mentat/tolstoy/src/tx_mapper.rs

142 lines
4.7 KiB
Rust
Raw Normal View History

// Copyright 2018 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.
use std::convert::TryInto;
use uuid::Uuid;
use core_traits::Entid;
use public_traits::errors::Result;
Basic sync support (#563) r=nalexander * Pre: remove remnants of 'open_empty' * Pre: Cleanup 'datoms' table after a timeline move Since timeline move operations use a transactor, they generate a "phantom" 'tx' and a 'txInstant' assertion. It is "phantom" in a sense that it was never present in the 'transactions' table, and is entirely synthetic as far as our database is concerned. It's an implementational artifact, and we were not cleaning it up. It becomes a problem when we start inserting transactions after a move. Once the transactor clashes with the phantom 'tx', it will retract the phantom 'txInstant' value, leaving the transactions log in an incorrect state. This patch adds a test for this scenario and elects the easy way out: simply remove the offending 'txInstant' datom. * Part 1: Sync without support for side-effects A "side-effect" is defined here as a mutation of a remote state as part of the sync. If, during a sync we determine that a remote state needs to be changed, bail out. This generally supports different variations of "baton-passing" syncing, where clients will succeed syncing if each change is non-conflicting. * Part 2: Support basic "side-effects" syncing This patch introduces a concept of a follow-up sync. If a sync generated a "merge transaction" (a regular transaction that contains assertions necessary for local and remote transaction logs to converge), then this transaction needs to be uploaded in a follow-up sync. Generated SyncReport indicates if a follow-up sync is required. Follow-up sync itself is just a regular sync. If remote state did not change, it will result in a simple RemoteFastForward. Otherwise, we'll continue merging and requesting a follow-up. Schema alterations are explicitly not supported. As local transactions are rebased on top of remote, following changes happen: - entids are changed into tempids, letting transactor upsert :db/unique values - entids for retractions are changed into lookup-refs if we're confident they'll succeed -- otherwise, retractions are dropped on the floor * Post: use a macro for more readable tests * Tolstoy README
2018-09-08 02:18:20 +00:00
use tolstoy_traits::errors::TolstoyError;
Basic sync support (#563) r=nalexander * Pre: remove remnants of 'open_empty' * Pre: Cleanup 'datoms' table after a timeline move Since timeline move operations use a transactor, they generate a "phantom" 'tx' and a 'txInstant' assertion. It is "phantom" in a sense that it was never present in the 'transactions' table, and is entirely synthetic as far as our database is concerned. It's an implementational artifact, and we were not cleaning it up. It becomes a problem when we start inserting transactions after a move. Once the transactor clashes with the phantom 'tx', it will retract the phantom 'txInstant' value, leaving the transactions log in an incorrect state. This patch adds a test for this scenario and elects the easy way out: simply remove the offending 'txInstant' datom. * Part 1: Sync without support for side-effects A "side-effect" is defined here as a mutation of a remote state as part of the sync. If, during a sync we determine that a remote state needs to be changed, bail out. This generally supports different variations of "baton-passing" syncing, where clients will succeed syncing if each change is non-conflicting. * Part 2: Support basic "side-effects" syncing This patch introduces a concept of a follow-up sync. If a sync generated a "merge transaction" (a regular transaction that contains assertions necessary for local and remote transaction logs to converge), then this transaction needs to be uploaded in a follow-up sync. Generated SyncReport indicates if a follow-up sync is required. Follow-up sync itself is just a regular sync. If remote state did not change, it will result in a simple RemoteFastForward. Otherwise, we'll continue merging and requesting a follow-up. Schema alterations are explicitly not supported. As local transactions are rebased on top of remote, following changes happen: - entids are changed into tempids, letting transactor upsert :db/unique values - entids for retractions are changed into lookup-refs if we're confident they'll succeed -- otherwise, retractions are dropped on the floor * Post: use a macro for more readable tests * Tolstoy README
2018-09-08 02:18:20 +00:00
use crate::types::LocalGlobalTxMapping;
// Exposes a tx<->uuid mapping interface.
pub struct TxMapper {}
impl TxMapper {
pub fn set_lg_mappings(
2020-08-06 03:03:58 +00:00
db_tx: &mut rusqlite::Transaction<'_>,
mappings: Vec<LocalGlobalTxMapping<'_>>,
) -> Result<()> {
let mut stmt =
db_tx.prepare_cached("INSERT OR REPLACE INTO tolstoy_tu (tx, uuid) VALUES (?, ?)")?;
Basic sync support (#563) r=nalexander * Pre: remove remnants of 'open_empty' * Pre: Cleanup 'datoms' table after a timeline move Since timeline move operations use a transactor, they generate a "phantom" 'tx' and a 'txInstant' assertion. It is "phantom" in a sense that it was never present in the 'transactions' table, and is entirely synthetic as far as our database is concerned. It's an implementational artifact, and we were not cleaning it up. It becomes a problem when we start inserting transactions after a move. Once the transactor clashes with the phantom 'tx', it will retract the phantom 'txInstant' value, leaving the transactions log in an incorrect state. This patch adds a test for this scenario and elects the easy way out: simply remove the offending 'txInstant' datom. * Part 1: Sync without support for side-effects A "side-effect" is defined here as a mutation of a remote state as part of the sync. If, during a sync we determine that a remote state needs to be changed, bail out. This generally supports different variations of "baton-passing" syncing, where clients will succeed syncing if each change is non-conflicting. * Part 2: Support basic "side-effects" syncing This patch introduces a concept of a follow-up sync. If a sync generated a "merge transaction" (a regular transaction that contains assertions necessary for local and remote transaction logs to converge), then this transaction needs to be uploaded in a follow-up sync. Generated SyncReport indicates if a follow-up sync is required. Follow-up sync itself is just a regular sync. If remote state did not change, it will result in a simple RemoteFastForward. Otherwise, we'll continue merging and requesting a follow-up. Schema alterations are explicitly not supported. As local transactions are rebased on top of remote, following changes happen: - entids are changed into tempids, letting transactor upsert :db/unique values - entids for retractions are changed into lookup-refs if we're confident they'll succeed -- otherwise, retractions are dropped on the floor * Post: use a macro for more readable tests * Tolstoy README
2018-09-08 02:18:20 +00:00
for mapping in mappings.iter() {
let uuid_bytes = mapping.remote.as_bytes().to_vec();
stmt.execute(rusqlite::params![&mapping.local, &uuid_bytes])?;
}
Ok(())
}
pub fn set_lg_mapping(
2020-08-06 03:03:58 +00:00
db_tx: &mut rusqlite::Transaction<'_>,
mapping: LocalGlobalTxMapping<'_>,
) -> Result<()> {
Basic sync support (#563) r=nalexander * Pre: remove remnants of 'open_empty' * Pre: Cleanup 'datoms' table after a timeline move Since timeline move operations use a transactor, they generate a "phantom" 'tx' and a 'txInstant' assertion. It is "phantom" in a sense that it was never present in the 'transactions' table, and is entirely synthetic as far as our database is concerned. It's an implementational artifact, and we were not cleaning it up. It becomes a problem when we start inserting transactions after a move. Once the transactor clashes with the phantom 'tx', it will retract the phantom 'txInstant' value, leaving the transactions log in an incorrect state. This patch adds a test for this scenario and elects the easy way out: simply remove the offending 'txInstant' datom. * Part 1: Sync without support for side-effects A "side-effect" is defined here as a mutation of a remote state as part of the sync. If, during a sync we determine that a remote state needs to be changed, bail out. This generally supports different variations of "baton-passing" syncing, where clients will succeed syncing if each change is non-conflicting. * Part 2: Support basic "side-effects" syncing This patch introduces a concept of a follow-up sync. If a sync generated a "merge transaction" (a regular transaction that contains assertions necessary for local and remote transaction logs to converge), then this transaction needs to be uploaded in a follow-up sync. Generated SyncReport indicates if a follow-up sync is required. Follow-up sync itself is just a regular sync. If remote state did not change, it will result in a simple RemoteFastForward. Otherwise, we'll continue merging and requesting a follow-up. Schema alterations are explicitly not supported. As local transactions are rebased on top of remote, following changes happen: - entids are changed into tempids, letting transactor upsert :db/unique values - entids for retractions are changed into lookup-refs if we're confident they'll succeed -- otherwise, retractions are dropped on the floor * Post: use a macro for more readable tests * Tolstoy README
2018-09-08 02:18:20 +00:00
TxMapper::set_lg_mappings(db_tx, vec![mapping])
}
// TODO for when we're downloading, right?
2020-08-06 03:03:58 +00:00
pub fn get_or_set_uuid_for_tx(
db_tx: &mut rusqlite::Transaction<'_>,
tx: Entid,
) -> Result<Uuid> {
match TxMapper::get(db_tx, tx)? {
Some(uuid) => Ok(uuid),
None => {
let uuid = Uuid::new_v4();
let uuid_bytes = uuid.as_bytes().to_vec();
db_tx.execute(
"INSERT INTO tolstoy_tu (tx, uuid) VALUES (?, ?)",
rusqlite::params![&tx, &uuid_bytes],
)?;
2020-08-06 03:03:58 +00:00
Ok(uuid)
}
}
}
2020-08-06 03:03:58 +00:00
pub fn get_tx_for_uuid(
db_tx: &rusqlite::Transaction<'_>,
uuid: &Uuid,
) -> Result<Option<Entid>> {
let mut stmt = db_tx.prepare_cached("SELECT tx FROM tolstoy_tu WHERE uuid = ?")?;
let uuid_bytes = uuid.as_bytes().to_vec();
let results = stmt.query_map(&[&uuid_bytes], |r| r.get(0))?;
let mut txs = vec![];
txs.extend(results);
2020-08-06 03:03:58 +00:00
if txs.is_empty() {
return Ok(None);
} else if txs.len() > 1 {
2018-06-07 18:28:46 +00:00
bail!(TolstoyError::TxIncorrectlyMapped(txs.len()));
}
Ok(Some(txs.remove(0)?))
}
2020-08-06 03:03:58 +00:00
pub fn get(db_tx: &rusqlite::Transaction<'_>, tx: Entid) -> Result<Option<Uuid>> {
let mut stmt = db_tx.prepare_cached("SELECT uuid FROM tolstoy_tu WHERE tx = ?")?;
let results = stmt.query_and_then(&[&tx], |r| -> Result<Uuid> {
2019-07-22 12:58:19 +00:00
let bytes: Vec<u8> = r.get(0).unwrap();
Ok(Uuid::from_bytes(bytes.as_slice().try_into().unwrap()))
})?;
let mut uuids = vec![];
uuids.extend(results);
2020-08-06 03:03:58 +00:00
if uuids.is_empty() {
return Ok(None);
} else if uuids.len() > 1 {
2018-06-07 18:28:46 +00:00
bail!(TolstoyError::TxIncorrectlyMapped(uuids.len()));
}
Ok(Some(uuids.remove(0)?))
}
}
#[cfg(test)]
pub mod tests {
use super::*;
use crate::schema;
#[test]
fn test_getters() {
Basic sync support (#563) r=nalexander * Pre: remove remnants of 'open_empty' * Pre: Cleanup 'datoms' table after a timeline move Since timeline move operations use a transactor, they generate a "phantom" 'tx' and a 'txInstant' assertion. It is "phantom" in a sense that it was never present in the 'transactions' table, and is entirely synthetic as far as our database is concerned. It's an implementational artifact, and we were not cleaning it up. It becomes a problem when we start inserting transactions after a move. Once the transactor clashes with the phantom 'tx', it will retract the phantom 'txInstant' value, leaving the transactions log in an incorrect state. This patch adds a test for this scenario and elects the easy way out: simply remove the offending 'txInstant' datom. * Part 1: Sync without support for side-effects A "side-effect" is defined here as a mutation of a remote state as part of the sync. If, during a sync we determine that a remote state needs to be changed, bail out. This generally supports different variations of "baton-passing" syncing, where clients will succeed syncing if each change is non-conflicting. * Part 2: Support basic "side-effects" syncing This patch introduces a concept of a follow-up sync. If a sync generated a "merge transaction" (a regular transaction that contains assertions necessary for local and remote transaction logs to converge), then this transaction needs to be uploaded in a follow-up sync. Generated SyncReport indicates if a follow-up sync is required. Follow-up sync itself is just a regular sync. If remote state did not change, it will result in a simple RemoteFastForward. Otherwise, we'll continue merging and requesting a follow-up. Schema alterations are explicitly not supported. As local transactions are rebased on top of remote, following changes happen: - entids are changed into tempids, letting transactor upsert :db/unique values - entids for retractions are changed into lookup-refs if we're confident they'll succeed -- otherwise, retractions are dropped on the floor * Post: use a macro for more readable tests * Tolstoy README
2018-09-08 02:18:20 +00:00
let mut conn = schema::tests::setup_conn_bare();
let mut tx = schema::tests::setup_tx(&mut conn);
2020-08-06 03:03:58 +00:00
assert_eq!(None, TxMapper::get(&tx, 1).expect("success"));
let set_uuid = TxMapper::get_or_set_uuid_for_tx(&mut tx, 1).expect("success");
2020-08-06 03:03:58 +00:00
assert_eq!(Some(set_uuid), TxMapper::get(&tx, 1).expect("success"));
}
#[test]
fn test_bulk_setter() {
Basic sync support (#563) r=nalexander * Pre: remove remnants of 'open_empty' * Pre: Cleanup 'datoms' table after a timeline move Since timeline move operations use a transactor, they generate a "phantom" 'tx' and a 'txInstant' assertion. It is "phantom" in a sense that it was never present in the 'transactions' table, and is entirely synthetic as far as our database is concerned. It's an implementational artifact, and we were not cleaning it up. It becomes a problem when we start inserting transactions after a move. Once the transactor clashes with the phantom 'tx', it will retract the phantom 'txInstant' value, leaving the transactions log in an incorrect state. This patch adds a test for this scenario and elects the easy way out: simply remove the offending 'txInstant' datom. * Part 1: Sync without support for side-effects A "side-effect" is defined here as a mutation of a remote state as part of the sync. If, during a sync we determine that a remote state needs to be changed, bail out. This generally supports different variations of "baton-passing" syncing, where clients will succeed syncing if each change is non-conflicting. * Part 2: Support basic "side-effects" syncing This patch introduces a concept of a follow-up sync. If a sync generated a "merge transaction" (a regular transaction that contains assertions necessary for local and remote transaction logs to converge), then this transaction needs to be uploaded in a follow-up sync. Generated SyncReport indicates if a follow-up sync is required. Follow-up sync itself is just a regular sync. If remote state did not change, it will result in a simple RemoteFastForward. Otherwise, we'll continue merging and requesting a follow-up. Schema alterations are explicitly not supported. As local transactions are rebased on top of remote, following changes happen: - entids are changed into tempids, letting transactor upsert :db/unique values - entids for retractions are changed into lookup-refs if we're confident they'll succeed -- otherwise, retractions are dropped on the floor * Post: use a macro for more readable tests * Tolstoy README
2018-09-08 02:18:20 +00:00
let mut conn = schema::tests::setup_conn_bare();
let mut tx = schema::tests::setup_tx(&mut conn);
Basic sync support (#563) r=nalexander * Pre: remove remnants of 'open_empty' * Pre: Cleanup 'datoms' table after a timeline move Since timeline move operations use a transactor, they generate a "phantom" 'tx' and a 'txInstant' assertion. It is "phantom" in a sense that it was never present in the 'transactions' table, and is entirely synthetic as far as our database is concerned. It's an implementational artifact, and we were not cleaning it up. It becomes a problem when we start inserting transactions after a move. Once the transactor clashes with the phantom 'tx', it will retract the phantom 'txInstant' value, leaving the transactions log in an incorrect state. This patch adds a test for this scenario and elects the easy way out: simply remove the offending 'txInstant' datom. * Part 1: Sync without support for side-effects A "side-effect" is defined here as a mutation of a remote state as part of the sync. If, during a sync we determine that a remote state needs to be changed, bail out. This generally supports different variations of "baton-passing" syncing, where clients will succeed syncing if each change is non-conflicting. * Part 2: Support basic "side-effects" syncing This patch introduces a concept of a follow-up sync. If a sync generated a "merge transaction" (a regular transaction that contains assertions necessary for local and remote transaction logs to converge), then this transaction needs to be uploaded in a follow-up sync. Generated SyncReport indicates if a follow-up sync is required. Follow-up sync itself is just a regular sync. If remote state did not change, it will result in a simple RemoteFastForward. Otherwise, we'll continue merging and requesting a follow-up. Schema alterations are explicitly not supported. As local transactions are rebased on top of remote, following changes happen: - entids are changed into tempids, letting transactor upsert :db/unique values - entids for retractions are changed into lookup-refs if we're confident they'll succeed -- otherwise, retractions are dropped on the floor * Post: use a macro for more readable tests * Tolstoy README
2018-09-08 02:18:20 +00:00
TxMapper::set_lg_mappings(&mut tx, vec![]).expect("empty map success");
let uuid1 = Uuid::new_v4();
let uuid2 = Uuid::new_v4();
TxMapper::set_lg_mappings(&mut tx, vec![(1, &uuid1).into(), (2, &uuid2).into()])
.expect("map success");
2020-08-06 03:03:58 +00:00
assert_eq!(Some(uuid1), TxMapper::get(&tx, 1).expect("success"));
assert_eq!(Some(uuid2), TxMapper::get(&tx, 2).expect("success"));
Basic sync support (#563) r=nalexander * Pre: remove remnants of 'open_empty' * Pre: Cleanup 'datoms' table after a timeline move Since timeline move operations use a transactor, they generate a "phantom" 'tx' and a 'txInstant' assertion. It is "phantom" in a sense that it was never present in the 'transactions' table, and is entirely synthetic as far as our database is concerned. It's an implementational artifact, and we were not cleaning it up. It becomes a problem when we start inserting transactions after a move. Once the transactor clashes with the phantom 'tx', it will retract the phantom 'txInstant' value, leaving the transactions log in an incorrect state. This patch adds a test for this scenario and elects the easy way out: simply remove the offending 'txInstant' datom. * Part 1: Sync without support for side-effects A "side-effect" is defined here as a mutation of a remote state as part of the sync. If, during a sync we determine that a remote state needs to be changed, bail out. This generally supports different variations of "baton-passing" syncing, where clients will succeed syncing if each change is non-conflicting. * Part 2: Support basic "side-effects" syncing This patch introduces a concept of a follow-up sync. If a sync generated a "merge transaction" (a regular transaction that contains assertions necessary for local and remote transaction logs to converge), then this transaction needs to be uploaded in a follow-up sync. Generated SyncReport indicates if a follow-up sync is required. Follow-up sync itself is just a regular sync. If remote state did not change, it will result in a simple RemoteFastForward. Otherwise, we'll continue merging and requesting a follow-up. Schema alterations are explicitly not supported. As local transactions are rebased on top of remote, following changes happen: - entids are changed into tempids, letting transactor upsert :db/unique values - entids for retractions are changed into lookup-refs if we're confident they'll succeed -- otherwise, retractions are dropped on the floor * Post: use a macro for more readable tests * Tolstoy README
2018-09-08 02:18:20 +00:00
// Now let's replace one of the mappings.
let new_uuid2 = Uuid::new_v4();
TxMapper::set_lg_mappings(&mut tx, vec![(1, &uuid1).into(), (2, &new_uuid2).into()])
.expect("map success");
2020-08-06 03:03:58 +00:00
assert_eq!(Some(uuid1), TxMapper::get(&tx, 1).expect("success"));
assert_eq!(Some(new_uuid2), TxMapper::get(&tx, 2).expect("success"));
}
}