PgnExporter now extends GameContextExport and implements exportGameContext,
which replays moves from GameContext.initial to reconstruct HistoryMove records
with proper castling/promotion info before generating PGN.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Added three test cases to validate importGameContext method:
1. Valid PGN with moves returns Right with populated GameContext
2. Invalid PGN (illegal move) returns Left with error message
3. PGN with no moves returns Right with initial position
Tests use coordinate notation (e.g., e2e4) compatible with importGameContext's move replay mechanism.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Remove unnecessary if/else check (lines 49-50) that could never be true.
DefaultRules.applyMove always calls .withMove(move) on the context,
so finalCtx.moves is always populated when game.moves.nonEmpty.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Implement the importGameContext method to validate PGN input via validatePgn()
and replay each move using DefaultRules.applyMove to populate GameContext.moves.
Returns final GameContext with all moves applied or error message on failure.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Added 3 new tests for missing error paths:
- Invalid en passant square parsing (e.g., "x5")
- Half-move clock non-integer (e.g., "abc")
- Full-move number non-integer (e.g., "abc")
Updated 3 existing error tests to verify error messages:
- "expected 6" for invalid parts count
- "color" for invalid active color
- "castling" for invalid castling rights
All error assertions now use .fold() to verify message content, improving test robustness. Test count increased from 19 to 22.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
FenParser.parseFen now returns Either[String, GameContext] instead of Option.
Updated all tests to use Either combinators (isRight, isLeft, fold) instead
of Option methods (isDefined, get). Both FenParserTest and FenExporterTest
now properly handle the Either-based return type.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Update parseFen to return Either[String, GameContext] with specific error messages for each validation failure:
- Invalid parts count: reports expected 6 fields
- Invalid board: clear message about board position
- Invalid color: explains expected 'w' or 'b'
- Invalid castling, en passant, and move counts with clear descriptions
Simplify importGameContext to delegate directly to parseFen.
Keep helper methods (parseColor, parseCastling, parseEnPassant, etc.) returning Option as before.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Update GameContextExport trait to use Scala 3 syntax (colon-based) to match GameContextImport
- Add test coverage for FenParser.importGameContext method:
* Valid FEN string returns Right[GameContext] with correct context data
* Invalid FEN string returns Left[String] with error message
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Update GameContextImport trait to return Either[String, GameContext] instead of
Option[GameContext] to provide better error context during parsing. Update FenParser
implementation to convert Option to Either using toRight().
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Janis