- Allow parsing tuple values with or without name prefixes
- Supports both [x: 10; y: 20] and [10; 20] formats
- Useful for CSV data where names are schema-only metadata
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
- Add NamedSchema interface with optional name property
- Update TupleSchema.elements to use NamedSchema[]
- Add parseNamedSchema() to support [x: number; y: number] syntax
- Update validator to parse named members in tuple values
- Fix schemaToTypeString in csv-loader for NamedSchema
- Fix createValidator to handle NamedSchema.schema
- Ensure single named element stays as tuple (not array)
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
- Add writeToDisk option to write .d.ts files directly to disk
- Useful for rsbuild/rspack dev server which uses in-memory FS
- Set writeToDisk: true in dev mode to see generated types
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
- Fix leading slash in relative path
- Use .d.ts extension instead of .csv.d.ts
- Support typesOutputDir option properly
- Use ./ prefix for module declaration path
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
- Include dist/ in git for npm package distribution
- Add .npmignore to exclude source files from npm package
- This ensures npm install from git works without requiring build
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
hypercross
hyper