sqlsift-lsp 0.1.0 → 0.1.3

package/CHANGELOG.md

@@ -7,151 +7,39 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
-## [0.1.0-alpha.8] - 2026-02-13
-
-### Changed
-- **Project renamed from sqlsurge to sqlsift** to avoid name conflict with existing [senkenn/sqlsurge](https://github.com/senkenn/sqlsurge) VS Code extension
-  - crate names: `sqlsift-core`, `sqlsift-cli`, `sqlsift-lsp`
-  - npm package: `sqlsift-cli`
-  - binary: `sqlsift`
-  - config file: `sqlsift.toml`
-
-### Fixed
-- Fix `auto-tag.yml` release automation: add `actions: write` permission to enable `gh workflow run`
-
-## [0.1.0-alpha.7] - 2026-02-11
+## [0.1.3](https://github.com/yukikotani231/sqlsift/compare/v0.1.2...v0.1.3) - 2026-03-04
 
 ### Added
-- **LSP server** (`sqlsift-lsp`): Language Server Protocol support for real-time SQL diagnostics in editors
-  - textDocument/didOpen, didChange, didSave, didClose
-  - Automatic schema loading from `sqlsift.toml`
-  - Schema file change detection with catalog rebuild
-  - Diagnostic severity, error codes, and help messages
-  - 12 unit tests (diagnostic conversion, state management)
-- **VS Code extension** (`editors/vscode/`): Thin LSP client for VS Code
-  - Configurable server path via `sqlsift.serverPath` setting
-  - SQL language configuration (comments, brackets)
-- **Neovim support**: Works out of the box with built-in LSP client
-
-### Fixed
-- `SELECT *` diagnostic now points to the `SELECT` keyword instead of file start
-
-## [0.1.0-alpha.6] - 2026-02-08
 
-### Added
-- **Type inference engine**: SQL expression type checking for WHERE clauses and JOIN conditions
-  - E0003 (type-mismatch): Detect incompatible type comparisons (e.g., `WHERE id = 'text'`)
-  - E0007 (join-type-mismatch): Detect JOIN condition type incompatibilities (e.g., `ON users.id = orders.name`)
-  - Binary operator type validation: comparisons (=, !=, <, >, <=, >=) and arithmetic (+, -, *, /)
-  - Nested expression type inference: `(a + b) * 2 = c`
-  - Numeric type compatibility: implicit casts between TINYINT, SMALLINT, INTEGER, BIGINT
-
-### Changed
-- Reorganized test suite: moved integration tests to `tests/analyzer_tests.rs` (74 tests)
-- Improved API documentation with doc-test examples
-- Replaced `unwrap()` with `expect()` in catalog code for better error messages
+- validate set operations and wire CLI runtime flags ([#65](https://github.com/yukikotani231/sqlsift/pull/65))
 
-## [0.1.0-alpha.5] - 2026-02-08
+### Other
 
-### Added
-- **MySQL dialect support**: Full schema parsing and query validation for MySQL
-- MySQL-specific types: `TINYINT`, `MEDIUMINT`, `UNSIGNED` integer variants, `DATETIME`, inline `ENUM`
-- `AUTO_INCREMENT` handling with implicit NOT NULL inference
-- 10 MySQL unit tests covering schema parsing, SELECT, JOIN, INSERT, UPDATE, DELETE, subquery, CTE, and error detection
-- Real-world MySQL test fixtures:
-  - **Sakila** (BSD): 16 tables, 40 valid queries, 12 error detection tests
-  - **Chinook MySQL** (MIT): 11 tables, 40 valid queries, 12 error detection tests
+- refresh outdated capability notes ([#63](https://github.com/yukikotani231/sqlsift/pull/63))
 
-## [0.1.0-alpha.4] - 2026-02-08
+## [0.1.2](https://github.com/yukikotani231/sqlsift/compare/v0.1.1...v0.1.2) - 2026-02-19
 
 ### Added
-- **Derived table (subquery in FROM) support**: Resolve aliases and validate column references for `FROM (SELECT ...) AS sub`
-- **LATERAL vs non-LATERAL scope isolation**: Non-LATERAL subqueries correctly cannot see outer FROM tables
-- **UPDATE ... FROM / DELETE ... USING**: PostgreSQL-specific multi-table update/delete syntax
-- **Recursive CTE support**: CTEs can reference themselves in recursive queries
-- **Table-valued functions in FROM**: `generate_series()`, `unnest()` etc. recognized as table sources
-- **UNION/INTERSECT/EXCEPT column inference**: Infer output columns from set operations for CTE/derived table validation
-- **Comprehensive expression resolution**: AtTimeZone, Collate, Ceil/Floor, Overlay, IsDistinctFrom, IsUnknown, SimilarTo, Tuple, Array, Subscript, Method, GroupingSets/Cube/Rollup
-- **Function FILTER/OVER clause resolution**: Validate column references in `COUNT(*) FILTER (WHERE ...)` and `OVER (PARTITION BY ... ORDER BY ...)`
-- **ORDER BY column resolution**: Validate ORDER BY references including SELECT alias support
-- **Named function argument resolution**: Handle `func(name => value)` syntax
-- 72 PostgreSQL pattern test fixtures (basic, advanced, and expression coverage)
-
-### Fixed
-- WHERE subquery scope leak: subqueries in IN/EXISTS no longer pollute outer table scope
-- VALUES derived table column aliases now correctly applied
-- Empty derived_columns (table-valued functions) no longer cause false column-not-found errors
-
-## [0.1.0-alpha.3] - 2026-02-08
 
-### Added
-- **`--dialect` flag wired up**: CLI `--dialect` option now correctly configures the SQL parser dialect (previously ignored)
-- **Real-world schema test fixtures**: Chinook, Pagila, Northwind schemas with comprehensive valid/invalid query tests covering SELECT, JOIN, INSERT, UPDATE, DELETE, subqueries, and CTEs
-- Third-party license file for test fixtures
-
-### Fixed
-- `--dialect` CLI flag was completely ignored; PostgreSQL dialect was hardcoded throughout
-- ALTER TABLE warnings for non-schema-affecting operations (e.g., `OWNER TO`) are now suppressed
+- add LSP auto-completion for table/column/view names ([#53](https://github.com/yukikotani231/sqlsift/pull/53))
+- add inline comment directives for diagnostic suppression ([#52](https://github.com/yukikotani231/sqlsift/pull/52))
+- add textDocument/hover for table, view, and column info ([#49](https://github.com/yukikotani231/sqlsift/pull/49))
 
-## [0.1.0-alpha.2] - 2026-02-08
+### Other
 
-### Added
-- **CHECK constraints**: Column-level and table-level CHECK constraint parsing and storage
-- **CREATE TYPE AS ENUM**: Enum type definitions with value storage in catalog
-- **GENERATED AS IDENTITY**: ALWAYS and BY DEFAULT identity columns with implicit NOT NULL
-- **CREATE VIEW**: View definitions with column inference from SELECT projection, wildcard expansion, and query-time resolution
-- **ALTER TABLE**: ADD COLUMN, DROP COLUMN, RENAME COLUMN, RENAME TABLE, ADD CONSTRAINT support
-- **Resilient SQL parsing**: Gracefully skip unsupported DDL statements (CREATE FUNCTION, CREATE TRIGGER, CREATE DOMAIN, etc.) instead of failing the entire schema file
-- Real-world test fixtures from Sakila and webknossos schemas
+- add GitHub templates and CI integration examples ([#54](https://github.com/yukikotani231/sqlsift/pull/54))
 
-### Fixed
-- Schema files with mixed supported/unsupported SQL statements now parse correctly
-- Comments preceding DDL statements no longer cause statement-by-statement parsing to skip valid statements
+## [0.1.1](https://github.com/yukikotani231/sqlsift/compare/sqlsift-lsp-v0.1.0...sqlsift-lsp-v0.1.1) - 2026-02-14
 
-### Changed
-- Known Limitations updated: VIEWs are now supported; ALTER TABLE is now supported
+### Other
 
-## [0.1.0-alpha.1] - 2026-02-07
-
-### Added
-- Initial release of sqlsift
-- SQL static analysis against schema definitions
-- Support for PostgreSQL dialect
-- Schema parsing from CREATE TABLE statements
-- Query validation for SELECT, INSERT, UPDATE, DELETE statements
-- Error detection:
-  - E0001: Table not found
-  - E0002: Column not found
-  - E0003: Type mismatch (reserved)
-  - E0004: Potential NULL violation (reserved)
-  - E0005: Column count mismatch in INSERT
-  - E0006: Ambiguous column reference
-  - E0007: JOIN type mismatch (reserved)
-  - E1000: Parse error
-- JOIN condition validation
-- Subquery support (including correlated subqueries)
-- CTE (Common Table Expressions) support
-- Error position reporting (line and column numbers)
-- Multiple output formats: human-readable, JSON, SARIF
-- Configuration file support (sqlsift.toml)
-- Rule disabling via CLI (--disable) or config file
-- CLI with check, schema, and parse commands
-- Typo suggestions using Levenshtein distance
-- CI/CD integration support via exit codes and SARIF output
-- Framework integration examples (Rails, Prisma)
-
-### Known Limitations
-- Only PostgreSQL dialect fully supported
-- Type checking is basic (existence only, not full type inference)
-- No support for VIEWs, functions, or stored procedures
-- Derived table (subquery in FROM) column resolution is incomplete
-
-[Unreleased]: https://github.com/yukikotani231/sqlsift/compare/v0.1.0-alpha.8...HEAD
-[0.1.0-alpha.8]: https://github.com/yukikotani231/sqlsift/compare/v0.1.0-alpha.7...v0.1.0-alpha.8
-[0.1.0-alpha.7]: https://github.com/yukikotani231/sqlsift/compare/v0.1.0-alpha.6...v0.1.0-alpha.7
-[0.1.0-alpha.6]: https://github.com/yukikotani231/sqlsift/compare/v0.1.0-alpha.5...v0.1.0-alpha.6
-[0.1.0-alpha.5]: https://github.com/yukikotani231/sqlsift/compare/v0.1.0-alpha.4...v0.1.0-alpha.5
-[0.1.0-alpha.4]: https://github.com/yukikotani231/sqlsift/compare/v0.1.0-alpha.3...v0.1.0-alpha.4
-[0.1.0-alpha.3]: https://github.com/yukikotani231/sqlsift/compare/v0.1.0-alpha.2...v0.1.0-alpha.3
-[0.1.0-alpha.2]: https://github.com/yukikotani231/sqlsift/compare/v0.1.0-alpha.1...v0.1.0-alpha.2
-[0.1.0-alpha.1]: https://github.com/yukikotani231/sqlsift/releases/tag/v0.1.0-alpha.1
+- update README and CLAUDE.md for current state
+- Rename project from sqlsurge to sqlsift
+- add comprehensive TODO documentation for type inference
+- Prepare v0.1.0-alpha.5 release
+- Update docs to reflect current PostgreSQL support level
+- Update README and CLAUDE.md to reflect current features
+- Prepare v0.1.0-alpha.1 release
+- Add npm package distribution via cargo-dist
+- Fix GitHub username in URLs
+- Add README, CLAUDE.md, and CI workflow

package/README.md

@@ -134,18 +134,39 @@ sqlsift check --schema schema/*.sql queries/**/*.sql
 - ✅ WHERE clause comparisons (`WHERE id = 'text'`)
 - ✅ Arithmetic operations (`SELECT name + 10`)
 - ✅ JOIN conditions (`ON users.id = orders.user_name`)
+- ✅ Set operations column validation (`UNION` / `INTERSECT` / `EXCEPT` column count and type compatibility)
 - ✅ INSERT value type mismatches (`INSERT INTO users (id) VALUES ('text')`)
 - ✅ UPDATE assignment type mismatches (`UPDATE users SET id = 'text'`)
+- ✅ CAST expression type inference (`CAST(name AS INTEGER)`)
+- ✅ Function return type inference (e.g., `COUNT`, `SUM`, `UPPER`, `LENGTH`, `COALESCE`)
 - ✅ Nested expressions (`WHERE (a + b) * 2 = 'text'`)
 - ✅ All comparison operators (=, !=, <, >, <=, >=)
 - ✅ Numeric type compatibility (INTEGER, BIGINT, DECIMAL, etc.)
 
 **Not Yet Detected:**
-- ⏳ CAST expression type inference
-- ⏳ Function return types (COUNT, SUM, AVG, etc.)
 - ⏳ CASE expression type consistency
 - ⏳ Subquery/CTE column type inference
 
+### Inline Suppression
+
+Suppress diagnostics on specific lines using SQL comments:
+
+```sql
+-- Suppress a specific rule on the next line
+-- sqlsift:disable E0002
+SELECT legacy_col FROM users;
+
+-- Suppress on the same line
+SELECT legacy_col FROM users; -- sqlsift:disable E0002
+
+-- Suppress multiple rules
+SELECT bad_col FROM missing_table; -- sqlsift:disable E0001, E0002
+
+-- Suppress all rules on the next line
+-- sqlsift:disable
+SELECT bad_col FROM missing_table;
+```
+
 ## CLI Reference
 
 ```
@@ -161,9 +182,9 @@ Options:
       --disable <RULE>      Disable specific rules (e.g., E0001, E0002)
   -d, --dialect <NAME>      SQL dialect [default: postgresql]
   -f, --format <FORMAT>     Output format: human, json, sarif [default: human]
-      --max-errors <N>      Maximum number of errors before stopping [default: 100]
-  -v, --verbose             Enable verbose output
-  -q, --quiet               Suppress non-error output
+      --max-errors <N>      Maximum number of errors before stopping [default: 100, 0 = unlimited]
+  -v, --verbose             Enable verbose logging (-vv for debug)
+  -q, --quiet               Suppress summary/non-error output
   -h, --help                Print help
 ```
 
@@ -193,6 +214,46 @@ sqlsift check -s schema.sql -f json queries/*.sql
 sqlsift check -s schema.sql -f sarif queries/*.sql > results.sarif
 ```
 
+## CI Integration
+
+### GitHub Actions
+
+Add sqlsift to your CI pipeline to catch SQL errors in pull requests:
+
+```yaml
+# .github/workflows/sqlsift.yml
+name: SQL Lint
+on: [push, pull_request]
+jobs:
+  sqlsift:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+      - run: npx sqlsift-cli check --schema schema.sql queries/*.sql
+```
+
+### GitHub Code Scanning (SARIF)
+
+Upload results to GitHub's Security tab:
+
+```yaml
+# .github/workflows/sqlsift.yml
+name: SQL Lint
+on: [push, pull_request]
+jobs:
+  sqlsift:
+    runs-on: ubuntu-latest
+    permissions:
+      security-events: write
+    steps:
+      - uses: actions/checkout@v4
+      - run: npx sqlsift-cli check -s schema.sql -f sarif queries/*.sql > results.sarif
+        continue-on-error: true
+      - uses: github/codeql-action/upload-sarif@v3
+        with:
+          sarif_file: results.sarif
+```
+
 ## Supported SQL Queries
 
 - SELECT, INSERT, UPDATE, DELETE with full column/table validation
@@ -235,8 +296,6 @@ Use the `--dialect` flag to specify the dialect.
 - [x] LSP server for editor integration (VS Code extension)
 
 ### Planned
-- [ ] CAST expression type inference
-- [ ] Function return type inference (COUNT, SUM, AVG, etc.)
 - [ ] CASE expression type consistency checking
 - [ ] Subquery/CTE column type inference
 - [ ] Custom rule plugins

package/npm-shrinkwrap.json

@@ -23,7 +23,7 @@
       "hasInstallScript": true,
       "license": "MIT",
       "name": "sqlsift-lsp",
-      "version": "0.1.0"
+      "version": "0.1.3"
     },
     "node_modules/@isaacs/balanced-match": {
       "engines": {
@@ -515,5 +515,5 @@
     }
   },
   "requires": true,
-  "version": "0.1.0"
+  "version": "0.1.3"
 }

package/package.json

@@ -1,5 +1,5 @@
 {
-  "artifactDownloadUrl": "https://github.com/yukikotani231/sqlsift/releases/download/v0.1.0",
+  "artifactDownloadUrl": "https://github.com/yukikotani231/sqlsift/releases/download/v0.1.3",
   "bin": {
     "sqlsift-lsp": "run-sqlsift-lsp.js"
   },
@@ -91,7 +91,7 @@
       "zipExt": ".tar.xz"
     }
   },
-  "version": "0.1.0",
+  "version": "0.1.3",
   "volta": {
     "node": "18.14.1",
     "npm": "9.5.0"