v8.0.0

Migrate to Pest-based parser, allowing us to parse far greater variety of the .dbc files. Add support for REL parsing.

Added

• use AttributeDefinition for BA_DEF_REL_ (#67)
• [breaking] parse attribute definition parts (#22)
• implement rel parsing (#64)
• skip serializing empty vectors in various structs (#55)

Fixed

• fix Windows CI (#69)
• fix failing tests (#66)
• [breaking] change EnvType variants: 0-int, 1-float, 2-string (#61)
• [breaking] trim whitespace from attribute definitions (#58)
• correctly feature gate default features (#53)

Other

• update parser library reference from nom to Pest
• simplify signal parsing (#68)
• split the attribute_values per type (#63)
• Try to parse all attributes as integer first (#65)
• [breaking] clearer errors, stricter integer types, and API tweaks (#62)
• migrate to can-dbc-pest lexer, parse many new types, parse many new .dbc tests (#48)
• [breaking] remove getters because all fields are public (#60)
• make all fields pub and move structs/enums to individual ast/* files (#59) (#59)
• rework testing framework (#57)
• [breaking] remove deprecated from_str method from Dbc implementation (#50)
• [breaking] change Node to hold a single string, i.e. nodes: Vec<Node> (#54)
• move tests to separate files (#52)
• reorganize tests (#49)

v7.0.0

NOTE: can-dbc has moved to the OxiBUS GitHub org - a community focused on Rust development for automotive and industrial communication. If you maintain an open-source Rust project in this area, consider joining us - collaboration makes maintenance easier!

Breaking

• major struct and enum naming refactoring. See (#45) for the full list of changes.
• remove from_slice - use cp1252 decoder and pass in a &str instead (#44)
• rename DBC→Dbc and feature with-serde→serde (#42)

Added

• added cp1252 decoding support (#44)

Other

• relicense as MIT OR Apache-2.0 (#38)
• update README with usage examples and license information (#46)
• move tests to the end (#43)
• move test files to submodule, default with serde feature (#40)
• allow space after message ID (#25)
• upgrade to nom 8 (#36)
• clippy lints
• use insta to test all parsing results
• use clap-derive in example
• run cargo fmt
• bump dependencies and minor cleanup
• Simplify tests, test for other escaped characters
• Remove obsolete is_quote function
• Add support for escaped strings within comments
• Derive the Deserialize trait when serde is enabled
• consolidate docs with readme

CI improvements

• auto-release and cargo deny (#39)
• update README to oxibus org (#37)
• add automatic validation with precommit on CI side (#30)
• modernize CI (#33)

6.0.0

6.0.0 (2024-02-08)

• Update dependencies.
• Breaking: PR #13 and #14. Thank you @erzoe and @kistenklaus.
  Replace u32 in MessageId with enum type to raise awareness that the raw dbc message ids contain the extended bit.
  Add MessageId.raw() which returns the previous raw identifier including the extended id!

5.0.0

5.0.0 (2022-06-22)

• Add support for dbcs with extended multiplex messages. Thank you @pbert519!
• Breaking: fn message_multiplexor_switch now returns Result<Option<&Signal>, Error> instead of Option<&Signal>.
  This is necessary due to the possibility of a message having multiple multiplexor switches.

4.0.0

4.0.0 (2021-11-15)

• Migrate from nom 4.2 to nom 7.1.
• Allow "random" order of elements in DBC. They still have to be in a block e.g. all messages or comments.
• Remove verbose error in returned error
• Return &str of remaining unparsed content instead of Vec<u8>.
• Mark DBC::from_str as deprecated and replace it with TryFrom::from_str

3.0.2

3.0.2 (2020-10-28)

• Fix: Allow trailing whitespaces in dbc. Thanks @killercup.

3.0.1

• Fix: Allow zero time occurence of BU name
• Remove Cargo.lock file

3.0.0

• BREAKING: Pass MessageId's by value
• Cleanup clippy warnings

2.2.1

Update dependencies

2.1.1

Expose Symbol, Version and Node values