rust/src/librustc/hir/lowering.rs

3615 lines
149 KiB
Rust
Raw Normal View History

2015-07-31 00:04:06 -07:00
// Copyright 2015 The Rust Project Developers. See the COPYRIGHT
// file at the top-level directory of this distribution and at
// http://rust-lang.org/COPYRIGHT.
//
// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or
// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license
// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your
// option. This file may not be copied, modified, or distributed
// except according to those terms.
//! Lowers the AST to the HIR.
//!
//! Since the AST and HIR are fairly similar, this is mostly a simple procedure,
//! much like a fold. Where lowering involves a bit more work things get more
//! interesting and there are some invariants you should know about. These mostly
//! concern spans and ids.
//!
//! Spans are assigned to AST nodes during parsing and then are modified during
//! expansion to indicate the origin of a node and the process it went through
//! being expanded. Ids are assigned to AST nodes just before lowering.
//!
//! For the simpler lowering steps, ids and spans should be preserved. Unlike
//! expansion we do not preserve the process of lowering in the spans, so spans
//! should not be modified here. When creating a new node (as opposed to
//! 'folding' an existing one), then you create a new id using `next_id()`.
//!
//! You must ensure that ids are unique. That means that you should only use the
//! id from an AST node in a single HIR node (you can assume that AST node ids
//! are unique). Every new node must have a unique id. Avoid cloning HIR nodes.
//! If you do, you must then set the new node's id to a fresh one.
//!
//! Spans are used for error messages and for tools to map semantics back to
//! source code. It is therefore not as important with spans as ids to be strict
//! about use (you can't break the compiler by screwing up a span). Obviously, a
//! HIR node can only have a single span. But multiple nodes can have the same
//! span and spans don't need to be kept in order, etc. Where code is preserved
//! by lowering, it should have the same span as in the AST. Where HIR nodes are
//! new it is probably best to give a span for the whole AST node being lowered.
//! All nodes should have real spans, don't use dummy spans. Tools are likely to
//! get confused if the spans from leaf AST nodes occur in multiple places
//! in the HIR, especially for multiple identifiers.
2015-07-31 00:04:06 -07:00
use dep_graph::DepGraph;
2015-07-31 00:04:06 -07:00
use hir;
use hir::HirVec;
use hir::map::{Definitions, DefKey, DefPathData};
use hir::def_id::{DefIndex, DefId, CRATE_DEF_INDEX, DefIndexAddressSpace};
use hir::def::{Def, PathResolution};
use lint::builtin::PARENTHESIZED_PARAMS_IN_TYPES_AND_MODULES;
use middle::cstore::CrateStore;
use rustc_data_structures::indexed_vec::IndexVec;
use session::Session;
use util::common::FN_OUTPUT_NAME;
2017-03-24 23:03:15 +00:00
use util::nodemap::{DefIdMap, FxHashMap, NodeMap};
2015-07-31 00:04:06 -07:00
use std::collections::{BTreeMap, HashSet};
use std::fmt::Debug;
use std::iter;
2017-02-15 14:52:27 -08:00
use std::mem;
use syntax::attr;
2015-07-31 00:04:06 -07:00
use syntax::ast::*;
use syntax::errors;
2017-03-17 04:04:41 +00:00
use syntax::ext::hygiene::{Mark, SyntaxContext};
2015-07-31 00:04:06 -07:00
use syntax::ptr::P;
use syntax::codemap::{self, respan, Spanned, CompilerDesugaringKind};
2015-09-28 15:00:15 +13:00
use syntax::std_inject;
use syntax::symbol::{Symbol, keywords};
rustc: Forbid interpolated tokens in the HIR Right now the HIR contains raw `syntax::ast::Attribute` structure but nowadays these can contain arbitrary tokens. One variant of the `Token` enum is an "interpolated" token which basically means to shove all the tokens for a nonterminal in this position. A "nonterminal" in this case is roughly analagous to a macro argument: macro_rules! foo { ($a:expr) => { // $a is a nonterminal as an expression } } Currently nonterminals contain namely items and expressions, and this poses a problem for incremental compilation! With incremental we want a stable hash of all HIR items, but this means we may transitively need a stable hash *of the entire AST*, which is certainly not stable w/ node ids and whatnot. Hence today there's a "bug" where the "stable hash" of an AST is just the raw hash value of the AST, and this only arises with interpolated nonterminals. The downside of this approach, however, is that a bunch of errors get spewed out during compilation about how this isn't a great idea. This PR is focused at fixing these warnings, basically deleting them from the compiler. The implementation here is to alter attributes as they're lowered from the AST to HIR, expanding all nonterminals in-place as we see them. This code for expanding a nonterminal to a token stream already exists for the `proc_macro` crate, so we basically just reuse the same implementation there. After this PR it's considered a bug to have an `Interpolated` token and hence the stable hash implementation simply uses `bug!` in this location. Closes #40946
2017-09-15 08:28:34 -07:00
use syntax::tokenstream::{TokenStream, TokenTree, Delimited};
use syntax::parse::token::Token;
use syntax::util::small_vector::SmallVector;
use syntax::visit::{self, Visitor};
use syntax_pos::Span;
2015-07-31 00:04:06 -07:00
const HIR_ID_COUNTER_LOCKED: u32 = 0xFFFFFFFF;
pub struct LoweringContext<'a> {
2015-09-28 15:00:15 +13:00
crate_root: Option<&'static str>,
2017-06-04 18:15:42 +02:00
// Use to assign ids to hir nodes that do not directly correspond to an ast node
sess: &'a Session,
2017-06-04 18:15:42 +02:00
cstore: &'a CrateStore,
2016-04-18 10:30:55 +12:00
// As we walk the AST we must keep track of the current 'parent' def id (in
// the form of a DefIndex) so that if we create a new node which introduces
// a definition, then we can properly create the def id.
parent_def: Option<DefIndex>,
resolver: &'a mut Resolver,
2017-03-24 23:03:15 +00:00
name_map: FxHashMap<Ident, Name>,
/// The items being lowered are collected here.
items: BTreeMap<NodeId, hir::Item>,
trait_items: BTreeMap<hir::TraitItemId, hir::TraitItem>,
impl_items: BTreeMap<hir::ImplItemId, hir::ImplItem>,
2017-02-21 10:55:40 -05:00
bodies: BTreeMap<hir::BodyId, hir::Body>,
exported_macros: Vec<hir::MacroDef>,
trait_impls: BTreeMap<DefId, Vec<NodeId>>,
trait_auto_impl: BTreeMap<DefId, NodeId>,
is_generator: bool,
2016-12-26 14:34:03 +01:00
catch_scopes: Vec<NodeId>,
2017-02-15 14:52:27 -08:00
loop_scopes: Vec<NodeId>,
is_in_loop_condition: bool,
is_in_trait_impl: bool,
2017-02-15 14:52:27 -08:00
2017-11-16 22:59:45 -08:00
// Used to create lifetime definitions from in-band lifetime usages.
// e.g. `fn foo(x: &'x u8) -> &'x u8` to `fn foo<'x>(x: &'x u8) -> &'x u8`
// When a named lifetime is encountered in a function or impl header and
// has not been defined
// (i.e. it doesn't appear in the in_scope_lifetimes list), it is added
// to this list. The results of this list are then added to the list of
// lifetime definitions in the corresponding impl or function generics.
lifetimes_to_define: Vec<(Span, Name)>,
// Whether or not in-band lifetimes are being collected. This is used to
// indicate whether or not we're in a place where new lifetimes will result
// in in-band lifetime definitions, such a function or an impl header.
// This will always be false unless the `in_band_lifetimes` feature is
// enabled.
is_collecting_in_band_lifetimes: bool,
// Currently in-scope lifetimes defined in impl headers, fn headers, or HRTB.
// When `is_collectin_in_band_lifetimes` is true, each lifetime is checked
// against this list to see if it is already in-scope, or if a definition
// needs to be created for it.
in_scope_lifetimes: Vec<Name>,
type_def_lifetime_params: DefIdMap<usize>,
current_hir_id_owner: Vec<(DefIndex, u32)>,
item_local_id_counters: NodeMap<u32>,
node_id_to_hir_id: IndexVec<NodeId, hir::HirId>,
}
pub trait Resolver {
/// Resolve a hir path generated by the lowerer when expanding `for`, `if let`, etc.
2016-11-25 06:07:21 +00:00
fn resolve_hir_path(&mut self, path: &mut hir::Path, is_value: bool);
/// Obtain the resolution for a node id
fn get_resolution(&mut self, id: NodeId) -> Option<PathResolution>;
/// We must keep the set of definitions up to date as we add nodes that weren't in the AST.
/// This should only return `None` during testing.
fn definitions(&mut self) -> &mut Definitions;
2015-09-25 16:03:28 +12:00
}
#[derive(Clone, Copy, Debug)]
enum ImplTraitContext {
/// Treat `impl Trait` as shorthand for a new universal generic parameter.
/// Example: `fn foo(x: impl Debug)`, where `impl Debug` is conceptually
/// equivalent to a fresh universal parameter like `fn foo<T: Debug>(x: T)`.
///
/// We store a DefId here so we can look up necessary information later
Universal(DefId),
/// Treat `impl Trait` as shorthand for a new universal existential parameter.
/// Example: `fn foo() -> impl Debug`, where `impl Debug` is conceptually
/// equivalent to a fresh existential parameter like `abstract type T; fn foo() -> T`.
Existential,
/// `impl Trait` is not accepted in this position.
Disallowed,
}
pub fn lower_crate(sess: &Session,
cstore: &CrateStore,
dep_graph: &DepGraph,
krate: &Crate,
resolver: &mut Resolver)
2016-05-10 05:29:13 +00:00
-> hir::Crate {
// We're constructing the HIR here; we don't care what we will
// read, since we haven't even constructed the *input* to
// incr. comp. yet.
let _ignore = dep_graph.in_ignore();
2016-05-10 05:29:13 +00:00
LoweringContext {
crate_root: std_inject::injected_crate_name(),
sess,
cstore,
2016-05-10 05:29:13 +00:00
parent_def: None,
resolver,
2017-03-24 23:03:15 +00:00
name_map: FxHashMap(),
items: BTreeMap::new(),
trait_items: BTreeMap::new(),
impl_items: BTreeMap::new(),
2017-02-21 10:55:40 -05:00
bodies: BTreeMap::new(),
trait_impls: BTreeMap::new(),
trait_auto_impl: BTreeMap::new(),
exported_macros: Vec::new(),
catch_scopes: Vec::new(),
2017-02-15 14:52:27 -08:00
loop_scopes: Vec::new(),
is_in_loop_condition: false,
type_def_lifetime_params: DefIdMap(),
current_hir_id_owner: vec![(CRATE_DEF_INDEX, 0)],
item_local_id_counters: NodeMap(),
node_id_to_hir_id: IndexVec::new(),
is_generator: false,
is_in_trait_impl: false,
2017-11-16 22:59:45 -08:00
lifetimes_to_define: Vec::new(),
is_collecting_in_band_lifetimes: false,
in_scope_lifetimes: Vec::new(),
2016-05-10 05:29:13 +00:00
}.lower_crate(krate)
}
2015-09-28 15:00:15 +13:00
#[derive(Copy, Clone, PartialEq, Eq)]
enum ParamMode {
/// Any path in a type context.
Explicit,
/// The `module::Type` in `module::Type::method` in an expression.
Optional
}
struct LoweredNodeId {
node_id: NodeId,
hir_id: hir::HirId,
}
enum ParenthesizedGenericArgs {
Ok,
Warn,
Err,
}
2016-05-10 05:29:13 +00:00
impl<'a> LoweringContext<'a> {
fn lower_crate(mut self, c: &Crate) -> hir::Crate {
/// Full-crate AST visitor that inserts into a fresh
/// `LoweringContext` any information that may be
/// needed from arbitrary locations in the crate.
/// E.g. The number of lifetime generic parameters
/// declared for every type and trait definition.
struct MiscCollector<'lcx, 'interner: 'lcx> {
lctx: &'lcx mut LoweringContext<'interner>,
}
impl<'lcx, 'interner> Visitor<'lcx> for MiscCollector<'lcx, 'interner> {
fn visit_item(&mut self, item: &'lcx Item) {
self.lctx.allocate_hir_id_counter(item.id, item);
match item.node {
ItemKind::Struct(_, ref generics) |
ItemKind::Union(_, ref generics) |
ItemKind::Enum(_, ref generics) |
ItemKind::Ty(_, ref generics) |
ItemKind::Trait(_, _, ref generics, ..) => {
let def_id = self.lctx.resolver.definitions().local_def_id(item.id);
let count = generics.lifetimes.len();
self.lctx.type_def_lifetime_params.insert(def_id, count);
}
_ => {}
}
visit::walk_item(self, item);
}
fn visit_trait_item(&mut self, item: &'lcx TraitItem) {
self.lctx.allocate_hir_id_counter(item.id, item);
visit::walk_trait_item(self, item);
}
fn visit_impl_item(&mut self, item: &'lcx ImplItem) {
self.lctx.allocate_hir_id_counter(item.id, item);
visit::walk_impl_item(self, item);
}
}
struct ItemLowerer<'lcx, 'interner: 'lcx> {
lctx: &'lcx mut LoweringContext<'interner>,
}
impl<'lcx, 'interner> ItemLowerer<'lcx, 'interner> {
fn with_trait_impl_ref<F>(&mut self, trait_impl_ref: &Option<TraitRef>, f: F)
where F: FnOnce(&mut Self)
{
let old = self.lctx.is_in_trait_impl;
self.lctx.is_in_trait_impl = if let &None = trait_impl_ref {
false
} else {
true
};
f(self);
self.lctx.is_in_trait_impl = old;
}
}
impl<'lcx, 'interner> Visitor<'lcx> for ItemLowerer<'lcx, 'interner> {
fn visit_item(&mut self, item: &'lcx Item) {
let mut item_lowered = true;
self.lctx.with_hir_id_owner(item.id, |lctx| {
if let Some(hir_item) = lctx.lower_item(item) {
lctx.items.insert(item.id, hir_item);
} else {
item_lowered = false;
}
});
if item_lowered {
2017-11-16 22:59:45 -08:00
let item_lifetimes = match self.lctx.items.get(&item.id).unwrap().node {
hir::Item_::ItemImpl(_,_,_,ref generics,..) |
hir::Item_::ItemTrait(_,_,ref generics,..) =>
generics.lifetimes.clone(),
_ => Vec::new().into(),
};
self.lctx.with_parent_impl_lifetime_defs(&item_lifetimes, |this| {
let this = &mut ItemLowerer { lctx: this };
if let ItemKind::Impl(_,_,_,_,ref opt_trait_ref,_,_) = item.node {
this.with_trait_impl_ref(opt_trait_ref, |this| {
visit::walk_item(this, item)
});
} else {
visit::walk_item(this, item);
}
});
}
}
fn visit_trait_item(&mut self, item: &'lcx TraitItem) {
self.lctx.with_hir_id_owner(item.id, |lctx| {
let id = hir::TraitItemId { node_id: item.id };
let hir_item = lctx.lower_trait_item(item);
lctx.trait_items.insert(id, hir_item);
});
visit::walk_trait_item(self, item);
}
fn visit_impl_item(&mut self, item: &'lcx ImplItem) {
self.lctx.with_hir_id_owner(item.id, |lctx| {
let id = hir::ImplItemId { node_id: item.id };
let hir_item = lctx.lower_impl_item(item);
lctx.impl_items.insert(id, hir_item);
});
visit::walk_impl_item(self, item);
}
}
self.lower_node_id(CRATE_NODE_ID);
debug_assert!(self.node_id_to_hir_id[CRATE_NODE_ID] == hir::CRATE_HIR_ID);
visit::walk_crate(&mut MiscCollector { lctx: &mut self }, c);
visit::walk_crate(&mut ItemLowerer { lctx: &mut self }, c);
let module = self.lower_mod(&c.module);
let attrs = self.lower_attrs(&c.attrs);
2017-02-21 12:23:47 -05:00
let body_ids = body_ids(&self.bodies);
self.resolver
.definitions()
.init_node_id_to_hir_id_mapping(self.node_id_to_hir_id);
hir::Crate {
module,
attrs,
span: c.span,
exported_macros: hir::HirVec::from(self.exported_macros),
items: self.items,
trait_items: self.trait_items,
impl_items: self.impl_items,
bodies: self.bodies,
body_ids,
trait_impls: self.trait_impls,
trait_auto_impl: self.trait_auto_impl,
}
}
fn allocate_hir_id_counter<T: Debug>(&mut self,
owner: NodeId,
debug: &T) {
if self.item_local_id_counters.insert(owner, 0).is_some() {
bug!("Tried to allocate item_local_id_counter for {:?} twice", debug);
}
// Always allocate the first HirId for the owner itself
self.lower_node_id_with_owner(owner, owner);
}
fn lower_node_id_generic<F>(&mut self,
ast_node_id: NodeId,
alloc_hir_id: F)
-> LoweredNodeId
where F: FnOnce(&mut Self) -> hir::HirId
{
if ast_node_id == DUMMY_NODE_ID {
return LoweredNodeId {
node_id: DUMMY_NODE_ID,
hir_id: hir::DUMMY_HIR_ID,
}
}
let min_size = ast_node_id.as_usize() + 1;
if min_size > self.node_id_to_hir_id.len() {
self.node_id_to_hir_id.resize(min_size, hir::DUMMY_HIR_ID);
}
let existing_hir_id = self.node_id_to_hir_id[ast_node_id];
if existing_hir_id == hir::DUMMY_HIR_ID {
// Generate a new HirId
let hir_id = alloc_hir_id(self);
self.node_id_to_hir_id[ast_node_id] = hir_id;
LoweredNodeId {
node_id: ast_node_id,
hir_id,
}
} else {
LoweredNodeId {
node_id: ast_node_id,
hir_id: existing_hir_id,
}
}
}
fn with_hir_id_owner<F>(&mut self, owner: NodeId, f: F)
where F: FnOnce(&mut Self)
{
let counter = self.item_local_id_counters
.insert(owner, HIR_ID_COUNTER_LOCKED)
.unwrap();
let def_index = self.resolver.definitions().opt_def_index(owner).unwrap();
self.current_hir_id_owner.push((def_index, counter));
f(self);
let (new_def_index, new_counter) = self.current_hir_id_owner.pop().unwrap();
debug_assert!(def_index == new_def_index);
debug_assert!(new_counter >= counter);
let prev = self.item_local_id_counters.insert(owner, new_counter).unwrap();
debug_assert!(prev == HIR_ID_COUNTER_LOCKED);
}
/// This method allocates a new HirId for the given NodeId and stores it in
/// the LoweringContext's NodeId => HirId map.
/// Take care not to call this method if the resulting HirId is then not
/// actually used in the HIR, as that would trigger an assertion in the
/// HirIdValidator later on, which makes sure that all NodeIds got mapped
/// properly. Calling the method twice with the same NodeId is fine though.
fn lower_node_id(&mut self, ast_node_id: NodeId) -> LoweredNodeId {
self.lower_node_id_generic(ast_node_id, |this| {
let &mut (def_index, ref mut local_id_counter) = this.current_hir_id_owner
.last_mut()
.unwrap();
let local_id = *local_id_counter;
*local_id_counter += 1;
hir::HirId {
owner: def_index,
local_id: hir::ItemLocalId(local_id),
}
})
}
fn lower_node_id_with_owner(&mut self,
ast_node_id: NodeId,
owner: NodeId)
-> LoweredNodeId {
self.lower_node_id_generic(ast_node_id, |this| {
let local_id_counter = this.item_local_id_counters
.get_mut(&owner)
.unwrap();
let local_id = *local_id_counter;
// We want to be sure not to modify the counter in the map while it
// is also on the stack. Otherwise we'll get lost updates when writing
// back from the stack to the map.
debug_assert!(local_id != HIR_ID_COUNTER_LOCKED);
*local_id_counter += 1;
let def_index = this.resolver.definitions().opt_def_index(owner).unwrap();
hir::HirId {
owner: def_index,
local_id: hir::ItemLocalId(local_id),
}
})
}
fn record_body(&mut self, value: hir::Expr, decl: Option<&FnDecl>)
-> hir::BodyId {
let body = hir::Body {
arguments: decl.map_or(hir_vec![], |decl| {
decl.inputs.iter().map(|x| self.lower_arg(x)).collect()
}),
is_generator: self.is_generator,
value,
};
let id = body.id();
self.bodies.insert(id, body);
id
}
fn next_id(&mut self) -> LoweredNodeId {
self.lower_node_id(self.sess.next_node_id())
}
fn expect_full_def(&mut self, id: NodeId) -> Def {
self.resolver.get_resolution(id).map_or(Def::Err, |pr| {
if pr.unresolved_segments() != 0 {
bug!("path not fully resolved: {:?}", pr);
}
pr.base_def()
})
}
fn diagnostic(&self) -> &errors::Handler {
self.sess.diagnostic()
2015-09-28 15:00:15 +13:00
}
2015-10-07 08:26:22 +13:00
2016-03-06 15:54:44 +03:00
fn str_to_ident(&self, s: &'static str) -> Name {
Symbol::gensym(s)
2015-10-07 08:26:22 +13:00
}
2017-07-31 23:04:34 +03:00
fn allow_internal_unstable(&self, reason: CompilerDesugaringKind, span: Span) -> Span
{
2017-03-22 08:39:51 +00:00
let mark = Mark::fresh(Mark::root());
2017-03-17 04:04:41 +00:00
mark.set_expn_info(codemap::ExpnInfo {
call_site: span,
callee: codemap::NameAndSpan {
2017-08-12 19:43:43 -05:00
format: codemap::CompilerDesugaring(reason),
span: Some(span),
allow_internal_unstable: true,
allow_internal_unsafe: false,
},
});
2017-07-31 23:04:34 +03:00
span.with_ctxt(SyntaxContext::empty().apply_mark(mark))
}
2017-11-16 22:59:45 -08:00
// Creates a new hir::LifetimeDef for every new lifetime encountered
// while evaluating `f`. Definitions are created with the parent provided.
// If no `parent_id` is provided, no definitions will be returned.
fn collect_in_band_lifetime_defs<T, F>(
&mut self,
parent_id: Option<DefId>,
f: F
) -> (Vec<hir::LifetimeDef>, T) where F: FnOnce(&mut LoweringContext) -> T
{
assert!(!self.is_collecting_in_band_lifetimes);
assert!(self.lifetimes_to_define.is_empty());
self.is_collecting_in_band_lifetimes = self.sess.features.borrow().in_band_lifetimes;
let res = f(self);
self.is_collecting_in_band_lifetimes = false;
let lifetimes_to_define = self.lifetimes_to_define.split_off(0);
let lifetime_defs = match parent_id {
Some(parent_id) => lifetimes_to_define.into_iter().map(|(span, name)| {
let def_node_id = self.next_id().node_id;
// Add a definition for the in-band lifetime def
self.resolver.definitions().create_def_with_parent(
parent_id.index,
def_node_id,
DefPathData::LifetimeDef(name.as_str()),
DefIndexAddressSpace::High,
Mark::root()
);
hir::LifetimeDef {
lifetime: hir::Lifetime {
id: def_node_id,
span,
name: hir::LifetimeName::Name(name),
},
bounds: Vec::new().into(),
pure_wrt_drop: false,
in_band: true,
}
}).collect(),
None => Vec::new(),
};
(lifetime_defs, res)
}
// Evaluates `f` with the lifetimes in `lt_defs` in-scope.
// This is used to track which lifetimes have already been defined, and
// which are new in-band lifetimes that need to have a definition created
// for them.
fn with_in_scope_lifetime_defs<T, F>(
&mut self,
lt_defs: &[LifetimeDef],
f: F
) -> T where F: FnOnce(&mut LoweringContext) -> T
{
let old_len = self.in_scope_lifetimes.len();
let lt_def_names = lt_defs.iter().map(|lt_def| lt_def.lifetime.ident.name);
self.in_scope_lifetimes.extend(lt_def_names);
let res = f(self);
self.in_scope_lifetimes.truncate(old_len);
res
}
// Same as the method above, but accepts `hir::LifetimeDef`s
// instead of `ast::LifetimeDef`s.
// This should only be used with generics that have already had their
// in-band lifetimes added. In practice, this means that this function is
// only used when lowering a child item of a trait or impl.
fn with_parent_impl_lifetime_defs<T, F>(
&mut self,
lt_defs: &[hir::LifetimeDef],
f: F
) -> T where F: FnOnce(&mut LoweringContext) -> T
{
let old_len = self.in_scope_lifetimes.len();
let lt_def_names = lt_defs.iter().map(|lt_def| lt_def.lifetime.name.name());
self.in_scope_lifetimes.extend(lt_def_names);
let res = f(self);
self.in_scope_lifetimes.truncate(old_len);
res
}
// Appends in-band lifetime defs to the existing set of out-of-band lifetime defs.
// Evaluates all within the context of the out-of-band defs.
// If provided, `impl_item_id` is used to find the parent impls of impl items so
// that their generics are not duplicated.
fn add_in_band_lifetime_defs<F, T>(
&mut self,
generics: &Generics,
parent_id: Option<DefId>,
f: F
) -> (hir::Generics, T)
where F: FnOnce(&mut LoweringContext) -> T
{
let (in_band_defs, (mut lowered_generics, res)) =
self.with_in_scope_lifetime_defs(&generics.lifetimes, |this| {
this.collect_in_band_lifetime_defs(parent_id, |this| {
(this.lower_generics(generics), f(this))
})
});
lowered_generics.lifetimes =
lowered_generics.lifetimes
.iter().cloned()
.chain(in_band_defs.into_iter())
.collect();
(lowered_generics, res)
}
fn with_catch_scope<T, F>(&mut self, catch_id: NodeId, f: F) -> T
where F: FnOnce(&mut LoweringContext) -> T
{
let len = self.catch_scopes.len();
self.catch_scopes.push(catch_id);
let result = f(self);
assert_eq!(len + 1, self.catch_scopes.len(),
"catch scopes should be added and removed in stack order");
self.catch_scopes.pop().unwrap();
result
}
2016-12-26 14:34:03 +01:00
fn lower_body<F>(&mut self, decl: Option<&FnDecl>, f: F) -> hir::BodyId
where F: FnOnce(&mut LoweringContext) -> hir::Expr
{
let prev = mem::replace(&mut self.is_generator, false);
2016-12-26 14:34:03 +01:00
let result = f(self);
let r = self.record_body(result, decl);
2017-07-11 12:57:05 -07:00
self.is_generator = prev;
return r
2016-12-26 14:34:03 +01:00
}
2017-02-15 14:52:27 -08:00
fn with_loop_scope<T, F>(&mut self, loop_id: NodeId, f: F) -> T
where F: FnOnce(&mut LoweringContext) -> T
{
// We're no longer in the base loop's condition; we're in another loop.
let was_in_loop_condition = self.is_in_loop_condition;
self.is_in_loop_condition = false;
2017-02-15 14:52:27 -08:00
let len = self.loop_scopes.len();
self.loop_scopes.push(loop_id);
2017-02-15 14:52:27 -08:00
let result = f(self);
assert_eq!(len + 1, self.loop_scopes.len(),
"Loop scopes should be added and removed in stack order");
2017-02-15 14:52:27 -08:00
self.loop_scopes.pop().unwrap();
self.is_in_loop_condition = was_in_loop_condition;
result
}
fn with_loop_condition_scope<T, F>(&mut self, f: F) -> T
where F: FnOnce(&mut LoweringContext) -> T
{
let was_in_loop_condition = self.is_in_loop_condition;
self.is_in_loop_condition = true;
let result = f(self);
self.is_in_loop_condition = was_in_loop_condition;
2017-02-15 14:52:27 -08:00
result
}
fn with_new_scopes<T, F>(&mut self, f: F) -> T
2017-02-15 14:52:27 -08:00
where F: FnOnce(&mut LoweringContext) -> T
{
let was_in_loop_condition = self.is_in_loop_condition;
self.is_in_loop_condition = false;
let catch_scopes = mem::replace(&mut self.catch_scopes, Vec::new());
2017-02-15 14:52:27 -08:00
let loop_scopes = mem::replace(&mut self.loop_scopes, Vec::new());
let result = f(self);
self.catch_scopes = catch_scopes;
self.loop_scopes = loop_scopes;
self.is_in_loop_condition = was_in_loop_condition;
2017-02-15 14:52:27 -08:00
result
}
fn with_parent_def<T, F>(&mut self, parent_id: NodeId, f: F) -> T
where F: FnOnce(&mut LoweringContext) -> T
{
let old_def = self.parent_def;
self.parent_def = {
let defs = self.resolver.definitions();
Some(defs.opt_def_index(parent_id).unwrap())
};
2016-05-06 08:24:04 +00:00
let result = f(self);
2016-04-18 10:30:55 +12:00
self.parent_def = old_def;
2016-04-18 10:30:55 +12:00
result
}
2015-07-31 00:04:06 -07:00
fn def_key(&mut self, id: DefId) -> DefKey {
if id.is_local() {
self.resolver.definitions().def_key(id.index)
} else {
self.cstore.def_key(id)
}
}
2017-03-24 23:03:15 +00:00
fn lower_ident(&mut self, ident: Ident) -> Name {
let ident = ident.modern();
if ident.ctxt == SyntaxContext::empty() {
return ident.name;
}
*self.name_map.entry(ident).or_insert_with(|| Symbol::from_ident(ident))
}
fn lower_opt_sp_ident(&mut self, o_id: Option<Spanned<Ident>>) -> Option<Spanned<Name>> {
o_id.map(|sp_ident| respan(sp_ident.span, sp_ident.node.name))
}
fn lower_loop_destination(&mut self, destination: Option<(NodeId, Spanned<Ident>)>)
2017-02-17 18:55:28 -08:00
-> hir::Destination
{
match destination {
Some((id, label_ident)) => {
let target = if let Def::Label(loop_id) = self.expect_full_def(id) {
hir::LoopIdResult::Ok(self.lower_node_id(loop_id).node_id)
} else {
hir::LoopIdResult::Err(hir::LoopIdError::UnresolvedLabel)
};
hir::Destination {
ident: Some(label_ident),
target_id: hir::ScopeTarget::Loop(target),
}
2017-02-15 14:52:27 -08:00
},
None => {
let loop_id = self.loop_scopes
.last()
.map(|innermost_loop_id| *innermost_loop_id);
hir::Destination {
ident: None,
target_id: hir::ScopeTarget::Loop(
loop_id.map(|id| Ok(self.lower_node_id(id).node_id))
.unwrap_or(Err(hir::LoopIdError::OutsideLoopScope))
.into())
}
}
2017-02-15 14:52:27 -08:00
}
}
fn lower_attrs(&mut self, attrs: &Vec<Attribute>) -> hir::HirVec<Attribute> {
rustc: Forbid interpolated tokens in the HIR Right now the HIR contains raw `syntax::ast::Attribute` structure but nowadays these can contain arbitrary tokens. One variant of the `Token` enum is an "interpolated" token which basically means to shove all the tokens for a nonterminal in this position. A "nonterminal" in this case is roughly analagous to a macro argument: macro_rules! foo { ($a:expr) => { // $a is a nonterminal as an expression } } Currently nonterminals contain namely items and expressions, and this poses a problem for incremental compilation! With incremental we want a stable hash of all HIR items, but this means we may transitively need a stable hash *of the entire AST*, which is certainly not stable w/ node ids and whatnot. Hence today there's a "bug" where the "stable hash" of an AST is just the raw hash value of the AST, and this only arises with interpolated nonterminals. The downside of this approach, however, is that a bunch of errors get spewed out during compilation about how this isn't a great idea. This PR is focused at fixing these warnings, basically deleting them from the compiler. The implementation here is to alter attributes as they're lowered from the AST to HIR, expanding all nonterminals in-place as we see them. This code for expanding a nonterminal to a token stream already exists for the `proc_macro` crate, so we basically just reuse the same implementation there. After this PR it's considered a bug to have an `Interpolated` token and hence the stable hash implementation simply uses `bug!` in this location. Closes #40946
2017-09-15 08:28:34 -07:00
attrs.iter().map(|a| self.lower_attr(a)).collect::<Vec<_>>().into()
}
fn lower_attr(&mut self, attr: &Attribute) -> Attribute {
Attribute {
id: attr.id,
style: attr.style,
path: attr.path.clone(),
tokens: self.lower_token_stream(attr.tokens.clone()),
is_sugared_doc: attr.is_sugared_doc,
span: attr.span,
}
}
fn lower_token_stream(&mut self, tokens: TokenStream) -> TokenStream {
tokens.into_trees()
.flat_map(|tree| self.lower_token_tree(tree).into_trees())
.collect()
rustc: Forbid interpolated tokens in the HIR Right now the HIR contains raw `syntax::ast::Attribute` structure but nowadays these can contain arbitrary tokens. One variant of the `Token` enum is an "interpolated" token which basically means to shove all the tokens for a nonterminal in this position. A "nonterminal" in this case is roughly analagous to a macro argument: macro_rules! foo { ($a:expr) => { // $a is a nonterminal as an expression } } Currently nonterminals contain namely items and expressions, and this poses a problem for incremental compilation! With incremental we want a stable hash of all HIR items, but this means we may transitively need a stable hash *of the entire AST*, which is certainly not stable w/ node ids and whatnot. Hence today there's a "bug" where the "stable hash" of an AST is just the raw hash value of the AST, and this only arises with interpolated nonterminals. The downside of this approach, however, is that a bunch of errors get spewed out during compilation about how this isn't a great idea. This PR is focused at fixing these warnings, basically deleting them from the compiler. The implementation here is to alter attributes as they're lowered from the AST to HIR, expanding all nonterminals in-place as we see them. This code for expanding a nonterminal to a token stream already exists for the `proc_macro` crate, so we basically just reuse the same implementation there. After this PR it's considered a bug to have an `Interpolated` token and hence the stable hash implementation simply uses `bug!` in this location. Closes #40946
2017-09-15 08:28:34 -07:00
}
fn lower_token_tree(&mut self, tree: TokenTree) -> TokenStream {
rustc: Forbid interpolated tokens in the HIR Right now the HIR contains raw `syntax::ast::Attribute` structure but nowadays these can contain arbitrary tokens. One variant of the `Token` enum is an "interpolated" token which basically means to shove all the tokens for a nonterminal in this position. A "nonterminal" in this case is roughly analagous to a macro argument: macro_rules! foo { ($a:expr) => { // $a is a nonterminal as an expression } } Currently nonterminals contain namely items and expressions, and this poses a problem for incremental compilation! With incremental we want a stable hash of all HIR items, but this means we may transitively need a stable hash *of the entire AST*, which is certainly not stable w/ node ids and whatnot. Hence today there's a "bug" where the "stable hash" of an AST is just the raw hash value of the AST, and this only arises with interpolated nonterminals. The downside of this approach, however, is that a bunch of errors get spewed out during compilation about how this isn't a great idea. This PR is focused at fixing these warnings, basically deleting them from the compiler. The implementation here is to alter attributes as they're lowered from the AST to HIR, expanding all nonterminals in-place as we see them. This code for expanding a nonterminal to a token stream already exists for the `proc_macro` crate, so we basically just reuse the same implementation there. After this PR it's considered a bug to have an `Interpolated` token and hence the stable hash implementation simply uses `bug!` in this location. Closes #40946
2017-09-15 08:28:34 -07:00
match tree {
TokenTree::Token(span, token) => {
self.lower_token(token, span)
}
TokenTree::Delimited(span, delimited) => {
TokenTree::Delimited(span, Delimited {
delim: delimited.delim,
tts: self.lower_token_stream(delimited.tts.into()).into(),
}).into()
rustc: Forbid interpolated tokens in the HIR Right now the HIR contains raw `syntax::ast::Attribute` structure but nowadays these can contain arbitrary tokens. One variant of the `Token` enum is an "interpolated" token which basically means to shove all the tokens for a nonterminal in this position. A "nonterminal" in this case is roughly analagous to a macro argument: macro_rules! foo { ($a:expr) => { // $a is a nonterminal as an expression } } Currently nonterminals contain namely items and expressions, and this poses a problem for incremental compilation! With incremental we want a stable hash of all HIR items, but this means we may transitively need a stable hash *of the entire AST*, which is certainly not stable w/ node ids and whatnot. Hence today there's a "bug" where the "stable hash" of an AST is just the raw hash value of the AST, and this only arises with interpolated nonterminals. The downside of this approach, however, is that a bunch of errors get spewed out during compilation about how this isn't a great idea. This PR is focused at fixing these warnings, basically deleting them from the compiler. The implementation here is to alter attributes as they're lowered from the AST to HIR, expanding all nonterminals in-place as we see them. This code for expanding a nonterminal to a token stream already exists for the `proc_macro` crate, so we basically just reuse the same implementation there. After this PR it's considered a bug to have an `Interpolated` token and hence the stable hash implementation simply uses `bug!` in this location. Closes #40946
2017-09-15 08:28:34 -07:00
}
}
}
fn lower_token(&mut self, token: Token, span: Span) -> TokenStream {
rustc: Forbid interpolated tokens in the HIR Right now the HIR contains raw `syntax::ast::Attribute` structure but nowadays these can contain arbitrary tokens. One variant of the `Token` enum is an "interpolated" token which basically means to shove all the tokens for a nonterminal in this position. A "nonterminal" in this case is roughly analagous to a macro argument: macro_rules! foo { ($a:expr) => { // $a is a nonterminal as an expression } } Currently nonterminals contain namely items and expressions, and this poses a problem for incremental compilation! With incremental we want a stable hash of all HIR items, but this means we may transitively need a stable hash *of the entire AST*, which is certainly not stable w/ node ids and whatnot. Hence today there's a "bug" where the "stable hash" of an AST is just the raw hash value of the AST, and this only arises with interpolated nonterminals. The downside of this approach, however, is that a bunch of errors get spewed out during compilation about how this isn't a great idea. This PR is focused at fixing these warnings, basically deleting them from the compiler. The implementation here is to alter attributes as they're lowered from the AST to HIR, expanding all nonterminals in-place as we see them. This code for expanding a nonterminal to a token stream already exists for the `proc_macro` crate, so we basically just reuse the same implementation there. After this PR it's considered a bug to have an `Interpolated` token and hence the stable hash implementation simply uses `bug!` in this location. Closes #40946
2017-09-15 08:28:34 -07:00
match token {
Token::Interpolated(_) => {}
other => return TokenTree::Token(span, other).into(),
rustc: Forbid interpolated tokens in the HIR Right now the HIR contains raw `syntax::ast::Attribute` structure but nowadays these can contain arbitrary tokens. One variant of the `Token` enum is an "interpolated" token which basically means to shove all the tokens for a nonterminal in this position. A "nonterminal" in this case is roughly analagous to a macro argument: macro_rules! foo { ($a:expr) => { // $a is a nonterminal as an expression } } Currently nonterminals contain namely items and expressions, and this poses a problem for incremental compilation! With incremental we want a stable hash of all HIR items, but this means we may transitively need a stable hash *of the entire AST*, which is certainly not stable w/ node ids and whatnot. Hence today there's a "bug" where the "stable hash" of an AST is just the raw hash value of the AST, and this only arises with interpolated nonterminals. The downside of this approach, however, is that a bunch of errors get spewed out during compilation about how this isn't a great idea. This PR is focused at fixing these warnings, basically deleting them from the compiler. The implementation here is to alter attributes as they're lowered from the AST to HIR, expanding all nonterminals in-place as we see them. This code for expanding a nonterminal to a token stream already exists for the `proc_macro` crate, so we basically just reuse the same implementation there. After this PR it's considered a bug to have an `Interpolated` token and hence the stable hash implementation simply uses `bug!` in this location. Closes #40946
2017-09-15 08:28:34 -07:00
}
let tts = token.interpolated_to_tokenstream(&self.sess.parse_sess, span);
self.lower_token_stream(tts)
}
2015-07-31 00:04:06 -07:00
fn lower_arm(&mut self, arm: &Arm) -> hir::Arm {
hir::Arm {
attrs: self.lower_attrs(&arm.attrs),
pats: arm.pats.iter().map(|x| self.lower_pat(x)).collect(),
guard: arm.guard.as_ref().map(|ref x| P(self.lower_expr(x))),
body: P(self.lower_expr(&arm.body)),
}
2015-07-31 00:04:06 -07:00
}
fn lower_ty_binding(&mut self, b: &TypeBinding, itctx: ImplTraitContext) -> hir::TypeBinding {
hir::TypeBinding {
id: self.lower_node_id(b.id).node_id,
2017-03-24 23:03:15 +00:00
name: self.lower_ident(b.ident),
ty: self.lower_ty(&b.ty, itctx),
span: b.span,
}
2015-07-31 00:04:06 -07:00
}
fn lower_ty(&mut self, t: &Ty, itctx: ImplTraitContext) -> P<hir::Ty> {
let kind = match t.node {
TyKind::Infer => hir::TyInfer,
TyKind::Err => hir::TyErr,
TyKind::Slice(ref ty) => hir::TySlice(self.lower_ty(ty, itctx)),
TyKind::Ptr(ref mt) => hir::TyPtr(self.lower_mt(mt, itctx)),
TyKind::Rptr(ref region, ref mt) => {
2017-07-31 23:04:34 +03:00
let span = t.span.with_hi(t.span.lo());
let lifetime = match *region {
Some(ref lt) => self.lower_lifetime(lt),
None => self.elided_lifetime(span)
};
hir::TyRptr(lifetime, self.lower_mt(mt, itctx))
}
TyKind::BareFn(ref f) => {
2017-11-16 22:59:45 -08:00
self.with_in_scope_lifetime_defs(&f.lifetimes, |this|
hir::TyBareFn(P(hir::BareFnTy {
lifetimes: this.lower_lifetime_defs(&f.lifetimes),
unsafety: this.lower_unsafety(f.unsafety),
abi: f.abi,
decl: this.lower_fn_decl(&f.decl, None, false),
arg_names: this.lower_fn_args_to_names(&f.decl),
})))
}
TyKind::Never => hir::TyNever,
TyKind::Tup(ref tys) => {
hir::TyTup(tys.iter().map(|ty| self.lower_ty(ty, itctx)).collect())
}
TyKind::Paren(ref ty) => {
return self.lower_ty(ty, itctx);
}
TyKind::Path(ref qself, ref path) => {
2017-09-13 18:29:59 +03:00
let id = self.lower_node_id(t.id);
let qpath = self.lower_qpath(t.id, qself, path, ParamMode::Explicit, itctx);
return self.ty_path(id, t.span, qpath);
}
TyKind::ImplicitSelf => {
hir::TyPath(hir::QPath::Resolved(None, P(hir::Path {
def: self.expect_full_def(t.id),
segments: hir_vec![
hir::PathSegment::from_name(keywords::SelfType.name())
],
span: t.span,
})))
}
TyKind::Array(ref ty, ref length) => {
let length = self.lower_body(None, |this| this.lower_expr(length));
hir::TyArray(self.lower_ty(ty, itctx), length)
}
TyKind::Typeof(ref expr) => {
let expr = self.lower_body(None, |this| this.lower_expr(expr));
hir::TyTypeof(expr)
}
2017-10-10 17:33:19 +03:00
TyKind::TraitObject(ref bounds, ..) => {
let mut lifetime_bound = None;
let bounds = bounds.iter().filter_map(|bound| {
match *bound {
TraitTyParamBound(ref ty, TraitBoundModifier::None) => {
Some(self.lower_poly_trait_ref(ty, itctx))
}
TraitTyParamBound(_, TraitBoundModifier::Maybe) => None,
RegionTyParamBound(ref lifetime) => {
if lifetime_bound.is_none() {
lifetime_bound = Some(self.lower_lifetime(lifetime));
}
None
}
}
}).collect();
let lifetime_bound = lifetime_bound.unwrap_or_else(|| {
self.elided_lifetime(t.span)
});
hir::TyTraitObject(bounds, lifetime_bound)
}
TyKind::ImplTrait(ref bounds) => {
use syntax::feature_gate::{emit_feature_err, GateIssue};
match itctx {
ImplTraitContext::Existential => {
let has_feature = self.sess.features.borrow().conservative_impl_trait;
if !t.span.allows_unstable() && !has_feature {
emit_feature_err(&self.sess.parse_sess, "conservative_impl_trait",
t.span, GateIssue::Language,
"`impl Trait` in return position is experimental");
}
let def_index = self.resolver.definitions().opt_def_index(t.id).unwrap();
let hir_bounds = self.lower_bounds(bounds, itctx);
let (lifetimes, lifetime_defs) =
self.lifetimes_from_impl_trait_bounds(def_index, &hir_bounds);
hir::TyImplTraitExistential(hir::ExistTy {
generics: hir::Generics {
lifetimes: lifetime_defs,
// Type parameters are taken from environment:
ty_params: Vec::new().into(),
where_clause: hir::WhereClause {
id: self.next_id().node_id,
predicates: Vec::new().into(),
},
span: t.span,
},
bounds: hir_bounds,
}, lifetimes)
},
ImplTraitContext::Universal(def_id) => {
let has_feature = self.sess.features.borrow().universal_impl_trait;
if !t.span.allows_unstable() && !has_feature {
emit_feature_err(&self.sess.parse_sess, "universal_impl_trait",
t.span, GateIssue::Language,
"`impl Trait` in argument position is experimental");
}
hir::TyImplTraitUniversal(def_id, self.lower_bounds(bounds, itctx))
},
ImplTraitContext::Disallowed => {
span_err!(self.sess, t.span, E0562,
"`impl Trait` not allowed outside of function \
and inherent method return types");
hir::TyErr
}
}
}
TyKind::Mac(_) => panic!("TyMac should have been expanded by now."),
};
2017-09-13 18:29:59 +03:00
let LoweredNodeId { node_id, hir_id } = self.lower_node_id(t.id);
P(hir::Ty {
2017-09-13 18:29:59 +03:00
id: node_id,
node: kind,
span: t.span,
2017-09-13 18:29:59 +03:00
hir_id,
})
}
2015-07-31 00:04:06 -07:00
fn lifetimes_from_impl_trait_bounds(
&mut self,
parent_index: DefIndex,
bounds: &hir::TyParamBounds
) -> (HirVec<hir::Lifetime>, HirVec<hir::LifetimeDef>) {
// This visitor walks over impl trait bounds and creates defs for all lifetimes which
// appear in the bounds, excluding lifetimes that are created within the bounds.
// e.g. 'a, 'b, but not 'c in `impl for<'c> SomeTrait<'a, 'b, 'c>`
struct ImplTraitLifetimeCollector<'r, 'a: 'r> {
context: &'r mut LoweringContext<'a>,
parent: DefIndex,
2017-12-09 21:54:58 -08:00
collect_elided_lifetimes: bool,
currently_bound_lifetimes: Vec<hir::LifetimeName>,
already_defined_lifetimes: HashSet<hir::LifetimeName>,
output_lifetimes: Vec<hir::Lifetime>,
output_lifetime_defs: Vec<hir::LifetimeDef>,
}
impl<'r, 'a: 'r, 'v> hir::intravisit::Visitor<'v> for ImplTraitLifetimeCollector<'r, 'a> {
fn nested_visit_map<'this>(&'this mut self)
-> hir::intravisit::NestedVisitorMap<'this, 'v> {
hir::intravisit::NestedVisitorMap::None
}
2017-12-09 21:54:58 -08:00
fn visit_path_parameters(&mut self, span: Span, parameters: &'v hir::PathParameters) {
// Don't collect elided lifetimes used inside of `Fn()` syntax.
if parameters.parenthesized {
let old_collect_elided_lifetimes = self.collect_elided_lifetimes;
self.collect_elided_lifetimes = false;
hir::intravisit::walk_path_parameters(self, span, parameters);
self.collect_elided_lifetimes = old_collect_elided_lifetimes;
} else {
hir::intravisit::walk_path_parameters(self, span, parameters);
}
}
fn visit_ty(&mut self, t: &'v hir::Ty) {
// Don't collect elided lifetimes used inside of `fn()` syntax
if let &hir::Ty_::TyBareFn(_) = &t.node {
let old_collect_elided_lifetimes = self.collect_elided_lifetimes;
self.collect_elided_lifetimes = false;
hir::intravisit::walk_ty(self, t);
self.collect_elided_lifetimes = old_collect_elided_lifetimes;
} else {
hir::intravisit::walk_ty(self, t);
}
}
fn visit_poly_trait_ref(&mut self,
polytr: &'v hir::PolyTraitRef,
_: hir::TraitBoundModifier) {
let old_len = self.currently_bound_lifetimes.len();
// Record the introduction of 'a in `for<'a> ...`
for lt_def in &polytr.bound_lifetimes {
// Introduce lifetimes one at a time so that we can handle
2017-12-09 21:54:58 -08:00
// cases like `fn foo<'d>() -> impl for<'a, 'b: 'a, 'c: 'b + 'd>`
self.currently_bound_lifetimes.push(lt_def.lifetime.name);
// Visit the lifetime bounds
for lt_bound in &lt_def.bounds {
self.visit_lifetime(&lt_bound);
}
}
hir::intravisit::walk_trait_ref(self, &polytr.trait_ref);
self.currently_bound_lifetimes.truncate(old_len);
}
fn visit_lifetime(&mut self, lifetime: &'v hir::Lifetime) {
2017-12-09 21:54:58 -08:00
let name = match lifetime.name {
hir::LifetimeName::Implicit |
hir::LifetimeName::Underscore =>
if self.collect_elided_lifetimes {
// Use `'_` for both implicit and underscore lifetimes in
// `abstract type Foo<'_>: SomeTrait<'_>;`
hir::LifetimeName::Underscore
} else {
return
}
name @ hir::LifetimeName::Name(_) => name,
hir::LifetimeName::Static => return,
};
if !self.currently_bound_lifetimes.contains(&name) &&
!self.already_defined_lifetimes.contains(&name)
{
self.already_defined_lifetimes.insert(name);
self.output_lifetimes.push(hir::Lifetime {
id: self.context.next_id().node_id,
span: lifetime.span,
name,
});
let def_node_id = self.context.next_id().node_id;
self.context.resolver.definitions().create_def_with_parent(
self.parent,
def_node_id,
DefPathData::LifetimeDef(name.name().as_str()),
DefIndexAddressSpace::High,
Mark::root()
);
let def_lifetime = hir::Lifetime {
id: def_node_id,
span: lifetime.span,
name: name,
};
self.output_lifetime_defs.push(hir::LifetimeDef {
lifetime: def_lifetime,
bounds: Vec::new().into(),
pure_wrt_drop: false,
in_band: false,
});
}
}
}
let mut lifetime_collector = ImplTraitLifetimeCollector {
context: self,
parent: parent_index,
2017-12-09 21:54:58 -08:00
collect_elided_lifetimes: true,
currently_bound_lifetimes: Vec::new(),
already_defined_lifetimes: HashSet::new(),
output_lifetimes: Vec::new(),
output_lifetime_defs: Vec::new(),
};
for bound in bounds {
hir::intravisit::walk_ty_param_bound(&mut lifetime_collector, &bound);
}
(
lifetime_collector.output_lifetimes.into(),
lifetime_collector.output_lifetime_defs.into()
)
}
fn lower_foreign_mod(&mut self, fm: &ForeignMod) -> hir::ForeignMod {
hir::ForeignMod {
abi: fm.abi,
items: fm.items.iter().map(|x| self.lower_foreign_item(x)).collect(),
}
2015-07-31 00:04:06 -07:00
}
2017-03-15 21:27:40 -05:00
fn lower_global_asm(&mut self, ga: &GlobalAsm) -> P<hir::GlobalAsm> {
P(hir::GlobalAsm {
asm: ga.asm,
ctxt: ga.ctxt,
})
}
fn lower_variant(&mut self, v: &Variant) -> hir::Variant {
Spanned {
node: hir::Variant_ {
name: v.node.name.name,
attrs: self.lower_attrs(&v.node.attrs),
data: self.lower_variant_data(&v.node.data),
disr_expr: v.node.disr_expr.as_ref().map(|e| {
2016-12-26 14:34:03 +01:00
self.lower_body(None, |this| this.lower_expr(e))
}),
},
span: v.span,
}
2015-07-31 00:04:06 -07:00
}
fn lower_qpath(&mut self,
id: NodeId,
qself: &Option<QSelf>,
p: &Path,
param_mode: ParamMode,
itctx: ImplTraitContext)
-> hir::QPath {
let qself_position = qself.as_ref().map(|q| q.position);
let qself = qself.as_ref().map(|q| self.lower_ty(&q.ty, itctx));
let resolution = self.resolver.get_resolution(id)
.unwrap_or(PathResolution::new(Def::Err));
let proj_start = p.segments.len() - resolution.unresolved_segments();
let path = P(hir::Path {
def: resolution.base_def(),
segments: p.segments[..proj_start].iter().enumerate().map(|(i, segment)| {
let param_mode = match (qself_position, param_mode) {
(Some(j), ParamMode::Optional) if i < j => {
// This segment is part of the trait path in a
// qualified path - one of `a`, `b` or `Trait`
// in `<X as a::b::Trait>::T::U::method`.
ParamMode::Explicit
}
_ => param_mode
};
// Figure out if this is a type/trait segment,
// which may need lifetime elision performed.
let parent_def_id = |this: &mut Self, def_id: DefId| {
DefId {
krate: def_id.krate,
index: this.def_key(def_id).parent.expect("missing parent")
}
};
let type_def_id = match resolution.base_def() {
Def::AssociatedTy(def_id) if i + 2 == proj_start => {
Some(parent_def_id(self, def_id))
}
Def::Variant(def_id) if i + 1 == proj_start => {
Some(parent_def_id(self, def_id))
}
Def::Struct(def_id) |
Def::Union(def_id) |
Def::Enum(def_id) |
Def::TyAlias(def_id) |
Def::Trait(def_id) if i + 1 == proj_start => Some(def_id),
_ => None
};
let parenthesized_generic_args = match resolution.base_def() {
// `a::b::Trait(Args)`
Def::Trait(..) if i + 1 == proj_start => ParenthesizedGenericArgs::Ok,
// `a::b::Trait(Args)::TraitItem`
Def::Method(..) |
Def::AssociatedConst(..) |
Def::AssociatedTy(..) if i + 2 == proj_start => ParenthesizedGenericArgs::Ok,
// Avoid duplicated errors
Def::Err => ParenthesizedGenericArgs::Ok,
// An error
Def::Struct(..) | Def::Enum(..) | Def::Union(..) | Def::TyAlias(..) |
Def::Variant(..) if i + 1 == proj_start => ParenthesizedGenericArgs::Err,
// A warning for now, for compatibility reasons
_ => ParenthesizedGenericArgs::Warn,
};
let num_lifetimes = type_def_id.map_or(0, |def_id| {
if let Some(&n) = self.type_def_lifetime_params.get(&def_id) {
return n;
}
assert!(!def_id.is_local());
let n = self.cstore
.item_generics_cloned_untracked(def_id, self.sess)
.regions
.len();
self.type_def_lifetime_params.insert(def_id, n);
n
});
self.lower_path_segment(p.span, segment, param_mode, num_lifetimes,
parenthesized_generic_args, itctx)
}).collect(),
span: p.span,
});
// Simple case, either no projections, or only fully-qualified.
// E.g. `std::mem::size_of` or `<I as Iterator>::Item`.
if resolution.unresolved_segments() == 0 {
return hir::QPath::Resolved(qself, path);
}
// Create the innermost type that we're projecting from.
let mut ty = if path.segments.is_empty() {
// If the base path is empty that means there exists a
// syntactical `Self`, e.g. `&i32` in `<&i32>::clone`.
qself.expect("missing QSelf for <T>::...")
} else {
// Otherwise, the base path is an implicit `Self` type path,
// e.g. `Vec` in `Vec::new` or `<I as Iterator>::Item` in
// `<I as Iterator>::Item::default`.
2017-09-13 18:29:59 +03:00
let new_id = self.next_id();
self.ty_path(new_id, p.span, hir::QPath::Resolved(qself, path))
};
// Anything after the base path are associated "extensions",
// out of which all but the last one are associated types,
// e.g. for `std::vec::Vec::<T>::IntoIter::Item::clone`:
// * base path is `std::vec::Vec<T>`
// * "extensions" are `IntoIter`, `Item` and `clone`
// * type nodes are:
// 1. `std::vec::Vec<T>` (created above)
// 2. `<std::vec::Vec<T>>::IntoIter`
// 3. `<<std::vec::Vec<T>>::IntoIter>::Item`
// * final path is `<<<std::vec::Vec<T>>::IntoIter>::Item>::clone`
for (i, segment) in p.segments.iter().enumerate().skip(proj_start) {
let segment = P(self.lower_path_segment(p.span, segment, param_mode, 0,
ParenthesizedGenericArgs::Warn,
itctx));
let qpath = hir::QPath::TypeRelative(ty, segment);
// It's finished, return the extension of the right node type.
if i == p.segments.len() - 1 {
return qpath;
}
// Wrap the associated extension in another type node.
2017-09-13 18:29:59 +03:00
let new_id = self.next_id();
ty = self.ty_path(new_id, p.span, qpath);
}
// Should've returned in the for loop above.
span_bug!(p.span, "lower_qpath: no final extension segment in {}..{}",
proj_start, p.segments.len())
2015-07-31 00:04:06 -07:00
}
fn lower_path_extra(&mut self,
id: NodeId,
p: &Path,
name: Option<Name>,
param_mode: ParamMode,
defaults_to_global: bool)
-> hir::Path {
let mut segments = p.segments.iter();
if defaults_to_global && p.is_global() {
segments.next();
}
hir::Path {
def: self.expect_full_def(id),
segments: segments.map(|segment| {
self.lower_path_segment(p.span, segment, param_mode, 0,
ParenthesizedGenericArgs::Err,
ImplTraitContext::Disallowed)
}).chain(name.map(|name| hir::PathSegment::from_name(name)))
.collect(),
span: p.span,
}
}
fn lower_path(&mut self,
id: NodeId,
p: &Path,
param_mode: ParamMode,
defaults_to_global: bool)
-> hir::Path {
self.lower_path_extra(id, p, None, param_mode, defaults_to_global)
}
fn lower_path_segment(&mut self,
path_span: Span,
segment: &PathSegment,
param_mode: ParamMode,
expected_lifetimes: usize,
parenthesized_generic_args: ParenthesizedGenericArgs,
itctx: ImplTraitContext)
-> hir::PathSegment {
let (mut parameters, infer_types) = if let Some(ref parameters) = segment.parameters {
let msg = "parenthesized parameters may only be used with a trait";
2016-12-10 06:45:58 +00:00
match **parameters {
PathParameters::AngleBracketed(ref data) => {
self.lower_angle_bracketed_parameter_data(data, param_mode, itctx)
2016-12-10 06:45:58 +00:00
}
PathParameters::Parenthesized(ref data) => match parenthesized_generic_args {
ParenthesizedGenericArgs::Ok =>
self.lower_parenthesized_parameter_data(data),
ParenthesizedGenericArgs::Warn => {
self.sess.buffer_lint(PARENTHESIZED_PARAMS_IN_TYPES_AND_MODULES,
CRATE_NODE_ID, data.span, msg.into());
(hir::PathParameters::none(), true)
}
ParenthesizedGenericArgs::Err => {
struct_span_err!(self.sess, data.span, E0214, "{}", msg)
.span_label(data.span, "only traits may use parentheses").emit();
(hir::PathParameters::none(), true)
}
2016-12-10 06:45:58 +00:00
}
}
2016-12-10 06:45:58 +00:00
} else {
self.lower_angle_bracketed_parameter_data(&Default::default(), param_mode, itctx)
};
if !parameters.parenthesized && parameters.lifetimes.is_empty() {
parameters.lifetimes = (0..expected_lifetimes).map(|_| {
self.elided_lifetime(path_span)
}).collect();
}
hir::PathSegment::new(
self.lower_ident(segment.identifier),
parameters,
infer_types
)
2015-07-31 00:04:06 -07:00
}
fn lower_angle_bracketed_parameter_data(&mut self,
data: &AngleBracketedParameterData,
param_mode: ParamMode,
itctx: ImplTraitContext)
-> (hir::PathParameters, bool) {
let &AngleBracketedParameterData { ref lifetimes, ref types, ref bindings, .. } = data;
(hir::PathParameters {
lifetimes: self.lower_lifetimes(lifetimes),
types: types.iter().map(|ty| self.lower_ty(ty, itctx)).collect(),
bindings: bindings.iter().map(|b| self.lower_ty_binding(b, itctx)).collect(),
parenthesized: false,
}, types.is_empty() && param_mode == ParamMode::Optional)
2015-07-31 00:04:06 -07:00
}
fn lower_parenthesized_parameter_data(&mut self,
data: &ParenthesizedParameterData)
-> (hir::PathParameters, bool) {
const DISALLOWED: ImplTraitContext = ImplTraitContext::Disallowed;
let &ParenthesizedParameterData { ref inputs, ref output, span } = data;
let inputs = inputs.iter().map(|ty| self.lower_ty(ty, DISALLOWED)).collect();
let mk_tup = |this: &mut Self, tys, span| {
2017-09-13 18:29:59 +03:00
let LoweredNodeId { node_id, hir_id } = this.next_id();
P(hir::Ty { node: hir::TyTup(tys), id: node_id, hir_id, span })
};
(hir::PathParameters {
lifetimes: hir::HirVec::new(),
types: hir_vec![mk_tup(self, inputs, span)],
bindings: hir_vec![hir::TypeBinding {
id: self.next_id().node_id,
name: Symbol::intern(FN_OUTPUT_NAME),
ty: output.as_ref().map(|ty| self.lower_ty(&ty, DISALLOWED))
.unwrap_or_else(|| mk_tup(self, hir::HirVec::new(), span)),
span: output.as_ref().map_or(span, |ty| ty.span),
}],
parenthesized: true,
}, false)
2015-10-06 16:03:56 +13:00
}
2015-07-31 00:04:06 -07:00
fn lower_local(&mut self, l: &Local) -> P<hir::Local> {
let LoweredNodeId { node_id, hir_id } = self.lower_node_id(l.id);
P(hir::Local {
id: node_id,
hir_id,
ty: l.ty.as_ref().map(|t| self.lower_ty(t, ImplTraitContext::Disallowed)),
pat: self.lower_pat(&l.pat),
init: l.init.as_ref().map(|e| P(self.lower_expr(e))),
span: l.span,
attrs: l.attrs.clone(),
source: hir::LocalSource::Normal,
})
2015-10-06 16:03:56 +13:00
}
2015-07-31 00:04:06 -07:00
fn lower_mutability(&mut self, m: Mutability) -> hir::Mutability {
match m {
Mutability::Mutable => hir::MutMutable,
Mutability::Immutable => hir::MutImmutable,
}
2015-07-31 00:04:06 -07:00
}
fn lower_arg(&mut self, arg: &Arg) -> hir::Arg {
let LoweredNodeId { node_id, hir_id } = self.lower_node_id(arg.id);
hir::Arg {
id: node_id,
hir_id,
pat: self.lower_pat(&arg.pat),
}
2015-09-28 08:23:31 +13:00
}
2015-07-31 00:04:06 -07:00
fn lower_fn_args_to_names(&mut self, decl: &FnDecl)
-> hir::HirVec<Spanned<Name>> {
decl.inputs.iter().map(|arg| {
match arg.pat.node {
PatKind::Ident(_, ident, None) => {
respan(ident.span, ident.node.name)
}
_ => respan(arg.pat.span, keywords::Invalid.name()),
}
}).collect()
}
fn lower_fn_decl(&mut self,
decl: &FnDecl,
fn_def_id: Option<DefId>,
impl_trait_return_allow: bool)
-> P<hir::FnDecl> {
// NOTE: The two last paramters here have to do with impl Trait. If fn_def_id is Some,
// then impl Trait arguments are lowered into generic paramters on the given
// fn_def_id, otherwise impl Trait is disallowed. (for now)
//
// Furthermore, if impl_trait_return_allow is true, then impl Trait may be used in
// return positions as well. This guards against trait declarations and their impls
// where impl Trait is disallowed. (again for now)
P(hir::FnDecl {
inputs: decl.inputs.iter()
.map(|arg| if let Some(def_id) = fn_def_id {
self.lower_ty(&arg.ty, ImplTraitContext::Universal(def_id))
} else {
self.lower_ty(&arg.ty, ImplTraitContext::Disallowed)
}).collect(),
output: match decl.output {
FunctionRetTy::Ty(ref ty) => match fn_def_id {
Some(_) if impl_trait_return_allow =>
hir::Return(self.lower_ty(ty, ImplTraitContext::Existential)),
_ => hir::Return(self.lower_ty(ty, ImplTraitContext::Disallowed)),
},
FunctionRetTy::Default(span) => hir::DefaultReturn(span),
},
variadic: decl.variadic,
2017-03-27 01:35:50 +03:00
has_implicit_self: decl.inputs.get(0).map_or(false, |arg| {
match arg.ty.node {
TyKind::ImplicitSelf => true,
TyKind::Rptr(_, ref mt) => mt.ty.node == TyKind::ImplicitSelf,
_ => false
}
})
})
}
2015-07-31 00:04:06 -07:00
fn lower_ty_param_bound(&mut self, tpb: &TyParamBound, itctx: ImplTraitContext)
-> hir::TyParamBound {
match *tpb {
TraitTyParamBound(ref ty, modifier) => {
hir::TraitTyParamBound(self.lower_poly_trait_ref(ty, itctx),
self.lower_trait_bound_modifier(modifier))
}
RegionTyParamBound(ref lifetime) => {
hir::RegionTyParamBound(self.lower_lifetime(lifetime))
}
}
}
2015-07-31 00:04:06 -07:00
2016-10-22 03:33:36 +03:00
fn lower_ty_param(&mut self, tp: &TyParam, add_bounds: &[TyParamBound]) -> hir::TyParam {
2017-03-24 23:03:15 +00:00
let mut name = self.lower_ident(tp.ident);
// Don't expose `Self` (recovered "keyword used as ident" parse error).
// `rustc::ty` expects `Self` to be only used for a trait's `Self`.
// Instead, use gensym("Self") to create a distinct name that looks the same.
if name == keywords::SelfType.name() {
name = Symbol::gensym("Self");
}
let itctx = ImplTraitContext::Universal(self.resolver.definitions().local_def_id(tp.id));
let mut bounds = self.lower_bounds(&tp.bounds, itctx);
2016-10-22 03:33:36 +03:00
if !add_bounds.is_empty() {
bounds = bounds.into_iter().chain(
self.lower_bounds(add_bounds, itctx).into_iter()
).collect();
2016-10-22 03:33:36 +03:00
}
hir::TyParam {
id: self.lower_node_id(tp.id).node_id,
name,
bounds,
default: tp.default.as_ref().map(|x| self.lower_ty(x, ImplTraitContext::Disallowed)),
span: tp.span,
pure_wrt_drop: tp.attrs.iter().any(|attr| attr.check_name("may_dangle")),
synthetic: tp.attrs.iter()
.filter(|attr| attr.check_name("rustc_synthetic"))
.map(|_| hir::SyntheticTyParamKind::ImplTrait)
.nth(0),
}
2015-07-31 00:04:06 -07:00
}
fn lower_ty_params(&mut self, tps: &Vec<TyParam>, add_bounds: &NodeMap<Vec<TyParamBound>>)
2016-10-22 03:33:36 +03:00
-> hir::HirVec<hir::TyParam> {
tps.iter().map(|tp| {
self.lower_ty_param(tp, add_bounds.get(&tp.id).map_or(&[][..], |x| &x))
}).collect()
2015-07-31 00:04:06 -07:00
}
fn lower_lifetime(&mut self, l: &Lifetime) -> hir::Lifetime {
2017-11-16 22:59:45 -08:00
let name = match self.lower_ident(l.ident) {
x if x == "'_" => hir::LifetimeName::Underscore,
x if x == "'static" => hir::LifetimeName::Static,
name => {
if self.is_collecting_in_band_lifetimes &&
!self.in_scope_lifetimes.contains(&name) &&
self.lifetimes_to_define.iter()
.find(|&&(_, lt_name)| lt_name == name)
.is_none()
{
self.lifetimes_to_define.push((l.span, name));
}
hir::LifetimeName::Name(name)
}
};
hir::Lifetime {
id: self.lower_node_id(l.id).node_id,
2017-11-16 22:59:45 -08:00
name,
span: l.span,
2015-07-31 00:04:06 -07:00
}
}
fn lower_lifetime_def(&mut self, l: &LifetimeDef) -> hir::LifetimeDef {
2017-11-16 22:59:45 -08:00
let was_collecting_in_band = self.is_collecting_in_band_lifetimes;
self.is_collecting_in_band_lifetimes = false;
let def = hir::LifetimeDef {
lifetime: self.lower_lifetime(&l.lifetime),
bounds: self.lower_lifetimes(&l.bounds),
pure_wrt_drop: l.attrs.iter().any(|attr| attr.check_name("may_dangle")),
2017-11-16 22:59:45 -08:00
in_band: false,
};
self.is_collecting_in_band_lifetimes = was_collecting_in_band;
def
}
2015-07-31 00:04:06 -07:00
fn lower_lifetimes(&mut self, lts: &Vec<Lifetime>) -> hir::HirVec<hir::Lifetime> {
lts.iter().map(|l| self.lower_lifetime(l)).collect()
2015-10-06 16:03:56 +13:00
}
2015-07-31 00:04:06 -07:00
fn lower_lifetime_defs(&mut self, lts: &Vec<LifetimeDef>) -> hir::HirVec<hir::LifetimeDef> {
lts.iter().map(|l| self.lower_lifetime_def(l)).collect()
2015-07-31 00:04:06 -07:00
}
fn lower_generics(&mut self, g: &Generics) -> hir::Generics {
2016-10-22 03:33:36 +03:00
// Collect `?Trait` bounds in where clause and move them to parameter definitions.
let mut add_bounds = NodeMap();
for pred in &g.where_clause.predicates {
if let WherePredicate::BoundPredicate(ref bound_pred) = *pred {
'next_bound: for bound in &bound_pred.bounds {
if let TraitTyParamBound(_, TraitBoundModifier::Maybe) = *bound {
let report_error = |this: &mut Self| {
this.diagnostic().span_err(bound_pred.bounded_ty.span,
"`?Trait` bounds are only permitted at the \
point where a type parameter is declared");
};
// Check if the where clause type is a plain type parameter.
match bound_pred.bounded_ty.node {
TyKind::Path(None, ref path)
if path.segments.len() == 1 &&
bound_pred.bound_lifetimes.is_empty() => {
2016-10-22 03:33:36 +03:00
if let Some(Def::TyParam(def_id)) =
self.resolver.get_resolution(bound_pred.bounded_ty.id)
.map(|d| d.base_def()) {
2016-10-22 03:33:36 +03:00
if let Some(node_id) =
self.resolver.definitions().as_local_node_id(def_id) {
for ty_param in &g.ty_params {
if node_id == ty_param.id {
add_bounds.entry(ty_param.id).or_insert(Vec::new())
.push(bound.clone());
continue 'next_bound;
}
}
}
}
report_error(self)
}
_ => report_error(self)
}
}
}
}
}
hir::Generics {
2016-10-22 03:33:36 +03:00
ty_params: self.lower_ty_params(&g.ty_params, &add_bounds),
lifetimes: self.lower_lifetime_defs(&g.lifetimes),
where_clause: self.lower_where_clause(&g.where_clause),
2016-08-10 19:39:12 +02:00
span: g.span,
}
}
2015-07-31 00:04:06 -07:00
fn lower_where_clause(&mut self, wc: &WhereClause) -> hir::WhereClause {
hir::WhereClause {
id: self.lower_node_id(wc.id).node_id,
predicates: wc.predicates
.iter()
.map(|predicate| self.lower_where_predicate(predicate))
.collect(),
}
2015-10-06 16:03:56 +13:00
}
2015-07-31 00:04:06 -07:00
fn lower_where_predicate(&mut self, pred: &WherePredicate) -> hir::WherePredicate {
match *pred {
WherePredicate::BoundPredicate(WhereBoundPredicate{ ref bound_lifetimes,
ref bounded_ty,
ref bounds,
span}) => {
2017-11-16 22:59:45 -08:00
self.with_in_scope_lifetime_defs(bound_lifetimes, |this| {
hir::WherePredicate::BoundPredicate(hir::WhereBoundPredicate {
bound_lifetimes: this.lower_lifetime_defs(bound_lifetimes),
bounded_ty: this.lower_ty(bounded_ty, ImplTraitContext::Disallowed),
bounds: bounds.iter().filter_map(|bound| match *bound {
// Ignore `?Trait` bounds.
// Tthey were copied into type parameters already.
TraitTyParamBound(_, TraitBoundModifier::Maybe) => None,
_ => Some(this.lower_ty_param_bound(
bound, ImplTraitContext::Disallowed))
}).collect(),
span,
})
})
}
WherePredicate::RegionPredicate(WhereRegionPredicate{ ref lifetime,
ref bounds,
span}) => {
hir::WherePredicate::RegionPredicate(hir::WhereRegionPredicate {
span,
lifetime: self.lower_lifetime(lifetime),
bounds: bounds.iter().map(|bound| self.lower_lifetime(bound)).collect(),
})
}
WherePredicate::EqPredicate(WhereEqPredicate{ id,
ref lhs_ty,
ref rhs_ty,
span}) => {
hir::WherePredicate::EqPredicate(hir::WhereEqPredicate {
id: self.lower_node_id(id).node_id,
lhs_ty: self.lower_ty(lhs_ty, ImplTraitContext::Disallowed),
rhs_ty: self.lower_ty(rhs_ty, ImplTraitContext::Disallowed),
span,
})
}
2015-07-31 00:04:06 -07:00
}
}
fn lower_variant_data(&mut self, vdata: &VariantData) -> hir::VariantData {
match *vdata {
VariantData::Struct(ref fields, id) => {
hir::VariantData::Struct(fields.iter()
.enumerate()
.map(|f| self.lower_struct_field(f))
.collect(),
self.lower_node_id(id).node_id)
}
VariantData::Tuple(ref fields, id) => {
hir::VariantData::Tuple(fields.iter()
.enumerate()
.map(|f| self.lower_struct_field(f))
.collect(),
self.lower_node_id(id).node_id)
}
VariantData::Unit(id) => hir::VariantData::Unit(self.lower_node_id(id).node_id),
2015-07-31 00:04:06 -07:00
}
}
fn lower_trait_ref(&mut self, p: &TraitRef, itctx: ImplTraitContext) -> hir::TraitRef {
let path = match self.lower_qpath(p.ref_id, &None, &p.path, ParamMode::Explicit, itctx) {
hir::QPath::Resolved(None, path) => path.and_then(|path| path),
qpath => bug!("lower_trait_ref: unexpected QPath `{:?}`", qpath)
};
hir::TraitRef {
path,
ref_id: self.lower_node_id(p.ref_id).node_id,
2015-07-31 00:04:06 -07:00
}
}
fn lower_poly_trait_ref(&mut self,
p: &PolyTraitRef,
itctx: ImplTraitContext)
-> hir::PolyTraitRef {
2017-11-16 22:59:45 -08:00
let bound_lifetimes = self.lower_lifetime_defs(&p.bound_lifetimes);
let trait_ref = self.with_parent_impl_lifetime_defs(&bound_lifetimes,
|this| this.lower_trait_ref(&p.trait_ref, itctx));
hir::PolyTraitRef {
2017-11-16 22:59:45 -08:00
bound_lifetimes,
trait_ref,
span: p.span,
2015-07-31 00:04:06 -07:00
}
}
fn lower_struct_field(&mut self, (index, f): (usize, &StructField)) -> hir::StructField {
hir::StructField {
span: f.span,
id: self.lower_node_id(f.id).node_id,
2017-03-24 23:03:15 +00:00
name: self.lower_ident(match f.ident {
Some(ident) => ident,
// FIXME(jseyfried) positional field hygiene
2017-07-31 23:04:34 +03:00
None => Ident { name: Symbol::intern(&index.to_string()), ctxt: f.span.ctxt() },
2017-03-24 23:03:15 +00:00
}),
vis: self.lower_visibility(&f.vis, None),
ty: self.lower_ty(&f.ty, ImplTraitContext::Disallowed),
attrs: self.lower_attrs(&f.attrs),
2015-07-31 00:04:06 -07:00
}
}
fn lower_field(&mut self, f: &Field) -> hir::Field {
hir::Field {
2017-03-24 23:03:15 +00:00
name: respan(f.ident.span, self.lower_ident(f.ident.node)),
expr: P(self.lower_expr(&f.expr)),
span: f.span,
is_shorthand: f.is_shorthand,
2016-04-18 10:30:55 +12:00
}
}
2015-07-31 00:04:06 -07:00
fn lower_mt(&mut self, mt: &MutTy, itctx: ImplTraitContext) -> hir::MutTy {
hir::MutTy {
ty: self.lower_ty(&mt.ty, itctx),
mutbl: self.lower_mutability(mt.mutbl),
2016-04-18 10:30:55 +12:00
}
}
2015-07-31 00:04:06 -07:00
fn lower_bounds(&mut self, bounds: &[TyParamBound], itctx: ImplTraitContext)
-> hir::TyParamBounds {
bounds.iter().map(|bound| self.lower_ty_param_bound(bound, itctx)).collect()
}
fn lower_block(&mut self, b: &Block, targeted_by_break: bool) -> P<hir::Block> {
2016-06-23 09:51:18 +00:00
let mut expr = None;
let mut stmts = vec![];
for (index, stmt) in b.stmts.iter().enumerate() {
if index == b.stmts.len() - 1 {
if let StmtKind::Expr(ref e) = stmt.node {
expr = Some(P(self.lower_expr(e)));
} else {
stmts.extend(self.lower_stmt(stmt));
}
2016-06-23 09:51:18 +00:00
} else {
stmts.extend(self.lower_stmt(stmt));
2016-06-23 09:51:18 +00:00
}
}
let LoweredNodeId { node_id, hir_id } = self.lower_node_id(b.id);
P(hir::Block {
id: node_id,
hir_id,
2016-06-23 09:51:18 +00:00
stmts: stmts.into(),
expr,
rules: self.lower_block_check_mode(&b.rules),
span: b.span,
targeted_by_break,
})
}
fn lower_item_kind(&mut self,
id: NodeId,
name: &mut Name,
attrs: &hir::HirVec<Attribute>,
vis: &mut hir::Visibility,
i: &ItemKind)
-> hir::Item_ {
match *i {
ItemKind::ExternCrate(string) => hir::ItemExternCrate(string),
ItemKind::Use(ref use_tree) => {
// Start with an empty prefix
let prefix = Path {
segments: vec![],
span: use_tree.span,
};
self.lower_use_tree(use_tree, &prefix, id, vis, name, attrs)
}
ItemKind::Static(ref t, m, ref e) => {
let value = self.lower_body(None, |this| this.lower_expr(e));
hir::ItemStatic(self.lower_ty(t, ImplTraitContext::Disallowed),
self.lower_mutability(m),
value)
}
ItemKind::Const(ref t, ref e) => {
let value = self.lower_body(None, |this| this.lower_expr(e));
hir::ItemConst(self.lower_ty(t, ImplTraitContext::Disallowed), value)
}
ItemKind::Fn(ref decl, unsafety, constness, abi, ref generics, ref body) => {
let fn_def_id = self.resolver.definitions().opt_local_def_id(id);
self.with_new_scopes(|this| {
2016-12-26 14:34:03 +01:00
let body_id = this.lower_body(Some(decl), |this| {
let body = this.lower_block(body, false);
this.expr_block(body, ThinVec::new())
});
2017-11-16 22:59:45 -08:00
let (generics, fn_decl) =
this.add_in_band_lifetime_defs(generics, fn_def_id, |this|
this.lower_fn_decl(decl, fn_def_id, true));
hir::ItemFn(fn_decl,
this.lower_unsafety(unsafety),
this.lower_constness(constness),
abi,
2017-11-16 22:59:45 -08:00
generics,
body_id)
2017-02-15 14:52:27 -08:00
})
}
ItemKind::Mod(ref m) => hir::ItemMod(self.lower_mod(m)),
ItemKind::ForeignMod(ref nm) => hir::ItemForeignMod(self.lower_foreign_mod(nm)),
2017-03-15 21:27:40 -05:00
ItemKind::GlobalAsm(ref ga) => hir::ItemGlobalAsm(self.lower_global_asm(ga)),
ItemKind::Ty(ref t, ref generics) => {
hir::ItemTy(self.lower_ty(t, ImplTraitContext::Disallowed),
self.lower_generics(generics))
}
ItemKind::Enum(ref enum_definition, ref generics) => {
hir::ItemEnum(hir::EnumDef {
variants: enum_definition.variants
.iter()
.map(|x| self.lower_variant(x))
.collect(),
},
self.lower_generics(generics))
}
ItemKind::Struct(ref struct_def, ref generics) => {
let struct_def = self.lower_variant_data(struct_def);
hir::ItemStruct(struct_def, self.lower_generics(generics))
}
ItemKind::Union(ref vdata, ref generics) => {
let vdata = self.lower_variant_data(vdata);
hir::ItemUnion(vdata, self.lower_generics(generics))
}
ItemKind::AutoImpl(unsafety, ref trait_ref) => {
let trait_ref = self.lower_trait_ref(trait_ref, ImplTraitContext::Disallowed);
if let Def::Trait(def_id) = trait_ref.path.def {
self.trait_auto_impl.insert(def_id, id);
}
hir::ItemAutoImpl(self.lower_unsafety(unsafety),
trait_ref)
}
ItemKind::Impl(unsafety,
polarity,
defaultness,
2017-11-16 22:59:45 -08:00
ref ast_generics,
ref ifce,
ref ty,
ref impl_items) => {
2017-11-16 22:59:45 -08:00
let def_id = self.resolver.definitions().opt_local_def_id(id);
let (generics, (ifce, lowered_ty)) =
self.add_in_band_lifetime_defs(ast_generics, def_id, |this| {
let ifce = ifce.as_ref().map(|trait_ref| {
this.lower_trait_ref(trait_ref, ImplTraitContext::Disallowed)
});
if let Some(ref trait_ref) = ifce {
if let Def::Trait(def_id) = trait_ref.path.def {
this.trait_impls.entry(def_id).or_insert(vec![]).push(id);
}
}
let lowered_ty = this.lower_ty(ty, ImplTraitContext::Disallowed);
(ifce, lowered_ty)
});
let new_impl_items = self.with_in_scope_lifetime_defs(
&ast_generics.lifetimes, |this| {
impl_items.iter()
.map(|item| this.lower_impl_item_ref(item))
.collect()
});
hir::ItemImpl(self.lower_unsafety(unsafety),
self.lower_impl_polarity(polarity),
self.lower_defaultness(defaultness, true /* [1] */),
2017-11-16 22:59:45 -08:00
generics,
ifce,
2017-11-16 22:59:45 -08:00
lowered_ty,
new_impl_items)
}
ItemKind::Trait(is_auto, unsafety, ref generics, ref bounds, ref items) => {
let bounds = self.lower_bounds(bounds, ImplTraitContext::Disallowed);
let items = items.iter().map(|item| self.lower_trait_item_ref(item)).collect();
hir::ItemTrait(self.lower_is_auto(is_auto),
self.lower_unsafety(unsafety),
self.lower_generics(generics),
bounds,
items)
}
ItemKind::MacroDef(..) | ItemKind::Mac(..) => {
panic!("Shouldn't still be around")
}
}
// [1] `defaultness.has_value()` is never called for an `impl`, always `true` in order to
// not cause an assertion failure inside the `lower_defaultness` function
2015-10-06 16:03:56 +13:00
}
2015-07-31 00:04:06 -07:00
fn lower_use_tree(&mut self,
tree: &UseTree,
prefix: &Path,
id: NodeId,
vis: &mut hir::Visibility,
name: &mut Name,
attrs: &hir::HirVec<Attribute>)
-> hir::Item_ {
let path = &tree.prefix;
match tree.kind {
UseTreeKind::Simple(ident) => {
*name = ident.name;
// First apply the prefix to the path
let mut path = Path {
segments: prefix.segments
.iter()
.chain(path.segments.iter())
.cloned()
.collect(),
span: path.span.to(prefix.span),
};
// Correctly resolve `self` imports
if path.segments.last().unwrap().identifier.name == keywords::SelfValue.name() {
let _ = path.segments.pop();
if ident.name == keywords::SelfValue.name() {
*name = path.segments.last().unwrap().identifier.name;
}
}
let path = P(self.lower_path(id, &path, ParamMode::Explicit, true));
hir::ItemUse(path, hir::UseKind::Single)
}
UseTreeKind::Glob => {
let path = P(self.lower_path(id, &Path {
segments: prefix.segments
.iter()
.chain(path.segments.iter())
.cloned()
.collect(),
span: path.span,
}, ParamMode::Explicit, true));
hir::ItemUse(path, hir::UseKind::Glob)
}
UseTreeKind::Nested(ref trees) => {
let prefix = Path {
segments: prefix.segments
.iter()
.chain(path.segments.iter())
.cloned()
.collect(),
span: prefix.span.to(path.span),
};
// Add all the nested PathListItems in the HIR
for &(ref use_tree, id) in trees {
self.allocate_hir_id_counter(id, &use_tree);
let LoweredNodeId {
node_id: new_id,
hir_id: new_hir_id,
} = self.lower_node_id(id);
let mut vis = vis.clone();
let mut name = name.clone();
let item = self.lower_use_tree(
use_tree, &prefix, new_id, &mut vis, &mut name, &attrs,
);
self.with_hir_id_owner(new_id, |this| {
let vis = match vis {
hir::Visibility::Public => hir::Visibility::Public,
hir::Visibility::Crate => hir::Visibility::Crate,
hir::Visibility::Inherited => hir::Visibility::Inherited,
hir::Visibility::Restricted { ref path, id: _ } => {
hir::Visibility::Restricted {
path: path.clone(),
// We are allocating a new NodeId here
id: this.next_id().node_id,
}
}
};
this.items.insert(new_id, hir::Item {
id: new_id,
hir_id: new_hir_id,
name: name,
attrs: attrs.clone(),
node: item,
vis,
span: use_tree.span,
});
});
}
// Privatize the degenerate import base, used only to check
// the stability of `use a::{};`, to avoid it showing up as
// a reexport by accident when `pub`, e.g. in documentation.
let path = P(self.lower_path(id, &prefix, ParamMode::Explicit, true));
*vis = hir::Inherited;
hir::ItemUse(path, hir::UseKind::ListStem)
}
}
}
fn lower_trait_item(&mut self, i: &TraitItem) -> hir::TraitItem {
self.with_parent_def(i.id, |this| {
let LoweredNodeId { node_id, hir_id } = this.lower_node_id(i.id);
let fn_def_id = this.resolver.definitions().opt_local_def_id(node_id);
2017-11-16 22:59:45 -08:00
let (generics, node) = match i.node {
TraitItemKind::Const(ref ty, ref default) => {
(
this.lower_generics(&i.generics),
hir::TraitItemKind::Const(
this.lower_ty(ty, ImplTraitContext::Disallowed),
default.as_ref().map(|x| {
this.lower_body(None, |this| this.lower_expr(x))
}))
)
}
TraitItemKind::Method(ref sig, None) => {
let names = this.lower_fn_args_to_names(&sig.decl);
this.add_in_band_lifetime_defs(&i.generics, fn_def_id, |this|
hir::TraitItemKind::Method(
this.lower_method_sig(sig, fn_def_id, false),
hir::TraitMethod::Required(names)))
}
TraitItemKind::Method(ref sig, Some(ref body)) => {
let body_id = this.lower_body(Some(&sig.decl), |this| {
let body = this.lower_block(body, false);
this.expr_block(body, ThinVec::new())
});
this.add_in_band_lifetime_defs(&i.generics, fn_def_id, |this|
hir::TraitItemKind::Method(
this.lower_method_sig(sig, fn_def_id, false),
hir::TraitMethod::Provided(body_id)))
}
TraitItemKind::Type(ref bounds, ref default) => {
(
this.lower_generics(&i.generics),
hir::TraitItemKind::Type(
this.lower_bounds(bounds, ImplTraitContext::Disallowed),
default.as_ref().map(|x| {
this.lower_ty(x, ImplTraitContext::Disallowed)
}))
)
}
TraitItemKind::Macro(..) => panic!("Shouldn't exist any more"),
};
hir::TraitItem {
id: node_id,
hir_id,
2017-03-24 23:03:15 +00:00
name: this.lower_ident(i.ident),
attrs: this.lower_attrs(&i.attrs),
2017-11-16 22:59:45 -08:00
generics,
node,
span: i.span,
}
})
2015-07-31 00:04:06 -07:00
}
fn lower_trait_item_ref(&mut self, i: &TraitItem) -> hir::TraitItemRef {
let (kind, has_default) = match i.node {
TraitItemKind::Const(_, ref default) => {
(hir::AssociatedItemKind::Const, default.is_some())
}
TraitItemKind::Type(_, ref default) => {
(hir::AssociatedItemKind::Type, default.is_some())
}
TraitItemKind::Method(ref sig, ref default) => {
(hir::AssociatedItemKind::Method {
has_self: sig.decl.has_self(),
}, default.is_some())
}
TraitItemKind::Macro(..) => unimplemented!(),
};
hir::TraitItemRef {
id: hir::TraitItemId { node_id: i.id },
2017-03-24 23:03:15 +00:00
name: self.lower_ident(i.ident),
span: i.span,
defaultness: self.lower_defaultness(Defaultness::Default, has_default),
kind,
}
}
fn lower_impl_item(&mut self, i: &ImplItem) -> hir::ImplItem {
self.with_parent_def(i.id, |this| {
let LoweredNodeId { node_id, hir_id } = this.lower_node_id(i.id);
let fn_def_id = this.resolver.definitions().opt_local_def_id(node_id);
2017-11-16 22:59:45 -08:00
let (generics, node) = match i.node {
ImplItemKind::Const(ref ty, ref expr) => {
let body_id = this.lower_body(None, |this| this.lower_expr(expr));
(
this.lower_generics(&i.generics),
hir::ImplItemKind::Const(
this.lower_ty(ty, ImplTraitContext::Disallowed),
body_id
)
)
}
ImplItemKind::Method(ref sig, ref body) => {
let body_id = this.lower_body(Some(&sig.decl), |this| {
let body = this.lower_block(body, false);
this.expr_block(body, ThinVec::new())
});
let impl_trait_return_allow = !this.is_in_trait_impl;
this.add_in_band_lifetime_defs(&i.generics, fn_def_id, |this|
hir::ImplItemKind::Method(
this.lower_method_sig(sig, fn_def_id, impl_trait_return_allow),
body_id))
}
ImplItemKind::Type(ref ty) => (
this.lower_generics(&i.generics),
hir::ImplItemKind::Type(
this.lower_ty(ty, ImplTraitContext::Disallowed)),
),
ImplItemKind::Macro(..) => panic!("Shouldn't exist any more"),
};
hir::ImplItem {
id: node_id,
hir_id,
2017-03-24 23:03:15 +00:00
name: this.lower_ident(i.ident),
attrs: this.lower_attrs(&i.attrs),
2017-11-16 22:59:45 -08:00
generics,
vis: this.lower_visibility(&i.vis, None),
defaultness: this.lower_defaultness(i.defaultness, true /* [1] */),
2017-11-16 22:59:45 -08:00
node,
span: i.span,
}
})
// [1] since `default impl` is not yet implemented, this is always true in impls
2015-07-31 00:04:06 -07:00
}
fn lower_impl_item_ref(&mut self, i: &ImplItem) -> hir::ImplItemRef {
hir::ImplItemRef {
id: hir::ImplItemId { node_id: i.id },
2017-03-24 23:03:15 +00:00
name: self.lower_ident(i.ident),
span: i.span,
vis: self.lower_visibility(&i.vis, Some(i.id)),
defaultness: self.lower_defaultness(i.defaultness, true /* [1] */),
kind: match i.node {
ImplItemKind::Const(..) => hir::AssociatedItemKind::Const,
ImplItemKind::Type(..) => hir::AssociatedItemKind::Type,
ImplItemKind::Method(ref sig, _) => {
hir::AssociatedItemKind::Method {
has_self: sig.decl.has_self(),
}
},
ImplItemKind::Macro(..) => unimplemented!(),
},
}
// [1] since `default impl` is not yet implemented, this is always true in impls
}
fn lower_mod(&mut self, m: &Mod) -> hir::Mod {
hir::Mod {
inner: m.inner,
item_ids: m.items.iter().flat_map(|x| self.lower_item_id(x)).collect(),
}
}
2015-07-31 00:04:06 -07:00
fn lower_item_id(&mut self, i: &Item) -> SmallVector<hir::ItemId> {
match i.node {
ItemKind::Use(ref use_tree) => {
let mut vec = SmallVector::one(hir::ItemId { id: i.id });
self.lower_item_id_use_tree(use_tree, &mut vec);
return vec;
}
ItemKind::MacroDef(..) => return SmallVector::new(),
_ => {}
}
SmallVector::one(hir::ItemId { id: i.id })
2015-07-31 00:04:06 -07:00
}
fn lower_item_id_use_tree(&self, tree: &UseTree, vec: &mut SmallVector<hir::ItemId>) {
match tree.kind {
UseTreeKind::Nested(ref nested_vec) => {
for &(ref nested, id) in nested_vec {
vec.push(hir::ItemId { id, });
self.lower_item_id_use_tree(nested, vec);
}
}
UseTreeKind::Glob => {}
UseTreeKind::Simple(..) => {}
}
}
pub fn lower_item(&mut self, i: &Item) -> Option<hir::Item> {
let mut name = i.ident.name;
2017-03-25 01:46:38 +00:00
let mut vis = self.lower_visibility(&i.vis, None);
let attrs = self.lower_attrs(&i.attrs);
if let ItemKind::MacroDef(ref def) = i.node {
if !def.legacy || i.attrs.iter().any(|attr| attr.path == "macro_export") {
rustc: Forbid interpolated tokens in the HIR Right now the HIR contains raw `syntax::ast::Attribute` structure but nowadays these can contain arbitrary tokens. One variant of the `Token` enum is an "interpolated" token which basically means to shove all the tokens for a nonterminal in this position. A "nonterminal" in this case is roughly analagous to a macro argument: macro_rules! foo { ($a:expr) => { // $a is a nonterminal as an expression } } Currently nonterminals contain namely items and expressions, and this poses a problem for incremental compilation! With incremental we want a stable hash of all HIR items, but this means we may transitively need a stable hash *of the entire AST*, which is certainly not stable w/ node ids and whatnot. Hence today there's a "bug" where the "stable hash" of an AST is just the raw hash value of the AST, and this only arises with interpolated nonterminals. The downside of this approach, however, is that a bunch of errors get spewed out during compilation about how this isn't a great idea. This PR is focused at fixing these warnings, basically deleting them from the compiler. The implementation here is to alter attributes as they're lowered from the AST to HIR, expanding all nonterminals in-place as we see them. This code for expanding a nonterminal to a token stream already exists for the `proc_macro` crate, so we basically just reuse the same implementation there. After this PR it's considered a bug to have an `Interpolated` token and hence the stable hash implementation simply uses `bug!` in this location. Closes #40946
2017-09-15 08:28:34 -07:00
let body = self.lower_token_stream(def.stream());
self.exported_macros.push(hir::MacroDef {
name,
vis,
attrs,
2017-03-25 01:46:38 +00:00
id: i.id,
span: i.span,
rustc: Forbid interpolated tokens in the HIR Right now the HIR contains raw `syntax::ast::Attribute` structure but nowadays these can contain arbitrary tokens. One variant of the `Token` enum is an "interpolated" token which basically means to shove all the tokens for a nonterminal in this position. A "nonterminal" in this case is roughly analagous to a macro argument: macro_rules! foo { ($a:expr) => { // $a is a nonterminal as an expression } } Currently nonterminals contain namely items and expressions, and this poses a problem for incremental compilation! With incremental we want a stable hash of all HIR items, but this means we may transitively need a stable hash *of the entire AST*, which is certainly not stable w/ node ids and whatnot. Hence today there's a "bug" where the "stable hash" of an AST is just the raw hash value of the AST, and this only arises with interpolated nonterminals. The downside of this approach, however, is that a bunch of errors get spewed out during compilation about how this isn't a great idea. This PR is focused at fixing these warnings, basically deleting them from the compiler. The implementation here is to alter attributes as they're lowered from the AST to HIR, expanding all nonterminals in-place as we see them. This code for expanding a nonterminal to a token stream already exists for the `proc_macro` crate, so we basically just reuse the same implementation there. After this PR it's considered a bug to have an `Interpolated` token and hence the stable hash implementation simply uses `bug!` in this location. Closes #40946
2017-09-15 08:28:34 -07:00
body,
2017-03-25 01:46:38 +00:00
legacy: def.legacy,
});
}
return None;
}
let node = self.with_parent_def(i.id, |this| {
this.lower_item_kind(i.id, &mut name, &attrs, &mut vis, &i.node)
});
let LoweredNodeId { node_id, hir_id } = self.lower_node_id(i.id);
Some(hir::Item {
id: node_id,
hir_id,
name,
attrs,
node,
vis,
2016-04-18 10:30:55 +12:00
span: i.span,
})
}
2015-07-31 00:04:06 -07:00
fn lower_foreign_item(&mut self, i: &ForeignItem) -> hir::ForeignItem {
self.with_parent_def(i.id, |this| {
2017-11-16 22:59:45 -08:00
let node_id = this.lower_node_id(i.id).node_id;
let def_id = this.resolver.definitions().local_def_id(node_id);
hir::ForeignItem {
2017-11-16 22:59:45 -08:00
id: node_id,
name: i.ident.name,
attrs: this.lower_attrs(&i.attrs),
node: match i.node {
ForeignItemKind::Fn(ref fdec, ref generics) => {
// Disallow impl Trait in foreign items
2017-11-16 22:59:45 -08:00
let (generics, (fn_dec, fn_args)) =
this.add_in_band_lifetime_defs(
generics,
Some(def_id),
|this| (
this.lower_fn_decl(fdec, None, false),
this.lower_fn_args_to_names(fdec)
)
);
hir::ForeignItemFn(fn_dec, fn_args, generics)
}
ForeignItemKind::Static(ref t, m) => {
hir::ForeignItemStatic(this.lower_ty(t, ImplTraitContext::Disallowed), m)
}
2017-09-03 19:53:58 +01:00
ForeignItemKind::Ty => {
hir::ForeignItemType
}
},
vis: this.lower_visibility(&i.vis, None),
span: i.span,
}
})
2015-07-31 00:04:06 -07:00
}
fn lower_method_sig(&mut self,
sig: &MethodSig,
fn_def_id: Option<DefId>,
impl_trait_return_allow: bool)
-> hir::MethodSig {
hir::MethodSig {
abi: sig.abi,
unsafety: self.lower_unsafety(sig.unsafety),
constness: self.lower_constness(sig.constness),
decl: self.lower_fn_decl(&sig.decl, fn_def_id, impl_trait_return_allow),
}
2015-07-31 00:04:06 -07:00
}
2017-10-12 11:18:55 -03:00
fn lower_is_auto(&mut self, a: IsAuto) -> hir::IsAuto {
match a {
IsAuto::Yes => hir::IsAuto::Yes,
IsAuto::No => hir::IsAuto::No,
}
}
fn lower_unsafety(&mut self, u: Unsafety) -> hir::Unsafety {
match u {
Unsafety::Unsafe => hir::Unsafety::Unsafe,
Unsafety::Normal => hir::Unsafety::Normal,
}
2015-07-31 00:04:06 -07:00
}
2016-08-10 16:20:12 -07:00
fn lower_constness(&mut self, c: Spanned<Constness>) -> hir::Constness {
match c.node {
Constness::Const => hir::Constness::Const,
Constness::NotConst => hir::Constness::NotConst,
}
2015-07-31 00:04:06 -07:00
}
fn lower_unop(&mut self, u: UnOp) -> hir::UnOp {
match u {
UnOp::Deref => hir::UnDeref,
UnOp::Not => hir::UnNot,
UnOp::Neg => hir::UnNeg,
}
2015-07-31 00:04:06 -07:00
}
fn lower_binop(&mut self, b: BinOp) -> hir::BinOp {
Spanned {
node: match b.node {
BinOpKind::Add => hir::BiAdd,
BinOpKind::Sub => hir::BiSub,
BinOpKind::Mul => hir::BiMul,
BinOpKind::Div => hir::BiDiv,
BinOpKind::Rem => hir::BiRem,
BinOpKind::And => hir::BiAnd,
BinOpKind::Or => hir::BiOr,
BinOpKind::BitXor => hir::BiBitXor,
BinOpKind::BitAnd => hir::BiBitAnd,
BinOpKind::BitOr => hir::BiBitOr,
BinOpKind::Shl => hir::BiShl,
BinOpKind::Shr => hir::BiShr,
BinOpKind::Eq => hir::BiEq,
BinOpKind::Lt => hir::BiLt,
BinOpKind::Le => hir::BiLe,
BinOpKind::Ne => hir::BiNe,
BinOpKind::Ge => hir::BiGe,
BinOpKind::Gt => hir::BiGt,
},
span: b.span,
}
}
2015-07-31 00:04:06 -07:00
fn lower_pat(&mut self, p: &Pat) -> P<hir::Pat> {
let LoweredNodeId { node_id, hir_id } = self.lower_node_id(p.id);
P(hir::Pat {
id: node_id,
hir_id,
node: match p.node {
PatKind::Wild => hir::PatKind::Wild,
PatKind::Ident(ref binding_mode, pth1, ref sub) => {
match self.resolver.get_resolution(p.id).map(|d| d.base_def()) {
// `None` can occur in body-less function signatures
def @ None | def @ Some(Def::Local(_)) => {
let canonical_id = match def {
Some(Def::Local(id)) => id,
_ => p.id
};
hir::PatKind::Binding(self.lower_binding_mode(binding_mode),
canonical_id,
respan(pth1.span, pth1.node.name),
sub.as_ref().map(|x| self.lower_pat(x)))
}
Some(def) => {
hir::PatKind::Path(hir::QPath::Resolved(None, P(hir::Path {
span: pth1.span,
def,
segments: hir_vec![
hir::PathSegment::from_name(pth1.node.name)
],
})))
}
}
}
PatKind::Lit(ref e) => hir::PatKind::Lit(P(self.lower_expr(e))),
PatKind::TupleStruct(ref path, ref pats, ddpos) => {
let qpath = self.lower_qpath(p.id, &None, path, ParamMode::Optional,
ImplTraitContext::Disallowed);
hir::PatKind::TupleStruct(qpath,
pats.iter().map(|x| self.lower_pat(x)).collect(),
ddpos)
}
PatKind::Path(ref qself, ref path) => {
hir::PatKind::Path(self.lower_qpath(p.id, qself, path, ParamMode::Optional,
ImplTraitContext::Disallowed))
}
PatKind::Struct(ref path, ref fields, etc) => {
let qpath = self.lower_qpath(p.id, &None, path, ParamMode::Optional,
ImplTraitContext::Disallowed);
let fs = fields.iter()
.map(|f| {
Spanned {
span: f.span,
node: hir::FieldPat {
2017-03-24 23:03:15 +00:00
name: self.lower_ident(f.node.ident),
pat: self.lower_pat(&f.node.pat),
is_shorthand: f.node.is_shorthand,
},
}
})
.collect();
hir::PatKind::Struct(qpath, fs, etc)
}
PatKind::Tuple(ref elts, ddpos) => {
hir::PatKind::Tuple(elts.iter().map(|x| self.lower_pat(x)).collect(), ddpos)
}
PatKind::Box(ref inner) => hir::PatKind::Box(self.lower_pat(inner)),
PatKind::Ref(ref inner, mutbl) => {
hir::PatKind::Ref(self.lower_pat(inner), self.lower_mutability(mutbl))
}
PatKind::Range(ref e1, ref e2, ref end) => {
hir::PatKind::Range(P(self.lower_expr(e1)),
P(self.lower_expr(e2)),
self.lower_range_end(end))
}
PatKind::Slice(ref before, ref slice, ref after) => {
hir::PatKind::Slice(before.iter().map(|x| self.lower_pat(x)).collect(),
slice.as_ref().map(|x| self.lower_pat(x)),
after.iter().map(|x| self.lower_pat(x)).collect())
}
PatKind::Mac(_) => panic!("Shouldn't exist here"),
},
span: p.span,
})
}
2015-10-06 16:03:56 +13:00
fn lower_range_end(&mut self, e: &RangeEnd) -> hir::RangeEnd {
match *e {
RangeEnd::Included(_) => hir::RangeEnd::Included,
RangeEnd::Excluded => hir::RangeEnd::Excluded,
}
}
fn lower_expr(&mut self, e: &Expr) -> hir::Expr {
let kind = match e.node {
// Issue #22181:
// Eventually a desugaring for `box EXPR`
// (similar to the desugaring above for `in PLACE BLOCK`)
// should go here, desugaring
//
// to:
//
// let mut place = BoxPlace::make_place();
// let raw_place = Place::pointer(&mut place);
// let value = $value;
// unsafe {
// ::std::ptr::write(raw_place, value);
// Boxed::finalize(place)
// }
//
// But for now there are type-inference issues doing that.
ExprKind::Box(ref inner) => {
hir::ExprBox(P(self.lower_expr(inner)))
}
// Desugar ExprBox: `in (PLACE) EXPR`
ExprKind::InPlace(ref placer, ref value_expr) => {
2015-09-29 13:46:01 +13:00
// to:
//
// let p = PLACE;
// let mut place = Placer::make_place(p);
2015-09-29 13:46:01 +13:00
// let raw_place = Place::pointer(&mut place);
// push_unsafe!({
// std::intrinsics::move_val_init(raw_place, pop_unsafe!( EXPR ));
// InPlace::finalize(place)
// })
let placer_expr = P(self.lower_expr(placer));
let value_expr = P(self.lower_expr(value_expr));
let placer_ident = self.str_to_ident("placer");
let place_ident = self.str_to_ident("place");
let p_ptr_ident = self.str_to_ident("p_ptr");
let make_place = ["ops", "Placer", "make_place"];
let place_pointer = ["ops", "Place", "pointer"];
let move_val_init = ["intrinsics", "move_val_init"];
let inplace_finalize = ["ops", "InPlace", "finalize"];
let unstable_span =
self.allow_internal_unstable(CompilerDesugaringKind::BackArrow, e.span);
let make_call = |this: &mut LoweringContext, p, args| {
let path = P(this.expr_std_path(unstable_span, p, ThinVec::new()));
P(this.expr_call(e.span, path, args))
};
2015-09-28 17:24:42 +13:00
let mk_stmt_let = |this: &mut LoweringContext, bind, expr| {
this.stmt_let(e.span, false, bind, expr)
};
let mk_stmt_let_mut = |this: &mut LoweringContext, bind, expr| {
this.stmt_let(e.span, true, bind, expr)
};
// let placer = <placer_expr> ;
let (s1, placer_binding) = {
mk_stmt_let(self, placer_ident, placer_expr)
};
// let mut place = Placer::make_place(placer);
let (s2, place_binding) = {
let placer = self.expr_ident(e.span, placer_ident, placer_binding);
let call = make_call(self, &make_place, hir_vec![placer]);
mk_stmt_let_mut(self, place_ident, call)
};
// let p_ptr = Place::pointer(&mut place);
let (s3, p_ptr_binding) = {
let agent = P(self.expr_ident(e.span, place_ident, place_binding));
let args = hir_vec![self.expr_mut_addr_of(e.span, agent)];
let call = make_call(self, &place_pointer, args);
mk_stmt_let(self, p_ptr_ident, call)
};
// pop_unsafe!(EXPR));
let pop_unsafe_expr = {
self.signal_block_expr(hir_vec![],
value_expr,
e.span,
hir::PopUnsafeBlock(hir::CompilerGenerated),
ThinVec::new())
};
// push_unsafe!({
// std::intrinsics::move_val_init(raw_place, pop_unsafe!( EXPR ));
// InPlace::finalize(place)
// })
let expr = {
let ptr = self.expr_ident(e.span, p_ptr_ident, p_ptr_binding);
let call_move_val_init =
hir::StmtSemi(
make_call(self, &move_val_init, hir_vec![ptr, pop_unsafe_expr]),
self.next_id().node_id);
let call_move_val_init = respan(e.span, call_move_val_init);
let place = self.expr_ident(e.span, place_ident, place_binding);
let call = make_call(self, &inplace_finalize, hir_vec![place]);
P(self.signal_block_expr(hir_vec![call_move_val_init],
call,
e.span,
hir::PushUnsafeBlock(hir::CompilerGenerated),
ThinVec::new()))
};
let block = self.block_all(e.span, hir_vec![s1, s2, s3], Some(expr));
hir::ExprBlock(P(block))
}
ExprKind::Array(ref exprs) => {
hir::ExprArray(exprs.iter().map(|x| self.lower_expr(x)).collect())
}
ExprKind::Repeat(ref expr, ref count) => {
let expr = P(self.lower_expr(expr));
let count = self.lower_body(None, |this| this.lower_expr(count));
hir::ExprRepeat(expr, count)
}
ExprKind::Tup(ref elts) => {
hir::ExprTup(elts.iter().map(|x| self.lower_expr(x)).collect())
}
ExprKind::Call(ref f, ref args) => {
let f = P(self.lower_expr(f));
hir::ExprCall(f, args.iter().map(|x| self.lower_expr(x)).collect())
}
ExprKind::MethodCall(ref seg, ref args) => {
let hir_seg = self.lower_path_segment(e.span, seg, ParamMode::Optional, 0,
ParenthesizedGenericArgs::Err,
ImplTraitContext::Disallowed);
let args = args.iter().map(|x| self.lower_expr(x)).collect();
hir::ExprMethodCall(hir_seg, seg.span, args)
}
ExprKind::Binary(binop, ref lhs, ref rhs) => {
let binop = self.lower_binop(binop);
let lhs = P(self.lower_expr(lhs));
let rhs = P(self.lower_expr(rhs));
hir::ExprBinary(binop, lhs, rhs)
}
ExprKind::Unary(op, ref ohs) => {
let op = self.lower_unop(op);
let ohs = P(self.lower_expr(ohs));
hir::ExprUnary(op, ohs)
}
ExprKind::Lit(ref l) => hir::ExprLit(P((**l).clone())),
ExprKind::Cast(ref expr, ref ty) => {
let expr = P(self.lower_expr(expr));
hir::ExprCast(expr, self.lower_ty(ty, ImplTraitContext::Disallowed))
}
ExprKind::Type(ref expr, ref ty) => {
let expr = P(self.lower_expr(expr));
hir::ExprType(expr, self.lower_ty(ty, ImplTraitContext::Disallowed))
}
ExprKind::AddrOf(m, ref ohs) => {
let m = self.lower_mutability(m);
let ohs = P(self.lower_expr(ohs));
hir::ExprAddrOf(m, ohs)
}
// More complicated than you might expect because the else branch
// might be `if let`.
ExprKind::If(ref cond, ref blk, ref else_opt) => {
let else_opt = else_opt.as_ref().map(|els| {
match els.node {
ExprKind::IfLet(..) => {
// wrap the if-let expr in a block
let span = els.span;
let els = P(self.lower_expr(els));
let LoweredNodeId {
node_id,
hir_id,
} = self.next_id();
let blk = P(hir::Block {
stmts: hir_vec![],
expr: Some(els),
id: node_id,
hir_id,
rules: hir::DefaultBlock,
span,
targeted_by_break: false,
});
P(self.expr_block(blk, ThinVec::new()))
}
_ => P(self.lower_expr(els)),
}
});
let then_blk = self.lower_block(blk, false);
let then_expr = self.expr_block(then_blk, ThinVec::new());
hir::ExprIf(P(self.lower_expr(cond)), P(then_expr), else_opt)
}
ExprKind::While(ref cond, ref body, opt_ident) => {
self.with_loop_scope(e.id, |this|
hir::ExprWhile(
this.with_loop_condition_scope(|this| P(this.lower_expr(cond))),
this.lower_block(body, false),
this.lower_opt_sp_ident(opt_ident)))
}
ExprKind::Loop(ref body, opt_ident) => {
self.with_loop_scope(e.id, |this|
hir::ExprLoop(this.lower_block(body, false),
this.lower_opt_sp_ident(opt_ident),
hir::LoopSource::Loop))
}
ExprKind::Catch(ref body) => {
self.with_catch_scope(body.id, |this|
hir::ExprBlock(this.lower_block(body, true)))
}
ExprKind::Match(ref expr, ref arms) => {
hir::ExprMatch(P(self.lower_expr(expr)),
arms.iter().map(|x| self.lower_arm(x)).collect(),
hir::MatchSource::Normal)
}
ExprKind::Closure(capture_clause, ref decl, ref body, fn_decl_span) => {
self.with_new_scopes(|this| {
this.with_parent_def(e.id, |this| {
let mut is_generator = false;
2016-12-26 14:34:03 +01:00
let body_id = this.lower_body(Some(decl), |this| {
let e = this.lower_expr(body);
is_generator = this.is_generator;
2016-12-26 14:34:03 +01:00
e
});
if is_generator && !decl.inputs.is_empty() {
2017-08-14 19:38:32 -07:00
span_err!(this.sess, fn_decl_span, E0628,
2017-07-20 08:06:21 +02:00
"generators cannot have explicit arguments");
this.sess.abort_if_errors();
2017-07-06 22:08:45 +02:00
}
hir::ExprClosure(this.lower_capture_clause(capture_clause),
this.lower_fn_decl(decl, None, false),
2016-12-26 14:34:03 +01:00
body_id,
fn_decl_span,
is_generator)
})
})
}
ExprKind::Block(ref blk) => hir::ExprBlock(self.lower_block(blk, false)),
ExprKind::Assign(ref el, ref er) => {
hir::ExprAssign(P(self.lower_expr(el)), P(self.lower_expr(er)))
}
ExprKind::AssignOp(op, ref el, ref er) => {
hir::ExprAssignOp(self.lower_binop(op),
P(self.lower_expr(el)),
P(self.lower_expr(er)))
}
ExprKind::Field(ref el, ident) => {
2017-03-24 23:03:15 +00:00
hir::ExprField(P(self.lower_expr(el)),
respan(ident.span, self.lower_ident(ident.node)))
}
ExprKind::TupField(ref el, ident) => {
hir::ExprTupField(P(self.lower_expr(el)), ident)
}
ExprKind::Index(ref el, ref er) => {
hir::ExprIndex(P(self.lower_expr(el)), P(self.lower_expr(er)))
}
ExprKind::Range(ref e1, ref e2, lims) => {
use syntax::ast::RangeLimits::*;
let path = match (e1, e2, lims) {
(&None, &None, HalfOpen) => "RangeFull",
(&Some(..), &None, HalfOpen) => "RangeFrom",
(&None, &Some(..), HalfOpen) => "RangeTo",
(&Some(..), &Some(..), HalfOpen) => "Range",
(&None, &Some(..), Closed) => "RangeToInclusive",
(&Some(..), &Some(..), Closed) => "RangeInclusive",
(_, &None, Closed) =>
panic!(self.diagnostic().span_fatal(
e.span, "inclusive range with no end")),
};
let fields =
e1.iter().map(|e| ("start", e)).chain(e2.iter().map(|e| ("end", e)))
.map(|(s, e)| {
let expr = P(self.lower_expr(&e));
let unstable_span =
self.allow_internal_unstable(CompilerDesugaringKind::DotFill, e.span);
self.field(Symbol::intern(s), expr, unstable_span)
}).collect::<P<[hir::Field]>>();
let is_unit = fields.is_empty();
let unstable_span =
self.allow_internal_unstable(CompilerDesugaringKind::DotFill, e.span);
let struct_path =
iter::once("ops").chain(iter::once(path))
.collect::<Vec<_>>();
let struct_path = self.std_path(unstable_span, &struct_path, is_unit);
let struct_path = hir::QPath::Resolved(None, P(struct_path));
let LoweredNodeId { node_id, hir_id } = self.lower_node_id(e.id);
return hir::Expr {
id: node_id,
hir_id,
node: if is_unit {
hir::ExprPath(struct_path)
} else {
hir::ExprStruct(struct_path, fields, None)
},
span: unstable_span,
attrs: e.attrs.clone(),
};
}
ExprKind::Path(ref qself, ref path) => {
hir::ExprPath(self.lower_qpath(e.id, qself, path, ParamMode::Optional,
ImplTraitContext::Disallowed))
}
ExprKind::Break(opt_ident, ref opt_expr) => {
let label_result = if self.is_in_loop_condition && opt_ident.is_none() {
hir::Destination {
ident: opt_ident,
target_id: hir::ScopeTarget::Loop(
Err(hir::LoopIdError::UnlabeledCfInWhileCondition).into()),
}
} else {
self.lower_loop_destination(opt_ident.map(|ident| (e.id, ident)))
};
hir::ExprBreak(
label_result,
opt_expr.as_ref().map(|x| P(self.lower_expr(x))))
}
ExprKind::Continue(opt_ident) =>
hir::ExprAgain(
if self.is_in_loop_condition && opt_ident.is_none() {
2017-02-17 18:55:28 -08:00
hir::Destination {
ident: opt_ident,
target_id: hir::ScopeTarget::Loop(Err(
hir::LoopIdError::UnlabeledCfInWhileCondition).into()),
}
} else {
self.lower_loop_destination(opt_ident.map( |ident| (e.id, ident)))
}),
ExprKind::Ret(ref e) => hir::ExprRet(e.as_ref().map(|x| P(self.lower_expr(x)))),
ExprKind::InlineAsm(ref asm) => {
let hir_asm = hir::InlineAsm {
inputs: asm.inputs.iter().map(|&(ref c, _)| c.clone()).collect(),
outputs: asm.outputs.iter().map(|out| {
hir::InlineAsmOutput {
constraint: out.constraint.clone(),
is_rw: out.is_rw,
is_indirect: out.is_indirect,
}
}).collect(),
asm: asm.asm.clone(),
asm_str_style: asm.asm_str_style,
clobbers: asm.clobbers.clone().into(),
volatile: asm.volatile,
alignstack: asm.alignstack,
dialect: asm.dialect,
2017-03-17 04:04:41 +00:00
ctxt: asm.ctxt,
};
let outputs =
asm.outputs.iter().map(|out| self.lower_expr(&out.expr)).collect();
let inputs =
asm.inputs.iter().map(|&(_, ref input)| self.lower_expr(input)).collect();
hir::ExprInlineAsm(P(hir_asm), outputs, inputs)
}
ExprKind::Struct(ref path, ref fields, ref maybe_expr) => {
hir::ExprStruct(self.lower_qpath(e.id, &None, path, ParamMode::Optional,
ImplTraitContext::Disallowed),
fields.iter().map(|x| self.lower_field(x)).collect(),
maybe_expr.as_ref().map(|x| P(self.lower_expr(x))))
}
ExprKind::Paren(ref ex) => {
let mut ex = self.lower_expr(ex);
// include parens in span, but only if it is a super-span.
if e.span.contains(ex.span) {
ex.span = e.span;
}
// merge attributes into the inner expression.
let mut attrs = e.attrs.clone();
attrs.extend::<Vec<_>>(ex.attrs.into());
ex.attrs = attrs;
return ex;
}
2016-12-26 14:34:03 +01:00
ExprKind::Yield(ref opt_expr) => {
self.is_generator = true;
2016-12-26 14:34:03 +01:00
let expr = opt_expr.as_ref().map(|x| self.lower_expr(x)).unwrap_or_else(|| {
self.expr(e.span, hir::ExprTup(hir_vec![]), ThinVec::new())
});
2017-07-10 21:11:31 +02:00
hir::ExprYield(P(expr))
2016-12-26 14:34:03 +01:00
}
// Desugar ExprIfLet
// From: `if let <pat> = <sub_expr> <body> [<else_opt>]`
ExprKind::IfLet(ref pat, ref sub_expr, ref body, ref else_opt) => {
// to:
//
// match <sub_expr> {
// <pat> => <body>,
// _ => [<else_opt> | ()]
// }
let mut arms = vec![];
// `<pat> => <body>`
{
let body = self.lower_block(body, false);
let body_expr = P(self.expr_block(body, ThinVec::new()));
let pat = self.lower_pat(pat);
arms.push(self.arm(hir_vec![pat], body_expr));
}
// _ => [<else_opt>|()]
{
let wildcard_arm: Option<&Expr> = else_opt.as_ref().map(|p| &**p);
let wildcard_pattern = self.pat_wild(e.span);
let body = if let Some(else_expr) = wildcard_arm {
2017-04-15 17:40:54 +05:30
P(self.lower_expr(else_expr))
2017-04-15 17:35:30 +05:30
} else {
2017-04-15 17:40:54 +05:30
self.expr_tuple(e.span, hir_vec![])
2017-04-15 17:35:30 +05:30
};
arms.push(self.arm(hir_vec![wildcard_pattern], body));
}
2015-09-28 17:24:42 +13:00
let contains_else_clause = else_opt.is_some();
2015-09-28 17:24:42 +13:00
let sub_expr = P(self.lower_expr(sub_expr));
2015-09-28 15:00:15 +13:00
hir::ExprMatch(
sub_expr,
arms.into(),
hir::MatchSource::IfLetDesugar {
contains_else_clause,
})
}
2015-09-28 15:00:15 +13:00
// Desugar ExprWhileLet
// From: `[opt_ident]: while let <pat> = <sub_expr> <body>`
ExprKind::WhileLet(ref pat, ref sub_expr, ref body, opt_ident) => {
// to:
//
// [opt_ident]: loop {
// match <sub_expr> {
// <pat> => <body>,
// _ => break
// }
// }
// Note that the block AND the condition are evaluated in the loop scope.
// This is done to allow `break` from inside the condition of the loop.
let (body, break_expr, sub_expr) = self.with_loop_scope(e.id, |this| (
this.lower_block(body, false),
this.expr_break(e.span, ThinVec::new()),
this.with_loop_condition_scope(|this| P(this.lower_expr(sub_expr))),
));
// `<pat> => <body>`
let pat_arm = {
let body_expr = P(self.expr_block(body, ThinVec::new()));
let pat = self.lower_pat(pat);
self.arm(hir_vec![pat], body_expr)
};
2015-09-28 15:00:15 +13:00
// `_ => break`
let break_arm = {
let pat_under = self.pat_wild(e.span);
self.arm(hir_vec![pat_under], break_expr)
};
2016-05-02 00:02:38 +00:00
// `match <sub_expr> { ... }`
let arms = hir_vec![pat_arm, break_arm];
let match_expr = self.expr(e.span,
hir::ExprMatch(sub_expr,
arms,
hir::MatchSource::WhileLetDesugar),
ThinVec::new());
// `[opt_ident]: loop { ... }`
let loop_block = P(self.block_expr(P(match_expr)));
let loop_expr = hir::ExprLoop(loop_block, self.lower_opt_sp_ident(opt_ident),
hir::LoopSource::WhileLet);
// add attributes to the outer returned expr node
loop_expr
}
// Desugar ExprForLoop
// From: `[opt_ident]: for <pat> in <head> <body>`
ExprKind::ForLoop(ref pat, ref head, ref body, opt_ident) => {
// to:
//
// {
// let result = match ::std::iter::IntoIterator::into_iter(<head>) {
// mut iter => {
// [opt_ident]: loop {
// let mut __next;
// match ::std::iter::Iterator::next(&mut iter) {
// ::std::option::Option::Some(val) => __next = val,
// ::std::option::Option::None => break
// };
// let <pat> = __next;
// StmtExpr(<body>);
// }
// }
// };
// result
// }
// expand <head>
let head = self.lower_expr(head);
let iter = self.str_to_ident("iter");
let next_ident = self.str_to_ident("__next");
2017-06-17 14:46:37 +02:00
let next_pat = self.pat_ident_binding_mode(e.span,
next_ident,
hir::BindingAnnotation::Mutable);
// `::std::option::Option::Some(val) => next = val`
let pat_arm = {
let val_ident = self.str_to_ident("val");
let val_pat = self.pat_ident(e.span, val_ident);
let val_expr = P(self.expr_ident(e.span, val_ident, val_pat.id));
let next_expr = P(self.expr_ident(e.span, next_ident, next_pat.id));
let assign = P(self.expr(e.span,
hir::ExprAssign(next_expr, val_expr),
ThinVec::new()));
let some_pat = self.pat_some(e.span, val_pat);
self.arm(hir_vec![some_pat], assign)
};
2016-05-02 00:02:38 +00:00
// `::std::option::Option::None => break`
let break_arm = {
let break_expr = self.with_loop_scope(e.id, |this|
this.expr_break(e.span, ThinVec::new()));
let pat = self.pat_none(e.span);
self.arm(hir_vec![pat], break_expr)
};
2016-05-02 00:02:38 +00:00
// `mut iter`
let iter_pat = self.pat_ident_binding_mode(e.span,
iter,
hir::BindingAnnotation::Mutable);
// `match ::std::iter::Iterator::next(&mut iter) { ... }`
let match_expr = {
let iter = P(self.expr_ident(e.span, iter, iter_pat.id));
let ref_mut_iter = self.expr_mut_addr_of(e.span, iter);
let next_path = &["iter", "Iterator", "next"];
let next_path = P(self.expr_std_path(e.span, next_path, ThinVec::new()));
let next_expr = P(self.expr_call(e.span, next_path,
hir_vec![ref_mut_iter]));
let arms = hir_vec![pat_arm, break_arm];
P(self.expr(e.span,
hir::ExprMatch(next_expr, arms,
hir::MatchSource::ForLoopDesugar),
ThinVec::new()))
};
let match_stmt = respan(e.span, hir::StmtExpr(match_expr, self.next_id().node_id));
let next_expr = P(self.expr_ident(e.span, next_ident, next_pat.id));
// `let mut __next`
let next_let = self.stmt_let_pat(e.span,
None,
next_pat,
hir::LocalSource::ForLoopDesugar);
// `let <pat> = __next`
let pat = self.lower_pat(pat);
let pat_let = self.stmt_let_pat(e.span,
Some(next_expr),
pat,
hir::LocalSource::ForLoopDesugar);
let body_block = self.with_loop_scope(e.id,
|this| this.lower_block(body, false));
let body_expr = P(self.expr_block(body_block, ThinVec::new()));
let body_stmt = respan(e.span, hir::StmtExpr(body_expr, self.next_id().node_id));
let loop_block = P(self.block_all(e.span,
hir_vec![next_let,
match_stmt,
pat_let,
body_stmt],
None));
// `[opt_ident]: loop { ... }`
let loop_expr = hir::ExprLoop(loop_block, self.lower_opt_sp_ident(opt_ident),
hir::LoopSource::ForLoop);
let LoweredNodeId { node_id, hir_id } = self.lower_node_id(e.id);
let loop_expr = P(hir::Expr {
id: node_id,
hir_id,
node: loop_expr,
span: e.span,
attrs: ThinVec::new(),
});
2015-09-28 15:00:15 +13:00
// `mut iter => { ... }`
let iter_arm = self.arm(hir_vec![iter_pat], loop_expr);
// `match ::std::iter::IntoIterator::into_iter(<head>) { ... }`
let into_iter_expr = {
let into_iter_path = &["iter", "IntoIterator", "into_iter"];
let into_iter = P(self.expr_std_path(e.span, into_iter_path,
ThinVec::new()));
P(self.expr_call(e.span, into_iter, hir_vec![head]))
};
let match_expr = P(self.expr_match(e.span,
into_iter_expr,
hir_vec![iter_arm],
hir::MatchSource::ForLoopDesugar));
// `{ let _result = ...; _result }`
// underscore prevents an unused_variables lint if the head diverges
let result_ident = self.str_to_ident("_result");
let (let_stmt, let_stmt_binding) =
self.stmt_let(e.span, false, result_ident, match_expr);
let result = P(self.expr_ident(e.span, result_ident, let_stmt_binding));
let block = P(self.block_all(e.span, hir_vec![let_stmt], Some(result)));
// add the attributes to the outer returned expr node
return self.expr_block(block, e.attrs.clone());
}
// Desugar ExprKind::Try
// From: `<expr>?`
ExprKind::Try(ref sub_expr) => {
// to:
//
// match Try::into_result(<expr>) {
// Ok(val) => #[allow(unreachable_code)] val,
// Err(err) => #[allow(unreachable_code)]
// // If there is an enclosing `catch {...}`
// break 'catch_target Try::from_error(From::from(err)),
// // Otherwise
// return Try::from_error(From::from(err)),
// }
let unstable_span =
self.allow_internal_unstable(CompilerDesugaringKind::QuestionMark, e.span);
// Try::into_result(<expr>)
let discr = {
// expand <expr>
let sub_expr = self.lower_expr(sub_expr);
let path = &["ops", "Try", "into_result"];
let path = P(self.expr_std_path(unstable_span, path, ThinVec::new()));
P(self.expr_call(e.span, path, hir_vec![sub_expr]))
};
// #[allow(unreachable_code)]
let attr = {
// allow(unreachable_code)
let allow = {
let allow_ident = self.str_to_ident("allow");
let uc_ident = self.str_to_ident("unreachable_code");
let uc_meta_item = attr::mk_spanned_word_item(e.span, uc_ident);
let uc_nested = NestedMetaItemKind::MetaItem(uc_meta_item);
let uc_spanned = respan(e.span, uc_nested);
attr::mk_spanned_list_item(e.span, allow_ident, vec![uc_spanned])
};
attr::mk_spanned_attr_outer(e.span, attr::mk_attr_id(), allow)
};
let attrs = vec![attr];
// Ok(val) => #[allow(unreachable_code)] val,
let ok_arm = {
let val_ident = self.str_to_ident("val");
let val_pat = self.pat_ident(e.span, val_ident);
let val_expr = P(self.expr_ident_with_attrs(e.span,
val_ident,
val_pat.id,
ThinVec::from(attrs.clone())));
let ok_pat = self.pat_ok(e.span, val_pat);
self.arm(hir_vec![ok_pat], val_expr)
};
// Err(err) => #[allow(unreachable_code)]
// return Try::from_error(From::from(err)),
let err_arm = {
let err_ident = self.str_to_ident("err");
let err_local = self.pat_ident(e.span, err_ident);
let from_expr = {
let path = &["convert", "From", "from"];
let from = P(self.expr_std_path(e.span, path, ThinVec::new()));
let err_expr = self.expr_ident(e.span, err_ident, err_local.id);
self.expr_call(e.span, from, hir_vec![err_expr])
};
let from_err_expr = {
let path = &["ops", "Try", "from_error"];
let from_err = P(self.expr_std_path(unstable_span, path,
ThinVec::new()));
P(self.expr_call(e.span, from_err, hir_vec![from_expr]))
};
let thin_attrs = ThinVec::from(attrs);
let catch_scope = self.catch_scopes.last().map(|x| *x);
let ret_expr = if let Some(catch_node) = catch_scope {
P(self.expr(
e.span,
hir::ExprBreak(
hir::Destination {
ident: None,
target_id: hir::ScopeTarget::Block(catch_node),
},
Some(from_err_expr)
),
thin_attrs))
} else {
P(self.expr(e.span,
hir::Expr_::ExprRet(Some(from_err_expr)),
thin_attrs))
};
let err_pat = self.pat_err(e.span, err_local);
self.arm(hir_vec![err_pat], ret_expr)
};
2016-05-02 00:02:38 +00:00
hir::ExprMatch(discr,
hir_vec![err_arm, ok_arm],
hir::MatchSource::TryDesugar)
}
ExprKind::Mac(_) => panic!("Shouldn't exist here"),
};
let LoweredNodeId { node_id, hir_id } = self.lower_node_id(e.id);
hir::Expr {
id: node_id,
hir_id,
node: kind,
span: e.span,
attrs: e.attrs.clone(),
}
}
2015-07-31 00:04:06 -07:00
fn lower_stmt(&mut self, s: &Stmt) -> SmallVector<hir::Stmt> {
SmallVector::one(match s.node {
2016-06-17 02:30:01 +00:00
StmtKind::Local(ref l) => Spanned {
node: hir::StmtDecl(P(Spanned {
node: hir::DeclLocal(self.lower_local(l)),
span: s.span,
}), self.lower_node_id(s.id).node_id),
2016-06-17 02:30:01 +00:00
span: s.span,
},
StmtKind::Item(ref it) => {
// Can only use the ID once.
let mut id = Some(s.id);
return self.lower_item_id(it).into_iter().map(|item_id| Spanned {
node: hir::StmtDecl(P(Spanned {
node: hir::DeclItem(item_id),
span: s.span,
}), id.take()
.map(|id| self.lower_node_id(id).node_id)
.unwrap_or_else(|| self.next_id().node_id)),
2016-06-17 02:30:01 +00:00
span: s.span,
}).collect();
}
2016-06-17 02:30:01 +00:00
StmtKind::Expr(ref e) => {
Spanned {
node: hir::StmtExpr(P(self.lower_expr(e)),
self.lower_node_id(s.id).node_id),
span: s.span,
}
}
2016-06-17 02:30:01 +00:00
StmtKind::Semi(ref e) => {
Spanned {
node: hir::StmtSemi(P(self.lower_expr(e)),
self.lower_node_id(s.id).node_id),
span: s.span,
}
}
StmtKind::Mac(..) => panic!("Shouldn't exist here"),
})
2015-07-31 00:04:06 -07:00
}
fn lower_capture_clause(&mut self, c: CaptureBy) -> hir::CaptureClause {
match c {
CaptureBy::Value => hir::CaptureByValue,
CaptureBy::Ref => hir::CaptureByRef,
}
2015-07-31 00:04:06 -07:00
}
/// If an `explicit_owner` is given, this method allocates the `HirId` in
/// the address space of that item instead of the item currently being
/// lowered. This can happen during `lower_impl_item_ref()` where we need to
/// lower a `Visibility` value although we haven't lowered the owning
/// `ImplItem` in question yet.
fn lower_visibility(&mut self,
v: &Visibility,
explicit_owner: Option<NodeId>)
-> hir::Visibility {
match *v {
Visibility::Public => hir::Public,
Visibility::Crate(..) => hir::Visibility::Crate,
Visibility::Restricted { ref path, id } => {
hir::Visibility::Restricted {
path: P(self.lower_path(id, path, ParamMode::Explicit, true)),
id: if let Some(owner) = explicit_owner {
self.lower_node_id_with_owner(id, owner).node_id
} else {
self.lower_node_id(id).node_id
}
}
}
Visibility::Inherited => hir::Inherited,
}
2015-07-31 00:04:06 -07:00
}
fn lower_defaultness(&mut self, d: Defaultness, has_value: bool) -> hir::Defaultness {
match d {
Defaultness::Default => hir::Defaultness::Default { has_value: has_value },
Defaultness::Final => {
assert!(has_value);
hir::Defaultness::Final
}
}
}
fn lower_block_check_mode(&mut self, b: &BlockCheckMode) -> hir::BlockCheckMode {
match *b {
BlockCheckMode::Default => hir::DefaultBlock,
BlockCheckMode::Unsafe(u) => hir::UnsafeBlock(self.lower_unsafe_source(u)),
}
2015-07-31 00:04:06 -07:00
}
fn lower_binding_mode(&mut self, b: &BindingMode) -> hir::BindingAnnotation {
match *b {
BindingMode::ByValue(Mutability::Immutable) =>
hir::BindingAnnotation::Unannotated,
BindingMode::ByRef(Mutability::Immutable) => hir::BindingAnnotation::Ref,
BindingMode::ByValue(Mutability::Mutable) => hir::BindingAnnotation::Mutable,
BindingMode::ByRef(Mutability::Mutable) => hir::BindingAnnotation::RefMut,
}
2015-07-31 00:04:06 -07:00
}
fn lower_unsafe_source(&mut self, u: UnsafeSource) -> hir::UnsafeSource {
match u {
CompilerGenerated => hir::CompilerGenerated,
UserProvided => hir::UserProvided,
}
2015-07-31 00:04:06 -07:00
}
fn lower_impl_polarity(&mut self, i: ImplPolarity) -> hir::ImplPolarity {
match i {
ImplPolarity::Positive => hir::ImplPolarity::Positive,
ImplPolarity::Negative => hir::ImplPolarity::Negative,
}
2015-07-31 00:04:06 -07:00
}
fn lower_trait_bound_modifier(&mut self, f: TraitBoundModifier) -> hir::TraitBoundModifier {
match f {
TraitBoundModifier::None => hir::TraitBoundModifier::None,
TraitBoundModifier::Maybe => hir::TraitBoundModifier::Maybe,
}
2015-07-31 00:04:06 -07:00
}
2015-09-28 15:00:15 +13:00
// Helper methods for building HIR.
2015-09-28 15:00:15 +13:00
fn arm(&mut self, pats: hir::HirVec<P<hir::Pat>>, expr: P<hir::Expr>) -> hir::Arm {
hir::Arm {
attrs: hir_vec![],
pats,
guard: None,
body: expr,
}
2015-09-28 15:00:15 +13:00
}
fn field(&mut self, name: Name, expr: P<hir::Expr>, span: Span) -> hir::Field {
hir::Field {
name: Spanned {
node: name,
span,
},
span,
expr,
is_shorthand: false,
}
}
fn expr_break(&mut self, span: Span, attrs: ThinVec<Attribute>) -> P<hir::Expr> {
let expr_break = hir::ExprBreak(self.lower_loop_destination(None), None);
2017-02-15 14:52:27 -08:00
P(self.expr(span, expr_break, attrs))
}
2015-09-28 15:00:15 +13:00
fn expr_call(&mut self, span: Span, e: P<hir::Expr>, args: hir::HirVec<hir::Expr>)
-> hir::Expr {
self.expr(span, hir::ExprCall(e, args), ThinVec::new())
}
2015-09-28 15:00:15 +13:00
fn expr_ident(&mut self, span: Span, id: Name, binding: NodeId) -> hir::Expr {
self.expr_ident_with_attrs(span, id, binding, ThinVec::new())
}
fn expr_ident_with_attrs(&mut self, span: Span,
id: Name,
binding: NodeId,
attrs: ThinVec<Attribute>) -> hir::Expr {
let expr_path = hir::ExprPath(hir::QPath::Resolved(None, P(hir::Path {
span,
def: Def::Local(binding),
segments: hir_vec![hir::PathSegment::from_name(id)],
})));
2016-05-06 08:24:04 +00:00
self.expr(span, expr_path, attrs)
}
2015-09-28 15:00:15 +13:00
fn expr_mut_addr_of(&mut self, span: Span, e: P<hir::Expr>) -> hir::Expr {
self.expr(span, hir::ExprAddrOf(hir::MutMutable, e), ThinVec::new())
}
2015-09-28 15:00:15 +13:00
fn expr_std_path(&mut self,
span: Span,
components: &[&str],
attrs: ThinVec<Attribute>)
-> hir::Expr {
let path = self.std_path(span, components, true);
2016-11-29 00:33:59 +00:00
self.expr(span, hir::ExprPath(hir::QPath::Resolved(None, P(path))), attrs)
}
2015-09-28 15:00:15 +13:00
fn expr_match(&mut self,
span: Span,
arg: P<hir::Expr>,
arms: hir::HirVec<hir::Arm>,
source: hir::MatchSource)
-> hir::Expr {
self.expr(span, hir::ExprMatch(arg, arms, source), ThinVec::new())
}
2015-09-28 15:00:15 +13:00
fn expr_block(&mut self, b: P<hir::Block>, attrs: ThinVec<Attribute>) -> hir::Expr {
self.expr(b.span, hir::ExprBlock(b), attrs)
}
2015-09-28 15:00:15 +13:00
fn expr_tuple(&mut self, sp: Span, exprs: hir::HirVec<hir::Expr>) -> P<hir::Expr> {
P(self.expr(sp, hir::ExprTup(exprs), ThinVec::new()))
}
2015-09-28 17:24:42 +13:00
fn expr(&mut self, span: Span, node: hir::Expr_, attrs: ThinVec<Attribute>) -> hir::Expr {
let LoweredNodeId { node_id, hir_id } = self.next_id();
hir::Expr {
id: node_id,
hir_id,
node,
span,
attrs,
}
}
fn stmt_let_pat(&mut self,
sp: Span,
ex: Option<P<hir::Expr>>,
pat: P<hir::Pat>,
source: hir::LocalSource)
-> hir::Stmt {
let LoweredNodeId { node_id, hir_id } = self.next_id();
let local = P(hir::Local {
pat,
ty: None,
init: ex,
id: node_id,
hir_id,
span: sp,
attrs: ThinVec::new(),
source,
});
let decl = respan(sp, hir::DeclLocal(local));
respan(sp, hir::StmtDecl(P(decl), self.next_id().node_id))
}
fn stmt_let(&mut self, sp: Span, mutbl: bool, ident: Name, ex: P<hir::Expr>)
-> (hir::Stmt, NodeId) {
let pat = if mutbl {
self.pat_ident_binding_mode(sp, ident, hir::BindingAnnotation::Mutable)
} else {
self.pat_ident(sp, ident)
};
let pat_id = pat.id;
(self.stmt_let_pat(sp, Some(ex), pat, hir::LocalSource::Normal), pat_id)
}
2015-09-28 15:00:15 +13:00
fn block_expr(&mut self, expr: P<hir::Expr>) -> hir::Block {
self.block_all(expr.span, hir::HirVec::new(), Some(expr))
}
2015-09-28 15:00:15 +13:00
fn block_all(&mut self, span: Span, stmts: hir::HirVec<hir::Stmt>, expr: Option<P<hir::Expr>>)
-> hir::Block {
let LoweredNodeId { node_id, hir_id } = self.next_id();
hir::Block {
stmts,
expr,
id: node_id,
hir_id,
rules: hir::DefaultBlock,
span,
targeted_by_break: false,
}
}
2015-09-28 15:00:15 +13:00
fn pat_ok(&mut self, span: Span, pat: P<hir::Pat>) -> P<hir::Pat> {
self.pat_std_enum(span, &["result", "Result", "Ok"], hir_vec![pat])
}
2015-09-28 15:00:15 +13:00
fn pat_err(&mut self, span: Span, pat: P<hir::Pat>) -> P<hir::Pat> {
self.pat_std_enum(span, &["result", "Result", "Err"], hir_vec![pat])
}
fn pat_some(&mut self, span: Span, pat: P<hir::Pat>) -> P<hir::Pat> {
self.pat_std_enum(span, &["option", "Option", "Some"], hir_vec![pat])
}
fn pat_none(&mut self, span: Span) -> P<hir::Pat> {
self.pat_std_enum(span, &["option", "Option", "None"], hir_vec![])
}
2015-09-28 15:00:15 +13:00
fn pat_std_enum(&mut self,
span: Span,
components: &[&str],
subpats: hir::HirVec<P<hir::Pat>>)
-> P<hir::Pat> {
let path = self.std_path(span, components, true);
let qpath = hir::QPath::Resolved(None, P(path));
let pt = if subpats.is_empty() {
hir::PatKind::Path(qpath)
} else {
hir::PatKind::TupleStruct(qpath, subpats, None)
};
2016-11-29 00:33:59 +00:00
self.pat(span, pt)
}
2015-09-28 15:00:15 +13:00
2016-03-06 15:54:44 +03:00
fn pat_ident(&mut self, span: Span, name: Name) -> P<hir::Pat> {
self.pat_ident_binding_mode(span, name, hir::BindingAnnotation::Unannotated)
}
2015-09-28 15:00:15 +13:00
fn pat_ident_binding_mode(&mut self, span: Span, name: Name, bm: hir::BindingAnnotation)
-> P<hir::Pat> {
let LoweredNodeId { node_id, hir_id } = self.next_id();
2015-09-28 17:24:42 +13:00
P(hir::Pat {
id: node_id,
hir_id,
node: hir::PatKind::Binding(bm,
node_id,
Spanned {
span,
node: name,
},
None),
span,
})
}
2015-09-28 15:00:15 +13:00
fn pat_wild(&mut self, span: Span) -> P<hir::Pat> {
self.pat(span, hir::PatKind::Wild)
}
2015-09-28 15:00:15 +13:00
fn pat(&mut self, span: Span, pat: hir::PatKind) -> P<hir::Pat> {
let LoweredNodeId { node_id, hir_id } = self.next_id();
P(hir::Pat {
id: node_id,
hir_id,
node: pat,
span,
})
}
2015-09-28 15:00:15 +13:00
/// Given suffix ["b","c","d"], returns path `::std::b::c::d` when
/// `fld.cx.use_std`, and `::core::b::c::d` otherwise.
/// The path is also resolved according to `is_value`.
fn std_path(&mut self, span: Span, components: &[&str], is_value: bool) -> hir::Path {
let mut path = hir::Path {
span,
def: Def::Err,
segments: iter::once(keywords::CrateRoot.name()).chain({
self.crate_root.into_iter().chain(components.iter().cloned()).map(Symbol::intern)
}).map(hir::PathSegment::from_name).collect(),
};
2016-11-25 06:07:21 +00:00
self.resolver.resolve_hir_path(&mut path, is_value);
path
}
fn signal_block_expr(&mut self,
stmts: hir::HirVec<hir::Stmt>,
expr: P<hir::Expr>,
span: Span,
rule: hir::BlockCheckMode,
attrs: ThinVec<Attribute>)
-> hir::Expr {
let LoweredNodeId { node_id, hir_id } = self.next_id();
let block = P(hir::Block {
rules: rule,
span,
id: node_id,
hir_id,
stmts,
expr: Some(expr),
targeted_by_break: false,
});
self.expr_block(block, attrs)
}
2017-09-13 18:29:59 +03:00
fn ty_path(&mut self, id: LoweredNodeId, span: Span, qpath: hir::QPath) -> P<hir::Ty> {
let mut id = id;
let node = match qpath {
hir::QPath::Resolved(None, path) => {
// Turn trait object paths into `TyTraitObject` instead.
if let Def::Trait(_) = path.def {
let principal = hir::PolyTraitRef {
bound_lifetimes: hir_vec![],
trait_ref: hir::TraitRef {
path: path.and_then(|path| path),
2017-09-13 18:29:59 +03:00
ref_id: id.node_id,
},
span,
};
// The original ID is taken by the `PolyTraitRef`,
// so the `Ty` itself needs a different one.
2017-09-13 18:29:59 +03:00
id = self.next_id();
hir::TyTraitObject(hir_vec![principal], self.elided_lifetime(span))
} else {
hir::TyPath(hir::QPath::Resolved(None, path))
}
}
_ => hir::TyPath(qpath)
};
2017-09-13 18:29:59 +03:00
P(hir::Ty { id: id.node_id, hir_id: id.hir_id, node, span })
}
fn elided_lifetime(&mut self, span: Span) -> hir::Lifetime {
hir::Lifetime {
id: self.next_id().node_id,
span,
2017-09-19 16:36:54 -07:00
name: hir::LifetimeName::Implicit,
}
}
}
2017-02-21 12:23:47 -05:00
fn body_ids(bodies: &BTreeMap<hir::BodyId, hir::Body>) -> Vec<hir::BodyId> {
// Sorting by span ensures that we get things in order within a
// file, and also puts the files in a sensible order.
let mut body_ids: Vec<_> = bodies.keys().cloned().collect();
body_ids.sort_by_key(|b| bodies[b].value.span);
body_ids
}