Crate nom_derive[−][src]
Expand description
nom-derive
Overview
nom-derive is a custom derive attribute, to derive nom parsers automatically from the structure definition.
It is not meant to replace nom, but to provide a quick and easy way to generate parsers for structures, especially for simple structures. This crate aims at simplifying common cases. In some cases, writing the parser manually will remain more efficient.
- API documentation
- The docs::Nom pseudo-module. This is the main
documentation for the
Nomattribute, with all possible options and many examples.
Feedback welcome !
#[derive(Nom)]
This crate exposes a single custom-derive macro Nom which
implements parse for the struct it is applied to.
The goal of this project is that:
derive(Nom)should be enough for you to derive nom parsers for simple structures easily, without having to write it manually- it allows overriding any parsing method by your own
- it allows using generated parsing functions along with handwritten parsers and combining them without efforts
- it remains as fast as nom
nom-derive adds declarative parsing to nom. It also allows mixing with
procedural parsing easily, making writing parsers for byte-encoded formats
very easy.
For example:
use nom_derive::*; #[derive(Nom)] struct S { a: u32, b: u16, c: u16 }
This generates an implementation of the Parse trait to S. The generated code looks
like (code simplified):
impl<'a> Parse<&'a> for S { pub fn parse_be(i: &'a [u8]) -> nom::IResult(&'a [u8], S) { let (i, a) = be_u32(i)?; let (i, b) = be_u16(i)?; let (i, c) = be_u16(i)?; Ok((i, S{ a, b, c })) } pub fn parse_le(i: &'a [u8]) -> nom::IResult(&'a [u8], S) { let (i, a) = le_u32(i)?; let (i, b) = le_u16(i)?; let (i, c) = le_u16(i)?; Ok((i, S{ a, b, c })) } pub fn parse(i: &'a [u8]) -> nom::IResult(&'a [u8], S) { S::parse_be(i) } }
To parse input, just call let res = S::parse_be(input);.
If the endianness of the struct is fixed (for ex. using the top-level BigEndian or
LittleEndian attributes, or the NomBE and NomLE custom derive), then the implementation
always uses this endianness, and all 3 functions are equivalent.
For extensive documentation of all attributes and examples, see the documentation of docs::Nom custom derive attribute.
Many examples are provided, and more can be found in the project tests.
Combinators visibility
All inferred parsers will generate code with absolute type path, so there is no need
to add use statements for them. However, if you use any combinator directly (or in a Parse
statement, for ex.), it has to be imported as usual.
That is probably not going to change, since
- a proc_macro cannot export items other than functions tagged with
#[proc_macro_derive] - there are variants of combinators with the same names (complete/streaming, bits/bytes), so re-exporting them would create side-effects.
Debug tips
- If the generated parser does not compile, add
#[nom(DebugDerive)]to the structure. It will dump the generated parser tostderr. - If the generated parser fails at runtime, try adding
#[nom(Debug)]to the structure or to fields. It wraps subparsers indbg_dmpand will print the field name and input tostderrif the parser fails.
Re-exports
pub use nom;Modules
The docs pseudo-module contains nom-derive documentation. Objects from this module
are only used to add documentation, and are not used in the crate.
Structs
Traits
Abstract method to calculate the input length
Common trait for all parsers in nom-derive
Slicing operations using ranges.
Type Definitions
Derive Macros
The Nom derive automatically generates a parse function for the structure
using nom parsers. It will try to infer parsers for primitive of known
types, but also allows you to specify parsers using custom attributes.
The NomBE acts like the Nom attribute, but sets the endianness to big-endian for the
current object. This can be overriden locally at the field-level.
This derive macro behaves exactly like Nom derive, except it prints the generated parser on stderr. This is helpful for debugging generated parsers.
The NomLE acts like the Nom attribute, but sets the endianness to little-endian for the
current object. This can be overriden locally at the field-level.