mentat/core/src/intern_set.rs

69 lines
2.3 KiB
Rust
Raw Normal View History

Implement upsert resolution algorithm. (#186, #283). r=rnewman, f=jsantell * Pre: Implement batch [a v] pair lookup. * Pre: Add InternSet for sharing ref-counted handles to large values. * Pre: Derive more for Entity. * Pre: Return DB from creating; return TxReport from transact. I explicitly am not supporting opening existing databases yet, let alone upgrading databases from earlier versions. That can follow fast once basic transactions are supported. * Pre: Parse string temporary ID entities; remove ValueOrLookupRef. This adds TempId entities, but we can't disambiguate String temporary IDs from values without the use of the schema, so there's no new value branch. Similarly, we can't disambiguate lookup-ref values from two element list values without a schema, so we remove this entirely. We'll handle the ambiguity later in the transactor. * Persist partitions to SQL store; allocate transaction ID. (#186) * Post: Test upserting with vectors. This converts an existing test to EDN: https://github.com/mozilla/mentat/blob/84a80f40f5c888f8452d07bd15f3b5fba49d3963/test/datomish/db_test.cljc#L193. * Implement tempid upsert resolution algorithm. (#184) * Post: Separate Tx out of DB. This is very preliminary, since we don't have a real connection type to manage transactions and their metadata yet. * Post: Comment on implementation choices in the transactor. * Review comment: Put long use lists on separate lines. * Review comment: Accept String: Borrow<S> instead of just String. * Review comment: Address nits.
2017-02-15 00:50:40 +00:00
// Copyright 2016 Mozilla
//
// Licensed under the Apache License, Version 2.0 (the "License"); you may not use
// this file except in compliance with the License. You may obtain a copy of the
// License at http://www.apache.org/licenses/LICENSE-2.0
// Unless required by applicable law or agreed to in writing, software distributed
// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR
// CONDITIONS OF ANY KIND, either express or implied. See the License for the
// specific language governing permissions and limitations under the License.
#![allow(dead_code)]
use std::collections::HashSet;
use std::hash::Hash;
use std::rc::Rc;
/// An `InternSet` allows to "intern" some potentially large values, maintaining a single value
/// instance owned by the `InternSet` and leaving consumers with lightweight ref-counted handles to
/// the large owned value. This can avoid expensive clone() operations.
///
/// In Mentat, such large values might be strings or arbitrary [a v] pairs.
///
/// See https://en.wikipedia.org/wiki/String_interning for discussion.
#[derive(Clone, Debug, Default, Eq, PartialEq)]
pub struct InternSet<T> where T: Eq + Hash {
pub inner: HashSet<Rc<T>>,
}
impl<T> InternSet<T> where T: Eq + Hash {
pub fn new() -> InternSet<T> {
InternSet {
inner: HashSet::new(),
}
}
pub fn len(&self) -> usize {
self.inner.len()
}
Implement upsert resolution algorithm. (#186, #283). r=rnewman, f=jsantell * Pre: Implement batch [a v] pair lookup. * Pre: Add InternSet for sharing ref-counted handles to large values. * Pre: Derive more for Entity. * Pre: Return DB from creating; return TxReport from transact. I explicitly am not supporting opening existing databases yet, let alone upgrading databases from earlier versions. That can follow fast once basic transactions are supported. * Pre: Parse string temporary ID entities; remove ValueOrLookupRef. This adds TempId entities, but we can't disambiguate String temporary IDs from values without the use of the schema, so there's no new value branch. Similarly, we can't disambiguate lookup-ref values from two element list values without a schema, so we remove this entirely. We'll handle the ambiguity later in the transactor. * Persist partitions to SQL store; allocate transaction ID. (#186) * Post: Test upserting with vectors. This converts an existing test to EDN: https://github.com/mozilla/mentat/blob/84a80f40f5c888f8452d07bd15f3b5fba49d3963/test/datomish/db_test.cljc#L193. * Implement tempid upsert resolution algorithm. (#184) * Post: Separate Tx out of DB. This is very preliminary, since we don't have a real connection type to manage transactions and their metadata yet. * Post: Comment on implementation choices in the transactor. * Review comment: Put long use lists on separate lines. * Review comment: Accept String: Borrow<S> instead of just String. * Review comment: Address nits.
2017-02-15 00:50:40 +00:00
/// Intern a value, providing a ref-counted handle to the interned value.
///
/// ```
/// use std::rc::Rc;
/// use mentat_core::intern_set::InternSet;
///
/// let mut s = InternSet::new();
///
/// let one = "foo".to_string();
/// let two = Rc::new("foo".to_string());
///
/// let out_one = s.intern(one);
/// assert_eq!(out_one, two);
/// // assert!(!&out_one.ptr_eq(&two)); // Nightly-only.
///
/// let out_two = s.intern(two);
/// assert_eq!(out_one, out_two);
/// assert_eq!(1, s.inner.len());
/// // assert!(&out_one.ptr_eq(&out_two)); // Nightly-only.
/// ```
pub fn intern<R: Into<Rc<T>>>(&mut self, value: R) -> Rc<T> {
let key: Rc<T> = value.into();
Implement upsert resolution algorithm. (#186, #283). r=rnewman, f=jsantell * Pre: Implement batch [a v] pair lookup. * Pre: Add InternSet for sharing ref-counted handles to large values. * Pre: Derive more for Entity. * Pre: Return DB from creating; return TxReport from transact. I explicitly am not supporting opening existing databases yet, let alone upgrading databases from earlier versions. That can follow fast once basic transactions are supported. * Pre: Parse string temporary ID entities; remove ValueOrLookupRef. This adds TempId entities, but we can't disambiguate String temporary IDs from values without the use of the schema, so there's no new value branch. Similarly, we can't disambiguate lookup-ref values from two element list values without a schema, so we remove this entirely. We'll handle the ambiguity later in the transactor. * Persist partitions to SQL store; allocate transaction ID. (#186) * Post: Test upserting with vectors. This converts an existing test to EDN: https://github.com/mozilla/mentat/blob/84a80f40f5c888f8452d07bd15f3b5fba49d3963/test/datomish/db_test.cljc#L193. * Implement tempid upsert resolution algorithm. (#184) * Post: Separate Tx out of DB. This is very preliminary, since we don't have a real connection type to manage transactions and their metadata yet. * Post: Comment on implementation choices in the transactor. * Review comment: Put long use lists on separate lines. * Review comment: Accept String: Borrow<S> instead of just String. * Review comment: Address nits.
2017-02-15 00:50:40 +00:00
if self.inner.insert(key.clone()) {
key
} else {
self.inner.get(&key).unwrap().clone()
}
}
}