nom8 changelog

[epage]

Background:

• nom8 was a short-lived fork of nom to experiment with active proposals.
• winnow is a long-term fork of nom that used nom8 as its base and has inherited nom8s versioning.

See also #94

winnow 0.3.0

Breaking Changes

General

• winnow::multi are now generic over the container via the Accumulate trait
  • In most cases, type inference will catch this, but not always
• winnow::error::Error's code field was renamed to kind
• winnow::error::ErrorKind for infinite loops, like in winnow::multi, was changed to ErrorKind::Assert
• winnow::error::Err was renamed to ErrMode
  • Error variant was renamed to Backtrack
  • Fatal variant was renamed to Cut

From 0.2.0

• Renamed winnow::input to winnow::stream
• Merged most stream traits into winnow::stream::Stream
• Renamed winnow::stream::InputIsStreaming to winnow::stream::StreamIsPartial
  • Changed it from taking a const STREAMING: bool to being is_partial/is_partial_supported
• Renamed winnow::stream::Streaming to winnow::stream::Partial
  • Now uses Partial::new instead of being directly constructable
• Parser::map_opt was renamed to Parser::verify_map
• Parser::debug_err was removed in favor of compiling with --features debug
• Unsigned parsers like winnow::character::u8 were merge into winnow::character::dec_uint
• Unsigned parsers like winnow::character::i8 were merge into winnow::character::dec_int
• winnow::character::hex_u32 was replaced with winnow::character::hex_uint
• winnow::bytes::take_till was renamed to winnow::bytes::take_till0
• winnow::bytes::take_while was renamed to winnow::bytes::take_while0
• winnow::bytes::take_until was renamed to winnow::bytes::take_until0
• winnow::bytes::escaped was moved to winnow::character::escaped
• winnow::bytes::escaped_transform was moved to winnow::character::escaped_transform

nom8 0.2.0 - 2022-12-27

Breaking Changes

From nom8

• Removed pub use bits::*; into root namespace (fix(bits)!: Don't re-export in root namespace  #52)
• Removed error::error_to_u32 (fix(error)!: Remove error_to_u32 #54)a

From 0.1.0

• Removed nom::character::char in favor of nom::bytes::one_of (fix(bytes): Deprecate one-off char error reporting #45)
• Changed IntoOutput to allow reconstructing state (fix(input): Allow reconstructing input state #57)

Deprecations

• Deprecated ParseError::from_char in favor of ContextError (fix(bytes): Deprecate one-off char error reporting #45)
• Deprecated error::make_error and error::append_error (fix(error): Deprecate make_error, append_eror #55)

Feature

• nom::input::Stateful for attaching state to the input type (feat(input): Thread state through parsing #58)
• nom::input::Located for tracking the input's location, along with Parser::span and Parser::with_span to capture it (feat(input): Thread state through parsing #58)

Fixes

From 0.1.0

• Generalize InputTakeAtPosition for Streaming

nom8 0.1.0 - 2022-12-23

Breaking Changes

• Moved input traits from nom to nom::input (refactor!: Clarify module names #13)
• Renamed InputTakeAtPosition functions to be explicit about complete vs streaming policy
• Removed impl ExtendInto for char (fix(input)!: Remove impl ExtendInto for char #19)
• Combinator structs returned by Parser were moved to combinator module (fix!: Move core combinators to combinators mod #4)
• Tweaks were made to what input traits are needed for various types

Deprecations

• character::is_* functions in favor of AsChar (fix: Deprecate character::is_* functions #25)
• *::complete::* and *::streaming::* parsers in favor of merged ones (feat: Auto-detect whether to use complete/streaming parser #28)
• Freestanding map, flat_map, map_parser in favor of Parser functions (fix: Aim for one way of doing a parser #37)
• Freestanding map_res, map_opt, complete, verify, value, recognize_with_value, context, dbg_dmp in favor of Parser methods (feat: Move non-grammar combinators to Parser #40)
• Freestanding into in favor of Parser::output_into and Parser::err_into (fix: Aim for one way of doing a parser #37, fix(parser)!: Split Parser::into into Parser::output_into and Parser::err_into #48)
• is_a with take_while1 thanks to new FindToken impls (fix(bytes): Generalize byte/char APIs #44)
• is_not with take_till1 thanks to new FindToken impls (fix(bytes): Generalize byte/char APIs #44)
• satisfy with one_of thanks to new FindToken impls (fix(bytes): Generalize byte/char APIs #44)
• Parser::and, Parser::or in favor of impl Parser for <tuples> and alt (fix: Aim for one way of doing a parser #37)
• tuple, pair parser in favor of using tuples (feat(parser): impl Parser for tuples  #42, fix(seq): Deprecate pair #46)

Features

• Merged streaming/complete parsers, opting in to streaming with the Streaming<I> input type (feat: Auto-detect whether to use complete/streaming parser #28)
  • Merged ASCII float parsers moved to character and renamed to their rust types (fix: Move float parsing to character #38)
  • Moved one_of, none_of from character to bytes as they are more general (fix(bytes): Generalize byte/char APIs #44)
  • Moved character::anychar to bytes::any with a more generic return type (fix(bytes): Generalize byte/char APIs #44)
• Allow generic context in errors (feat(error)!: Make Context content generic #49)
• Parser::by_ref to allow using &mut impl Parser as a Parser (refactor(multi): Reuse some length_data logic #34, fix(parser): Rename as_mut_parser to by_ref  #41)
• impl Parser for <tuples> as an alternative to tuple(()) (feat(parser): impl Parser for tuples  #42)
• impl Parser for (char|u8|&str|&[u8]) as an alternative to one_of, char, and tag (feat(parser): impl Parsers for literals #47, Revert "feat(parser): impl Parser for range types" #50)
• Add Parser::map_res, Parser::map_opt, Parser::complete, Parser::verify, Parser::value, Parser::with_recognized, Parser::context, Parser::dbg_err (feat: Move non-grammar combinators to Parser #40)
• Allow u8 / char / ranges of the prior, functions, and tuples of the prior wherever FindToken is accepted (fix(bytes): Generalize byte/char APIs #44)
• Generalize take_while, take_till with FindToken trait (fix(bytes): Generalize byte/char APIs #44)
• Add character::is_* functions to AsChar that were missing (fix!: Remove character::is_* in favor of AsChar  #2)

Fixes

• Change fill, iterator, bits::bits, and bits::byutes from taking a function to taking a Parser (fix(multi): Switch 'fill' from taking Fn to Parser #10, fix(comb)!: Allow ParserIterator on parsers #11, feat(error): Centralize backtracking policy #120)
• Allow byte arrays of any size (fix: Support byte arrays of any size #14)
• length_data and length_value now auto-detect complete/streaming, before were streaming only (feat: Auto-detect whether to use complete/streaming parser #28)
• hex_u32 is now generic over input (fix(number): Prep for merging complete/streaming #32)
• Clamp allocations in many_m_n, count (Clamp Vec::with_capacity to 64KiB to avoid OOM rust-bakery/nom#1545)

Documentation

• Pulled loose markdown into rustdoc (docs: Pull loose markdown into docs.rs #1)

nom 7.1.0