mentat/tx/src/entities.rs

107 lines
3.2 KiB
Rust
Raw Normal View History

// 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.
//! This module defines core types that support the transaction processor.
extern crate edn;
Lookup refs, nested vector values, map notation. Fixes #180, fixes #183, fixes #284. (#382) r=rnewman * Pre: Fix error in parser macros. * Pre: Make test unwrapping more verbose. * Pre: Make lookup refs be (lookup-ref a v) in the entity position. This has the advantage of being explicit in all situations and unambiguous at parse-time. This choice agrees with the Clojure implementation but not with Datomic. Datomic treats [a v] as a lookup ref, is ambiguous at parse-time, and is disambiguated in ways I do not understand at transaction time. We mooted making lookup refs [[a v]] and outlawing nested value vectors in transactions, but after implementing that approach I decided it was better to handle lookup refs at parse time and therefore outlawing nested value vectors is not necessary. * Handle lookup refs in the entity and value columns. Fixes #183. * Pre 0a: Use a stack instead of into_iter. * Pre 0b: Dedent. * Pre 0c: Handle `e` after `v`. This allows to use the original `e` while handling `v`. * Explode value lists for :db.cardinality/many attributes. Fixes #284. * Parse and accept map notation. Fixes #180. * Pre: Modernize add() and retract() into one add_or_retract(). * Pre: Add is_collection and is_atom to edn::Value. * Pre: Differentiate atoms from lookup-refs in value position. Initially, I expected to accept arbitrary edn::Value instances in the value position, and to differentiate in the transactor. However, the implementation quickly became a two-stage parser, since we always wanted to parse the resulting value position into some other known thing using the tx-parser. To save calls into the parser and to allow the parser to move forward with a smaller API surface, I push as much of this parsing as possible into the initial parse. * Pre: Modernize entities(). * Pre: Quote edn::Value::Text in Display. * Review comment: Add and use edn::Value::into_atom. * Review comment: Use skip(eof()) throughout. * Review comment: VecDeque instead of Vec. * Review comment: Part 0: Rename TempId to TempIdHandle. * Review comment: Part 1: Differentiate internal and external tempids. This breaks an abstraction boundary by pushing the Internal/External split up to the Entity level in tx/ and tx-parser/. This just makes it easier to explode Entity map notation instances into Entity instances, taking an existing External tempid :db/id or generating a new Internal tempid as appropriate. To do this without breaking the abstraction boundary would require adding flexibility to the transaction processor: we'd need to be able to turn Entity instances into some internal enum and handle the two cases independently. It wouldn't be too hard, but this reduces the combinatorial type explosion.
2017-03-27 23:30:04 +00:00
use std::collections::BTreeMap;
use std::fmt;
use self::edn::symbols::NamespacedKeyword;
Lookup refs, nested vector values, map notation. Fixes #180, fixes #183, fixes #284. (#382) r=rnewman * Pre: Fix error in parser macros. * Pre: Make test unwrapping more verbose. * Pre: Make lookup refs be (lookup-ref a v) in the entity position. This has the advantage of being explicit in all situations and unambiguous at parse-time. This choice agrees with the Clojure implementation but not with Datomic. Datomic treats [a v] as a lookup ref, is ambiguous at parse-time, and is disambiguated in ways I do not understand at transaction time. We mooted making lookup refs [[a v]] and outlawing nested value vectors in transactions, but after implementing that approach I decided it was better to handle lookup refs at parse time and therefore outlawing nested value vectors is not necessary. * Handle lookup refs in the entity and value columns. Fixes #183. * Pre 0a: Use a stack instead of into_iter. * Pre 0b: Dedent. * Pre 0c: Handle `e` after `v`. This allows to use the original `e` while handling `v`. * Explode value lists for :db.cardinality/many attributes. Fixes #284. * Parse and accept map notation. Fixes #180. * Pre: Modernize add() and retract() into one add_or_retract(). * Pre: Add is_collection and is_atom to edn::Value. * Pre: Differentiate atoms from lookup-refs in value position. Initially, I expected to accept arbitrary edn::Value instances in the value position, and to differentiate in the transactor. However, the implementation quickly became a two-stage parser, since we always wanted to parse the resulting value position into some other known thing using the tx-parser. To save calls into the parser and to allow the parser to move forward with a smaller API surface, I push as much of this parsing as possible into the initial parse. * Pre: Modernize entities(). * Pre: Quote edn::Value::Text in Display. * Review comment: Add and use edn::Value::into_atom. * Review comment: Use skip(eof()) throughout. * Review comment: VecDeque instead of Vec. * Review comment: Part 0: Rename TempId to TempIdHandle. * Review comment: Part 1: Differentiate internal and external tempids. This breaks an abstraction boundary by pushing the Internal/External split up to the Entity level in tx/ and tx-parser/. This just makes it easier to explode Entity map notation instances into Entity instances, taking an existing External tempid :db/id or generating a new Internal tempid as appropriate. To do this without breaking the abstraction boundary would require adding flexibility to the transaction processor: we'd need to be able to turn Entity instances into some internal enum and handle the two cases independently. It wouldn't be too hard, but this reduces the combinatorial type explosion.
2017-03-27 23:30:04 +00:00
/// A tempid, either an external tempid given in a transaction (usually as an `edn::Value::Text`),
/// or an internal tempid allocated by Mentat itself.
#[derive(Clone, Debug, Eq, Hash, Ord, PartialOrd, PartialEq)]
pub enum TempId {
External(String),
Internal(i64),
Tx, // Special identifier used to refer to the current transaction.
Lookup refs, nested vector values, map notation. Fixes #180, fixes #183, fixes #284. (#382) r=rnewman * Pre: Fix error in parser macros. * Pre: Make test unwrapping more verbose. * Pre: Make lookup refs be (lookup-ref a v) in the entity position. This has the advantage of being explicit in all situations and unambiguous at parse-time. This choice agrees with the Clojure implementation but not with Datomic. Datomic treats [a v] as a lookup ref, is ambiguous at parse-time, and is disambiguated in ways I do not understand at transaction time. We mooted making lookup refs [[a v]] and outlawing nested value vectors in transactions, but after implementing that approach I decided it was better to handle lookup refs at parse time and therefore outlawing nested value vectors is not necessary. * Handle lookup refs in the entity and value columns. Fixes #183. * Pre 0a: Use a stack instead of into_iter. * Pre 0b: Dedent. * Pre 0c: Handle `e` after `v`. This allows to use the original `e` while handling `v`. * Explode value lists for :db.cardinality/many attributes. Fixes #284. * Parse and accept map notation. Fixes #180. * Pre: Modernize add() and retract() into one add_or_retract(). * Pre: Add is_collection and is_atom to edn::Value. * Pre: Differentiate atoms from lookup-refs in value position. Initially, I expected to accept arbitrary edn::Value instances in the value position, and to differentiate in the transactor. However, the implementation quickly became a two-stage parser, since we always wanted to parse the resulting value position into some other known thing using the tx-parser. To save calls into the parser and to allow the parser to move forward with a smaller API surface, I push as much of this parsing as possible into the initial parse. * Pre: Modernize entities(). * Pre: Quote edn::Value::Text in Display. * Review comment: Add and use edn::Value::into_atom. * Review comment: Use skip(eof()) throughout. * Review comment: VecDeque instead of Vec. * Review comment: Part 0: Rename TempId to TempIdHandle. * Review comment: Part 1: Differentiate internal and external tempids. This breaks an abstraction boundary by pushing the Internal/External split up to the Entity level in tx/ and tx-parser/. This just makes it easier to explode Entity map notation instances into Entity instances, taking an existing External tempid :db/id or generating a new Internal tempid as appropriate. To do this without breaking the abstraction boundary would require adding flexibility to the transaction processor: we'd need to be able to turn Entity instances into some internal enum and handle the two cases independently. It wouldn't be too hard, but this reduces the combinatorial type explosion.
2017-03-27 23:30:04 +00:00
}
impl TempId {
pub fn into_external(self) -> Option<String> {
match self {
TempId::External(s) => Some(s),
TempId::Tx |
Lookup refs, nested vector values, map notation. Fixes #180, fixes #183, fixes #284. (#382) r=rnewman * Pre: Fix error in parser macros. * Pre: Make test unwrapping more verbose. * Pre: Make lookup refs be (lookup-ref a v) in the entity position. This has the advantage of being explicit in all situations and unambiguous at parse-time. This choice agrees with the Clojure implementation but not with Datomic. Datomic treats [a v] as a lookup ref, is ambiguous at parse-time, and is disambiguated in ways I do not understand at transaction time. We mooted making lookup refs [[a v]] and outlawing nested value vectors in transactions, but after implementing that approach I decided it was better to handle lookup refs at parse time and therefore outlawing nested value vectors is not necessary. * Handle lookup refs in the entity and value columns. Fixes #183. * Pre 0a: Use a stack instead of into_iter. * Pre 0b: Dedent. * Pre 0c: Handle `e` after `v`. This allows to use the original `e` while handling `v`. * Explode value lists for :db.cardinality/many attributes. Fixes #284. * Parse and accept map notation. Fixes #180. * Pre: Modernize add() and retract() into one add_or_retract(). * Pre: Add is_collection and is_atom to edn::Value. * Pre: Differentiate atoms from lookup-refs in value position. Initially, I expected to accept arbitrary edn::Value instances in the value position, and to differentiate in the transactor. However, the implementation quickly became a two-stage parser, since we always wanted to parse the resulting value position into some other known thing using the tx-parser. To save calls into the parser and to allow the parser to move forward with a smaller API surface, I push as much of this parsing as possible into the initial parse. * Pre: Modernize entities(). * Pre: Quote edn::Value::Text in Display. * Review comment: Add and use edn::Value::into_atom. * Review comment: Use skip(eof()) throughout. * Review comment: VecDeque instead of Vec. * Review comment: Part 0: Rename TempId to TempIdHandle. * Review comment: Part 1: Differentiate internal and external tempids. This breaks an abstraction boundary by pushing the Internal/External split up to the Entity level in tx/ and tx-parser/. This just makes it easier to explode Entity map notation instances into Entity instances, taking an existing External tempid :db/id or generating a new Internal tempid as appropriate. To do this without breaking the abstraction boundary would require adding flexibility to the transaction processor: we'd need to be able to turn Entity instances into some internal enum and handle the two cases independently. It wouldn't be too hard, but this reduces the combinatorial type explosion.
2017-03-27 23:30:04 +00:00
TempId::Internal(_) => None,
}
}
}
impl fmt::Display for TempId {
fn fmt(&self, f: &mut fmt::Formatter) -> Result<(), fmt::Error> {
match self {
&TempId::External(ref s) => write!(f, "{}", s),
&TempId::Internal(x) => write!(f, "<tempid {}>", x),
&TempId::Tx => write!(f, "<Tx>"),
Lookup refs, nested vector values, map notation. Fixes #180, fixes #183, fixes #284. (#382) r=rnewman * Pre: Fix error in parser macros. * Pre: Make test unwrapping more verbose. * Pre: Make lookup refs be (lookup-ref a v) in the entity position. This has the advantage of being explicit in all situations and unambiguous at parse-time. This choice agrees with the Clojure implementation but not with Datomic. Datomic treats [a v] as a lookup ref, is ambiguous at parse-time, and is disambiguated in ways I do not understand at transaction time. We mooted making lookup refs [[a v]] and outlawing nested value vectors in transactions, but after implementing that approach I decided it was better to handle lookup refs at parse time and therefore outlawing nested value vectors is not necessary. * Handle lookup refs in the entity and value columns. Fixes #183. * Pre 0a: Use a stack instead of into_iter. * Pre 0b: Dedent. * Pre 0c: Handle `e` after `v`. This allows to use the original `e` while handling `v`. * Explode value lists for :db.cardinality/many attributes. Fixes #284. * Parse and accept map notation. Fixes #180. * Pre: Modernize add() and retract() into one add_or_retract(). * Pre: Add is_collection and is_atom to edn::Value. * Pre: Differentiate atoms from lookup-refs in value position. Initially, I expected to accept arbitrary edn::Value instances in the value position, and to differentiate in the transactor. However, the implementation quickly became a two-stage parser, since we always wanted to parse the resulting value position into some other known thing using the tx-parser. To save calls into the parser and to allow the parser to move forward with a smaller API surface, I push as much of this parsing as possible into the initial parse. * Pre: Modernize entities(). * Pre: Quote edn::Value::Text in Display. * Review comment: Add and use edn::Value::into_atom. * Review comment: Use skip(eof()) throughout. * Review comment: VecDeque instead of Vec. * Review comment: Part 0: Rename TempId to TempIdHandle. * Review comment: Part 1: Differentiate internal and external tempids. This breaks an abstraction boundary by pushing the Internal/External split up to the Entity level in tx/ and tx-parser/. This just makes it easier to explode Entity map notation instances into Entity instances, taking an existing External tempid :db/id or generating a new Internal tempid as appropriate. To do this without breaking the abstraction boundary would require adding flexibility to the transaction processor: we'd need to be able to turn Entity instances into some internal enum and handle the two cases independently. It wouldn't be too hard, but this reduces the combinatorial type explosion.
2017-03-27 23:30:04 +00:00
}
}
}
#[derive(Clone, Debug, Eq, Hash, Ord, PartialOrd, PartialEq)]
pub enum Entid {
Entid(i64),
Ident(NamespacedKeyword),
}
impl Entid {
pub fn unreversed(&self) -> Option<Entid> {
match self {
&Entid::Entid(_) => None,
&Entid::Ident(ref a) => a.unreversed().map(Entid::Ident),
}
}
}
Lookup refs, nested vector values, map notation. Fixes #180, fixes #183, fixes #284. (#382) r=rnewman * Pre: Fix error in parser macros. * Pre: Make test unwrapping more verbose. * Pre: Make lookup refs be (lookup-ref a v) in the entity position. This has the advantage of being explicit in all situations and unambiguous at parse-time. This choice agrees with the Clojure implementation but not with Datomic. Datomic treats [a v] as a lookup ref, is ambiguous at parse-time, and is disambiguated in ways I do not understand at transaction time. We mooted making lookup refs [[a v]] and outlawing nested value vectors in transactions, but after implementing that approach I decided it was better to handle lookup refs at parse time and therefore outlawing nested value vectors is not necessary. * Handle lookup refs in the entity and value columns. Fixes #183. * Pre 0a: Use a stack instead of into_iter. * Pre 0b: Dedent. * Pre 0c: Handle `e` after `v`. This allows to use the original `e` while handling `v`. * Explode value lists for :db.cardinality/many attributes. Fixes #284. * Parse and accept map notation. Fixes #180. * Pre: Modernize add() and retract() into one add_or_retract(). * Pre: Add is_collection and is_atom to edn::Value. * Pre: Differentiate atoms from lookup-refs in value position. Initially, I expected to accept arbitrary edn::Value instances in the value position, and to differentiate in the transactor. However, the implementation quickly became a two-stage parser, since we always wanted to parse the resulting value position into some other known thing using the tx-parser. To save calls into the parser and to allow the parser to move forward with a smaller API surface, I push as much of this parsing as possible into the initial parse. * Pre: Modernize entities(). * Pre: Quote edn::Value::Text in Display. * Review comment: Add and use edn::Value::into_atom. * Review comment: Use skip(eof()) throughout. * Review comment: VecDeque instead of Vec. * Review comment: Part 0: Rename TempId to TempIdHandle. * Review comment: Part 1: Differentiate internal and external tempids. This breaks an abstraction boundary by pushing the Internal/External split up to the Entity level in tx/ and tx-parser/. This just makes it easier to explode Entity map notation instances into Entity instances, taking an existing External tempid :db/id or generating a new Internal tempid as appropriate. To do this without breaking the abstraction boundary would require adding flexibility to the transaction processor: we'd need to be able to turn Entity instances into some internal enum and handle the two cases independently. It wouldn't be too hard, but this reduces the combinatorial type explosion.
2017-03-27 23:30:04 +00:00
#[derive(Clone, Debug, Eq, Hash, Ord, PartialOrd, PartialEq)]
pub struct LookupRef {
pub a: Entid,
Lookup refs, nested vector values, map notation. Fixes #180, fixes #183, fixes #284. (#382) r=rnewman * Pre: Fix error in parser macros. * Pre: Make test unwrapping more verbose. * Pre: Make lookup refs be (lookup-ref a v) in the entity position. This has the advantage of being explicit in all situations and unambiguous at parse-time. This choice agrees with the Clojure implementation but not with Datomic. Datomic treats [a v] as a lookup ref, is ambiguous at parse-time, and is disambiguated in ways I do not understand at transaction time. We mooted making lookup refs [[a v]] and outlawing nested value vectors in transactions, but after implementing that approach I decided it was better to handle lookup refs at parse time and therefore outlawing nested value vectors is not necessary. * Handle lookup refs in the entity and value columns. Fixes #183. * Pre 0a: Use a stack instead of into_iter. * Pre 0b: Dedent. * Pre 0c: Handle `e` after `v`. This allows to use the original `e` while handling `v`. * Explode value lists for :db.cardinality/many attributes. Fixes #284. * Parse and accept map notation. Fixes #180. * Pre: Modernize add() and retract() into one add_or_retract(). * Pre: Add is_collection and is_atom to edn::Value. * Pre: Differentiate atoms from lookup-refs in value position. Initially, I expected to accept arbitrary edn::Value instances in the value position, and to differentiate in the transactor. However, the implementation quickly became a two-stage parser, since we always wanted to parse the resulting value position into some other known thing using the tx-parser. To save calls into the parser and to allow the parser to move forward with a smaller API surface, I push as much of this parsing as possible into the initial parse. * Pre: Modernize entities(). * Pre: Quote edn::Value::Text in Display. * Review comment: Add and use edn::Value::into_atom. * Review comment: Use skip(eof()) throughout. * Review comment: VecDeque instead of Vec. * Review comment: Part 0: Rename TempId to TempIdHandle. * Review comment: Part 1: Differentiate internal and external tempids. This breaks an abstraction boundary by pushing the Internal/External split up to the Entity level in tx/ and tx-parser/. This just makes it easier to explode Entity map notation instances into Entity instances, taking an existing External tempid :db/id or generating a new Internal tempid as appropriate. To do this without breaking the abstraction boundary would require adding flexibility to the transaction processor: we'd need to be able to turn Entity instances into some internal enum and handle the two cases independently. It wouldn't be too hard, but this reduces the combinatorial type explosion.
2017-03-27 23:30:04 +00:00
// In theory we could allow nested lookup-refs. In practice this would require us to process
// lookup-refs in multiple phases, like how we resolve tempids, which isn't worth the effort.
Improve parsing of nested `edn::ValueAndSpan` streams. r=rnewman (#393) * Pre: Expose more in edn. * Pre: Make it easier to work with ValueAndSpan. with_spans() is a temporary hack, needed only because I don't care to parse the bootstrap assertions from text right now. * Part 1a: Add `value_and_span` for parsing nested `edn::ValueAndSpan` instances. I wasn't able to abstract over `edn::Value` and `edn::ValueAndSpan`; there are multiple obstacles. I chose to roll with `edn::ValueAndSpan` since it exposes the additional span information that we will want to form good error messages in the future. * Part 1b: Add keyword_map() parsing an `edn::Value::Vector` into an `edn::Value::map`. * Part 1c: Add `Log`/`.log(...)` for logging parser progress. This is a terrible hack, but it sure helps to debug complicated nested parsers. I don't even know what a principled approach would look like; since our parser combinators are so frequently expressed in code, it's hard to imagine a data-driven interpreter that can help debug things. * Part 2: Use `value_and_span` apparatus in tx-parser/. I break an abstraction boundary by returning a value column `edn::ValueAndSpan` rather than just an `edn::Value`. That is, the transaction processor shouldn't care where the `edn::Value` it is processing arose -- even we care to track that information we should bake it into the `Entity` type. We do this because we need to dynamically parse the value column to support nested maps, and parsing requires a full `edn::ValueAndSpan`. Alternately, we could cheat and fake the spans when parsing nested maps, but that's potentially expensive. * Part 3: Use `value_and_span` apparatus in query-parser/. * Part 4: Use `value_and_span` apparatus in root crate. * Review comment: Make Span and SpanPosition Copy. * Review comment: nits. * Review comment: Make `or` be `or_exactly`. I baked the eof checking directly into the parser, rather than using the skip and eof parsers. I also took the time to restore some tests that were mistakenly commented out. * Review comment: Extract and use def_matches_* macros. * Review comment: .map() as late as possible.
2017-04-06 17:06:28 +00:00
pub v: edn::Value, // An atom.
}
Lookup refs, nested vector values, map notation. Fixes #180, fixes #183, fixes #284. (#382) r=rnewman * Pre: Fix error in parser macros. * Pre: Make test unwrapping more verbose. * Pre: Make lookup refs be (lookup-ref a v) in the entity position. This has the advantage of being explicit in all situations and unambiguous at parse-time. This choice agrees with the Clojure implementation but not with Datomic. Datomic treats [a v] as a lookup ref, is ambiguous at parse-time, and is disambiguated in ways I do not understand at transaction time. We mooted making lookup refs [[a v]] and outlawing nested value vectors in transactions, but after implementing that approach I decided it was better to handle lookup refs at parse time and therefore outlawing nested value vectors is not necessary. * Handle lookup refs in the entity and value columns. Fixes #183. * Pre 0a: Use a stack instead of into_iter. * Pre 0b: Dedent. * Pre 0c: Handle `e` after `v`. This allows to use the original `e` while handling `v`. * Explode value lists for :db.cardinality/many attributes. Fixes #284. * Parse and accept map notation. Fixes #180. * Pre: Modernize add() and retract() into one add_or_retract(). * Pre: Add is_collection and is_atom to edn::Value. * Pre: Differentiate atoms from lookup-refs in value position. Initially, I expected to accept arbitrary edn::Value instances in the value position, and to differentiate in the transactor. However, the implementation quickly became a two-stage parser, since we always wanted to parse the resulting value position into some other known thing using the tx-parser. To save calls into the parser and to allow the parser to move forward with a smaller API surface, I push as much of this parsing as possible into the initial parse. * Pre: Modernize entities(). * Pre: Quote edn::Value::Text in Display. * Review comment: Add and use edn::Value::into_atom. * Review comment: Use skip(eof()) throughout. * Review comment: VecDeque instead of Vec. * Review comment: Part 0: Rename TempId to TempIdHandle. * Review comment: Part 1: Differentiate internal and external tempids. This breaks an abstraction boundary by pushing the Internal/External split up to the Entity level in tx/ and tx-parser/. This just makes it easier to explode Entity map notation instances into Entity instances, taking an existing External tempid :db/id or generating a new Internal tempid as appropriate. To do this without breaking the abstraction boundary would require adding flexibility to the transaction processor: we'd need to be able to turn Entity instances into some internal enum and handle the two cases independently. It wouldn't be too hard, but this reduces the combinatorial type explosion.
2017-03-27 23:30:04 +00:00
pub type MapNotation = BTreeMap<Entid, AtomOrLookupRefOrVectorOrMapNotation>;
#[derive(Clone, Debug, Eq, Hash, Ord, PartialOrd, PartialEq)]
pub enum AtomOrLookupRefOrVectorOrMapNotation {
Improve parsing of nested `edn::ValueAndSpan` streams. r=rnewman (#393) * Pre: Expose more in edn. * Pre: Make it easier to work with ValueAndSpan. with_spans() is a temporary hack, needed only because I don't care to parse the bootstrap assertions from text right now. * Part 1a: Add `value_and_span` for parsing nested `edn::ValueAndSpan` instances. I wasn't able to abstract over `edn::Value` and `edn::ValueAndSpan`; there are multiple obstacles. I chose to roll with `edn::ValueAndSpan` since it exposes the additional span information that we will want to form good error messages in the future. * Part 1b: Add keyword_map() parsing an `edn::Value::Vector` into an `edn::Value::map`. * Part 1c: Add `Log`/`.log(...)` for logging parser progress. This is a terrible hack, but it sure helps to debug complicated nested parsers. I don't even know what a principled approach would look like; since our parser combinators are so frequently expressed in code, it's hard to imagine a data-driven interpreter that can help debug things. * Part 2: Use `value_and_span` apparatus in tx-parser/. I break an abstraction boundary by returning a value column `edn::ValueAndSpan` rather than just an `edn::Value`. That is, the transaction processor shouldn't care where the `edn::Value` it is processing arose -- even we care to track that information we should bake it into the `Entity` type. We do this because we need to dynamically parse the value column to support nested maps, and parsing requires a full `edn::ValueAndSpan`. Alternately, we could cheat and fake the spans when parsing nested maps, but that's potentially expensive. * Part 3: Use `value_and_span` apparatus in query-parser/. * Part 4: Use `value_and_span` apparatus in root crate. * Review comment: Make Span and SpanPosition Copy. * Review comment: nits. * Review comment: Make `or` be `or_exactly`. I baked the eof checking directly into the parser, rather than using the skip and eof parsers. I also took the time to restore some tests that were mistakenly commented out. * Review comment: Extract and use def_matches_* macros. * Review comment: .map() as late as possible.
2017-04-06 17:06:28 +00:00
Atom(edn::ValueAndSpan),
LookupRef(LookupRef),
Lookup refs, nested vector values, map notation. Fixes #180, fixes #183, fixes #284. (#382) r=rnewman * Pre: Fix error in parser macros. * Pre: Make test unwrapping more verbose. * Pre: Make lookup refs be (lookup-ref a v) in the entity position. This has the advantage of being explicit in all situations and unambiguous at parse-time. This choice agrees with the Clojure implementation but not with Datomic. Datomic treats [a v] as a lookup ref, is ambiguous at parse-time, and is disambiguated in ways I do not understand at transaction time. We mooted making lookup refs [[a v]] and outlawing nested value vectors in transactions, but after implementing that approach I decided it was better to handle lookup refs at parse time and therefore outlawing nested value vectors is not necessary. * Handle lookup refs in the entity and value columns. Fixes #183. * Pre 0a: Use a stack instead of into_iter. * Pre 0b: Dedent. * Pre 0c: Handle `e` after `v`. This allows to use the original `e` while handling `v`. * Explode value lists for :db.cardinality/many attributes. Fixes #284. * Parse and accept map notation. Fixes #180. * Pre: Modernize add() and retract() into one add_or_retract(). * Pre: Add is_collection and is_atom to edn::Value. * Pre: Differentiate atoms from lookup-refs in value position. Initially, I expected to accept arbitrary edn::Value instances in the value position, and to differentiate in the transactor. However, the implementation quickly became a two-stage parser, since we always wanted to parse the resulting value position into some other known thing using the tx-parser. To save calls into the parser and to allow the parser to move forward with a smaller API surface, I push as much of this parsing as possible into the initial parse. * Pre: Modernize entities(). * Pre: Quote edn::Value::Text in Display. * Review comment: Add and use edn::Value::into_atom. * Review comment: Use skip(eof()) throughout. * Review comment: VecDeque instead of Vec. * Review comment: Part 0: Rename TempId to TempIdHandle. * Review comment: Part 1: Differentiate internal and external tempids. This breaks an abstraction boundary by pushing the Internal/External split up to the Entity level in tx/ and tx-parser/. This just makes it easier to explode Entity map notation instances into Entity instances, taking an existing External tempid :db/id or generating a new Internal tempid as appropriate. To do this without breaking the abstraction boundary would require adding flexibility to the transaction processor: we'd need to be able to turn Entity instances into some internal enum and handle the two cases independently. It wouldn't be too hard, but this reduces the combinatorial type explosion.
2017-03-27 23:30:04 +00:00
Vector(Vec<AtomOrLookupRefOrVectorOrMapNotation>),
MapNotation(MapNotation),
}
Lookup refs, nested vector values, map notation. Fixes #180, fixes #183, fixes #284. (#382) r=rnewman * Pre: Fix error in parser macros. * Pre: Make test unwrapping more verbose. * Pre: Make lookup refs be (lookup-ref a v) in the entity position. This has the advantage of being explicit in all situations and unambiguous at parse-time. This choice agrees with the Clojure implementation but not with Datomic. Datomic treats [a v] as a lookup ref, is ambiguous at parse-time, and is disambiguated in ways I do not understand at transaction time. We mooted making lookup refs [[a v]] and outlawing nested value vectors in transactions, but after implementing that approach I decided it was better to handle lookup refs at parse time and therefore outlawing nested value vectors is not necessary. * Handle lookup refs in the entity and value columns. Fixes #183. * Pre 0a: Use a stack instead of into_iter. * Pre 0b: Dedent. * Pre 0c: Handle `e` after `v`. This allows to use the original `e` while handling `v`. * Explode value lists for :db.cardinality/many attributes. Fixes #284. * Parse and accept map notation. Fixes #180. * Pre: Modernize add() and retract() into one add_or_retract(). * Pre: Add is_collection and is_atom to edn::Value. * Pre: Differentiate atoms from lookup-refs in value position. Initially, I expected to accept arbitrary edn::Value instances in the value position, and to differentiate in the transactor. However, the implementation quickly became a two-stage parser, since we always wanted to parse the resulting value position into some other known thing using the tx-parser. To save calls into the parser and to allow the parser to move forward with a smaller API surface, I push as much of this parsing as possible into the initial parse. * Pre: Modernize entities(). * Pre: Quote edn::Value::Text in Display. * Review comment: Add and use edn::Value::into_atom. * Review comment: Use skip(eof()) throughout. * Review comment: VecDeque instead of Vec. * Review comment: Part 0: Rename TempId to TempIdHandle. * Review comment: Part 1: Differentiate internal and external tempids. This breaks an abstraction boundary by pushing the Internal/External split up to the Entity level in tx/ and tx-parser/. This just makes it easier to explode Entity map notation instances into Entity instances, taking an existing External tempid :db/id or generating a new Internal tempid as appropriate. To do this without breaking the abstraction boundary would require adding flexibility to the transaction processor: we'd need to be able to turn Entity instances into some internal enum and handle the two cases independently. It wouldn't be too hard, but this reduces the combinatorial type explosion.
2017-03-27 23:30:04 +00:00
#[derive(Clone, Debug, Eq, Hash, Ord, PartialOrd, PartialEq)]
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
pub enum EntidOrLookupRefOrTempId {
Entid(Entid),
LookupRef(LookupRef),
Lookup refs, nested vector values, map notation. Fixes #180, fixes #183, fixes #284. (#382) r=rnewman * Pre: Fix error in parser macros. * Pre: Make test unwrapping more verbose. * Pre: Make lookup refs be (lookup-ref a v) in the entity position. This has the advantage of being explicit in all situations and unambiguous at parse-time. This choice agrees with the Clojure implementation but not with Datomic. Datomic treats [a v] as a lookup ref, is ambiguous at parse-time, and is disambiguated in ways I do not understand at transaction time. We mooted making lookup refs [[a v]] and outlawing nested value vectors in transactions, but after implementing that approach I decided it was better to handle lookup refs at parse time and therefore outlawing nested value vectors is not necessary. * Handle lookup refs in the entity and value columns. Fixes #183. * Pre 0a: Use a stack instead of into_iter. * Pre 0b: Dedent. * Pre 0c: Handle `e` after `v`. This allows to use the original `e` while handling `v`. * Explode value lists for :db.cardinality/many attributes. Fixes #284. * Parse and accept map notation. Fixes #180. * Pre: Modernize add() and retract() into one add_or_retract(). * Pre: Add is_collection and is_atom to edn::Value. * Pre: Differentiate atoms from lookup-refs in value position. Initially, I expected to accept arbitrary edn::Value instances in the value position, and to differentiate in the transactor. However, the implementation quickly became a two-stage parser, since we always wanted to parse the resulting value position into some other known thing using the tx-parser. To save calls into the parser and to allow the parser to move forward with a smaller API surface, I push as much of this parsing as possible into the initial parse. * Pre: Modernize entities(). * Pre: Quote edn::Value::Text in Display. * Review comment: Add and use edn::Value::into_atom. * Review comment: Use skip(eof()) throughout. * Review comment: VecDeque instead of Vec. * Review comment: Part 0: Rename TempId to TempIdHandle. * Review comment: Part 1: Differentiate internal and external tempids. This breaks an abstraction boundary by pushing the Internal/External split up to the Entity level in tx/ and tx-parser/. This just makes it easier to explode Entity map notation instances into Entity instances, taking an existing External tempid :db/id or generating a new Internal tempid as appropriate. To do this without breaking the abstraction boundary would require adding flexibility to the transaction processor: we'd need to be able to turn Entity instances into some internal enum and handle the two cases independently. It wouldn't be too hard, but this reduces the combinatorial type explosion.
2017-03-27 23:30:04 +00:00
TempId(TempId),
}
#[derive(Clone, Copy, Debug, Eq, Hash, Ord, PartialOrd, PartialEq)]
pub enum OpType {
Add,
Retract,
}
Lookup refs, nested vector values, map notation. Fixes #180, fixes #183, fixes #284. (#382) r=rnewman * Pre: Fix error in parser macros. * Pre: Make test unwrapping more verbose. * Pre: Make lookup refs be (lookup-ref a v) in the entity position. This has the advantage of being explicit in all situations and unambiguous at parse-time. This choice agrees with the Clojure implementation but not with Datomic. Datomic treats [a v] as a lookup ref, is ambiguous at parse-time, and is disambiguated in ways I do not understand at transaction time. We mooted making lookup refs [[a v]] and outlawing nested value vectors in transactions, but after implementing that approach I decided it was better to handle lookup refs at parse time and therefore outlawing nested value vectors is not necessary. * Handle lookup refs in the entity and value columns. Fixes #183. * Pre 0a: Use a stack instead of into_iter. * Pre 0b: Dedent. * Pre 0c: Handle `e` after `v`. This allows to use the original `e` while handling `v`. * Explode value lists for :db.cardinality/many attributes. Fixes #284. * Parse and accept map notation. Fixes #180. * Pre: Modernize add() and retract() into one add_or_retract(). * Pre: Add is_collection and is_atom to edn::Value. * Pre: Differentiate atoms from lookup-refs in value position. Initially, I expected to accept arbitrary edn::Value instances in the value position, and to differentiate in the transactor. However, the implementation quickly became a two-stage parser, since we always wanted to parse the resulting value position into some other known thing using the tx-parser. To save calls into the parser and to allow the parser to move forward with a smaller API surface, I push as much of this parsing as possible into the initial parse. * Pre: Modernize entities(). * Pre: Quote edn::Value::Text in Display. * Review comment: Add and use edn::Value::into_atom. * Review comment: Use skip(eof()) throughout. * Review comment: VecDeque instead of Vec. * Review comment: Part 0: Rename TempId to TempIdHandle. * Review comment: Part 1: Differentiate internal and external tempids. This breaks an abstraction boundary by pushing the Internal/External split up to the Entity level in tx/ and tx-parser/. This just makes it easier to explode Entity map notation instances into Entity instances, taking an existing External tempid :db/id or generating a new Internal tempid as appropriate. To do this without breaking the abstraction boundary would require adding flexibility to the transaction processor: we'd need to be able to turn Entity instances into some internal enum and handle the two cases independently. It wouldn't be too hard, but this reduces the combinatorial type explosion.
2017-03-27 23:30:04 +00:00
#[derive(Clone, Debug, Eq, Hash, Ord, PartialOrd, PartialEq)]
pub enum Entity {
Lookup refs, nested vector values, map notation. Fixes #180, fixes #183, fixes #284. (#382) r=rnewman * Pre: Fix error in parser macros. * Pre: Make test unwrapping more verbose. * Pre: Make lookup refs be (lookup-ref a v) in the entity position. This has the advantage of being explicit in all situations and unambiguous at parse-time. This choice agrees with the Clojure implementation but not with Datomic. Datomic treats [a v] as a lookup ref, is ambiguous at parse-time, and is disambiguated in ways I do not understand at transaction time. We mooted making lookup refs [[a v]] and outlawing nested value vectors in transactions, but after implementing that approach I decided it was better to handle lookup refs at parse time and therefore outlawing nested value vectors is not necessary. * Handle lookup refs in the entity and value columns. Fixes #183. * Pre 0a: Use a stack instead of into_iter. * Pre 0b: Dedent. * Pre 0c: Handle `e` after `v`. This allows to use the original `e` while handling `v`. * Explode value lists for :db.cardinality/many attributes. Fixes #284. * Parse and accept map notation. Fixes #180. * Pre: Modernize add() and retract() into one add_or_retract(). * Pre: Add is_collection and is_atom to edn::Value. * Pre: Differentiate atoms from lookup-refs in value position. Initially, I expected to accept arbitrary edn::Value instances in the value position, and to differentiate in the transactor. However, the implementation quickly became a two-stage parser, since we always wanted to parse the resulting value position into some other known thing using the tx-parser. To save calls into the parser and to allow the parser to move forward with a smaller API surface, I push as much of this parsing as possible into the initial parse. * Pre: Modernize entities(). * Pre: Quote edn::Value::Text in Display. * Review comment: Add and use edn::Value::into_atom. * Review comment: Use skip(eof()) throughout. * Review comment: VecDeque instead of Vec. * Review comment: Part 0: Rename TempId to TempIdHandle. * Review comment: Part 1: Differentiate internal and external tempids. This breaks an abstraction boundary by pushing the Internal/External split up to the Entity level in tx/ and tx-parser/. This just makes it easier to explode Entity map notation instances into Entity instances, taking an existing External tempid :db/id or generating a new Internal tempid as appropriate. To do this without breaking the abstraction boundary would require adding flexibility to the transaction processor: we'd need to be able to turn Entity instances into some internal enum and handle the two cases independently. It wouldn't be too hard, but this reduces the combinatorial type explosion.
2017-03-27 23:30:04 +00:00
// Like [:db/add|:db/retract e a v].
AddOrRetract {
op: OpType,
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
e: EntidOrLookupRefOrTempId,
a: Entid,
Lookup refs, nested vector values, map notation. Fixes #180, fixes #183, fixes #284. (#382) r=rnewman * Pre: Fix error in parser macros. * Pre: Make test unwrapping more verbose. * Pre: Make lookup refs be (lookup-ref a v) in the entity position. This has the advantage of being explicit in all situations and unambiguous at parse-time. This choice agrees with the Clojure implementation but not with Datomic. Datomic treats [a v] as a lookup ref, is ambiguous at parse-time, and is disambiguated in ways I do not understand at transaction time. We mooted making lookup refs [[a v]] and outlawing nested value vectors in transactions, but after implementing that approach I decided it was better to handle lookup refs at parse time and therefore outlawing nested value vectors is not necessary. * Handle lookup refs in the entity and value columns. Fixes #183. * Pre 0a: Use a stack instead of into_iter. * Pre 0b: Dedent. * Pre 0c: Handle `e` after `v`. This allows to use the original `e` while handling `v`. * Explode value lists for :db.cardinality/many attributes. Fixes #284. * Parse and accept map notation. Fixes #180. * Pre: Modernize add() and retract() into one add_or_retract(). * Pre: Add is_collection and is_atom to edn::Value. * Pre: Differentiate atoms from lookup-refs in value position. Initially, I expected to accept arbitrary edn::Value instances in the value position, and to differentiate in the transactor. However, the implementation quickly became a two-stage parser, since we always wanted to parse the resulting value position into some other known thing using the tx-parser. To save calls into the parser and to allow the parser to move forward with a smaller API surface, I push as much of this parsing as possible into the initial parse. * Pre: Modernize entities(). * Pre: Quote edn::Value::Text in Display. * Review comment: Add and use edn::Value::into_atom. * Review comment: Use skip(eof()) throughout. * Review comment: VecDeque instead of Vec. * Review comment: Part 0: Rename TempId to TempIdHandle. * Review comment: Part 1: Differentiate internal and external tempids. This breaks an abstraction boundary by pushing the Internal/External split up to the Entity level in tx/ and tx-parser/. This just makes it easier to explode Entity map notation instances into Entity instances, taking an existing External tempid :db/id or generating a new Internal tempid as appropriate. To do this without breaking the abstraction boundary would require adding flexibility to the transaction processor: we'd need to be able to turn Entity instances into some internal enum and handle the two cases independently. It wouldn't be too hard, but this reduces the combinatorial type explosion.
2017-03-27 23:30:04 +00:00
v: AtomOrLookupRefOrVectorOrMapNotation,
},
Lookup refs, nested vector values, map notation. Fixes #180, fixes #183, fixes #284. (#382) r=rnewman * Pre: Fix error in parser macros. * Pre: Make test unwrapping more verbose. * Pre: Make lookup refs be (lookup-ref a v) in the entity position. This has the advantage of being explicit in all situations and unambiguous at parse-time. This choice agrees with the Clojure implementation but not with Datomic. Datomic treats [a v] as a lookup ref, is ambiguous at parse-time, and is disambiguated in ways I do not understand at transaction time. We mooted making lookup refs [[a v]] and outlawing nested value vectors in transactions, but after implementing that approach I decided it was better to handle lookup refs at parse time and therefore outlawing nested value vectors is not necessary. * Handle lookup refs in the entity and value columns. Fixes #183. * Pre 0a: Use a stack instead of into_iter. * Pre 0b: Dedent. * Pre 0c: Handle `e` after `v`. This allows to use the original `e` while handling `v`. * Explode value lists for :db.cardinality/many attributes. Fixes #284. * Parse and accept map notation. Fixes #180. * Pre: Modernize add() and retract() into one add_or_retract(). * Pre: Add is_collection and is_atom to edn::Value. * Pre: Differentiate atoms from lookup-refs in value position. Initially, I expected to accept arbitrary edn::Value instances in the value position, and to differentiate in the transactor. However, the implementation quickly became a two-stage parser, since we always wanted to parse the resulting value position into some other known thing using the tx-parser. To save calls into the parser and to allow the parser to move forward with a smaller API surface, I push as much of this parsing as possible into the initial parse. * Pre: Modernize entities(). * Pre: Quote edn::Value::Text in Display. * Review comment: Add and use edn::Value::into_atom. * Review comment: Use skip(eof()) throughout. * Review comment: VecDeque instead of Vec. * Review comment: Part 0: Rename TempId to TempIdHandle. * Review comment: Part 1: Differentiate internal and external tempids. This breaks an abstraction boundary by pushing the Internal/External split up to the Entity level in tx/ and tx-parser/. This just makes it easier to explode Entity map notation instances into Entity instances, taking an existing External tempid :db/id or generating a new Internal tempid as appropriate. To do this without breaking the abstraction boundary would require adding flexibility to the transaction processor: we'd need to be able to turn Entity instances into some internal enum and handle the two cases independently. It wouldn't be too hard, but this reduces the combinatorial type explosion.
2017-03-27 23:30:04 +00:00
// Like {:db/id "tempid" a1 v1 a2 v2}.
MapNotation(MapNotation),
}