- Fix dupl lint: add nolint:dupl to parseInlineTable and parseValArray
which have intentionally similar loop structures
- Fix gci lint: correct alignment in multiline basic string test entries
- Add unmarshaler tests for new TOML v1.1.0 features exercising public
APIs: multiline inline tables with comments, trailing commas, leading
commas, error cases (comma at start, missing separator, double comma,
incomplete table), escape sequences, and type mismatch errors
- Add parser test for inline table comment handling with KeepComments
- Coverage increases from 97.37% to 97.44% vs v2 base
https://claude.ai/code/session_01RdiWykFQdmwkQ2nbLwJwwP
TOML v1.1.0 introduces the \xHH escape notation for basic strings,
allowing two-digit hex escapes for Unicode code points U+0000 to
U+00FF.
We keep emitting \u00XX for backwards compatibility.
Removes all unsafe operations from go-toml, making the codebase
fully safe Go code. The internal/danger package that contained
unsafe operations has been deleted.
Changes:
- Replace pointer-based node navigation with index-based navigation
- Node.next and Node.child now store absolute indices into the
backing nodes slice instead of relative offsets
- Add nodes pointer to Node and Iterator for safe navigation
- Replace danger.TypeID with reflect.Type for cache keys
- Delete internal/danger package entirely
Performance overhead is under 10% compared to the unsafe version,
which is acceptable for the safety and maintainability benefits.
[Cursor][claude-sonnet-4-20250514]
Parser did not track the location of the faulty inline table in the
document, and unmarshaler tried to the use the non-raw data field of the
AST node, both resulting into a panic when generating the parser error.
Fixes#850
Claude