Add MSSQL engine for parse and analyze via teesql (#4571)

* Add MSSQL engine for parse and analyze via teesql

Add a new mssql engine backed by github.com/sqlc-dev/teesql, a T-SQL
parser producing a SqlScriptDOM-compatible AST. The engine converts
teesql's AST into sqlc's internal AST and analyzes queries through the
analysis core, seeded with SQL Server's type system.

- internal/engine/mssql: parser, AST converter, reserved keywords and
  dialect seed (types, functions, operator rules)
- Wire the engine into the compiler's core-analysis path and the parse
  and analyze commands as the mssql (alias sqlserver) dialect
- Named @parameters share a number across repeated uses; OUTPUT clauses
  map to returning lists with INSERTED/DELETED qualifiers stripped; the
  dbo schema maps to the catalog's default namespace
- End-to-end coverage under parse_basic, analyze_basic and
  analyze_params

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01FrnQ4EoZ3WCWWrSsm8bhCP

* mssql: fix DML alias targets, byte offsets, dead keyword

Address review findings on the initial engine:

- DELETE dropped its FROM clause, so the T-SQL join form
  (DELETE b FROM books b JOIN ...) failed to resolve. UPDATE and DELETE
  now resolve an alias target against the FROM clause: the matching
  relation becomes the statement's target, and the ON conditions of any
  inner join dissolved by pulling it out move to the WHERE clause.
  Relations under outer joins are left alone.
- AST Location fields carried teesql's UTF-16 code-unit offsets while
  StmtLocation/StmtLen are byte offsets; the converter now maps
  locations through the same byte-offset table as statement spans.
- Drop the unreachable "within group" reserved-keyword case: keywords
  are checked one identifier token at a time.
- Cover INSERT ... OUTPUT and the UPDATE/DELETE alias forms end to end
  under analyze_dml/mssql.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01FrnQ4EoZ3WCWWrSsm8bhCP

---------

Co-authored-by: Claude <noreply@anthropic.com>

Author: kyleconroy

docs/howto/analyze.md

@@ -26,7 +26,7 @@ provided. The schema is always read from the `--schema` file.
 ## Flags
 
 - `--dialect`, `-d` - The SQL dialect to use. One of `postgresql`, `mysql`,
-  `sqlite`, `clickhouse`, or `googlesql`. Required.
+  `sqlite`, `clickhouse`, `googlesql`, or `mssql`. Required.
 - `--schema`, `-s` - Path to the schema (DDL) file. Required.
 - `--ast` - Include each statement's AST in the output. Defaults to `false`.
 

docs/howto/parse.md

@@ -20,7 +20,7 @@ provided.
 ## Flags
 
 - `--dialect`, `-d` - The SQL dialect to use. One of `postgresql`, `mysql`,
-  `sqlite`, `clickhouse`, or `googlesql`. Required.
+  `sqlite`, `clickhouse`, `googlesql`, or `mssql`. Required.
 
 ## Examples
 

go.mod

@@ -22,6 +22,7 @@ require (
 	github.com/sqlc-dev/marino v0.1.0
 	github.com/sqlc-dev/meyer v0.1.1
 	github.com/sqlc-dev/oliphant v0.1.0
+	github.com/sqlc-dev/teesql v1.1.0
 	github.com/sqlc-dev/zetajones v0.1.0
 	github.com/tetratelabs/wazero v1.12.0
 	github.com/xeipuuv/gojsonschema v1.2.0

go.sum

@@ -73,6 +73,8 @@ github.com/sqlc-dev/meyer v0.1.1 h1:BAeZcfgLyTnk9f90DyGEKXPrHxtgvVD/DTM6awq2kUY=
 github.com/sqlc-dev/meyer v0.1.1/go.mod h1:pS4USCRf/SLjWtaMcnTo4YrEEFKBj8CyyqlxcVUJQH8=
 github.com/sqlc-dev/oliphant v0.1.0 h1:RAsO6BMitIzB2+swx/qzUR5nf6w4cQ1abgHIu+Fgppo=
 github.com/sqlc-dev/oliphant v0.1.0/go.mod h1:fRM/t4FutRddTIq2YCuS4O9o+2rRwSwELRvLMqtPloo=
+github.com/sqlc-dev/teesql v1.1.0 h1:3sVYQ9FGxQVcqrqQOQ27bk0aF4c4yN1H1zLL79uaSxQ=
+github.com/sqlc-dev/teesql v1.1.0/go.mod h1:WwOp9UtnxG17+eNFT5KXu/AGBQ8ucdGc8wIOpnj4XCI=
 github.com/sqlc-dev/zetajones v0.1.0 h1:VeG0atx6lNABr9V2bSI5vL9DvOKTHX0XjMqWUE/rv40=
 github.com/sqlc-dev/zetajones v0.1.0/go.mod h1:dU1DxwqC6Cahbpnw16KpH1J2waWRDMdwyDSvovMZR4I=
 github.com/stretchr/objx v0.1.0/go.mod h1:HFkY916IF+rwdDfMAkV7OtwuqBVzrE8GR6GFx+wExME=

internal/cmd/analyze.go

@@ -43,6 +43,9 @@ Examples:
   # Analyze a GoogleSQL (BigQuery, Spanner) query
   sqlc analyze --dialect googlesql --schema schema.sql query.sql
 
+  # Analyze a SQL Server (T-SQL) query
+  sqlc analyze --dialect mssql --schema schema.sql query.sql
+
   # Analyze a query piped via stdin
   echo "-- name: GetAuthor :one
   SELECT * FROM authors WHERE id = $1;" | sqlc analyze --dialect postgresql --schema schema.sql
@@ -56,7 +59,7 @@ Examples:
 				return err
 			}
 			if dialect == "" {
-				return fmt.Errorf("--dialect flag is required (postgresql, mysql, sqlite, clickhouse, or googlesql)")
+				return fmt.Errorf("--dialect flag is required (postgresql, mysql, sqlite, clickhouse, googlesql, or mssql)")
 			}
 
 			schemaPath, err := cmd.Flags().GetString("schema")
@@ -117,8 +120,10 @@ Examples:
 				engine = config.EngineClickHouse
 			case "googlesql":
 				engine = config.EngineGoogleSQL
+			case "mssql", "sqlserver":
+				engine = config.EngineMSSQL
 			default:
-				return fmt.Errorf("unsupported dialect: %s (use postgresql, mysql, sqlite, clickhouse, or googlesql)", dialect)
+				return fmt.Errorf("unsupported dialect: %s (use postgresql, mysql, sqlite, clickhouse, googlesql, or mssql)", dialect)
 			}
 
 			sql := config.SQL{
@@ -160,7 +165,7 @@ Examples:
 			return nil
 		},
 	}
-	cmd.Flags().StringP("dialect", "d", "", "SQL dialect to use (postgresql, mysql, sqlite, clickhouse, or googlesql)")
+	cmd.Flags().StringP("dialect", "d", "", "SQL dialect to use (postgresql, mysql, sqlite, clickhouse, googlesql, or mssql)")
 	cmd.Flags().StringP("schema", "s", "", "path to the schema file")
 	cmd.Flags().BoolP("ast", "", false, "include the statement AST in the output")
 	return cmd

internal/cmd/parse.go

@@ -12,6 +12,7 @@ import (
 	"github.com/sqlc-dev/sqlc/internal/engine/clickhouse"
 	"github.com/sqlc-dev/sqlc/internal/engine/dolphin"
 	"github.com/sqlc-dev/sqlc/internal/engine/googlesql"
+	"github.com/sqlc-dev/sqlc/internal/engine/mssql"
 	"github.com/sqlc-dev/sqlc/internal/engine/postgresql"
 	"github.com/sqlc-dev/sqlc/internal/engine/sqlite"
 	"github.com/sqlc-dev/sqlc/internal/metadata"
@@ -59,15 +60,18 @@ Examples:
   sqlc parse --dialect clickhouse queries.sql
 
   # Parse GoogleSQL (BigQuery, Spanner)
-  sqlc parse --dialect googlesql queries.sql`,
+  sqlc parse --dialect googlesql queries.sql
+
+  # Parse SQL Server (T-SQL) SQL
+  sqlc parse --dialect mssql queries.sql`,
 		Args: cobra.MaximumNArgs(1),
 		RunE: func(cmd *cobra.Command, args []string) error {
 			dialect, err := cmd.Flags().GetString("dialect")
 			if err != nil {
 				return err
 			}
 			if dialect == "" {
-				return fmt.Errorf("--dialect flag is required (postgresql, mysql, sqlite, clickhouse, or googlesql)")
+				return fmt.Errorf("--dialect flag is required (postgresql, mysql, sqlite, clickhouse, googlesql, or mssql)")
 			}
 
 			// Determine input source
@@ -104,8 +108,10 @@ Examples:
 				parser = clickhouse.NewParser()
 			case "googlesql":
 				parser = googlesql.NewParser()
+			case "mssql", "sqlserver":
+				parser = mssql.NewParser()
 			default:
-				return fmt.Errorf("unsupported dialect: %s (use postgresql, mysql, sqlite, clickhouse, or googlesql)", dialect)
+				return fmt.Errorf("unsupported dialect: %s (use postgresql, mysql, sqlite, clickhouse, googlesql, or mssql)", dialect)
 			}
 
 			// Read the full source so each statement's name and command can be
@@ -149,6 +155,6 @@ Examples:
 			return nil
 		},
 	}
-	cmd.Flags().StringP("dialect", "d", "", "SQL dialect to use (postgresql, mysql, sqlite, clickhouse, or googlesql)")
+	cmd.Flags().StringP("dialect", "d", "", "SQL dialect to use (postgresql, mysql, sqlite, clickhouse, googlesql, or mssql)")
 	return cmd
 }

internal/compiler/engine.go

@@ -11,6 +11,7 @@ import (
 	"github.com/sqlc-dev/sqlc/internal/engine/clickhouse"
 	"github.com/sqlc-dev/sqlc/internal/engine/dolphin"
 	"github.com/sqlc-dev/sqlc/internal/engine/googlesql"
+	"github.com/sqlc-dev/sqlc/internal/engine/mssql"
 	"github.com/sqlc-dev/sqlc/internal/engine/postgresql"
 	pganalyze "github.com/sqlc-dev/sqlc/internal/engine/postgresql/analyzer"
 	"github.com/sqlc-dev/sqlc/internal/engine/sqlite"
@@ -57,9 +58,10 @@ func NewCompiler(conf config.SQL, combo config.CombinedSettings, parserOpts opts
 		o(c)
 	}
 
-	// ClickHouse and GoogleSQL have no legacy analysis path to fall back to.
+	// ClickHouse, GoogleSQL and SQL Server have no legacy analysis path to
+	// fall back to.
 	switch conf.Engine {
-	case config.EngineClickHouse, config.EngineGoogleSQL:
+	case config.EngineClickHouse, config.EngineGoogleSQL, config.EngineMSSQL:
 		c.coreAnalysis = true
 	}
 	if c.coreAnalysis {
@@ -138,6 +140,10 @@ func (c *Compiler) initCore() error {
 		c.parser = googlesql.NewParser()
 		c.selector = newDefaultSelector()
 		dialect = googlesql.Dialect()
+	case config.EngineMSSQL:
+		c.parser = mssql.NewParser()
+		c.selector = newDefaultSelector()
+		dialect = mssql.Dialect()
 	default:
 		return fmt.Errorf("unknown engine: %s", c.conf.Engine)
 	}

internal/config/config.go

@@ -56,6 +56,7 @@ const (
 	EngineSQLite     Engine = "sqlite"
 	EngineClickHouse Engine = "clickhouse"
 	EngineGoogleSQL  Engine = "googlesql"
+	EngineMSSQL      Engine = "mssql"
 )
 
 type Config struct {

internal/endtoend/testdata/analyze_basic/mssql/exec.json

@@ -0,0 +1,5 @@
+{
+  "command": "analyze",
+  "args": ["--dialect", "mssql", "--schema", "schema.sql", "query.sql"],
+  "contexts": ["base"]
+}

internal/endtoend/testdata/analyze_basic/mssql/query.sql

@@ -0,0 +1,2 @@
+-- name: ListAuthors :many
+SELECT id, name, bio, royalties, created FROM authors;