mirror of
https://forgejo.ellis.link/continuwuation/continuwuity.git
synced 2025-06-30 15:34:29 +02:00
combine service/users data w/ mod unit split sliding sync related out of service/users instrument database entry points remove increment crap from database interface de-wrap all database get() calls de-wrap all database insert() calls de-wrap all database remove() calls refactor database interface for async streaming add query key serializer for database implement Debug for result handle add query deserializer for database add deserialization trait for option handle start a stream utils suite de-wrap/asyncify/type-query count_one_time_keys() de-wrap/asyncify users count add admin query users command suite de-wrap/asyncify users exists de-wrap/partially asyncify user filter related asyncify/de-wrap users device/keys related asyncify/de-wrap user auth/misc related asyncify/de-wrap users blurhash asyncify/de-wrap account_data get; merge Data into Service partial asyncify/de-wrap uiaa; merge Data into Service partially asyncify/de-wrap transaction_ids get; merge Data into Service partially asyncify/de-wrap key_backups; merge Data into Service asyncify/de-wrap pusher service getters; merge Data into Service asyncify/de-wrap rooms alias getters/some iterators asyncify/de-wrap rooms directory getters/iterator partially asyncify/de-wrap rooms lazy-loading partially asyncify/de-wrap rooms metadata asyncify/dewrap rooms outlier asyncify/dewrap rooms pdu_metadata dewrap/partially asyncify rooms read receipt de-wrap rooms search service de-wrap/partially asyncify rooms user service partial de-wrap rooms state_compressor de-wrap rooms state_cache de-wrap room state et al de-wrap rooms timeline service additional users device/keys related de-wrap/asyncify sender asyncify services refactor database to TryFuture/TryStream refactor services for TryFuture/TryStream asyncify api handlers additional asyncification for admin module abstract stream related; support reverse streams additional stream conversions asyncify state-res related Signed-off-by: Jason Volk <jason@zemos.net>
359 lines
10 KiB
Rust
359 lines
10 KiB
Rust
mod data;
|
|
|
|
use std::{
|
|
collections::HashSet,
|
|
fmt::Write,
|
|
mem::size_of,
|
|
sync::{Arc, Mutex as StdMutex, Mutex},
|
|
};
|
|
|
|
use conduit::{checked, utils, utils::math::usize_from_f64, Result};
|
|
use data::Data;
|
|
use lru_cache::LruCache;
|
|
use ruma::{EventId, RoomId};
|
|
|
|
use self::data::StateDiff;
|
|
use crate::{rooms, Dep};
|
|
|
|
type StateInfoLruCache = Mutex<
|
|
LruCache<
|
|
u64,
|
|
Vec<(
|
|
u64, // sstatehash
|
|
Arc<HashSet<CompressedStateEvent>>, // full state
|
|
Arc<HashSet<CompressedStateEvent>>, // added
|
|
Arc<HashSet<CompressedStateEvent>>, // removed
|
|
)>,
|
|
>,
|
|
>;
|
|
|
|
type ShortStateInfoResult = Vec<(
|
|
u64, // sstatehash
|
|
Arc<HashSet<CompressedStateEvent>>, // full state
|
|
Arc<HashSet<CompressedStateEvent>>, // added
|
|
Arc<HashSet<CompressedStateEvent>>, // removed
|
|
)>;
|
|
|
|
type ParentStatesVec = Vec<(
|
|
u64, // sstatehash
|
|
Arc<HashSet<CompressedStateEvent>>, // full state
|
|
Arc<HashSet<CompressedStateEvent>>, // added
|
|
Arc<HashSet<CompressedStateEvent>>, // removed
|
|
)>;
|
|
|
|
type HashSetCompressStateEvent = (u64, Arc<HashSet<CompressedStateEvent>>, Arc<HashSet<CompressedStateEvent>>);
|
|
pub type CompressedStateEvent = [u8; 2 * size_of::<u64>()];
|
|
|
|
pub struct Service {
|
|
db: Data,
|
|
services: Services,
|
|
pub stateinfo_cache: StateInfoLruCache,
|
|
}
|
|
|
|
struct Services {
|
|
short: Dep<rooms::short::Service>,
|
|
state: Dep<rooms::state::Service>,
|
|
}
|
|
|
|
impl crate::Service for Service {
|
|
fn build(args: crate::Args<'_>) -> Result<Arc<Self>> {
|
|
let config = &args.server.config;
|
|
let cache_capacity = f64::from(config.stateinfo_cache_capacity) * config.cache_capacity_modifier;
|
|
Ok(Arc::new(Self {
|
|
db: Data::new(args.db),
|
|
services: Services {
|
|
short: args.depend::<rooms::short::Service>("rooms::short"),
|
|
state: args.depend::<rooms::state::Service>("rooms::state"),
|
|
},
|
|
stateinfo_cache: StdMutex::new(LruCache::new(usize_from_f64(cache_capacity)?)),
|
|
}))
|
|
}
|
|
|
|
fn memory_usage(&self, out: &mut dyn Write) -> Result<()> {
|
|
let stateinfo_cache = self.stateinfo_cache.lock().expect("locked").len();
|
|
writeln!(out, "stateinfo_cache: {stateinfo_cache}")?;
|
|
|
|
Ok(())
|
|
}
|
|
|
|
fn clear_cache(&self) { self.stateinfo_cache.lock().expect("locked").clear(); }
|
|
|
|
fn name(&self) -> &str { crate::service::make_name(std::module_path!()) }
|
|
}
|
|
|
|
impl Service {
|
|
/// Returns a stack with info on shortstatehash, full state, added diff and
|
|
/// removed diff for the selected shortstatehash and each parent layer.
|
|
pub async fn load_shortstatehash_info(&self, shortstatehash: u64) -> Result<ShortStateInfoResult> {
|
|
if let Some(r) = self
|
|
.stateinfo_cache
|
|
.lock()
|
|
.expect("locked")
|
|
.get_mut(&shortstatehash)
|
|
{
|
|
return Ok(r.clone());
|
|
}
|
|
|
|
let StateDiff {
|
|
parent,
|
|
added,
|
|
removed,
|
|
} = self.db.get_statediff(shortstatehash).await?;
|
|
|
|
if let Some(parent) = parent {
|
|
let mut response = Box::pin(self.load_shortstatehash_info(parent)).await?;
|
|
let mut state = (*response.last().expect("at least one response").1).clone();
|
|
state.extend(added.iter().copied());
|
|
let removed = (*removed).clone();
|
|
for r in &removed {
|
|
state.remove(r);
|
|
}
|
|
|
|
response.push((shortstatehash, Arc::new(state), added, Arc::new(removed)));
|
|
|
|
self.stateinfo_cache
|
|
.lock()
|
|
.expect("locked")
|
|
.insert(shortstatehash, response.clone());
|
|
|
|
Ok(response)
|
|
} else {
|
|
let response = vec![(shortstatehash, added.clone(), added, removed)];
|
|
self.stateinfo_cache
|
|
.lock()
|
|
.expect("locked")
|
|
.insert(shortstatehash, response.clone());
|
|
|
|
Ok(response)
|
|
}
|
|
}
|
|
|
|
pub async fn compress_state_event(&self, shortstatekey: u64, event_id: &EventId) -> CompressedStateEvent {
|
|
let mut v = shortstatekey.to_be_bytes().to_vec();
|
|
v.extend_from_slice(
|
|
&self
|
|
.services
|
|
.short
|
|
.get_or_create_shorteventid(event_id)
|
|
.await
|
|
.to_be_bytes(),
|
|
);
|
|
|
|
v.try_into().expect("we checked the size above")
|
|
}
|
|
|
|
/// Returns shortstatekey, event id
|
|
#[inline]
|
|
pub async fn parse_compressed_state_event(
|
|
&self, compressed_event: &CompressedStateEvent,
|
|
) -> Result<(u64, Arc<EventId>)> {
|
|
use utils::u64_from_u8;
|
|
|
|
let shortstatekey = u64_from_u8(&compressed_event[0..size_of::<u64>()]);
|
|
let event_id = self
|
|
.services
|
|
.short
|
|
.get_eventid_from_short(u64_from_u8(&compressed_event[size_of::<u64>()..]))
|
|
.await?;
|
|
|
|
Ok((shortstatekey, event_id))
|
|
}
|
|
|
|
/// Creates a new shortstatehash that often is just a diff to an already
|
|
/// existing shortstatehash and therefore very efficient.
|
|
///
|
|
/// There are multiple layers of diffs. The bottom layer 0 always contains
|
|
/// the full state. Layer 1 contains diffs to states of layer 0, layer 2
|
|
/// diffs to layer 1 and so on. If layer n > 0 grows too big, it will be
|
|
/// combined with layer n-1 to create a new diff on layer n-1 that's
|
|
/// based on layer n-2. If that layer is also too big, it will recursively
|
|
/// fix above layers too.
|
|
///
|
|
/// * `shortstatehash` - Shortstatehash of this state
|
|
/// * `statediffnew` - Added to base. Each vec is shortstatekey+shorteventid
|
|
/// * `statediffremoved` - Removed from base. Each vec is
|
|
/// shortstatekey+shorteventid
|
|
/// * `diff_to_sibling` - Approximately how much the diff grows each time
|
|
/// for this layer
|
|
/// * `parent_states` - A stack with info on shortstatehash, full state,
|
|
/// added diff and removed diff for each parent layer
|
|
#[tracing::instrument(skip(self, statediffnew, statediffremoved, diff_to_sibling, parent_states), level = "debug")]
|
|
pub fn save_state_from_diff(
|
|
&self, shortstatehash: u64, statediffnew: Arc<HashSet<CompressedStateEvent>>,
|
|
statediffremoved: Arc<HashSet<CompressedStateEvent>>, diff_to_sibling: usize,
|
|
mut parent_states: ParentStatesVec,
|
|
) -> Result<()> {
|
|
let statediffnew_len = statediffnew.len();
|
|
let statediffremoved_len = statediffremoved.len();
|
|
let diffsum = checked!(statediffnew_len + statediffremoved_len)?;
|
|
|
|
if parent_states.len() > 3 {
|
|
// Number of layers
|
|
// To many layers, we have to go deeper
|
|
let parent = parent_states.pop().expect("parent must have a state");
|
|
|
|
let mut parent_new = (*parent.2).clone();
|
|
let mut parent_removed = (*parent.3).clone();
|
|
|
|
for removed in statediffremoved.iter() {
|
|
if !parent_new.remove(removed) {
|
|
// It was not added in the parent and we removed it
|
|
parent_removed.insert(*removed);
|
|
}
|
|
// Else it was added in the parent and we removed it again. We
|
|
// can forget this change
|
|
}
|
|
|
|
for new in statediffnew.iter() {
|
|
if !parent_removed.remove(new) {
|
|
// It was not touched in the parent and we added it
|
|
parent_new.insert(*new);
|
|
}
|
|
// Else it was removed in the parent and we added it again. We
|
|
// can forget this change
|
|
}
|
|
|
|
self.save_state_from_diff(
|
|
shortstatehash,
|
|
Arc::new(parent_new),
|
|
Arc::new(parent_removed),
|
|
diffsum,
|
|
parent_states,
|
|
)?;
|
|
|
|
return Ok(());
|
|
}
|
|
|
|
if parent_states.is_empty() {
|
|
// There is no parent layer, create a new state
|
|
self.db.save_statediff(
|
|
shortstatehash,
|
|
&StateDiff {
|
|
parent: None,
|
|
added: statediffnew,
|
|
removed: statediffremoved,
|
|
},
|
|
);
|
|
|
|
return Ok(());
|
|
};
|
|
|
|
// Else we have two options.
|
|
// 1. We add the current diff on top of the parent layer.
|
|
// 2. We replace a layer above
|
|
|
|
let parent = parent_states.pop().expect("parent must have a state");
|
|
let parent_2_len = parent.2.len();
|
|
let parent_3_len = parent.3.len();
|
|
let parent_diff = checked!(parent_2_len + parent_3_len)?;
|
|
|
|
if checked!(diffsum * diffsum)? >= checked!(2 * diff_to_sibling * parent_diff)? {
|
|
// Diff too big, we replace above layer(s)
|
|
let mut parent_new = (*parent.2).clone();
|
|
let mut parent_removed = (*parent.3).clone();
|
|
|
|
for removed in statediffremoved.iter() {
|
|
if !parent_new.remove(removed) {
|
|
// It was not added in the parent and we removed it
|
|
parent_removed.insert(*removed);
|
|
}
|
|
// Else it was added in the parent and we removed it again. We
|
|
// can forget this change
|
|
}
|
|
|
|
for new in statediffnew.iter() {
|
|
if !parent_removed.remove(new) {
|
|
// It was not touched in the parent and we added it
|
|
parent_new.insert(*new);
|
|
}
|
|
// Else it was removed in the parent and we added it again. We
|
|
// can forget this change
|
|
}
|
|
|
|
self.save_state_from_diff(
|
|
shortstatehash,
|
|
Arc::new(parent_new),
|
|
Arc::new(parent_removed),
|
|
diffsum,
|
|
parent_states,
|
|
)?;
|
|
} else {
|
|
// Diff small enough, we add diff as layer on top of parent
|
|
self.db.save_statediff(
|
|
shortstatehash,
|
|
&StateDiff {
|
|
parent: Some(parent.0),
|
|
added: statediffnew,
|
|
removed: statediffremoved,
|
|
},
|
|
);
|
|
}
|
|
|
|
Ok(())
|
|
}
|
|
|
|
/// Returns the new shortstatehash, and the state diff from the previous
|
|
/// room state
|
|
pub async fn save_state(
|
|
&self, room_id: &RoomId, new_state_ids_compressed: Arc<HashSet<CompressedStateEvent>>,
|
|
) -> Result<HashSetCompressStateEvent> {
|
|
let previous_shortstatehash = self
|
|
.services
|
|
.state
|
|
.get_room_shortstatehash(room_id)
|
|
.await
|
|
.ok();
|
|
|
|
let state_hash = utils::calculate_hash(
|
|
&new_state_ids_compressed
|
|
.iter()
|
|
.map(|bytes| &bytes[..])
|
|
.collect::<Vec<_>>(),
|
|
);
|
|
|
|
let (new_shortstatehash, already_existed) = self
|
|
.services
|
|
.short
|
|
.get_or_create_shortstatehash(&state_hash)
|
|
.await;
|
|
|
|
if Some(new_shortstatehash) == previous_shortstatehash {
|
|
return Ok((new_shortstatehash, Arc::new(HashSet::new()), Arc::new(HashSet::new())));
|
|
}
|
|
|
|
let states_parents = if let Some(p) = previous_shortstatehash {
|
|
self.load_shortstatehash_info(p).await.unwrap_or_default()
|
|
} else {
|
|
ShortStateInfoResult::new()
|
|
};
|
|
|
|
let (statediffnew, statediffremoved) = if let Some(parent_stateinfo) = states_parents.last() {
|
|
let statediffnew: HashSet<_> = new_state_ids_compressed
|
|
.difference(&parent_stateinfo.1)
|
|
.copied()
|
|
.collect();
|
|
|
|
let statediffremoved: HashSet<_> = parent_stateinfo
|
|
.1
|
|
.difference(&new_state_ids_compressed)
|
|
.copied()
|
|
.collect();
|
|
|
|
(Arc::new(statediffnew), Arc::new(statediffremoved))
|
|
} else {
|
|
(new_state_ids_compressed, Arc::new(HashSet::new()))
|
|
};
|
|
|
|
if !already_existed {
|
|
self.save_state_from_diff(
|
|
new_shortstatehash,
|
|
statediffnew.clone(),
|
|
statediffremoved.clone(),
|
|
2, // every state change is 2 event changes on average
|
|
states_parents,
|
|
)?;
|
|
};
|
|
|
|
Ok((new_shortstatehash, statediffnew, statediffremoved))
|
|
}
|
|
}
|