feat(sqlserver): add Microsoft SQL Server support #4243
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Add MSSQL/SQL Server engine support using the sqlc-dev/teesql parser. This implementation uses database-only mode (analyzer.database: only) which analyzes queries against a live SQL Server database using sp_describe_first_result_set for column metadata. Features: - T-SQL parser integration using teesql - Database analyzer using go-mssqldb driver - MSSQL catalog with dbo default schema - Example configuration in examples/authors/sqlserver The SQL Server engine requires a database connection and does not support static catalog analysis. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <[email protected]>
Change the engine identifier from "sqlserver" to "mssql" for consistency with the package naming and industry conventions. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <[email protected]>
dosubot
bot
added
size:XXL
Run go mod tidy to properly categorize direct vs indirect dependencies. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <[email protected]>
Move the sqlserver configuration to its own sqlc.yaml file to prevent test failures in CI. The MSSQL engine requires a database connection which isn't available in the standard CI environment. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <[email protected]>
Add native and Docker support for starting MSSQL Server instances in end-to-end tests, following the same pattern as PostgreSQL and MySQL. - Add internal/sqltest/native/mssql.go for native MSSQL service - Add internal/sqltest/docker/mssql.go for Docker-based MSSQL - Update endtoend_test.go to initialize and use MSSQL connections 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <[email protected]>
Consolidate the sqlserver example configuration back into the main examples/authors/sqlc.yaml file, using an environment variable for the database URI like the other examples. 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <[email protected]>
Add MSSQL variant of the column_as test to verify basic SELECT with column aliases works correctly with the MSSQL engine. 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <[email protected]>
Update MSSQL Docker startup code to create a new connection on each ping attempt, matching the pattern used by PostgreSQL. This ensures the connection is properly established even if MSSQL takes time to become ready. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <[email protected]>
kyleconroy
force-pushed
the
claude/add-mssql-support-MrXFp
branch
from
b2c8b18 to
1764695
Compare
Add PostgreSQL, MySQL, and MSSQL as GitHub Actions services so they're ready before tests run. This avoids the timeout issues from pulling the large MSSQL image (~1.5GB) during test execution. Set environment variables so tests use the pre-started service databases instead of trying to start their own containers. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <[email protected]>
Remove GitHub Actions services and let the tests start databases via the sqltest/docker package, matching the existing pattern for PostgreSQL and MySQL. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <[email protected]>
This comment was marked as spam.
This comment was marked as spam.
diffray-bot
reviewed
Dec 30, 2025
| } | ||
| } | ||
|
|
||
| uri := "sqlserver://sa:MySecretPassword1!@localhost:1433?database=master" |
This comment was marked as spam.
This comment was marked as spam.
Sorry, something went wrong.
| case *ast.ColumnReferenceExpression: | ||
| return convertColumnReference(e) | ||
| case *ast.IntegerLiteral: | ||
| val, _ := strconv.ParseInt(e.Value, 10, 64) |
This comment was marked as spam.
This comment was marked as spam.
Sorry, something went wrong.
Comment on lines
+140
to
+141
| col.Name = name.String | ||
| col.IsNullable = isNullable |
This comment was marked as spam.
This comment was marked as spam.
Sorry, something went wrong.
Comment on lines
+611
to
+619
| switch src := spec.InsertSource.(type) { | ||
| case *ast.ValuesInsertSource: | ||
| // Handle VALUES clauses | ||
| stmt.SelectStmt = &sqast.TODO{} | ||
| case *ast.SelectInsertSource: | ||
| if src.Select != nil { | ||
| stmt.SelectStmt = convertQueryExpression(src.Select) | ||
| } | ||
| } |
This comment was marked as spam.
This comment was marked as spam.
Sorry, something went wrong.
Comment on lines
+200
to
+209
| func countParameters(query string) int { | ||
| count := 0 | ||
| for i := 1; i <= 100; i++ { | ||
| param := fmt.Sprintf("@p%d", i) | ||
| if strings.Contains(query, param) { | ||
| count = i | ||
| } | ||
| } | ||
| return count | ||
| } |
This comment was marked as spam.
This comment was marked as spam.
Sorry, something went wrong.
Comment on lines
+422
to
+428
| scanArgs := make([]interface{}, 39) | ||
| scanArgs[0] = &isHidden | ||
| scanArgs[1] = &colOrdinal | ||
| scanArgs[2] = &name | ||
| for i := 3; i < 39; i++ { | ||
| scanArgs[i] = &dummy | ||
| } |
This comment was marked as spam.
This comment was marked as spam.
Sorry, something went wrong.
Comment on lines
+58
to
+190
| func (a *Analyzer) analyzeQuery(ctx context.Context, n ast.Node, query string, ps *named.ParamSet) (*core.Analysis, error) { | ||
| var result core.Analysis | ||
|
|
||
| // Use sp_describe_first_result_set to get column metadata | ||
| // This is MSSQL's equivalent of PostgreSQL's PREPARE for getting result set metadata | ||
| rows, err := a.conn.QueryContext(ctx, "EXEC sp_describe_first_result_set @tsql = @p1", query) | ||
| if err != nil { | ||
| return nil, a.extractSqlErr(n, err) | ||
| } | ||
| defer rows.Close() | ||
|
|
||
| for rows.Next() { | ||
| var col columnInfo | ||
| // sp_describe_first_result_set returns many columns, we only need a few | ||
| // Columns: is_hidden, column_ordinal, name, is_nullable, system_type_id, system_type_name, | ||
| // max_length, precision, scale, collation_name, user_type_id, user_type_database, | ||
| // user_type_schema, user_type_name, assembly_qualified_type_name, xml_collection_id, | ||
| // xml_collection_database, xml_collection_schema, xml_collection_name, is_xml_document, | ||
| // is_case_sensitive, is_fixed_length_clr_type, source_server, source_database, | ||
| // source_schema, source_table, source_column, is_identity_column, is_part_of_unique_key, | ||
| // is_updateable, is_computed_column, is_sparse_column_set, ordinal_in_order_by_list, | ||
| // order_by_is_descending, order_by_list_length, tds_type_id, tds_length, | ||
| // tds_collation_id, tds_collation_sort_id | ||
|
|
||
| var isHidden bool | ||
| var colOrdinal int | ||
| var name sql.NullString | ||
| var isNullable bool | ||
| var sysTypeId int | ||
| var sysTypeName sql.NullString | ||
| var maxLength int | ||
| var precision int | ||
| var scale int | ||
| var collationName sql.NullString | ||
| var userTypeId sql.NullInt64 | ||
| var userTypeDb sql.NullString | ||
| var userTypeSchema sql.NullString | ||
| var userTypeName sql.NullString | ||
| var assemblyQualTypeName sql.NullString | ||
| var xmlColId sql.NullInt64 | ||
| var xmlColDb sql.NullString | ||
| var xmlColSchema sql.NullString | ||
| var xmlColName sql.NullString | ||
| var isXmlDoc bool | ||
| var isCaseSensitive bool | ||
| var isFixedLenClr bool | ||
| var sourceServer sql.NullString | ||
| var sourceDb sql.NullString | ||
| var sourceSchema sql.NullString | ||
| var sourceTable sql.NullString | ||
| var sourceColumn sql.NullString | ||
| var isIdentity bool | ||
| var isPartOfUniqueKey sql.NullBool | ||
| var isUpdateable bool | ||
| var isComputed bool | ||
| var isSparseColSet bool | ||
| var ordinalInOrderBy sql.NullInt64 | ||
| var orderByDesc sql.NullBool | ||
| var orderByLen sql.NullInt64 | ||
| var tdsTypeId sql.NullInt64 | ||
| var tdsLength sql.NullInt64 | ||
| var tdsCollationId sql.NullInt64 | ||
| var tdsCollationSortId sql.NullInt64 | ||
|
|
||
| err := rows.Scan( | ||
| &isHidden, &colOrdinal, &name, &isNullable, &sysTypeId, &sysTypeName, | ||
| &maxLength, &precision, &scale, &collationName, &userTypeId, &userTypeDb, | ||
| &userTypeSchema, &userTypeName, &assemblyQualTypeName, &xmlColId, | ||
| &xmlColDb, &xmlColSchema, &xmlColName, &isXmlDoc, &isCaseSensitive, | ||
| &isFixedLenClr, &sourceServer, &sourceDb, &sourceSchema, &sourceTable, | ||
| &sourceColumn, &isIdentity, &isPartOfUniqueKey, &isUpdateable, | ||
| &isComputed, &isSparseColSet, &ordinalInOrderBy, &orderByDesc, | ||
| &orderByLen, &tdsTypeId, &tdsLength, &tdsCollationId, &tdsCollationSortId, | ||
| ) | ||
| if err != nil { | ||
| return nil, fmt.Errorf("scanning column info: %w", err) | ||
| } | ||
|
|
||
| if isHidden { | ||
| continue | ||
| } | ||
|
|
||
| col.Name = name.String | ||
| col.IsNullable = isNullable | ||
| col.DataType = normalizeTypeName(sysTypeName.String, maxLength, precision, scale) | ||
| col.Table = sourceTable.String | ||
| col.Schema = sourceSchema.String | ||
|
|
||
| coreCol := &core.Column{ | ||
| Name: col.Name, | ||
| OriginalName: col.Name, | ||
| DataType: col.DataType, | ||
| NotNull: !col.IsNullable, | ||
| } | ||
|
|
||
| if col.Table != "" { | ||
| coreCol.Table = &core.Identifier{ | ||
| Schema: col.Schema, | ||
| Name: col.Table, | ||
| } | ||
| } | ||
|
|
||
| result.Columns = append(result.Columns, coreCol) | ||
| } | ||
|
|
||
| if err := rows.Err(); err != nil { | ||
| return nil, fmt.Errorf("iterating column info: %w", err) | ||
| } | ||
|
|
||
| // Get parameter information | ||
| // MSSQL doesn't have a built-in way to get parameter metadata from a query string | ||
| // We'll count the @pN placeholders in the query and create parameters | ||
| paramCount := countParameters(query) | ||
| for i := 1; i <= paramCount; i++ { | ||
| paramName := "" | ||
| if ps != nil { | ||
| if n, ok := ps.NameFor(i); ok { | ||
| paramName = n | ||
| } | ||
| } | ||
|
|
||
| result.Params = append(result.Params, &core.Parameter{ | ||
| Number: int32(i), | ||
| Column: &core.Column{ | ||
| Name: paramName, | ||
| DataType: "any", // MSSQL doesn't provide parameter type info | ||
| NotNull: false, | ||
| }, | ||
| }) | ||
| } | ||
|
|
||
| return &result, nil | ||
| } |
This comment was marked as spam.
This comment was marked as spam.
Sorry, something went wrong.
Comment on lines
+122
to
+131
| err := rows.Scan( | ||
| &isHidden, &colOrdinal, &name, &isNullable, &sysTypeId, &sysTypeName, | ||
| &maxLength, &precision, &scale, &collationName, &userTypeId, &userTypeDb, | ||
| &userTypeSchema, &userTypeName, &assemblyQualTypeName, &xmlColId, | ||
| &xmlColDb, &xmlColSchema, &xmlColName, &isXmlDoc, &isCaseSensitive, | ||
| &isFixedLenClr, &sourceServer, &sourceDb, &sourceSchema, &sourceTable, | ||
| &sourceColumn, &isIdentity, &isPartOfUniqueKey, &isUpdateable, | ||
| &isComputed, &isSparseColSet, &ordinalInOrderBy, &orderByDesc, | ||
| &orderByLen, &tdsTypeId, &tdsLength, &tdsCollationId, &tdsCollationSortId, | ||
| ) |
This comment was marked as spam.
This comment was marked as spam.
Sorry, something went wrong.
Comment on lines
+350
to
+356
| if _, err := a.conn.ExecContext(ctx, batch); err != nil { | ||
| // Check if it's a "already exists" error and skip it | ||
| if !isObjectExistsError(err) { | ||
| a.conn.Close() | ||
| a.conn = nil | ||
| return fmt.Errorf("migration failed: %s: %w", batch, err) | ||
| } |
This comment was marked as spam.
This comment was marked as spam.
Sorry, something went wrong.
This comment was marked as spam.
This comment was marked as spam.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Add MSSQL/SQL Server engine support using the sqlc-dev/teesql parser.
This implementation uses database-only mode (analyzer.database: only)
which analyzes queries against a live SQL Server database using
sp_describe_first_result_set for column metadata.
Features:
The SQL Server engine requires a database connection and does not
support static catalog analysis.
🤖 Generated with Claude Code
Co-Authored-By: Claude [email protected]