1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476
use nohash_hasher::IntMap;
use re_log_types::{DataCell, EntityPath, RowId, StoreId, TimeInt, TimePoint, Timeline};
use re_types_core::ComponentName;
use crate::StoreGeneration;
// Used all over in docstrings.
#[allow(unused_imports)]
use crate::{DataStore, StoreSubscriber};
// ---
/// The atomic unit of change in the Rerun [`DataStore`].
///
/// A [`StoreEvent`] describes the changes caused by the addition or deletion of a
/// [`re_log_types::DataRow`] in the store.
///
/// Methods that mutate the [`DataStore`], such as [`DataStore::insert_row`] and [`DataStore::gc`],
/// return [`StoreEvent`]s that describe the changes.
/// You can also register your own [`StoreSubscriber`] in order to be notified of changes as soon as they
/// happen.
///
/// Refer to field-level documentation for more details and check out [`StoreDiff`] for a precise
/// definition of what an event involves.
#[derive(Debug, Clone, PartialEq)]
pub struct StoreEvent {
/// Which [`DataStore`] sent this event?
pub store_id: StoreId,
/// What was the store's generation when it sent that event?
pub store_generation: StoreGeneration,
/// Monotonically increasing ID of the event.
///
/// This is on a per-store basis.
///
/// When handling a [`StoreEvent`], if this is the first time you process this [`StoreId`] and
/// the associated `event_id` is not `1`, it means you registered late and missed some updates.
pub event_id: u64,
/// What actually changed?
///
/// Refer to [`StoreDiff`] for more information.
pub diff: StoreDiff,
}
impl std::ops::Deref for StoreEvent {
type Target = StoreDiff;
#[inline]
fn deref(&self) -> &Self::Target {
&self.diff
}
}
/// Is it an addition or a deletion?
///
/// Reminder: ⚠ Do not confuse _a deletion_ and _a clear_ ⚠.
///
/// A deletion is the result of a row being completely removed from the store as part of the
/// garbage collection process.
///
/// A clear, on the other hand, is the act of logging an empty [`re_types_core::ComponentBatch`],
/// either directly using the logging APIs, or indirectly through the use of a
/// [`re_types_core::archetypes::Clear`] archetype.
#[derive(Debug, Clone, Copy, PartialEq, Eq)]
pub enum StoreDiffKind {
Addition,
Deletion,
}
impl StoreDiffKind {
#[inline]
pub fn delta(&self) -> i64 {
match self {
StoreDiffKind::Addition => 1,
StoreDiffKind::Deletion => -1,
}
}
}
/// Describes an atomic change in the Rerun [`DataStore`]: a row has been added or deleted.
///
/// From a query model standpoint, the [`DataStore`] _always_ operates one row at a time:
/// - The contents of a row (i.e. its columns) are immutable past insertion, by virtue of
/// [`RowId`]s being unique and non-reusable.
/// - Similarly, garbage collection always removes _all the data_ associated with a row in one go:
/// there cannot be orphaned columns. When a row is gone, all data associated with it is gone too.
///
/// Refer to field-level documentation for more information.
#[derive(Debug, Clone, PartialEq)]
pub struct StoreDiff {
/// Addition or deletion?
///
/// The store's internals are opaque and don't necessarily reflect the query model (e.g. there
/// might be data in the store that cannot by reached by any query).
///
/// A [`StoreDiff`] answers a logical question: "does there exist a query path which can return
/// data from that row?".
///
/// An event of kind deletion only tells you that, from this point on, no query can return data from that row.
/// That doesn't necessarily mean that the data is actually gone, i.e. don't make assumptions of e.g. the size
/// in bytes of the store based on these events.
/// They are in "query-model space" and are not an accurate representation of what happens in storage space.
pub kind: StoreDiffKind,
/// What's the row's [`RowId`]?
///
/// [`RowId`]s are guaranteed to be unique within a single [`DataStore`].
///
/// Put another way, the same [`RowId`] can only appear twice in a [`StoreDiff`] event:
/// one addition and (optionally) one deletion (in that order!).
pub row_id: RowId,
/// The time data associated with that row.
///
/// Since insertions and deletions both work on a row-level basis, this is guaranteed to be the
/// same value for both the insertion and deletion events (if any).
///
/// This is not a [`TimePoint`] for performance reasons.
//
// NOTE: Empirical testing shows that a SmallVec isn't any better in the best case, and can be a
// significant performant drop at worst.
// pub times: SmallVec<[(Timeline, TimeInt); 5]>, // "5 timelines ought to be enough for anyone"
pub times: Vec<(Timeline, TimeInt)>,
/// The [`EntityPath`] associated with that row.
///
/// Since insertions and deletions both work on a row-level basis, this is guaranteed to be the
/// same value for both the insertion and deletion events (if any).
pub entity_path: EntityPath,
/// All the [`DataCell`]s associated with that row.
///
/// Since insertions and deletions both work on a row-level basis, this is guaranteed to be the
/// same set of values for both the insertion and deletion events (if any).
pub cells: IntMap<ComponentName, DataCell>,
}
impl StoreDiff {
#[inline]
pub fn addition(row_id: impl Into<RowId>, entity_path: impl Into<EntityPath>) -> Self {
Self {
kind: StoreDiffKind::Addition,
row_id: row_id.into(),
times: Default::default(),
entity_path: entity_path.into(),
cells: Default::default(),
}
}
#[inline]
pub fn deletion(row_id: impl Into<RowId>, entity_path: impl Into<EntityPath>) -> Self {
Self {
kind: StoreDiffKind::Deletion,
row_id: row_id.into(),
times: Default::default(),
entity_path: entity_path.into(),
cells: Default::default(),
}
}
#[inline]
pub fn at_timepoint(&mut self, timepoint: impl Into<TimePoint>) -> &mut Self {
self.times.extend(timepoint.into());
self
}
#[inline]
pub fn at_timestamp(
&mut self,
timeline: impl Into<Timeline>,
time: impl Into<TimeInt>,
) -> &mut Self {
self.times.push((timeline.into(), time.into()));
self
}
#[inline]
pub fn with_cells(&mut self, cells: impl IntoIterator<Item = DataCell>) -> &mut Self {
self.cells
.extend(cells.into_iter().map(|cell| (cell.component_name(), cell)));
self
}
#[inline]
pub fn timepoint(&self) -> TimePoint {
self.times.clone().into_iter().collect()
}
#[inline]
pub fn is_static(&self) -> bool {
self.times.is_empty()
}
/// `-1` for deletions, `+1` for additions.
#[inline]
pub fn delta(&self) -> i64 {
self.kind.delta()
}
#[inline]
pub fn num_components(&self) -> usize {
self.cells.len()
}
}
#[cfg(test)]
mod tests {
use std::collections::BTreeMap;
use re_log_types::{
example_components::{MyColor, MyIndex, MyPoint},
DataRow, RowId, TimePoint, Timeline,
};
use re_types_core::Loggable as _;
use crate::{DataStore, GarbageCollectionOptions};
use super::*;
/// A simple store subscriber for test purposes that keeps track of the quantity of data available
/// in the store at the lowest level of detail.
///
/// The counts represent numbers of rows: e.g. how many unique rows contain this entity path?
#[derive(Default, Debug, PartialEq, Eq)]
struct GlobalCounts {
row_ids: BTreeMap<RowId, i64>,
timelines: BTreeMap<Timeline, i64>,
entity_paths: BTreeMap<EntityPath, i64>,
component_names: BTreeMap<ComponentName, i64>,
times: BTreeMap<TimeInt, i64>,
num_static: i64,
}
impl GlobalCounts {
fn new(
row_ids: impl IntoIterator<Item = (RowId, i64)>, //
timelines: impl IntoIterator<Item = (Timeline, i64)>, //
entity_paths: impl IntoIterator<Item = (EntityPath, i64)>, //
component_names: impl IntoIterator<Item = (ComponentName, i64)>, //
times: impl IntoIterator<Item = (TimeInt, i64)>, //
num_static: i64,
) -> Self {
Self {
row_ids: row_ids.into_iter().collect(),
timelines: timelines.into_iter().collect(),
entity_paths: entity_paths.into_iter().collect(),
component_names: component_names.into_iter().collect(),
times: times.into_iter().collect(),
num_static,
}
}
}
impl GlobalCounts {
fn on_events(&mut self, events: &[StoreEvent]) {
for event in events {
let delta = event.delta();
*self.row_ids.entry(event.row_id).or_default() += delta;
*self
.entity_paths
.entry(event.entity_path.clone())
.or_default() += delta;
for component_name in event.cells.keys() {
*self.component_names.entry(*component_name).or_default() += delta;
}
if event.is_static() {
self.num_static += delta;
} else {
for &(timeline, time) in &event.times {
*self.timelines.entry(timeline).or_default() += delta;
*self.times.entry(time).or_default() += delta;
}
}
}
}
}
#[test]
fn store_events() -> anyhow::Result<()> {
let mut store = DataStore::new(
re_log_types::StoreId::random(re_log_types::StoreKind::Recording),
Default::default(),
);
let mut view = GlobalCounts::default();
let timeline_frame = Timeline::new_sequence("frame");
let timeline_other = Timeline::new_temporal("other");
let timeline_yet_another = Timeline::new_sequence("yet_another");
let row_id1 = RowId::new();
let timepoint1 = TimePoint::from_iter([
(timeline_frame, 42), //
(timeline_other, 666), //
(timeline_yet_another, 1), //
]);
let entity_path1: EntityPath = "entity_a".into();
let row1 = DataRow::from_component_batches(
row_id1,
timepoint1.clone(),
entity_path1.clone(),
[&MyIndex::from_iter(0..10) as _],
)?;
view.on_events(&[store.insert_row(&row1)?]);
similar_asserts::assert_eq!(
GlobalCounts::new(
[
(row_id1, 1), //
],
[
(timeline_frame, 1),
(timeline_other, 1),
(timeline_yet_another, 1),
],
[
(entity_path1.clone(), 1), //
],
[
(MyIndex::name(), 1), //
],
[
(42.try_into().unwrap(), 1), //
(666.try_into().unwrap(), 1),
(1.try_into().unwrap(), 1),
],
0,
),
view,
);
let row_id2 = RowId::new();
let timepoint2 = TimePoint::from_iter([
(timeline_frame, 42), //
(timeline_yet_another, 1), //
]);
let entity_path2: EntityPath = "entity_b".into();
let row2 = {
let num_instances = 3;
let points: Vec<_> = (0..num_instances)
.map(|i| MyPoint::new(0.0, i as f32))
.collect();
let colors = vec![MyColor::from(0xFF0000FF)];
DataRow::from_component_batches(
row_id2,
timepoint2.clone(),
entity_path2.clone(),
[&points as _, &colors as _],
)?
};
view.on_events(&[store.insert_row(&row2)?]);
similar_asserts::assert_eq!(
GlobalCounts::new(
[
(row_id1, 1), //
(row_id2, 1),
],
[
(timeline_frame, 2),
(timeline_other, 1),
(timeline_yet_another, 2),
],
[
(entity_path1.clone(), 1), //
(entity_path2.clone(), 1), //
],
[
(MyIndex::name(), 1), // autogenerated, doesn't change
(MyPoint::name(), 1), //
(MyColor::name(), 1), //
],
[
(42.try_into().unwrap(), 2), //
(666.try_into().unwrap(), 1),
(1.try_into().unwrap(), 2),
],
0,
),
view,
);
let row_id3 = RowId::new();
let timepoint3 = TimePoint::default();
let row3 = {
let num_instances = 6;
let colors = vec![MyColor::from(0x00DD00FF); num_instances];
DataRow::from_component_batches(
row_id3,
timepoint3.clone(),
entity_path2.clone(),
[
&MyIndex::from_iter(0..num_instances as _) as _,
&colors as _,
],
)?
};
view.on_events(&[store.insert_row(&row3)?]);
similar_asserts::assert_eq!(
GlobalCounts::new(
[
(row_id1, 1), //
(row_id2, 1),
(row_id3, 1),
],
[
(timeline_frame, 2),
(timeline_other, 1),
(timeline_yet_another, 2),
],
[
(entity_path1.clone(), 1), //
(entity_path2.clone(), 2), //
],
[
(MyIndex::name(), 2), //
(MyPoint::name(), 1), //
(MyColor::name(), 2), //
],
[
(42.try_into().unwrap(), 2), //
(666.try_into().unwrap(), 1),
(1.try_into().unwrap(), 2),
],
1,
),
view,
);
let events = store.gc(&GarbageCollectionOptions::gc_everything()).0;
view.on_events(&events);
similar_asserts::assert_eq!(
GlobalCounts::new(
[
(row_id1, 0), //
(row_id2, 0),
(row_id3, 1), // static -- no gc
],
[
(timeline_frame, 0),
(timeline_other, 0),
(timeline_yet_another, 0),
],
[
(entity_path1.clone(), 0), //
(entity_path2.clone(), 1), // static -- no gc
],
[
(MyIndex::name(), 1), // static -- no gc
(MyPoint::name(), 0), //
(MyColor::name(), 1), // static -- no gc
],
[
(42.try_into().unwrap(), 0), //
(666.try_into().unwrap(), 0),
(1.try_into().unwrap(), 0),
],
1, // static -- no gc
),
view,
);
Ok(())
}
}