Skip to content

BE-921: Calculate archival temporal changes before persistence #10125

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
montekki wants to merge 1 commit into from
Draft

BE-921: Calculate archival temporal changes before persistence #10125

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
34 changes: 34 additions & 0 deletions libs/@local/graph/postgres-store/src/store/postgres/knowledge/entity/batch/write.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -35,6 +35,40 @@ impl<C: AsClient> PostgresStore<C, InTransaction> {
&self,
history: &EntityHistory,
) -> Result<(), Report<UpdateError>> {
let removed: Vec<_> = history.removed().cloned().collect();
if !removed.is_empty() {
let parameters = EntityTemporalMetadataRow::columnar_parameters(&removed);
self.as_client()
.execute_raw(
"
DELETE FROM entity_temporal_metadata AS temporal
USING
unnest(
$1::UUID [],
$2::UUID [],
$3::UUID [],
$4::UUID [],
$5::TSTZRANGE [],
$6::TSTZRANGE []
)
AS previous (
web_id, entity_uuid, draft_id, edition_id,
decision_time, transaction_time
)
WHERE temporal.web_id = previous.web_id
AND temporal.entity_uuid = previous.entity_uuid
AND temporal.draft_id IS NOT DISTINCT FROM previous.draft_id
AND temporal.entity_edition_id = previous.edition_id
AND temporal.decision_time = previous.decision_time
AND temporal.transaction_time = previous.transaction_time
",
parameters
.into_iter()
.map(|(_, column)| column.into_values()),
)
.await
.change_context(UpdateError)?;
}
self.update_entity_temporal_rows(
history
.updated()
Expand Down
50 changes: 49 additions & 1 deletion libs/@local/graph/postgres-store/src/store/postgres/knowledge/entity/history.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15,16 +15,18 @@ struct TemporalEntry {
original: Option<EntityTemporalMetadataRow>,
}

/// Tracks the final replacement of each loaded row while retaining earlier history
/// Tracks the final replacement or removal of each loaded row while retaining earlier history
/// across both time axes.
pub(crate) struct EntityHistory {
current: HashMap<EntityId, Vec<TemporalEntry>>,
removed: Vec<EntityTemporalMetadataRow>,
}

impl EntityHistory {
pub(super) fn new(rows: impl IntoIterator<Item = EntityTemporalMetadataRow>) -> Self {
let mut history = Self {
current: HashMap::new(),
removed: Vec::new(),
};
for row in rows {
history.push(TemporalEntry {
Expand Down Expand Up @@ -138,6 +140,31 @@ impl EntityHistory {
self.push(entry);
}

/// Ends the live state that a published draft supersedes while retaining its past on both
/// time axes. A row is deleted only when neither earlier interval survives.
pub(super) fn close(
&mut self,
row: &EntityTemporalMetadataRow,
transaction_time: Timestamp<TransactionTime>,
decision_time: Timestamp<DecisionTime>,
) {
let mut entry = self.take(row);
let mut fragments = Self::split(row, transaction_time, decision_time);
if let Some(row) = fragments.pop() {
entry.row = row;
self.push(entry);
} else {
self.removed.extend(entry.original);
}
for fragment in fragments {
self.insert(fragment);
}
}

pub(super) fn removed(&self) -> impl Iterator<Item = &EntityTemporalMetadataRow> {
self.removed.iter()
}

/// Pairs each changed persisted row with its final replacement. Repeated edits are combined
/// so a bulk update never targets the same database row twice.
pub(super) fn updated(
Expand Down Expand Up @@ -287,4 +314,25 @@ mod tests {
);
}
}

#[test]
fn close_same_transaction_start() {
let original = row();
let mut history = EntityHistory::new([original.clone()]);
history.close(
&original,
Timestamp::from_unix_timestamp(0),
Timestamp::from_unix_timestamp(0),
);
assert_eq!(
history.inserted().count(),
0,
"should omit empty historical intervals"
);
assert_eq!(
history.removed().count(),
1,
"should remove the superseded row"
);
}
}
150 changes: 13 additions & 137 deletions libs/@local/graph/postgres-store/src/store/postgres/knowledge/entity/mod.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2898,157 +2898,33 @@ where
Ok(temporal_versioning)
}

/// Closes the superseded live interval and records the actor responsible for publication.
#[tracing::instrument(level = "info", skip(self))]
#[expect(clippy::too_many_lines)]
async fn archive_entity(
&self,
actor_id: ActorId,
locked_row: LockedEntityEdition,
transaction_time: Timestamp<TransactionTime>,
decision_time: Timestamp<DecisionTime>,
) -> Result<(), Report<UpdateError>> {
// Closing at the start would leave an empty decision interval
if locked_row.decision_time.start() == &ClosedTemporalBound::Inclusive(decision_time) {
self.as_client()
.query(
"DELETE FROM entity_temporal_metadata
WHERE web_id = $1
AND entity_uuid = $2
AND draft_id IS NOT DISTINCT FROM $3
AND transaction_time @> $4::timestamptz
AND decision_time @> $5::timestamptz;",
&[
&locked_row.entity_id.web_id,
&locked_row.entity_id.entity_uuid,
&locked_row.entity_id.draft_id,
&transaction_time,
&decision_time,
],
)
.instrument(tracing::info_span!(
"DELETE",
otel.kind = "client",
db.system = "postgresql",
peer.service = "Postgres",
))
.await
.change_context(UpdateError)?;
} else if let Some(draft_id) = locked_row.entity_id.draft_id {
self.client
.as_client()
.query(
"
UPDATE entity_temporal_metadata
SET transaction_time = tstzrange($4::timestamptz, NULL, '[)'),
decision_time = tstzrange(lower(decision_time), $5::timestamptz, '[)')
WHERE entity_temporal_metadata.web_id = $1
AND entity_temporal_metadata.entity_uuid = $2
AND entity_temporal_metadata.draft_id = $3
AND entity_temporal_metadata.transaction_time @> $4::timestamptz
AND entity_temporal_metadata.decision_time @> $5::timestamptz;",
&[
&locked_row.entity_id.web_id,
&locked_row.entity_id.entity_uuid,
&draft_id,
&transaction_time,
&decision_time,
],
)
.instrument(tracing::info_span!(
"UPDATE",
otel.kind = "client",
db.system = "postgresql",
peer.service = "Postgres"
))
.await
.change_context(UpdateError)?;
} else {
self.client
.as_client()
.query(
"
UPDATE entity_temporal_metadata
SET transaction_time = tstzrange($3::timestamptz, NULL, '[)'),
decision_time = tstzrange(lower(decision_time), $4::timestamptz, '[)')
WHERE entity_temporal_metadata.web_id = $1
AND entity_temporal_metadata.entity_uuid = $2
AND entity_temporal_metadata.draft_id IS NULL
AND entity_temporal_metadata.transaction_time @> $3::timestamptz
AND entity_temporal_metadata.decision_time @> $4::timestamptz;",
&[
&locked_row.entity_id.web_id,
&locked_row.entity_id.entity_uuid,
&transaction_time,
&decision_time,
],
)
.instrument(tracing::info_span!(
"UPDATE",
otel.kind = "client",
db.system = "postgresql",
peer.service = "Postgres"
))
.await
.change_context(UpdateError)?;
}

self.client
.as_client()
.query(
let previous = EntityTemporalMetadataRow::from(locked_row);
let mut history = EntityHistory::new([previous.clone()]);
history.close(&previous, transaction_time, decision_time);
self.update_entity_history(&history).await?;
self.as_client()
.execute(
"
INSERT INTO entity_temporal_metadata (
web_id,
entity_uuid,
draft_id,
entity_edition_id,
transaction_time,
decision_time
) SELECT
$1,
$2,
$3,
$4,
tstzrange(lower($6::tstzrange), $7, '[)'),
$5
WHERE lower($6::tstzrange) < $7::timestamptz;",
UPDATE entity_editions
SET provenance = provenance || jsonb_build_object('archivedById', $2::UUID)
WHERE entity_edition_id = $1
",
&[
&locked_row.entity_id.web_id,
&locked_row.entity_id.entity_uuid,
&locked_row.entity_id.draft_id,
&locked_row.entity_edition_id,
&locked_row.decision_time,
&locked_row.transaction_time,
&transaction_time,
&previous.entity_edition_id,
&ActorEntityUuid::from(actor_id),
],
)
.instrument(tracing::info_span!(
"INSERT",
otel.kind = "client",
db.system = "postgresql",
peer.service = "Postgres",
))
.await
.change_context(UpdateError)?;

self.as_client()
.query(
"
UPDATE entity_editions SET
provenance = provenance || JSONB_BUILD_OBJECT(
'archivedById', $2::UUID
)
WHERE entity_edition_id = $1",
&[&locked_row.entity_edition_id, &actor_id],
)
.instrument(tracing::info_span!(
"UPDATE",
otel.kind = "client",
db.system = "postgresql",
peer.service = "Postgres",
))
.await
.change_context(UpdateError)?;

Ok(())
}
}
Expand Down
Loading