All notable changes to this project will be documented in this file.
27f32 a float literals in docs.u128."foo"bla or 23px)suffix() method to all literal types except BoolLitIntegerBase::valuefrom_suffix and suffix methods to FloatType and IntegerTypeFromStr and Display impls to FloatType and IntegerTypeFloatType and IntegerType as #[non_exhaustive]27f32. Literal::parse and IntegerLit::parse will both identify this as an integer literal.27f32. A float literal must have a period OR an exponent part, according to the spec. Previously decimal integers were accepted in FloatLit::parse.OwnedLiteral and SharedLiteralraw_input and into_raw_input to non-bool *Lit typesimpl From<*Lit> for pm::Literal (for non-bool literals)impl From<BoolLit> for pm::Ident[into_]raw_input for integer and float literals, their internals were changed a bit so that they store the full input string now.\n\r into \nexpected value of the error returned from TryFrom<TokenTree> impls in some casesError to ParseError. That describes its purpose more closely and is particular useful now that other error types exist in the library.proc-macro feature and instead offer the corresponding impls unconditionally. Since the feature didn‘t enable/disable a dependency (proc-macro is a compiler provided crate) and since apparently it works fine in no_std environments, I dropped this feature. I don’t currently see a reason why the corresponding impls should be conditional.TryFrom<TokenTree> for litrs::Literal implsFrom<*Lit> for litrs::Literal implsTryFrom<proc_macro[2]::Literal> for *LitTryFrom<TokenTree> for *LitInvalidToken error type for all new TryFrom implsFrom impls to create a Literal from references to proc-macro literal types:From<&proc_macro::Literal>From<&proc_macro2::Literal>