Breaking Changes - DO NOT USE 0.8.0 please #

• API Refactoring: In preparation for RDF 1.1 Dataset support:
  • Renamed RdfCodec → RdfGraphCodec
  • Renamed RdfEncoder → RdfGraphEncoder
  • Renamed RdfDecoder → RdfGraphDecoder

Breaking Changes #

• API Refactoring: Changed terminology to align with dart:convert standards
  • Renamed parse → decode
  • Renamed serialize → encode
  • Renamed RdfFormat → RdfCodec
  • Renamed withStandardFormats → withStandardCodecs
  • Renamed withFormats → withCodecs

Added #

• Added additionalCodecs parameter to withStandardCodecs factory constructor
• Added global convenience variables (turtle, jsonld, ntriples, rdf) for easier API usage
• Improved documentation for registry access through the registry property

Added #

• RdfNamespaceMappings can now generate a prefix automatically if there was no matching prefix found

Fixed #

• IriTerm used to be case insensitive in equals, but that is actually wrong - for example: https://schema.org/Datatype != https://schema.org/datatype !

Changed #

• Another turtle tweak: suppress duplicate objects in multivalue predicates

Changed #

• Better sorting of turtle output

Changed #

• Removed unused code

Added #

• Added additional well-known namespaces to standard mappings:
  • Added geo (http://www.w3.org/2003/01/geo/wgs84_pos#)
  • Added contact (http://www.w3.org/2000/10/swap/pim/contact#)
  • Added time (http://www.w3.org/2006/time#)
  • Added vs (http://www.w3.org/2003/06/sw-vocab-status/ns#)
  • Added dcmitype (http://purl.org/dc/dcmitype/)
  • Added void (http://rdfs.org/ns/void#)
  • Added prov (http://www.w3.org/ns/prov#)
  • Added gr (http://purl.org/goodrelations/v1#)

Changed #

• Enhanced Turtle serialization for improved readability:
  • Implemented inline blank node serialization for nodes referenced exactly once
  • Added support for nested inline blank nodes in square bracket notation
  • Added proper whitespace formatting with double line breaks between subject groups
  • Optimized collection serialization to work with inline blank nodes
• Added linting configuration using package:lints/core.yaml
• Improved code quality with lint fixes and optimized Dart idioms

Changed #

• Triple, IriTerm, BlankNodeTerm and LiteralTerm classes now output in n-triple style in .toString() for better readability of debug output.
• Included information about json-ld in README and homepage.

Added #

• Added support for N-Triples format, including parser and serializer implementations
  • Implemented N-Triples as per W3C RDF 1.1 N-Triples specification
  • Added proper escaping and validation for N-Triples syntax
  • Integrated N-Triples format into the standard formats registry
  • Available via MIME type 'application/n-triples' and file extension '.nt'

Added #

• Added new factory methods LiteralTerm.integer(), LiteralTerm.decimal(), and LiteralTerm.boolean() for more concise and type-safe creation of common literal types.
• Updated TurtleParser to use these specialized factory methods for literal values.
• Enhanced TurtleSerializer to use native syntax for integer, decimal and boolean literals, resulting in more idiomatic and concise Turtle output.
• Enhanced TurtleSerializer to use native syntax for lists (aka rdf collections: ex:Person1 ex:someProperty ("foo" "bar" "blub");) and sets (aka multiple triple per subject/predicate pair: ex:Person1 ex:someProperty "foo", "bar", "blub";).

Fixed #

• Export exceptions

Fixed #

• Added back path and yaml to dev dependencies

Fixed #

• More dependency cleanups

Fixed #

• Fixed imports

Fixed #

• TurtleFormat and JsonLdFormat should have been exported
• unused http dependency was removed

Fixed #

• Fixed tests to use relative paths instead of absolute paths for better portability
• Improved examples for handling broken Turtle syntax

Added #

• Enhanced Turtle parsing flexibility with configurable parsing flags in TurtleTokenizer
  • Added TurtleParsingFlag enum with options for handling non-standard Turtle syntax
  • Implemented support for digits in local names (allowDigitInLocalName)
  • Added support for missing dots after prefix declarations (allowMissingDotAfterPrefix)
  • Implemented auto-addition of common prefixes (autoAddCommonPrefixes)
  • Added support for prefix declarations without @ symbol (allowPrefixWithoutAtSign)
  • Improved handling of missing final dots (allowMissingFinalDot)
  • Implemented support for identifiers without colons (allowIdentifiersWithoutColon)
• Improved error reporting with detailed position information and diagnostics
• Added comprehensive logging to help diagnose parsing issues in real-world Turtle files

Changed #

• Refactored TurtleTokenizer for better maintainability and modularity
• Improved documentation with practical examples of non-standard Turtle syntax handling

Fixed #

• Made Turtle parser more robust when processing real-world datasets with syntax variations
• Enhanced error messages to be more informative for debugging parsing issues

Fixed #

• Important fix: build.yaml should not have been committed, it breaks the build of user projects

Changed #

• Removed a deprecated annotation that was no longer needed after the vocabulary cleanup

Removed #

• BREAKING CHANGE: Removed all deprecated vocabulary classes from the vocab directory
  • Removed all previously deprecated vocabulary classes (SchemaProperties, SchemaTypes, SchemaPersonProperties, etc.)
  • Users should now use direct IriTerm instances instead of vocabulary classes
  • Example: replace SchemaProperties.name with IriTerm('https://schema.org/name')
• Updated examples to use direct IriTerm instances rather than vocabulary constants

Added #

• Improved documentation around IriTerm usage patterns
• Added more comprehensive test coverage for core RDF functionality
• Support for multiline strings in turtle files

Fixed #

• Fixed release script to correctly parse the changelog format with square brackets
• Added missing development dependencies required for the release tooling

• BREAKING CHANGE: Marked all vocabulary classes as deprecated
  • Added @deprecated annotation to all classes in vocabulary modules (acl, dc, dc_terms, foaf, ldp, etc.)
  • Classes will be removed in a future release due to API design concerns
  • Users should migrate to the upcoming new vocabulary API
• Improved code documentation
• Enhanced developer warnings for approaching breaking changes

• BREAKING CHANGE: Reorganized API structure for better modularity and usability
• Added RdfNamespaceMappings for improved namespace handling
• Enhanced serializers to support custom namespace mappings
• Improved documentation across the codebase
• Updated API documentation to reflect new structure

• Extended Schema.org vocabulary with postal address support
  • Added PostalAddress, ContactPoint, and Country classes
  • Added SchemaAddressProperties for complete address modeling
  • Added properties for street address, locality, region, postal code, country, etc.
• Added comprehensive tests for new postal address components
• Added example demonstrating postal address usage with Schema.org vocabulary

• Enhanced blank node handling in RDF serialization and parsing
• Refactored IriTerm usage with prevalidated constructor and improved validation logic
• Refactored RDF Core Vocabulary: replaced constants with vocabulary modules
• Added RDF vocabularies for RDFS, Schema.org, SKOS, Solid, and vCard
• Improved documentation and consistency across RDF vocabularies
• Added comprehensive tests for RDF exception handling and parsing
• Enhanced test coverage for TurtleTokenizer and JsonLdFormat functionality
• Fixed edge cases in blank node handling
• Improved test assertions and documentation for various RDF vocabularies

• Numerous improvements and new features since 0.1.2.
• See commit history for details on all enhancements and fixes.

• Cleaned up all library names for consistency (removed rdf_core. prefix, now use simple names like exceptions.base).
• Minor formatting and style improvements in core files.
• No breaking changes; all public APIs remain backward compatible.

• Example and documentation now consistently use canonical RDF vocabularies (e.g., http://xmlns.com/foaf/0.1/) with http://, not https://.
• Turtle and JSON-LD serializers automatically warn if non-canonical (https) namespaces are used when canonical (http) is available.
• Added documentation and comments on best practices for prefixes and vocabularies.
• Improved prefix handling and static analysis compliance.
• Added and refined tests for prefix usage and serialization correctness.

• Initial release of rdf_core: type-safe, extensible Dart library for RDF data manipulation.
• Implements RDF graph, triple, term, and serialization/parsing for Turtle and JSON-LD.
• Plugin architecture for formats and adapters.
• Comprehensive test coverage.