Anforderungen  |   Konzepte  |   Entwurf  |   Entwicklung  |   Qualitätssicherung  |   Lebenszyklus  |   Steuerung
 
 
 
 


Quelle  lib.rs   Sprache: unbekannt

 
//! A crate for low-level parsing of the WebAssembly text formats: WAT and WAST.
//!
//! This crate is intended to be a low-level detail of the `wat` crate,
//! providing a low-level parsing API for parsing WebAssembly text format
//! structures. The API provided by this crate is very similar to
//! [`syn`](https://docs.rs/syn) and provides the ability to write customized
//! parsers which may be an extension to the core WebAssembly text format. For
//! more documentation see the [`parser`] module.
//!
//! # High-level Overview
//!
//! This crate provides a few major pieces of functionality
//!
//! * [`lexer`] - this is a raw lexer for the wasm text format. This is not
//!   customizable, but if you'd like to iterate over raw tokens this is the
//!   module for you. You likely won't use this much.
//!
//! * [`parser`] - this is the workhorse of this crate. The [`parser`] module
//!   provides the [`Parse`][] trait primarily and utilities
//!   around working with a [`Parser`](`parser::Parser`) to parse streams of
//!   tokens.
//!
//! * [`Module`](crate::core::Module) - this contains an Abstract Syntax Tree
//!   (AST) of the WebAssembly Text format (WAT) as well as the unofficial WAST
//!   format. This also has a [`Module::encode`](crate::core::Module::encode)
//!   method to emit a module in its binary form.
//!
//! # Stability and WebAssembly Features
//!
//! This crate provides support for many in-progress WebAssembly features such
//! as reference types, multi-value, etc. Be sure to check out the documentation
//! of the [`wast` crate](https://docs.rs/wast) for policy information on crate
//! stability vs WebAssembly Features. The tl;dr; version is that this crate
//! will issue semver-non-breaking releases which will break the parsing of the
//! text format. This crate, unlike `wast`, is expected to have numerous Rust
//! public API changes, all of which will be accompanied with a semver-breaking
//! release.
//!
//! # Compile-time Cargo features
//!
//! This crate has a `wasm-module` feature which is turned on by default which
//! includes all necessary support to parse full WebAssembly modules. If you
//! don't need this (for example you're parsing your own s-expression format)
//! then this feature can be disabled.
//!
//! This crate also has an off-by-default `dwarf` feature which enables using
//! [`core::EncodeOptions::dwarf`] to embed DWARF debugging information in generated
//! binaries.
//!
//! [`Parse`]: parser::Parse
//! [`LexError`]: lexer::LexError

#![deny(missing_docs, rustdoc::broken_intra_doc_links)]
#![cfg_attr(docsrs, feature(doc_auto_cfg))]

/// A macro to create a custom keyword parser.
///
/// This macro is invoked in one of two forms:
///
/// ```
/// // keyword derived from the Rust identifier:
/// wast::custom_keyword!(foo);
///
/// // or an explicitly specified string representation of the keyword:
/// wast::custom_keyword!(my_keyword = "the-wasm-keyword");
/// ```
///
/// This can then be used to parse custom keyword for custom items, such as:
///
/// ```
/// use wast::parser::{Parser, Result, Parse};
///
/// struct InlineModule<'a> {
///     inline_text: &'a str,
/// }
///
/// mod kw {
///     wast::custom_keyword!(inline);
/// }
///
/// // Parse an inline string module of the form:
/// //
/// //    (inline "(module (func))")
/// impl<'a> Parse<'a> for InlineModule<'a> {
///     fn parse(parser: Parser<'a>) -> Result<Self> {
///         parser.parse::<kw::inline>()?;
///         Ok(InlineModule {
///             inline_text: parser.parse()?,
///         })
///     }
/// }
/// ```
///
/// Note that the keyword name can only start with a lower-case letter, i.e. 'a'..'z'.
#[macro_export]
macro_rules! custom_keyword {
    ($name:ident) => {
        $crate::custom_keyword!($name = stringify!($name));
    };
    ($name:ident = $kw:expr) => {
        #[allow(non_camel_case_types)]
        #[allow(missing_docs)]
        #[derive(Debug, Copy, Clone)]
        pub struct $name(#[allow(dead_code)] pub $crate::token::Span);

        impl<'a> $crate::parser::Parse<'a> for $name {
            fn parse(parser: $crate::parser::Parser<'a>) -> $crate::parser::Result<Self> {
                parser.step(|c| {
                    if let Some((kw, rest)) = c.keyword()? {
                        if kw == $kw {
                            return Ok(($name(c.cur_span()), rest));
                        }
                    }
                    Err(c.error(concat!("expected keyword `", $kw, "`")))
                })
            }
        }

        impl $crate::parser::Peek for $name {
            fn peek(cursor: $crate::parser::Cursor<'_>) -> $crate::parser::Result<bool> {
                Ok(if let Some((kw, _rest)) = cursor.keyword()? {
                    kw == $kw
                } else {
                    false
                })
            }

            fn display() -> &'static str {
                concat!("`", $kw, "`")
            }
        }
    };
}

/// A macro for defining custom reserved symbols.
///
/// This is like `custom_keyword!` but for reserved symbols (`Token::Reserved`)
/// instead of keywords (`Token::Keyword`).
///
/// ```
/// use wast::parser::{Parser, Result, Parse};
///
/// // Define a custom reserved symbol, the "spaceship" operator: `<=>`.
/// wast::custom_reserved!(spaceship = "<=>");
///
/// /// A "three-way comparison" like `(<=> a b)` that returns -1 if `a` is less
/// /// than `b`, 0 if they're equal, and 1 if `a` is greater than `b`.
/// struct ThreeWayComparison<'a> {
///     lhs: wast::core::Expression<'a>,
///     rhs: wast::core::Expression<'a>,
/// }
///
/// impl<'a> Parse<'a> for ThreeWayComparison<'a> {
///     fn parse(parser: Parser<'a>) -> Result<Self> {
///         parser.parse::<spaceship>()?;
///         let lhs = parser.parse()?;
///         let rhs = parser.parse()?;
///         Ok(ThreeWayComparison { lhs, rhs })
///     }
/// }
/// ```
#[macro_export]
macro_rules! custom_reserved {
    ($name:ident) => {
        $crate::custom_reserved!($name = stringify!($name));
    };
    ($name:ident = $rsv:expr) => {
        #[allow(non_camel_case_types)]
        #[allow(missing_docs)]
        #[derive(Debug)]
        pub struct $name(pub $crate::token::Span);

        impl<'a> $crate::parser::Parse<'a> for $name {
            fn parse(parser: $crate::parser::Parser<'a>) -> $crate::parser::Result<Self> {
                parser.step(|c| {
                    if let Some((rsv, rest)) = c.reserved()? {
                        if rsv == $rsv {
                            return Ok(($name(c.cur_span()), rest));
                        }
                    }
                    Err(c.error(concat!("expected reserved symbol `", $rsv, "`")))
                })
            }
        }

        impl $crate::parser::Peek for $name {
            fn peek(cursor: $crate::parser::Cursor<'_>) -> Result<bool> {
                if let Some((rsv, _rest)) = cursor.reserved()? {
                    Ok(rsv == $rsv)
                } else {
                    Ok(false)
                }
            }

            fn display() -> &'static str {
                concat!("`", $rsv, "`")
            }
        }
    };
}

/// A macro, like [`custom_keyword`], to create a type which can be used to
/// parse/peek annotation directives.
///
/// Note that when you're parsing custom annotations it can be somewhat tricky
/// due to the nature that most of them are skipped. You'll want to be sure to
/// consult the documentation of [`Parser::register_annotation`][register] when
/// using this macro.
///
/// # Examples
///
/// To see an example of how to use this macro, let's invent our own syntax for
/// the [producers section][section] which looks like:
///
/// ```wat
/// (@producer "wat" "1.0.2")
/// ```
///
/// Here, for simplicity, we'll assume everything is a `processed-by` directive,
/// but you could get much more fancy with this as well.
///
/// ```
/// # use wast::*;
/// # use wast::parser::*;
///
/// // First we define the custom annotation keyword we're using, and by
/// // convention we define it in an `annotation` module.
/// mod annotation {
///     wast::annotation!(producer);
/// }
///
/// struct Producer<'a> {
///     name: &'a str,
///     version: &'a str,
/// }
///
/// impl<'a> Parse<'a> for Producer<'a> {
///     fn parse(parser: Parser<'a>) -> Result<Self> {
///         // Remember that parser conventionally parse the *interior* of an
///         // s-expression, so we parse our `@producer` annotation and then we
///         // parse the payload of our annotation.
///         parser.parse::<annotation::producer>()?;
///         Ok(Producer {
///             name: parser.parse()?,
///             version: parser.parse()?,
///         })
///     }
/// }
/// ```
///
/// Note though that this is only half of the parser for annotations. The other
/// half is calling the [`register_annotation`][register] method at the right
/// time to ensure the parser doesn't automatically skip our `@producer`
/// directive. Note that we *can't* call it inside the `Parse for Producer`
/// definition because that's too late and the annotation would already have
/// been skipped.
///
/// Instead we'll need to call it from a higher level-parser before the
/// parenthesis have been parsed, like so:
///
/// ```
/// # use wast::*;
/// # use wast::parser::*;
/// struct Module<'a> {
///     fields: Vec<ModuleField<'a>>,
/// }
///
/// impl<'a> Parse<'a> for Module<'a> {
///     fn parse(parser: Parser<'a>) -> Result<Self> {
///         // .. parse module header here ...
///
///         // register our custom `@producer` annotation before we start
///         // parsing the parentheses of each field
///         let _r = parser.register_annotation("producer");
///
///         let mut fields = Vec::new();
///         while !parser.is_empty() {
///             fields.push(parser.parens(|p| p.parse())?);
///         }
///         Ok(Module { fields })
///     }
/// }
///
/// enum ModuleField<'a> {
///     Producer(Producer<'a>),
///     // ...
/// }
/// # struct Producer<'a>(&'a str);
/// # impl<'a> Parse<'a> for Producer<'a> {
/// #   fn parse(parser: Parser<'a>) -> Result<Self> { Ok(Producer(parser.parse()?)) }
/// # }
/// # mod annotation { wast::annotation!(producer); }
///
/// impl<'a> Parse<'a> for ModuleField<'a> {
///     fn parse(parser: Parser<'a>) -> Result<Self> {
///         // and here `peek` works and our delegated parsing works because the
///         // annotation has been registered.
///         if parser.peek::<annotation::producer>()? {
///             return Ok(ModuleField::Producer(parser.parse()?));
///         }
///
///         // .. typically we'd parse other module fields here...
///
///         Err(parser.error("unknown module field"))
///     }
/// }
/// ```
///
/// [register]: crate::parser::Parser::register_annotation
/// [section]: https://github.com/WebAssembly/tool-conventions/blob/master/ProducersSection.md
#[macro_export]
macro_rules! annotation {
    ($name:ident) => {
        $crate::annotation!($name = stringify!($name));
    };
    ($name:ident = $annotation:expr) => {
        #[allow(non_camel_case_types)]
        #[allow(missing_docs)]
        #[derive(Debug)]
        pub struct $name(pub $crate::token::Span);

        impl<'a> $crate::parser::Parse<'a> for $name {
            fn parse(parser: $crate::parser::Parser<'a>) -> $crate::parser::Result<Self> {
                parser.step(|c| {
                    if let Some((a, rest)) = c.annotation()? {
                        if a == $annotation {
                            return Ok(($name(c.cur_span()), rest));
                        }
                    }
                    Err(c.error(concat!("expected annotation `@", $annotation, "`")))
                })
            }
        }

        impl $crate::parser::Peek for $name {
            fn peek(cursor: $crate::parser::Cursor<'_>) -> $crate::parser::Result<bool> {
                Ok(if let Some((a, _rest)) = cursor.annotation()? {
                    a == $annotation
                } else {
                    false
                })
            }

            fn display() -> &'static str {
                concat!("`@", $annotation, "`")
            }
        }
    };
}

pub mod lexer;
pub mod parser;
pub mod token;

#[cfg(feature = "wasm-module")]
mod encode;
mod error;
#[cfg(feature = "wasm-module")]
mod gensym;
#[cfg(feature = "wasm-module")]
mod names;
pub use self::error::*;

#[cfg(feature = "wasm-module")]
macro_rules! id {
    ($($t:tt)*) => ($($t)*)
}

#[cfg(feature = "wasm-module")]
id! {
    mod wast;
    mod wat;
    pub use self::wast::*;
    pub use self::wat::*;

    // Support for core wasm parsing
    pub mod core;

    // Support for component model parsing
    #[cfg(feature = "component-model")]
    pub mod component;
    #[cfg(not(feature = "component-model"))]
    #[path = "component_disabled.rs"]
    pub mod component;
}

/// Common keyword used to parse WebAssembly text files.
pub mod kw {
    custom_keyword!(after);
    custom_keyword!(alias);
    custom_keyword!(any);
    custom_keyword!(anyref);
    custom_keyword!(arg);
    custom_keyword!(array);
    custom_keyword!(arrayref);
    custom_keyword!(assert_exception);
    custom_keyword!(assert_exhaustion);
    custom_keyword!(assert_invalid);
    custom_keyword!(assert_malformed);
    custom_keyword!(assert_return);
    custom_keyword!(assert_trap);
    custom_keyword!(assert_unlinkable);
    custom_keyword!(assert_suspension);
    custom_keyword!(before);
    custom_keyword!(binary);
    custom_keyword!(block);
    custom_keyword!(borrow);
    custom_keyword!(catch);
    custom_keyword!(catch_ref);
    custom_keyword!(catch_all);
    custom_keyword!(catch_all_ref);
    custom_keyword!(code);
    custom_keyword!(cont);
    custom_keyword!(contref);
    custom_keyword!(component);
    custom_keyword!(data);
    custom_keyword!(declare);
    custom_keyword!(delegate);
    custom_keyword!(r#do = "do");
    custom_keyword!(dtor);
    custom_keyword!(elem);
    custom_keyword!(end);
    custom_keyword!(tag);
    custom_keyword!(exn);
    custom_keyword!(exnref);
    custom_keyword!(export);
    custom_keyword!(r#extern = "extern");
    custom_keyword!(externref);
    custom_keyword!(eq);
    custom_keyword!(eqref);
    custom_keyword!(f32);
    custom_keyword!(f32x4);
    custom_keyword!(f64);
    custom_keyword!(f64x2);
    custom_keyword!(field);
    custom_keyword!(first);
    custom_keyword!(func);
    custom_keyword!(funcref);
    custom_keyword!(get);
    custom_keyword!(global);
    custom_keyword!(i16);
    custom_keyword!(i16x8);
    custom_keyword!(i31);
    custom_keyword!(i31ref);
    custom_keyword!(i32);
    custom_keyword!(i32x4);
    custom_keyword!(i64);
    custom_keyword!(i64x2);
    custom_keyword!(i8);
    custom_keyword!(i8x16);
    custom_keyword!(import);
    custom_keyword!(instance);
    custom_keyword!(instantiate);
    custom_keyword!(interface);
    custom_keyword!(invoke);
    custom_keyword!(item);
    custom_keyword!(last);
    custom_keyword!(local);
    custom_keyword!(memory);
    custom_keyword!(module);
    custom_keyword!(modulecode);
    custom_keyword!(nan_arithmetic = "nan:arithmetic");
    custom_keyword!(nan_canonical = "nan:canonical");
    custom_keyword!(nocont);
    custom_keyword!(nofunc);
    custom_keyword!(noextern);
    custom_keyword!(noexn);
    custom_keyword!(none);
    custom_keyword!(null);
    custom_keyword!(nullcontref);
    custom_keyword!(nullfuncref);
    custom_keyword!(nullexternref);
    custom_keyword!(nullexnref);
    custom_keyword!(nullref);
    custom_keyword!(offset);
    custom_keyword!(on);
    custom_keyword!(outer);
    custom_keyword!(own);
    custom_keyword!(pagesize);
    custom_keyword!(param);
    custom_keyword!(parent);
    custom_keyword!(passive);
    custom_keyword!(quote);
    custom_keyword!(r#else = "else");
    custom_keyword!(r#if = "if");
    custom_keyword!(r#loop = "loop");
    custom_keyword!(r#mut = "mut");
    custom_keyword!(r#type = "type");
    custom_keyword!(r#ref = "ref");
    custom_keyword!(ref_func = "ref.func");
    custom_keyword!(ref_null = "ref.null");
    custom_keyword!(register);
    custom_keyword!(rec);
    custom_keyword!(acq_rel);
    custom_keyword!(rep);
    custom_keyword!(resource);
    custom_keyword!(resource_new = "resource.new");
    custom_keyword!(resource_drop = "resource.drop");
    custom_keyword!(resource_rep = "resource.rep");
    custom_keyword!(result);
    custom_keyword!(seq_cst);
    custom_keyword!(shared);
    custom_keyword!(start);
    custom_keyword!(sub);
    custom_keyword!(switch);
    custom_keyword!(r#final = "final");
    custom_keyword!(table);
    custom_keyword!(then);
    custom_keyword!(r#try = "try");
    custom_keyword!(v128);
    custom_keyword!(value);
    custom_keyword!(s8);
    custom_keyword!(s16);
    custom_keyword!(s32);
    custom_keyword!(s64);
    custom_keyword!(u8);
    custom_keyword!(u16);
    custom_keyword!(u32);
    custom_keyword!(u64);
    custom_keyword!(char);
    custom_keyword!(case);
    custom_keyword!(refines);
    custom_keyword!(record);
    custom_keyword!(string);
    custom_keyword!(bool_ = "bool");
    custom_keyword!(float32);
    custom_keyword!(float64);
    custom_keyword!(variant);
    custom_keyword!(flags);
    custom_keyword!(option);
    custom_keyword!(tuple);
    custom_keyword!(list);
    custom_keyword!(error);
    custom_keyword!(canon);
    custom_keyword!(lift);
    custom_keyword!(lower);
    custom_keyword!(enum_ = "enum");
    custom_keyword!(string_utf8 = "string-encoding=utf8");
    custom_keyword!(string_utf16 = "string-encoding=utf16");
    custom_keyword!(string_latin1_utf16 = "string-encoding=latin1+utf16");
    custom_keyword!(r#struct = "struct");
    custom_keyword!(structref);
    custom_keyword!(realloc);
    custom_keyword!(post_return = "post-return");
    custom_keyword!(with);
    custom_keyword!(core);
    custom_keyword!(true_ = "true");
    custom_keyword!(false_ = "false");
    custom_keyword!(language);
    custom_keyword!(sdk);
    custom_keyword!(processed_by = "processed-by");
    custom_keyword!(mem_info = "mem-info");
    custom_keyword!(needed);
    custom_keyword!(export_info = "export-info");
    custom_keyword!(import_info = "import-info");
    custom_keyword!(thread);
    custom_keyword!(thread_spawn = "thread.spawn");
    custom_keyword!(thread_hw_concurrency = "thread.hw_concurrency");
    custom_keyword!(wait);
    custom_keyword!(definition);
}

/// Common annotations used to parse WebAssembly text files.
pub mod annotation {
    annotation!(custom);
    annotation!(name);
    annotation!(producers);
    annotation!(dylink_0 = "dylink.0");
    annotation!(metadata_code_branch_hint = "metadata.code.branch_hint");
}

[ Dauer der Verarbeitung: 0.46 Sekunden  (vorverarbeitet)  ]

                                                                                                                                                                                                                                                                                                                                                                                                     


Neuigkeiten

     Aktuelles
     Motto des Tages

Software

     Produkte
     Quellcodebibliothek

Aktivitäten

     Artikel über Sicherheit
     Anleitung zur Aktivierung von SSL

Muße

     Gedichte
     Musik
     Bilder

Jenseits des Üblichen ....

Besucherstatistik

Besucherstatistik

Monitoring

Montastic status badge