the column is specified.
+			// if dst.Schema().HasAutoIncrement() && (0 == len(insert.ColumnNames) || len(insert.ColumnNames) == len(dst.Schema())) {
+			if dst.Schema().HasAutoIncrement() {
+				return b.base.Build(ctx, root, r)
+			}
+			// For table with check constraints, we fallback to the framework.
+			if ct, ok := dst.(sql.CheckTable); ok {
+				if checks, err := ct.GetChecks(ctx); err == nil && len(checks) > 0 {
+					return b.base.Build(ctx, root, r)
+				}
+			}
+		}
 	}
 
 	// Fallback to the base builder if the plan contains system/user variables or is not a pure data query.

catalog/database.go

@@ -73,18 +73,29 @@ func (d *Database) GetTableInsensitive(ctx *sql.Context, tblName string) (sql.Ta
 func (d *Database) tablesInsensitive(ctx *sql.Context, pattern string) ([]*Table, error) {
 	tables, err := d.findTables(ctx, pattern)
 	if err != nil {
+		ctx.GetLogger().WithFields(logrus.Fields{
+			"catalog":  d.catalog,
+			"database": d.name,
+			"pattern":  pattern,
+		}).WithError(err).Error("Failed to find tables")
 		return nil, err
 	}
 	for _, t := range tables {
 		if err := t.withSchema(ctx); err != nil {
+			ctx.GetLogger().WithFields(logrus.Fields{
+				"catalog":  d.catalog,
+				"database": d.name,
+				"pattern":  pattern,
+				"table":    t.Name(),
+			}).WithError(err).Error("Failed to get table schema")
 			return nil, err
 		}
 	}
 	return tables, nil
 }
 
 func (d *Database) findTables(ctx *sql.Context, pattern string) ([]*Table, error) {
-	rows, err := adapter.QueryCatalog(ctx, "SELECT DISTINCT table_name, comment FROM duckdb_tables()  where (database_name = ? and schema_name = ? and table_name ILIKE ?) or (database_name = 'temp' and schema_name = 'main' and table_name ILIKE ?)", d.catalog, d.name, pattern, pattern)
+	rows, err := adapter.QueryCatalog(ctx, "SELECT DISTINCT table_name, comment FROM duckdb_tables() WHERE (database_name = ? AND schema_name = ? AND table_name ILIKE ?) OR (temporary IS TRUE AND table_name ILIKE ?)", d.catalog, d.name, pattern, pattern)
 	if err != nil {
 		return nil, ErrDuckDB.New(err)
 	}
@@ -113,7 +124,6 @@ func (d *Database) Name() string {
 }
 
 func (d *Database) createAllTable(ctx *sql.Context, name string, schema sql.PrimaryKeySchema, collation sql.CollationID, comment string, temporary bool) error {
-
 	var columns []string
 	var columnCommentSQLs []string
 	var fullTableName string
@@ -219,7 +229,7 @@ func (d *Database) createAllTable(ctx *sql.Context, name string, schema sql.Prim
 	b.WriteString(")")
 
 	// Add comment to the table
-	info := ExtraTableInfo{schema.PkOrdinals, withoutIndex, fullSequenceName}
+	info := ExtraTableInfo{schema.PkOrdinals, withoutIndex, fullSequenceName, nil}
 	b.WriteString(fmt.Sprintf(
 		"; COMMENT ON TABLE %s IS '%s'",
 		fullTableName,