Skip to content

Commit

Permalink
indexer-alt: event indices
Browse files Browse the repository at this point in the history
## Description

Adding pipelines to index all tables used to filter events. They differ
from the equivalent schemas in the existing indexer in the following
ways:

- They only mention the transaction sequence number, and not the event
  sequent number. To use these tables, we first filter down to the
  transaction containing the event, and then scan the events in that
  transaction.
- Struct instantiations are stored as a separate name field and then a
  BCS encoded type tag. This is to reduce their footprint (package IDs
  weight twice as much when stored as text compared to BCS), and because
  we only ever filter using an exact match, so we don't need to store
  the instantiation as text.

## Test plan

Ran the indexer locally and spot checked the events.
  • Loading branch information
amnn committed Oct 19, 2024
1 parent 31f4571 commit 59571f6
Show file tree
Hide file tree
Showing 13 changed files with 606 additions and 0 deletions.
Original file line number Diff line number Diff line change
@@ -0,0 +1,6 @@
DROP TABLE IF EXISTS ev_emit_pkg;
DROP TABLE IF EXISTS ev_emit_mod;
DROP TABLE IF EXISTS ev_struct_pkg;
DROP TABLE IF EXISTS ev_struct_mod;
DROP TABLE IF EXISTS ev_struct_name;
DROP TABLE IF EXISTS ev_struct_inst;
Original file line number Diff line number Diff line change
@@ -0,0 +1,91 @@
CREATE TABLE IF NOT EXISTS ev_emit_pkg
(
package BYTEA,
tx_sequence_number BIGINT,
sender BYTEA NOT NULL,
PRIMARY KEY(package, tx_sequence_number)
);

CREATE INDEX IF NOT EXISTS ev_emit_pkg_tx_sequence_number
ON ev_emit_pkg (tx_sequence_number);

CREATE INDEX IF NOT EXISTS ev_emit_pkg_sender
ON ev_emit_pkg (sender, package, tx_sequence_number);

CREATE TABLE IF NOT EXISTS ev_emit_mod
(
package BYTEA,
module TEXT,
tx_sequence_number BIGINT,
sender BYTEA NOT NULL,
PRIMARY KEY(package, module, tx_sequence_number)
);

CREATE INDEX IF NOT EXISTS ev_emit_mod_tx_sequence_number
ON ev_emit_mod (tx_sequence_number);

CREATE INDEX IF NOT EXISTS ev_emit_mod_sender
ON ev_emit_mod (sender, package, module, tx_sequence_number);

CREATE TABLE IF NOT EXISTS ev_struct_pkg
(
package BYTEA,
tx_sequence_number BIGINT,
sender BYTEA NOT NULL,
PRIMARY KEY(package, tx_sequence_number)
);

CREATE INDEX IF NOT EXISTS ev_struct_pkg_tx_sequence_number
ON ev_struct_pkg (tx_sequence_number);

CREATE INDEX IF NOT EXISTS ev_struct_pkg_sender
ON ev_struct_pkg (sender, package, tx_sequence_number);

CREATE TABLE IF NOT EXISTS ev_struct_mod
(
package BYTEA,
module TEXT,
tx_sequence_number BIGINT,
sender BYTEA NOT NULL,
PRIMARY KEY(package, module, tx_sequence_number)
);

CREATE INDEX IF NOT EXISTS ev_struct_mod_tx_sequence_number
ON ev_struct_mod (tx_sequence_number);

CREATE INDEX IF NOT EXISTS ev_struct_mod_sender
ON ev_struct_mod (sender, package, module, tx_sequence_number);

CREATE TABLE IF NOT EXISTS ev_struct_name
(
package BYTEA,
module TEXT,
name TEXT,
tx_sequence_number BIGINT,
sender BYTEA NOT NULL,
PRIMARY KEY(package, module, name, tx_sequence_number)
);

CREATE INDEX IF NOT EXISTS ev_struct_name_tx_sequence_number
ON ev_struct_name (tx_sequence_number);

CREATE INDEX IF NOT EXISTS ev_struct_name_sender
ON ev_struct_name (sender, package, module, name, tx_sequence_number);

CREATE TABLE IF NOT EXISTS ev_struct_inst
(
package BYTEA,
module TEXT,
name TEXT,
-- BCS encoded array of TypeTags for type parameters.
instantiation BYTEA,
tx_sequence_number BIGINT,
sender BYTEA NOT NULL,
PRIMARY KEY(package, module, instantiation, tx_sequence_number)
);

CREATE INDEX IF NOT EXISTS ev_struct_inst_tx_sequence_number
ON ev_struct_inst (tx_sequence_number);

CREATE INDEX IF NOT EXISTS ev_struct_inst_sender
ON ev_struct_inst (sender, package, module, instantiation, tx_sequence_number);
60 changes: 60 additions & 0 deletions crates/sui-indexer-alt/src/handlers/ev_emit_mod.rs
Original file line number Diff line number Diff line change
@@ -0,0 +1,60 @@
// Copyright (c) Mysten Labs, Inc.
// SPDX-License-Identifier: Apache-2.0

use std::{collections::BTreeSet, sync::Arc};

use anyhow::Result;
use diesel_async::RunQueryDsl;
use sui_types::full_checkpoint_content::CheckpointData;

use crate::{db, models::events::StoredEvEmitMod, schema::ev_emit_mod};

use super::Handler;

pub struct EvEmitMod;

#[async_trait::async_trait]
impl Handler for EvEmitMod {
const NAME: &'static str = "ev_emit_mod";

const BATCH_SIZE: usize = 100;
const CHUNK_SIZE: usize = 1000;
const MAX_PENDING_SIZE: usize = 10000;

type Value = StoredEvEmitMod;

fn handle(checkpoint: &Arc<CheckpointData>) -> Result<Vec<Self::Value>> {
let CheckpointData {
transactions,
checkpoint_summary,
..
} = checkpoint.as_ref();

let mut values = BTreeSet::new();
let first_tx = checkpoint_summary.network_total_transactions as usize - transactions.len();

for (i, tx) in transactions.iter().enumerate() {
values.extend(
tx.events
.iter()
.flat_map(|evs| &evs.data)
.map(|ev| StoredEvEmitMod {
package: ev.package_id.to_vec(),
module: ev.transaction_module.to_string(),
tx_sequence_number: (first_tx + i) as i64,
sender: ev.sender.to_vec(),
}),
);
}

Ok(values.into_iter().collect())
}

async fn commit(values: &[Self::Value], conn: &mut db::Connection<'_>) -> Result<usize> {
Ok(diesel::insert_into(ev_emit_mod::table)
.values(values)
.on_conflict_do_nothing()
.execute(conn)
.await?)
}
}
59 changes: 59 additions & 0 deletions crates/sui-indexer-alt/src/handlers/ev_emit_pkg.rs
Original file line number Diff line number Diff line change
@@ -0,0 +1,59 @@
// Copyright (c) Mysten Labs, Inc.
// SPDX-License-Identifier: Apache-2.0

use std::{collections::BTreeSet, sync::Arc};

use anyhow::Result;
use diesel_async::RunQueryDsl;
use sui_types::full_checkpoint_content::CheckpointData;

use crate::{db, models::events::StoredEvEmitPkg, schema::ev_emit_pkg};

use super::Handler;

pub struct EvEmitPkg;

#[async_trait::async_trait]
impl Handler for EvEmitPkg {
const NAME: &'static str = "ev_emit_pkg";

const BATCH_SIZE: usize = 100;
const CHUNK_SIZE: usize = 1000;
const MAX_PENDING_SIZE: usize = 10000;

type Value = StoredEvEmitPkg;

fn handle(checkpoint: &Arc<CheckpointData>) -> Result<Vec<Self::Value>> {
let CheckpointData {
transactions,
checkpoint_summary,
..
} = checkpoint.as_ref();

let mut values = BTreeSet::new();
let first_tx = checkpoint_summary.network_total_transactions as usize - transactions.len();

for (i, tx) in transactions.iter().enumerate() {
values.extend(
tx.events
.iter()
.flat_map(|evs| &evs.data)
.map(|ev| StoredEvEmitPkg {
package: ev.package_id.to_vec(),
tx_sequence_number: (first_tx + i) as i64,
sender: ev.sender.to_vec(),
}),
);
}

Ok(values.into_iter().collect())
}

async fn commit(values: &[Self::Value], conn: &mut db::Connection<'_>) -> Result<usize> {
Ok(diesel::insert_into(ev_emit_pkg::table)
.values(values)
.on_conflict_do_nothing()
.execute(conn)
.await?)
}
}
63 changes: 63 additions & 0 deletions crates/sui-indexer-alt/src/handlers/ev_struct_inst.rs
Original file line number Diff line number Diff line change
@@ -0,0 +1,63 @@
// Copyright (c) Mysten Labs, Inc.
// SPDX-License-Identifier: Apache-2.0

use std::{collections::BTreeSet, sync::Arc};

use anyhow::{Context, Result};
use diesel_async::RunQueryDsl;
use sui_types::full_checkpoint_content::CheckpointData;

use crate::{db, models::events::StoredEvStructInst, schema::ev_struct_inst};

use super::Handler;

pub struct EvStructInst;

#[async_trait::async_trait]
impl Handler for EvStructInst {
const NAME: &'static str = "ev_struct_inst";

const BATCH_SIZE: usize = 100;
const CHUNK_SIZE: usize = 1000;
const MAX_PENDING_SIZE: usize = 10000;

type Value = StoredEvStructInst;

fn handle(checkpoint: &Arc<CheckpointData>) -> Result<Vec<Self::Value>> {
let CheckpointData {
transactions,
checkpoint_summary,
..
} = checkpoint.as_ref();

let mut values = BTreeSet::new();
let first_tx = checkpoint_summary.network_total_transactions as usize - transactions.len();

for (i, tx) in transactions.iter().enumerate() {
let tx_sequence_number = (first_tx + i) as i64;
for (j, ev) in tx.events.iter().flat_map(|evs| evs.data.iter().enumerate()) {
values.insert(StoredEvStructInst {
package: ev.type_.address.to_vec(),
module: ev.type_.module.to_string(),
name: ev.type_.name.to_string(),
instantiation: bcs::to_bytes(&ev.type_.type_params)
.with_context(|| format!(
"Failed to serialize type parameters for event ({tx_sequence_number}, {j})"
))?,
tx_sequence_number: (first_tx + i) as i64,
sender: ev.sender.to_vec(),
});
}
}

Ok(values.into_iter().collect())
}

async fn commit(values: &[Self::Value], conn: &mut db::Connection<'_>) -> Result<usize> {
Ok(diesel::insert_into(ev_struct_inst::table)
.values(values)
.on_conflict_do_nothing()
.execute(conn)
.await?)
}
}
60 changes: 60 additions & 0 deletions crates/sui-indexer-alt/src/handlers/ev_struct_mod.rs
Original file line number Diff line number Diff line change
@@ -0,0 +1,60 @@
// Copyright (c) Mysten Labs, Inc.
// SPDX-License-Identifier: Apache-2.0

use std::{collections::BTreeSet, sync::Arc};

use anyhow::Result;
use diesel_async::RunQueryDsl;
use sui_types::full_checkpoint_content::CheckpointData;

use crate::{db, models::events::StoredEvStructMod, schema::ev_struct_mod};

use super::Handler;

pub struct EvStructMod;

#[async_trait::async_trait]
impl Handler for EvStructMod {
const NAME: &'static str = "ev_struct_mod";

const BATCH_SIZE: usize = 100;
const CHUNK_SIZE: usize = 1000;
const MAX_PENDING_SIZE: usize = 10000;

type Value = StoredEvStructMod;

fn handle(checkpoint: &Arc<CheckpointData>) -> Result<Vec<Self::Value>> {
let CheckpointData {
transactions,
checkpoint_summary,
..
} = checkpoint.as_ref();

let mut values = BTreeSet::new();
let first_tx = checkpoint_summary.network_total_transactions as usize - transactions.len();

for (i, tx) in transactions.iter().enumerate() {
values.extend(
tx.events
.iter()
.flat_map(|evs| &evs.data)
.map(|ev| StoredEvStructMod {
package: ev.type_.address.to_vec(),
module: ev.type_.module.to_string(),
tx_sequence_number: (first_tx + i) as i64,
sender: ev.sender.to_vec(),
}),
);
}

Ok(values.into_iter().collect())
}

async fn commit(values: &[Self::Value], conn: &mut db::Connection<'_>) -> Result<usize> {
Ok(diesel::insert_into(ev_struct_mod::table)
.values(values)
.on_conflict_do_nothing()
.execute(conn)
.await?)
}
}
Loading

0 comments on commit 59571f6

Please sign in to comment.