use std::fmt; use std::hash::Hash; use std::marker::PhantomData; use fxhash::FxHashMap; /// An index within a [`Interner`] structure. pub struct Interned { idx: u16, _phantom: PhantomData, } impl Interned { pub fn new(idx: u16) -> Self { Self { idx, _phantom: PhantomData } } pub fn into_inner(self) -> u16 { self.idx } } // TODO: the stable store should be replaced by a bump allocator // and the interned value should be a pointer wrapper // then we can get its value with `interned.get()` instead of `interner.get(interned)` // and as a bonus, its validity is tracked with Rust's lifetime system // one problem is that we need two lifetimes: one for the bump allocator, one for the // hashmap // but that's okay, we can use: // ``` // struct Interner<'bump> { // bump: &'bump Bump, // lookup: FxHashMap // } // ``` /// An [`Interner`] is used to store a unique copy of a value of type `T`. This value /// is then identified by a lightweight index of type [`Interned`], which can /// be copied, compared, and hashed efficiently. An immutable reference to the original value /// can be retrieved using `self.get(interned)`. #[derive(Clone)] pub struct DedupInterner { stable_store: Vec, lookup: FxHashMap>, } impl Default for DedupInterner { fn default() -> Self { Self { stable_store: Default::default(), lookup: Default::default() } } } impl DedupInterner { pub fn freeze(self) -> FixedSizeInterner { FixedSizeInterner { stable_store: self.stable_store } } } impl DedupInterner where T: Clone + Eq + Hash, { pub fn insert(&mut self, s: T) -> Interned { if let Some(interned) = self.lookup.get(&s) { *interned } else { assert!(self.stable_store.len() < u16::MAX as usize); self.stable_store.push(s.clone()); let interned = Interned::new(self.stable_store.len() as u16 - 1); self.lookup.insert(s, interned); interned } } pub fn get(&self, interned: Interned) -> &T { &self.stable_store[interned.idx as usize] } } #[derive(Clone)] pub struct Interner { stable_store: Vec, } impl Default for Interner { fn default() -> Self { Self { stable_store: Default::default() } } } impl Interner { pub fn freeze(self) -> FixedSizeInterner { FixedSizeInterner { stable_store: self.stable_store } } pub fn push(&mut self, s: T) -> Interned { assert!(self.stable_store.len() < u16::MAX as usize); self.stable_store.push(s); Interned::new(self.stable_store.len() as u16 - 1) } } #[derive(Clone)] pub struct FixedSizeInterner { stable_store: Vec, } impl FixedSizeInterner { pub fn new(length: u16, value: T) -> Self { Self { stable_store: vec![value; length as usize] } } } impl FixedSizeInterner { pub fn from_vec(store: Vec) -> Self { Self { stable_store: store } } pub fn get(&self, interned: Interned) -> &T { &self.stable_store[interned.idx as usize] } pub fn get_mut(&mut self, interned: Interned) -> &mut T { &mut self.stable_store[interned.idx as usize] } pub fn len(&self) -> u16 { self.stable_store.len() as u16 } pub fn map(&self, map_f: impl Fn(&T) -> U) -> MappedInterner { MappedInterner { stable_store: self.stable_store.iter().map(map_f).collect(), _phantom: PhantomData, } } pub fn indexes(&self) -> impl Iterator> { (0..self.stable_store.len()).map(|i| Interned::new(i as u16)) } pub fn iter(&self) -> impl Iterator, &T)> { self.stable_store.iter().enumerate().map(|(i, x)| (Interned::new(i as u16), x)) } pub fn iter_mut(&mut self) -> impl Iterator, &mut T)> { self.stable_store.iter_mut().enumerate().map(|(i, x)| (Interned::new(i as u16), x)) } } #[derive(Clone)] pub struct MappedInterner { stable_store: Vec, _phantom: PhantomData, } impl MappedInterner { pub fn get(&self, interned: Interned) -> &T { &self.stable_store[interned.idx as usize] } pub fn get_mut(&mut self, interned: Interned) -> &mut T { &mut self.stable_store[interned.idx as usize] } pub fn map(&self, map_f: impl Fn(&T) -> U) -> MappedInterner { MappedInterner { stable_store: self.stable_store.iter().map(map_f).collect(), _phantom: PhantomData, } } pub fn iter(&self) -> impl Iterator, &T)> { self.stable_store.iter().enumerate().map(|(i, x)| (Interned::new(i as u16), x)) } pub fn iter_mut(&mut self) -> impl Iterator, &mut T)> { self.stable_store.iter_mut().enumerate().map(|(i, x)| (Interned::new(i as u16), x)) } } // Interned boilerplate implementations impl Hash for Interned { fn hash(&self, state: &mut H) { self.idx.hash(state); } } impl Ord for Interned { fn cmp(&self, other: &Self) -> std::cmp::Ordering { self.idx.cmp(&other.idx) } } impl PartialOrd for Interned { fn partial_cmp(&self, other: &Self) -> Option { self.idx.partial_cmp(&other.idx) } } impl Eq for Interned {} impl PartialEq for Interned { fn eq(&self, other: &Self) -> bool { self.idx == other.idx } } impl Clone for Interned { fn clone(&self) -> Self { Self { idx: self.idx, _phantom: PhantomData } } } impl Copy for Interned {} impl fmt::Display for Interned { fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result { fmt::Display::fmt(&self.idx, f) } } impl fmt::Debug for Interned { fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result { fmt::Debug::fmt(&self.idx, f) } }