xref: /linux/rust/quote/lib.rs (revision 784faa8eca8270671e0ed6d9d21f04bbb80fc5f7)

// SPDX-License-Identifier: Apache-2.0 OR MIT

//! [![github]](https://github.com/dtolnay/quote) [![crates-io]](https://crates.io/crates/quote) [![docs-rs]](https://docs.rs/quote)
//!
//! [github]: https://img.shields.io/badge/github-8da0cb?style=for-the-badge&labelColor=555555&logo=github
//! [crates-io]: https://img.shields.io/badge/crates.io-fc8d62?style=for-the-badge&labelColor=555555&logo=rust
//! [docs-rs]: https://img.shields.io/badge/docs.rs-66c2a5?style=for-the-badge&labelColor=555555&logo=docs.rs
//!
//! <br>
//!
//! This crate provides the [`quote!`] macro for turning Rust syntax tree data
//! structures into tokens of source code.
//!
//! Procedural macros in Rust receive a stream of tokens as input, execute
//! arbitrary Rust code to determine how to manipulate those tokens, and produce
//! a stream of tokens to hand back to the compiler to compile into the caller's
//! crate. Quasi-quoting is a solution to one piece of that &mdash; producing
//! tokens to return to the compiler.
//!
//! The idea of quasi-quoting is that we write *code* that we treat as *data*.
//! Within the `quote!` macro, we can write what looks like code to our text
//! editor or IDE. We get all the benefits of the editor's brace matching,
//! syntax highlighting, indentation, and maybe autocompletion. But rather than
//! compiling that as code into the current crate, we can treat it as data, pass
//! it around, mutate it, and eventually hand it back to the compiler as tokens
//! to compile into the macro caller's crate.
//!
//! This crate is motivated by the procedural macro use case, but is a
//! general-purpose Rust quasi-quoting library and is not specific to procedural
//! macros.
//!
//! ```toml
//! [dependencies]
//! quote = "1.0"
//! ```
//!
//! <br>
//!
//! # Example
//!
//! The following quasi-quoted block of code is something you might find in [a]
//! procedural macro having to do with data structure serialization. The `#var`
//! syntax performs interpolation of runtime variables into the quoted tokens.
//! Check out the documentation of the [`quote!`] macro for more detail about
//! the syntax. See also the [`quote_spanned!`] macro which is important for
//! implementing hygienic procedural macros.
//!
//! [a]: https://serde.rs/
//!
//! ```
//! # use quote::quote;
//! #
//! # let generics = "";
//! # let where_clause = "";
//! # let field_ty = "";
//! # let item_ty = "";
//! # let path = "";
//! # let value = "";
//! #
//! let tokens = quote! {
//!     struct SerializeWith #generics #where_clause {
//!         value: &'a #field_ty,
//!         phantom: core::marker::PhantomData<#item_ty>,
//!     }
//!
//!     impl #generics serde::Serialize for SerializeWith #generics #where_clause {
//!         fn serialize<S>(&self, serializer: S) -> Result<S::Ok, S::Error>
//!         where
//!             S: serde::Serializer,
//!         {
//!             #path(self.value, serializer)
//!         }
//!     }
//!
//!     SerializeWith {
//!         value: #value,
//!         phantom: core::marker::PhantomData::<#item_ty>,
//!     }
//! };
//! ```
//!
//! <br>
//!
//! # Non-macro code generators
//!
//! When using `quote` in a build.rs or main.rs and writing the output out to a
//! file, consider having the code generator pass the tokens through
//! [prettyplease] before writing. This way if an error occurs in the generated
//! code it is convenient for a human to read and debug.
//!
//! [prettyplease]: https://github.com/dtolnay/prettyplease

// Quote types in rustdoc of other crates get linked to here.
#![doc(html_root_url = "https://docs.rs/quote/1.0.40")]
#![allow(
    clippy::doc_markdown,
    clippy::elidable_lifetime_names,
    clippy::missing_errors_doc,
    clippy::missing_panics_doc,
    clippy::module_name_repetitions,
    clippy::needless_lifetimes,
    // false positive https://github.com/rust-lang/rust-clippy/issues/6983
    clippy::wrong_self_convention,
)]

extern crate alloc;

#[cfg(feature = "proc-macro")]
extern crate proc_macro;

mod ext;
mod format;
mod ident_fragment;
mod to_tokens;

// Not public API.
#[doc(hidden)]
#[path = "runtime.rs"]
pub mod __private;

pub use crate::ext::TokenStreamExt;
pub use crate::ident_fragment::IdentFragment;
pub use crate::to_tokens::ToTokens;

// Not public API.
#[doc(hidden)]
pub mod spanned;

macro_rules! __quote {
    ($quote:item) => {
        /// The whole point.
        ///
        /// Performs variable interpolation against the input and produces it as
        /// [`proc_macro2::TokenStream`].
        ///
        /// Note: for returning tokens to the compiler in a procedural macro, use
        /// `.into()` on the result to convert to [`proc_macro::TokenStream`].
        ///
        /// <br>
        ///
        /// # Interpolation
        ///
        /// Variable interpolation is done with `#var` (similar to `$var` in
        /// `macro_rules!` macros). This grabs the `var` variable that is currently in
        /// scope and inserts it in that location in the output tokens. Any type
        /// implementing the [`ToTokens`] trait can be interpolated. This includes most
        /// Rust primitive types as well as most of the syntax tree types from the [Syn]
        /// crate.
        ///
        /// [Syn]: https://github.com/dtolnay/syn
        ///
        /// Repetition is done using `#(...)*` or `#(...),*` again similar to
        /// `macro_rules!`. This iterates through the elements of any variable
        /// interpolated within the repetition and inserts a copy of the repetition body
        /// for each one. The variables in an interpolation may be a `Vec`, slice,
        /// `BTreeSet`, or any `Iterator`.
        ///
        /// - `#(#var)*` — no separators
        /// - `#(#var),*` — the character before the asterisk is used as a separator
        /// - `#( struct #var; )*` — the repetition can contain other tokens
        /// - `#( #k => println!("{}", #v), )*` — even multiple interpolations
        ///
        /// <br>
        ///
        /// # Hygiene
        ///
        /// Any interpolated tokens preserve the `Span` information provided by their
        /// `ToTokens` implementation. Tokens that originate within the `quote!`
        /// invocation are spanned with [`Span::call_site()`].
        ///
        /// [`Span::call_site()`]: proc_macro2::Span::call_site
        ///
        /// A different span can be provided through the [`quote_spanned!`] macro.
        ///
        /// <br>
        ///
        /// # Return type
        ///
        /// The macro evaluates to an expression of type `proc_macro2::TokenStream`.
        /// Meanwhile Rust procedural macros are expected to return the type
        /// `proc_macro::TokenStream`.
        ///
        /// The difference between the two types is that `proc_macro` types are entirely
        /// specific to procedural macros and cannot ever exist in code outside of a
        /// procedural macro, while `proc_macro2` types may exist anywhere including
        /// tests and non-macro code like main.rs and build.rs. This is why even the
        /// procedural macro ecosystem is largely built around `proc_macro2`, because
        /// that ensures the libraries are unit testable and accessible in non-macro
        /// contexts.
        ///
        /// There is a [`From`]-conversion in both directions so returning the output of
        /// `quote!` from a procedural macro usually looks like `tokens.into()` or
        /// `proc_macro::TokenStream::from(tokens)`.
        ///
        /// <br>
        ///
        /// # Examples
        ///
        /// ### Procedural macro
        ///
        /// The structure of a basic procedural macro is as follows. Refer to the [Syn]
        /// crate for further useful guidance on using `quote!` as part of a procedural
        /// macro.
        ///
        /// [Syn]: https://github.com/dtolnay/syn
        ///
        /// ```
        /// # #[cfg(any())]
        /// extern crate proc_macro;
        /// # extern crate proc_macro2;
        ///
        /// # #[cfg(any())]
        /// use proc_macro::TokenStream;
        /// # use proc_macro2::TokenStream;
        /// use quote::quote;
        ///
        /// # const IGNORE_TOKENS: &'static str = stringify! {
        /// #[proc_macro_derive(HeapSize)]
        /// # };
        /// pub fn derive_heap_size(input: TokenStream) -> TokenStream {
        ///     // Parse the input and figure out what implementation to generate...
        ///     # const IGNORE_TOKENS: &'static str = stringify! {
        ///     let name = /* ... */;
        ///     let expr = /* ... */;
        ///     # };
        ///     #
        ///     # let name = 0;
        ///     # let expr = 0;
        ///
        ///     let expanded = quote! {
        ///         // The generated impl.
        ///         impl heapsize::HeapSize for #name {
        ///             fn heap_size_of_children(&self) -> usize {
        ///                 #expr
        ///             }
        ///         }
        ///     };
        ///
        ///     // Hand the output tokens back to the compiler.
        ///     TokenStream::from(expanded)
        /// }
        /// ```
        ///
        /// <p><br></p>
        ///
        /// ### Combining quoted fragments
        ///
        /// Usually you don't end up constructing an entire final `TokenStream` in one
        /// piece. Different parts may come from different helper functions. The tokens
        /// produced by `quote!` themselves implement `ToTokens` and so can be
        /// interpolated into later `quote!` invocations to build up a final result.
        ///
        /// ```
        /// # use quote::quote;
        /// #
        /// let type_definition = quote! {...};
        /// let methods = quote! {...};
        ///
        /// let tokens = quote! {
        ///     #type_definition
        ///     #methods
        /// };
        /// ```
        ///
        /// <p><br></p>
        ///
        /// ### Constructing identifiers
        ///
        /// Suppose we have an identifier `ident` which came from somewhere in a macro
        /// input and we need to modify it in some way for the macro output. Let's
        /// consider prepending the identifier with an underscore.
        ///
        /// Simply interpolating the identifier next to an underscore will not have the
        /// behavior of concatenating them. The underscore and the identifier will
        /// continue to be two separate tokens as if you had written `_ x`.
        ///
        /// ```
        /// # use proc_macro2::{self as syn, Span};
        /// # use quote::quote;
        /// #
        /// # let ident = syn::Ident::new("i", Span::call_site());
        /// #
        /// // incorrect
        /// quote! {
        ///     let mut _#ident = 0;
        /// }
        /// # ;
        /// ```
        ///
        /// The solution is to build a new identifier token with the correct value. As
        /// this is such a common case, the [`format_ident!`] macro provides a
        /// convenient utility for doing so correctly.
        ///
        /// ```
        /// # use proc_macro2::{Ident, Span};
        /// # use quote::{format_ident, quote};
        /// #
        /// # let ident = Ident::new("i", Span::call_site());
        /// #
        /// let varname = format_ident!("_{}", ident);
        /// quote! {
        ///     let mut #varname = 0;
        /// }
        /// # ;
        /// ```
        ///
        /// Alternatively, the APIs provided by Syn and proc-macro2 can be used to
        /// directly build the identifier. This is roughly equivalent to the above, but
        /// will not handle `ident` being a raw identifier.
        ///
        /// ```
        /// # use proc_macro2::{self as syn, Span};
        /// # use quote::quote;
        /// #
        /// # let ident = syn::Ident::new("i", Span::call_site());
        /// #
        /// let concatenated = format!("_{}", ident);
        /// let varname = syn::Ident::new(&concatenated, ident.span());
        /// quote! {
        ///     let mut #varname = 0;
        /// }
        /// # ;
        /// ```
        ///
        /// <p><br></p>
        ///
        /// ### Making method calls
        ///
        /// Let's say our macro requires some type specified in the macro input to have
        /// a constructor called `new`. We have the type in a variable called
        /// `field_type` of type `syn::Type` and want to invoke the constructor.
        ///
        /// ```
        /// # use quote::quote;
        /// #
        /// # let field_type = quote!(...);
        /// #
        /// // incorrect
        /// quote! {
        ///     let value = #field_type::new();
        /// }
        /// # ;
        /// ```
        ///
        /// This works only sometimes. If `field_type` is `String`, the expanded code
        /// contains `String::new()` which is fine. But if `field_type` is something
        /// like `Vec<i32>` then the expanded code is `Vec<i32>::new()` which is invalid
        /// syntax. Ordinarily in handwritten Rust we would write `Vec::<i32>::new()`
        /// but for macros often the following is more convenient.
        ///
        /// ```
        /// # use quote::quote;
        /// #
        /// # let field_type = quote!(...);
        /// #
        /// quote! {
        ///     let value = <#field_type>::new();
        /// }
        /// # ;
        /// ```
        ///
        /// This expands to `<Vec<i32>>::new()` which behaves correctly.
        ///
        /// A similar pattern is appropriate for trait methods.
        ///
        /// ```
        /// # use quote::quote;
        /// #
        /// # let field_type = quote!(...);
        /// #
        /// quote! {
        ///     let value = <#field_type as core::default::Default>::default();
        /// }
        /// # ;
        /// ```
        ///
        /// <p><br></p>
        ///
        /// ### Interpolating text inside of doc comments
        ///
        /// Neither doc comments nor string literals get interpolation behavior in
        /// quote:
        ///
        /// ```compile_fail
        /// quote! {
        ///     /// try to interpolate: #ident
        ///     ///
        ///     /// ...
        /// }
        /// ```
        ///
        /// ```compile_fail
        /// quote! {
        ///     #[doc = "try to interpolate: #ident"]
        /// }
        /// ```
        ///
        /// Instead the best way to build doc comments that involve variables is by
        /// formatting the doc string literal outside of quote.
        ///
        /// ```rust
        /// # use proc_macro2::{Ident, Span};
        /// # use quote::quote;
        /// #
        /// # const IGNORE: &str = stringify! {
        /// let msg = format!(...);
        /// # };
        /// #
        /// # let ident = Ident::new("var", Span::call_site());
        /// # let msg = format!("try to interpolate: {}", ident);
        /// quote! {
        ///     #[doc = #msg]
        ///     ///
        ///     /// ...
        /// }
        /// # ;
        /// ```
        ///
        /// <p><br></p>
        ///
        /// ### Indexing into a tuple struct
        ///
        /// When interpolating indices of a tuple or tuple struct, we need them not to
        /// appears suffixed as integer literals by interpolating them as [`syn::Index`]
        /// instead.
        ///
        /// [`syn::Index`]: https://docs.rs/syn/2.0/syn/struct.Index.html
        ///
        /// ```compile_fail
        /// let i = 0usize..self.fields.len();
        ///
        /// // expands to 0 + self.0usize.heap_size() + self.1usize.heap_size() + ...
        /// // which is not valid syntax
        /// quote! {
        ///     0 #( + self.#i.heap_size() )*
        /// }
        /// ```
        ///
        /// ```
        /// # use proc_macro2::{Ident, TokenStream};
        /// # use quote::quote;
        /// #
        /// # mod syn {
        /// #     use proc_macro2::{Literal, TokenStream};
        /// #     use quote::{ToTokens, TokenStreamExt};
        /// #
        /// #     pub struct Index(usize);
        /// #
        /// #     impl From<usize> for Index {
        /// #         fn from(i: usize) -> Self {
        /// #             Index(i)
        /// #         }
        /// #     }
        /// #
        /// #     impl ToTokens for Index {
        /// #         fn to_tokens(&self, tokens: &mut TokenStream) {
        /// #             tokens.append(Literal::usize_unsuffixed(self.0));
        /// #         }
        /// #     }
        /// # }
        /// #
        /// # struct Struct {
        /// #     fields: Vec<Ident>,
        /// # }
        /// #
        /// # impl Struct {
        /// #     fn example(&self) -> TokenStream {
        /// let i = (0..self.fields.len()).map(syn::Index::from);
        ///
        /// // expands to 0 + self.0.heap_size() + self.1.heap_size() + ...
        /// quote! {
        ///     0 #( + self.#i.heap_size() )*
        /// }
        /// #     }
        /// # }
        /// ```
        $quote
    };
}

#[cfg(doc)]
__quote![
    #[macro_export]
    macro_rules! quote {
        ($($tt:tt)*) => {
            ...
        };
    }
];

#[cfg(not(doc))]
__quote![
    #[macro_export]
    macro_rules! quote {
        () => {
            $crate::__private::TokenStream::new()
        };

        // Special case rule for a single tt, for performance.
        ($tt:tt) => {{
            let mut _s = $crate::__private::TokenStream::new();
            $crate::quote_token!{$tt _s}
            _s
        }};

        // Special case rules for two tts, for performance.
        (# $var:ident) => {{
            let mut _s = $crate::__private::TokenStream::new();
            $crate::ToTokens::to_tokens(&$var, &mut _s);
            _s
        }};
        ($tt1:tt $tt2:tt) => {{
            let mut _s = $crate::__private::TokenStream::new();
            $crate::quote_token!{$tt1 _s}
            $crate::quote_token!{$tt2 _s}
            _s
        }};

        // Rule for any other number of tokens.
        ($($tt:tt)*) => {{
            let mut _s = $crate::__private::TokenStream::new();
            $crate::quote_each_token!{_s $($tt)*}
            _s
        }};
    }
];

macro_rules! __quote_spanned {
    ($quote_spanned:item) => {
        /// Same as `quote!`, but applies a given span to all tokens originating within
        /// the macro invocation.
        ///
        /// <br>
        ///
        /// # Syntax
        ///
        /// A span expression of type [`Span`], followed by `=>`, followed by the tokens
        /// to quote. The span expression should be brief &mdash; use a variable for
        /// anything more than a few characters. There should be no space before the
        /// `=>` token.
        ///
        /// [`Span`]: proc_macro2::Span
        ///
        /// ```
        /// # use proc_macro2::Span;
        /// # use quote::quote_spanned;
        /// #
        /// # const IGNORE_TOKENS: &'static str = stringify! {
        /// let span = /* ... */;
        /// # };
        /// # let span = Span::call_site();
        /// # let init = 0;
        ///
        /// // On one line, use parentheses.
        /// let tokens = quote_spanned!(span=> Box::into_raw(Box::new(#init)));
        ///
        /// // On multiple lines, place the span at the top and use braces.
        /// let tokens = quote_spanned! {span=>
        ///     Box::into_raw(Box::new(#init))
        /// };
        /// ```
        ///
        /// The lack of space before the `=>` should look jarring to Rust programmers
        /// and this is intentional. The formatting is designed to be visibly
        /// off-balance and draw the eye a particular way, due to the span expression
        /// being evaluated in the context of the procedural macro and the remaining
        /// tokens being evaluated in the generated code.
        ///
        /// <br>
        ///
        /// # Hygiene
        ///
        /// Any interpolated tokens preserve the `Span` information provided by their
        /// `ToTokens` implementation. Tokens that originate within the `quote_spanned!`
        /// invocation are spanned with the given span argument.
        ///
        /// <br>
        ///
        /// # Example
        ///
        /// The following procedural macro code uses `quote_spanned!` to assert that a
        /// particular Rust type implements the [`Sync`] trait so that references can be
        /// safely shared between threads.
        ///
        /// ```
        /// # use quote::{quote_spanned, TokenStreamExt, ToTokens};
        /// # use proc_macro2::{Span, TokenStream};
        /// #
        /// # struct Type;
        /// #
        /// # impl Type {
        /// #     fn span(&self) -> Span {
        /// #         Span::call_site()
        /// #     }
        /// # }
        /// #
        /// # impl ToTokens for Type {
        /// #     fn to_tokens(&self, _tokens: &mut TokenStream) {}
        /// # }
        /// #
        /// # let ty = Type;
        /// # let call_site = Span::call_site();
        /// #
        /// let ty_span = ty.span();
        /// let assert_sync = quote_spanned! {ty_span=>
        ///     struct _AssertSync where #ty: Sync;
        /// };
        /// ```
        ///
        /// If the assertion fails, the user will see an error like the following. The
        /// input span of their type is highlighted in the error.
        ///
        /// ```text
        /// error[E0277]: the trait bound `*const (): std::marker::Sync` is not satisfied
        ///   --> src/main.rs:10:21
        ///    |
        /// 10 |     static ref PTR: *const () = &();
        ///    |                     ^^^^^^^^^ `*const ()` cannot be shared between threads safely
        /// ```
        ///
        /// In this example it is important for the where-clause to be spanned with the
        /// line/column information of the user's input type so that error messages are
        /// placed appropriately by the compiler.
        $quote_spanned
    };
}

#[cfg(doc)]
__quote_spanned![
    #[macro_export]
    macro_rules! quote_spanned {
        ($span:expr=> $($tt:tt)*) => {
            ...
        };
    }
];

#[cfg(not(doc))]
__quote_spanned![
    #[macro_export]
    macro_rules! quote_spanned {
        ($span:expr=>) => {{
            let _: $crate::__private::Span = $crate::__private::get_span($span).__into_span();
            $crate::__private::TokenStream::new()
        }};

        // Special case rule for a single tt, for performance.
        ($span:expr=> $tt:tt) => {{
            let mut _s = $crate::__private::TokenStream::new();
            let _span: $crate::__private::Span = $crate::__private::get_span($span).__into_span();
            $crate::quote_token_spanned!{$tt _s _span}
            _s
        }};

        // Special case rules for two tts, for performance.
        ($span:expr=> # $var:ident) => {{
            let mut _s = $crate::__private::TokenStream::new();
            let _: $crate::__private::Span = $crate::__private::get_span($span).__into_span();
            $crate::ToTokens::to_tokens(&$var, &mut _s);
            _s
        }};
        ($span:expr=> $tt1:tt $tt2:tt) => {{
            let mut _s = $crate::__private::TokenStream::new();
            let _span: $crate::__private::Span = $crate::__private::get_span($span).__into_span();
            $crate::quote_token_spanned!{$tt1 _s _span}
            $crate::quote_token_spanned!{$tt2 _s _span}
            _s
        }};

        // Rule for any other number of tokens.
        ($span:expr=> $($tt:tt)*) => {{
            let mut _s = $crate::__private::TokenStream::new();
            let _span: $crate::__private::Span = $crate::__private::get_span($span).__into_span();
            $crate::quote_each_token_spanned!{_s _span $($tt)*}
            _s
        }};
    }
];

// Extract the names of all #metavariables and pass them to the $call macro.
//
// in:   pounded_var_names!(then!(...) a #b c #( #d )* #e)
// out:  then!(... b);
//       then!(... d);
//       then!(... e);
#[macro_export]
#[doc(hidden)]
macro_rules! pounded_var_names {
    ($call:ident! $extra:tt $($tts:tt)*) => {
        $crate::pounded_var_names_with_context!{$call! $extra
            (@ $($tts)*)
            ($($tts)* @)
        }
    };
}

#[macro_export]
#[doc(hidden)]
macro_rules! pounded_var_names_with_context {
    ($call:ident! $extra:tt ($($b1:tt)*) ($($curr:tt)*)) => {
        $(
            $crate::pounded_var_with_context!{$call! $extra $b1 $curr}
        )*
    };
}

#[macro_export]
#[doc(hidden)]
macro_rules! pounded_var_with_context {
    ($call:ident! $extra:tt $b1:tt ( $($inner:tt)* )) => {
        $crate::pounded_var_names!{$call! $extra $($inner)*}
    };

    ($call:ident! $extra:tt $b1:tt [ $($inner:tt)* ]) => {
        $crate::pounded_var_names!{$call! $extra $($inner)*}
    };

    ($call:ident! $extra:tt $b1:tt { $($inner:tt)* }) => {
        $crate::pounded_var_names!{$call! $extra $($inner)*}
    };

    ($call:ident!($($extra:tt)*) # $var:ident) => {
        $crate::$call!($($extra)* $var);
    };

    ($call:ident! $extra:tt $b1:tt $curr:tt) => {};
}

#[macro_export]
#[doc(hidden)]
macro_rules! quote_bind_into_iter {
    ($has_iter:ident $var:ident) => {
        // `mut` may be unused if $var occurs multiple times in the list.
        #[allow(unused_mut)]
        let (mut $var, i) = $var.quote_into_iter();
        let $has_iter = $has_iter | i;
    };
}

#[macro_export]
#[doc(hidden)]
macro_rules! quote_bind_next_or_break {
    ($var:ident) => {
        let $var = match $var.next() {
            Some(_x) => $crate::__private::RepInterp(_x),
            None => break,
        };
    };
}

// The obvious way to write this macro is as a tt muncher. This implementation
// does something more complex for two reasons.
//
//   - With a tt muncher it's easy to hit Rust's built-in recursion_limit, which
//     this implementation avoids because it isn't tail recursive.
//
//   - Compile times for a tt muncher are quadratic relative to the length of
//     the input. This implementation is linear, so it will be faster
//     (potentially much faster) for big inputs. However, the constant factors
//     of this implementation are higher than that of a tt muncher, so it is
//     somewhat slower than a tt muncher if there are many invocations with
//     short inputs.
//
// An invocation like this:
//
//     quote_each_token!(_s a b c d e f g h i j);
//
// expands to this:
//
//     quote_tokens_with_context!(_s
//         (@  @  @  @   @   @   a   b   c   d   e   f   g  h  i  j)
//         (@  @  @  @   @   a   b   c   d   e   f   g   h  i  j  @)
//         (@  @  @  @   a   b   c   d   e   f   g   h   i  j  @  @)
//         (@  @  @ (a) (b) (c) (d) (e) (f) (g) (h) (i) (j) @  @  @)
//         (@  @  a  b   c   d   e   f   g   h   i   j   @  @  @  @)
//         (@  a  b  c   d   e   f   g   h   i   j   @   @  @  @  @)
//         (a  b  c  d   e   f   g   h   i   j   @   @   @  @  @  @)
//     );
//
// which gets transposed and expanded to this:
//
//     quote_token_with_context!(_s @ @ @  @  @ @ a);
//     quote_token_with_context!(_s @ @ @  @  @ a b);
//     quote_token_with_context!(_s @ @ @  @  a b c);
//     quote_token_with_context!(_s @ @ @ (a) b c d);
//     quote_token_with_context!(_s @ @ a (b) c d e);
//     quote_token_with_context!(_s @ a b (c) d e f);
//     quote_token_with_context!(_s a b c (d) e f g);
//     quote_token_with_context!(_s b c d (e) f g h);
//     quote_token_with_context!(_s c d e (f) g h i);
//     quote_token_with_context!(_s d e f (g) h i j);
//     quote_token_with_context!(_s e f g (h) i j @);
//     quote_token_with_context!(_s f g h (i) j @ @);
//     quote_token_with_context!(_s g h i (j) @ @ @);
//     quote_token_with_context!(_s h i j  @  @ @ @);
//     quote_token_with_context!(_s i j @  @  @ @ @);
//     quote_token_with_context!(_s j @ @  @  @ @ @);
//
// Without having used muncher-style recursion, we get one invocation of
// quote_token_with_context for each original tt, with three tts of context on
// either side. This is enough for the longest possible interpolation form (a
// repetition with separator, as in `# (#var) , *`) to be fully represented with
// the first or last tt in the middle.
//
// The middle tt (surrounded by parentheses) is the tt being processed.
//
//   - When it is a `#`, quote_token_with_context can do an interpolation. The
//     interpolation kind will depend on the three subsequent tts.
//
//   - When it is within a later part of an interpolation, it can be ignored
//     because the interpolation has already been done.
//
//   - When it is not part of an interpolation it can be pushed as a single
//     token into the output.
//
//   - When the middle token is an unparenthesized `@`, that call is one of the
//     first 3 or last 3 calls of quote_token_with_context and does not
//     correspond to one of the original input tokens, so turns into nothing.
#[macro_export]
#[doc(hidden)]
macro_rules! quote_each_token {
    ($tokens:ident $($tts:tt)*) => {
        $crate::quote_tokens_with_context!{$tokens
            (@ @ @ @ @ @ $($tts)*)
            (@ @ @ @ @ $($tts)* @)
            (@ @ @ @ $($tts)* @ @)
            (@ @ @ $(($tts))* @ @ @)
            (@ @ $($tts)* @ @ @ @)
            (@ $($tts)* @ @ @ @ @)
            ($($tts)* @ @ @ @ @ @)
        }
    };
}

// See the explanation on quote_each_token.
#[macro_export]
#[doc(hidden)]
macro_rules! quote_each_token_spanned {
    ($tokens:ident $span:ident $($tts:tt)*) => {
        $crate::quote_tokens_with_context_spanned!{$tokens $span
            (@ @ @ @ @ @ $($tts)*)
            (@ @ @ @ @ $($tts)* @)
            (@ @ @ @ $($tts)* @ @)
            (@ @ @ $(($tts))* @ @ @)
            (@ @ $($tts)* @ @ @ @)
            (@ $($tts)* @ @ @ @ @)
            ($($tts)* @ @ @ @ @ @)
        }
    };
}

// See the explanation on quote_each_token.
#[macro_export]
#[doc(hidden)]
macro_rules! quote_tokens_with_context {
    ($tokens:ident
        ($($b3:tt)*) ($($b2:tt)*) ($($b1:tt)*)
        ($($curr:tt)*)
        ($($a1:tt)*) ($($a2:tt)*) ($($a3:tt)*)
    ) => {
        $(
            $crate::quote_token_with_context!{$tokens $b3 $b2 $b1 $curr $a1 $a2 $a3}
        )*
    };
}

// See the explanation on quote_each_token.
#[macro_export]
#[doc(hidden)]
macro_rules! quote_tokens_with_context_spanned {
    ($tokens:ident $span:ident
        ($($b3:tt)*) ($($b2:tt)*) ($($b1:tt)*)
        ($($curr:tt)*)
        ($($a1:tt)*) ($($a2:tt)*) ($($a3:tt)*)
    ) => {
        $(
            $crate::quote_token_with_context_spanned!{$tokens $span $b3 $b2 $b1 $curr $a1 $a2 $a3}
        )*
    };
}

// See the explanation on quote_each_token.
#[macro_export]
#[doc(hidden)]
macro_rules! quote_token_with_context {
    // Unparenthesized `@` indicates this call does not correspond to one of the
    // original input tokens. Ignore it.
    ($tokens:ident $b3:tt $b2:tt $b1:tt @ $a1:tt $a2:tt $a3:tt) => {};

    // A repetition with no separator.
    ($tokens:ident $b3:tt $b2:tt $b1:tt (#) ( $($inner:tt)* ) * $a3:tt) => {{
        use $crate::__private::ext::*;
        let has_iter = $crate::__private::ThereIsNoIteratorInRepetition;
        $crate::pounded_var_names!{quote_bind_into_iter!(has_iter) () $($inner)*}
        let _: $crate::__private::HasIterator = has_iter;
        // This is `while true` instead of `loop` because if there are no
        // iterators used inside of this repetition then the body would not
        // contain any `break`, so the compiler would emit unreachable code
        // warnings on anything below the loop. We use has_iter to detect and
        // fail to compile when there are no iterators, so here we just work
        // around the unneeded extra warning.
        while true {
            $crate::pounded_var_names!{quote_bind_next_or_break!() () $($inner)*}
            $crate::quote_each_token!{$tokens $($inner)*}
        }
    }};
    // ... and one step later.
    ($tokens:ident $b3:tt $b2:tt # (( $($inner:tt)* )) * $a2:tt $a3:tt) => {};
    // ... and one step later.
    ($tokens:ident $b3:tt # ( $($inner:tt)* ) (*) $a1:tt $a2:tt $a3:tt) => {};

    // A repetition with separator.
    ($tokens:ident $b3:tt $b2:tt $b1:tt (#) ( $($inner:tt)* ) $sep:tt *) => {{
        use $crate::__private::ext::*;
        let mut _i = 0usize;
        let has_iter = $crate::__private::ThereIsNoIteratorInRepetition;
        $crate::pounded_var_names!{quote_bind_into_iter!(has_iter) () $($inner)*}
        let _: $crate::__private::HasIterator = has_iter;
        while true {
            $crate::pounded_var_names!{quote_bind_next_or_break!() () $($inner)*}
            if _i > 0 {
                $crate::quote_token!{$sep $tokens}
            }
            _i += 1;
            $crate::quote_each_token!{$tokens $($inner)*}
        }
    }};
    // ... and one step later.
    ($tokens:ident $b3:tt $b2:tt # (( $($inner:tt)* )) $sep:tt * $a3:tt) => {};
    // ... and one step later.
    ($tokens:ident $b3:tt # ( $($inner:tt)* ) ($sep:tt) * $a2:tt $a3:tt) => {};
    // (A special case for `#(var)**`, where the first `*` is treated as the
    // repetition symbol and the second `*` is treated as an ordinary token.)
    ($tokens:ident # ( $($inner:tt)* ) * (*) $a1:tt $a2:tt $a3:tt) => {
        // https://github.com/dtolnay/quote/issues/130
        $crate::quote_token!{* $tokens}
    };
    // ... and one step later.
    ($tokens:ident # ( $($inner:tt)* ) $sep:tt (*) $a1:tt $a2:tt $a3:tt) => {};

    // A non-repetition interpolation.
    ($tokens:ident $b3:tt $b2:tt $b1:tt (#) $var:ident $a2:tt $a3:tt) => {
        $crate::ToTokens::to_tokens(&$var, &mut $tokens);
    };
    // ... and one step later.
    ($tokens:ident $b3:tt $b2:tt # ($var:ident) $a1:tt $a2:tt $a3:tt) => {};

    // An ordinary token, not part of any interpolation.
    ($tokens:ident $b3:tt $b2:tt $b1:tt ($curr:tt) $a1:tt $a2:tt $a3:tt) => {
        $crate::quote_token!{$curr $tokens}
    };
}

// See the explanation on quote_each_token, and on the individual rules of
// quote_token_with_context.
#[macro_export]
#[doc(hidden)]
macro_rules! quote_token_with_context_spanned {
    ($tokens:ident $span:ident $b3:tt $b2:tt $b1:tt @ $a1:tt $a2:tt $a3:tt) => {};

    ($tokens:ident $span:ident $b3:tt $b2:tt $b1:tt (#) ( $($inner:tt)* ) * $a3:tt) => {{
        use $crate::__private::ext::*;
        let has_iter = $crate::__private::ThereIsNoIteratorInRepetition;
        $crate::pounded_var_names!{quote_bind_into_iter!(has_iter) () $($inner)*}
        let _: $crate::__private::HasIterator = has_iter;
        while true {
            $crate::pounded_var_names!{quote_bind_next_or_break!() () $($inner)*}
            $crate::quote_each_token_spanned!{$tokens $span $($inner)*}
        }
    }};
    ($tokens:ident $span:ident $b3:tt $b2:tt # (( $($inner:tt)* )) * $a2:tt $a3:tt) => {};
    ($tokens:ident $span:ident $b3:tt # ( $($inner:tt)* ) (*) $a1:tt $a2:tt $a3:tt) => {};

    ($tokens:ident $span:ident $b3:tt $b2:tt $b1:tt (#) ( $($inner:tt)* ) $sep:tt *) => {{
        use $crate::__private::ext::*;
        let mut _i = 0usize;
        let has_iter = $crate::__private::ThereIsNoIteratorInRepetition;
        $crate::pounded_var_names!{quote_bind_into_iter!(has_iter) () $($inner)*}
        let _: $crate::__private::HasIterator = has_iter;
        while true {
            $crate::pounded_var_names!{quote_bind_next_or_break!() () $($inner)*}
            if _i > 0 {
                $crate::quote_token_spanned!{$sep $tokens $span}
            }
            _i += 1;
            $crate::quote_each_token_spanned!{$tokens $span $($inner)*}
        }
    }};
    ($tokens:ident $span:ident $b3:tt $b2:tt # (( $($inner:tt)* )) $sep:tt * $a3:tt) => {};
    ($tokens:ident $span:ident $b3:tt # ( $($inner:tt)* ) ($sep:tt) * $a2:tt $a3:tt) => {};
    ($tokens:ident $span:ident # ( $($inner:tt)* ) * (*) $a1:tt $a2:tt $a3:tt) => {
        // https://github.com/dtolnay/quote/issues/130
        $crate::quote_token_spanned!{* $tokens $span}
    };
    ($tokens:ident $span:ident # ( $($inner:tt)* ) $sep:tt (*) $a1:tt $a2:tt $a3:tt) => {};

    ($tokens:ident $span:ident $b3:tt $b2:tt $b1:tt (#) $var:ident $a2:tt $a3:tt) => {
        $crate::ToTokens::to_tokens(&$var, &mut $tokens);
    };
    ($tokens:ident $span:ident $b3:tt $b2:tt # ($var:ident) $a1:tt $a2:tt $a3:tt) => {};

    ($tokens:ident $span:ident $b3:tt $b2:tt $b1:tt ($curr:tt) $a1:tt $a2:tt $a3:tt) => {
        $crate::quote_token_spanned!{$curr $tokens $span}
    };
}

// These rules are ordered by approximate token frequency, at least for the
// first 10 or so, to improve compile times. Having `ident` first is by far the
// most important because it's typically 2-3x more common than the next most
// common token.
//
// Separately, we put the token being matched in the very front so that failing
// rules may fail to match as quickly as possible.
#[macro_export]
#[doc(hidden)]
macro_rules! quote_token {
    ($ident:ident $tokens:ident) => {
        $crate::__private::push_ident(&mut $tokens, stringify!($ident));
    };

    (:: $tokens:ident) => {
        $crate::__private::push_colon2(&mut $tokens);
    };

    (( $($inner:tt)* ) $tokens:ident) => {
        $crate::__private::push_group(
            &mut $tokens,
            $crate::__private::Delimiter::Parenthesis,
            $crate::quote!($($inner)*),
        );
    };

    ([ $($inner:tt)* ] $tokens:ident) => {
        $crate::__private::push_group(
            &mut $tokens,
            $crate::__private::Delimiter::Bracket,
            $crate::quote!($($inner)*),
        );
    };

    ({ $($inner:tt)* } $tokens:ident) => {
        $crate::__private::push_group(
            &mut $tokens,
            $crate::__private::Delimiter::Brace,
            $crate::quote!($($inner)*),
        );
    };

    (# $tokens:ident) => {
        $crate::__private::push_pound(&mut $tokens);
    };

    (, $tokens:ident) => {
        $crate::__private::push_comma(&mut $tokens);
    };

    (. $tokens:ident) => {
        $crate::__private::push_dot(&mut $tokens);
    };

    (; $tokens:ident) => {
        $crate::__private::push_semi(&mut $tokens);
    };

    (: $tokens:ident) => {
        $crate::__private::push_colon(&mut $tokens);
    };

    (+ $tokens:ident) => {
        $crate::__private::push_add(&mut $tokens);
    };

    (+= $tokens:ident) => {
        $crate::__private::push_add_eq(&mut $tokens);
    };

    (& $tokens:ident) => {
        $crate::__private::push_and(&mut $tokens);
    };

    (&& $tokens:ident) => {
        $crate::__private::push_and_and(&mut $tokens);
    };

    (&= $tokens:ident) => {
        $crate::__private::push_and_eq(&mut $tokens);
    };

    (@ $tokens:ident) => {
        $crate::__private::push_at(&mut $tokens);
    };

    (! $tokens:ident) => {
        $crate::__private::push_bang(&mut $tokens);
    };

    (^ $tokens:ident) => {
        $crate::__private::push_caret(&mut $tokens);
    };

    (^= $tokens:ident) => {
        $crate::__private::push_caret_eq(&mut $tokens);
    };

    (/ $tokens:ident) => {
        $crate::__private::push_div(&mut $tokens);
    };

    (/= $tokens:ident) => {
        $crate::__private::push_div_eq(&mut $tokens);
    };

    (.. $tokens:ident) => {
        $crate::__private::push_dot2(&mut $tokens);
    };

    (... $tokens:ident) => {
        $crate::__private::push_dot3(&mut $tokens);
    };

    (..= $tokens:ident) => {
        $crate::__private::push_dot_dot_eq(&mut $tokens);
    };

    (= $tokens:ident) => {
        $crate::__private::push_eq(&mut $tokens);
    };

    (== $tokens:ident) => {
        $crate::__private::push_eq_eq(&mut $tokens);
    };

    (>= $tokens:ident) => {
        $crate::__private::push_ge(&mut $tokens);
    };

    (> $tokens:ident) => {
        $crate::__private::push_gt(&mut $tokens);
    };

    (<= $tokens:ident) => {
        $crate::__private::push_le(&mut $tokens);
    };

    (< $tokens:ident) => {
        $crate::__private::push_lt(&mut $tokens);
    };

    (*= $tokens:ident) => {
        $crate::__private::push_mul_eq(&mut $tokens);
    };

    (!= $tokens:ident) => {
        $crate::__private::push_ne(&mut $tokens);
    };

    (| $tokens:ident) => {
        $crate::__private::push_or(&mut $tokens);
    };

    (|= $tokens:ident) => {
        $crate::__private::push_or_eq(&mut $tokens);
    };

    (|| $tokens:ident) => {
        $crate::__private::push_or_or(&mut $tokens);
    };

    (? $tokens:ident) => {
        $crate::__private::push_question(&mut $tokens);
    };

    (-> $tokens:ident) => {
        $crate::__private::push_rarrow(&mut $tokens);
    };

    (<- $tokens:ident) => {
        $crate::__private::push_larrow(&mut $tokens);
    };

    (% $tokens:ident) => {
        $crate::__private::push_rem(&mut $tokens);
    };

    (%= $tokens:ident) => {
        $crate::__private::push_rem_eq(&mut $tokens);
    };

    (=> $tokens:ident) => {
        $crate::__private::push_fat_arrow(&mut $tokens);
    };

    (<< $tokens:ident) => {
        $crate::__private::push_shl(&mut $tokens);
    };

    (<<= $tokens:ident) => {
        $crate::__private::push_shl_eq(&mut $tokens);
    };

    (>> $tokens:ident) => {
        $crate::__private::push_shr(&mut $tokens);
    };

    (>>= $tokens:ident) => {
        $crate::__private::push_shr_eq(&mut $tokens);
    };

    (* $tokens:ident) => {
        $crate::__private::push_star(&mut $tokens);
    };

    (- $tokens:ident) => {
        $crate::__private::push_sub(&mut $tokens);
    };

    (-= $tokens:ident) => {
        $crate::__private::push_sub_eq(&mut $tokens);
    };

    ($lifetime:lifetime $tokens:ident) => {
        $crate::__private::push_lifetime(&mut $tokens, stringify!($lifetime));
    };

    (_ $tokens:ident) => {
        $crate::__private::push_underscore(&mut $tokens);
    };

    ($other:tt $tokens:ident) => {
        $crate::__private::parse(&mut $tokens, stringify!($other));
    };
}

// See the comment above `quote_token!` about the rule ordering.
#[macro_export]
#[doc(hidden)]
macro_rules! quote_token_spanned {
    ($ident:ident $tokens:ident $span:ident) => {
        $crate::__private::push_ident_spanned(&mut $tokens, $span, stringify!($ident));
    };

    (:: $tokens:ident $span:ident) => {
        $crate::__private::push_colon2_spanned(&mut $tokens, $span);
    };

    (( $($inner:tt)* ) $tokens:ident $span:ident) => {
        $crate::__private::push_group_spanned(
            &mut $tokens,
            $span,
            $crate::__private::Delimiter::Parenthesis,
            $crate::quote_spanned!($span=> $($inner)*),
        );
    };

    ([ $($inner:tt)* ] $tokens:ident $span:ident) => {
        $crate::__private::push_group_spanned(
            &mut $tokens,
            $span,
            $crate::__private::Delimiter::Bracket,
            $crate::quote_spanned!($span=> $($inner)*),
        );
    };

    ({ $($inner:tt)* } $tokens:ident $span:ident) => {
        $crate::__private::push_group_spanned(
            &mut $tokens,
            $span,
            $crate::__private::Delimiter::Brace,
            $crate::quote_spanned!($span=> $($inner)*),
        );
    };

    (# $tokens:ident $span:ident) => {
        $crate::__private::push_pound_spanned(&mut $tokens, $span);
    };

    (, $tokens:ident $span:ident) => {
        $crate::__private::push_comma_spanned(&mut $tokens, $span);
    };

    (. $tokens:ident $span:ident) => {
        $crate::__private::push_dot_spanned(&mut $tokens, $span);
    };

    (; $tokens:ident $span:ident) => {
        $crate::__private::push_semi_spanned(&mut $tokens, $span);
    };

    (: $tokens:ident $span:ident) => {
        $crate::__private::push_colon_spanned(&mut $tokens, $span);
    };

    (+ $tokens:ident $span:ident) => {
        $crate::__private::push_add_spanned(&mut $tokens, $span);
    };

    (+= $tokens:ident $span:ident) => {
        $crate::__private::push_add_eq_spanned(&mut $tokens, $span);
    };

    (& $tokens:ident $span:ident) => {
        $crate::__private::push_and_spanned(&mut $tokens, $span);
    };

    (&& $tokens:ident $span:ident) => {
        $crate::__private::push_and_and_spanned(&mut $tokens, $span);
    };

    (&= $tokens:ident $span:ident) => {
        $crate::__private::push_and_eq_spanned(&mut $tokens, $span);
    };

    (@ $tokens:ident $span:ident) => {
        $crate::__private::push_at_spanned(&mut $tokens, $span);
    };

    (! $tokens:ident $span:ident) => {
        $crate::__private::push_bang_spanned(&mut $tokens, $span);
    };

    (^ $tokens:ident $span:ident) => {
        $crate::__private::push_caret_spanned(&mut $tokens, $span);
    };

    (^= $tokens:ident $span:ident) => {
        $crate::__private::push_caret_eq_spanned(&mut $tokens, $span);
    };

    (/ $tokens:ident $span:ident) => {
        $crate::__private::push_div_spanned(&mut $tokens, $span);
    };

    (/= $tokens:ident $span:ident) => {
        $crate::__private::push_div_eq_spanned(&mut $tokens, $span);
    };

    (.. $tokens:ident $span:ident) => {
        $crate::__private::push_dot2_spanned(&mut $tokens, $span);
    };

    (... $tokens:ident $span:ident) => {
        $crate::__private::push_dot3_spanned(&mut $tokens, $span);
    };

    (..= $tokens:ident $span:ident) => {
        $crate::__private::push_dot_dot_eq_spanned(&mut $tokens, $span);
    };

    (= $tokens:ident $span:ident) => {
        $crate::__private::push_eq_spanned(&mut $tokens, $span);
    };

    (== $tokens:ident $span:ident) => {
        $crate::__private::push_eq_eq_spanned(&mut $tokens, $span);
    };

    (>= $tokens:ident $span:ident) => {
        $crate::__private::push_ge_spanned(&mut $tokens, $span);
    };

    (> $tokens:ident $span:ident) => {
        $crate::__private::push_gt_spanned(&mut $tokens, $span);
    };

    (<= $tokens:ident $span:ident) => {
        $crate::__private::push_le_spanned(&mut $tokens, $span);
    };

    (< $tokens:ident $span:ident) => {
        $crate::__private::push_lt_spanned(&mut $tokens, $span);
    };

    (*= $tokens:ident $span:ident) => {
        $crate::__private::push_mul_eq_spanned(&mut $tokens, $span);
    };

    (!= $tokens:ident $span:ident) => {
        $crate::__private::push_ne_spanned(&mut $tokens, $span);
    };

    (| $tokens:ident $span:ident) => {
        $crate::__private::push_or_spanned(&mut $tokens, $span);
    };

    (|= $tokens:ident $span:ident) => {
        $crate::__private::push_or_eq_spanned(&mut $tokens, $span);
    };

    (|| $tokens:ident $span:ident) => {
        $crate::__private::push_or_or_spanned(&mut $tokens, $span);
    };

    (? $tokens:ident $span:ident) => {
        $crate::__private::push_question_spanned(&mut $tokens, $span);
    };

    (-> $tokens:ident $span:ident) => {
        $crate::__private::push_rarrow_spanned(&mut $tokens, $span);
    };

    (<- $tokens:ident $span:ident) => {
        $crate::__private::push_larrow_spanned(&mut $tokens, $span);
    };

    (% $tokens:ident $span:ident) => {
        $crate::__private::push_rem_spanned(&mut $tokens, $span);
    };

    (%= $tokens:ident $span:ident) => {
        $crate::__private::push_rem_eq_spanned(&mut $tokens, $span);
    };

    (=> $tokens:ident $span:ident) => {
        $crate::__private::push_fat_arrow_spanned(&mut $tokens, $span);
    };

    (<< $tokens:ident $span:ident) => {
        $crate::__private::push_shl_spanned(&mut $tokens, $span);
    };

    (<<= $tokens:ident $span:ident) => {
        $crate::__private::push_shl_eq_spanned(&mut $tokens, $span);
    };

    (>> $tokens:ident $span:ident) => {
        $crate::__private::push_shr_spanned(&mut $tokens, $span);
    };

    (>>= $tokens:ident $span:ident) => {
        $crate::__private::push_shr_eq_spanned(&mut $tokens, $span);
    };

    (* $tokens:ident $span:ident) => {
        $crate::__private::push_star_spanned(&mut $tokens, $span);
    };

    (- $tokens:ident $span:ident) => {
        $crate::__private::push_sub_spanned(&mut $tokens, $span);
    };

    (-= $tokens:ident $span:ident) => {
        $crate::__private::push_sub_eq_spanned(&mut $tokens, $span);
    };

    ($lifetime:lifetime $tokens:ident $span:ident) => {
        $crate::__private::push_lifetime_spanned(&mut $tokens, $span, stringify!($lifetime));
    };

    (_ $tokens:ident $span:ident) => {
        $crate::__private::push_underscore_spanned(&mut $tokens, $span);
    };

    ($other:tt $tokens:ident $span:ident) => {
        $crate::__private::parse_spanned(&mut $tokens, $span, stringify!($other));
    };
}