imc 0.1.0

Trait-based, deduplicating, in-memory caching library for Rust
Documentation 
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

use std::any::TypeId;
use std::sync::mpsc::{self, Receiver, Sender};
use std::sync::Mutex;
#[cfg(not(feature = "tokio"))]
use std::thread::JoinHandle;
use std::time::Duration;
#[cfg(feature = "critical")]
use std::sync::OnceLock;

use crate::cache::global;
use crate::hasher::hash_value;
use crate::traits::ImcCacheable;

/// Global Redis URL used by the critical-key publisher thread.
#[cfg(feature = "critical")]
static REDIS_URL: OnceLock<String> = OnceLock::new();

#[cfg(feature = "critical")]
pub(crate) fn get_redis_url() -> Option<&'static str> {
    REDIS_URL.get().map(|s| s.as_str())
}

// ---------------------------------------------------------------------------
// Background worker command
// ---------------------------------------------------------------------------

pub(crate) enum CacheCmd {
    Remove(TypeId, u64),
    Clear(TypeId),
    ClearAll,
    Shutdown,
}

// ---------------------------------------------------------------------------
// Platform-agnostic thread/task handle
// ---------------------------------------------------------------------------

#[cfg(not(feature = "tokio"))]
type WorkerJoinHandle = JoinHandle<()>;
#[cfg(feature = "tokio")]
type WorkerJoinHandle = tokio::task::JoinHandle<()>;

#[cfg(not(feature = "tokio"))]
fn spawn_worker_thread(name: &'static str, f: impl FnOnce() + Send + 'static) -> WorkerJoinHandle {
    std::thread::Builder::new()
        .name(name.into())
        .spawn(f)
        .expect("failed to spawn imc worker thread")
}
#[cfg(feature = "tokio")]
fn spawn_worker_thread(_name: &'static str, f: impl FnOnce() + Send + 'static) -> WorkerJoinHandle {
    tokio::task::spawn_blocking(f)
}

// ---------------------------------------------------------------------------
// Global worker-active marker
// ---------------------------------------------------------------------------

pub(crate) static WORKER_TX: Mutex<Option<Sender<CacheCmd>>> = Mutex::new(None);

// ---------------------------------------------------------------------------
// Worker configuration
// ---------------------------------------------------------------------------

/// Configuration for the background maintenance worker.
#[derive(Clone, Debug)]
pub struct WorkerConfig {
    /// How often the background thread sweeps for expired & excess entries.
    pub sweep_interval: Duration,
    /// Optional listen address for the Prometheus metrics HTTP endpoint
    /// (e.g. `"127.0.0.1:9090"`). Requires the `metrics-prometheus` feature.
    #[cfg(feature = "metrics-prometheus")]
    pub metrics_listen_addr: Option<String>,
    /// Optional Redis connection string for cross-process cache invalidation
    /// via pub/sub. Requires the `invalidation-redis` feature.
    #[cfg(feature = "invalidation-redis")]
    pub redis_connection_string: Option<String>,
}

impl Default for WorkerConfig {
    fn default() -> Self {
        Self {
            sweep_interval: Duration::from_secs(10),
            #[cfg(feature = "metrics-prometheus")]
            metrics_listen_addr: None,
            #[cfg(feature = "invalidation-redis")]
            redis_connection_string: None,
        }
    }
}

// ---------------------------------------------------------------------------
// Worker handle
// ---------------------------------------------------------------------------

/// A background worker that periodically sweeps the cache for TTL-expired
/// entries and evicts over-capacity namespaces, and processes
/// remove/clear/shutdown commands.
///
/// When a worker is active the hot-path `through_imc` / `through_imc_async`
/// skip inline eviction so that the main thread is never blocked by an O(n)
/// eviction scan.
pub struct CacheWorker {
    tx: Sender<CacheCmd>,
    _handle: WorkerJoinHandle,
    #[cfg(feature = "invalidation-redis")]
    _redis_handle: Option<WorkerJoinHandle>,
    #[cfg(feature = "critical")]
    _critical_handle: Option<WorkerJoinHandle>,
    #[cfg(feature = "metrics-prometheus")]
    _metrics_handle: Option<WorkerJoinHandle>,
}

impl CacheWorker {
    /// Spawn a worker with default [`WorkerConfig`].
    pub fn spawn() -> Self {
        Self::spawn_with_config(WorkerConfig::default())
    }

    /// Spawn a worker with a custom configuration.
    ///
    /// # Panics
    ///
    /// Panics if a worker is already running (only one instance is allowed).
    pub fn spawn_with_config(config: WorkerConfig) -> Self {
        let (tx, rx) = mpsc::channel();

        let mut guard = WORKER_TX.lock().unwrap_or_else(|e| e.into_inner());
        assert!(
            guard.is_none(),
            "an imc background worker is already running"
        );
        *guard = Some(tx.clone());

        let cfg_for_loop = WorkerConfig {
            sweep_interval: config.sweep_interval,
            #[cfg(feature = "metrics-prometheus")]
            metrics_listen_addr: config.metrics_listen_addr.clone(),
            #[cfg(feature = "invalidation-redis")]
            redis_connection_string: config.redis_connection_string.clone(),
        };

        let handle = spawn_worker_thread("imc-worker", move || worker_loop(rx, cfg_for_loop));

        #[cfg(feature = "invalidation-redis")]
        let redis_handle = if let Some(ref redis_url) = config.redis_connection_string {
            let channels = crate::invalidation::snapshot_channels();
            if !channels.is_empty() {
                let url = redis_url.clone();
                Some(spawn_worker_thread("imc-redis", move || {
                    crate::invalidation::redis_subscriber_loop(&url, channels)
                }))
            } else {
                None
            }
        } else {
            None
        };

        #[cfg(feature = "critical")]
        let critical_handle = if let Some(ref redis_url) = config.redis_connection_string {
            let url = redis_url.clone();
            let _ = REDIS_URL.set(redis_url.clone());
            Some(spawn_worker_thread("imc-critical", move || {
                crate::critical::subscriber_loop(&url)
            }))
        } else {
            None
        };

        #[cfg(feature = "metrics-prometheus")]
        let metrics_handle = if let Some(ref addr) = config.metrics_listen_addr {
            let addr = addr.clone();
            Some(spawn_worker_thread("imc-metrics", move || {
                if let Err(e) = crate::metrics::serve(&addr) {
                    crate::log_event!(ERROR, crate::log::METRICS, crate::log::ERROR,
                        "metrics server on {} failed: {}", addr, e);
                }
            }))
        } else {
            None
        };

        crate::log_event!(INFO, crate::log::WORKER, crate::log::START,
            sweep_interval = ?config.sweep_interval);

        Self {
            tx,
            _handle: handle,
            #[cfg(feature = "invalidation-redis")]
            _redis_handle: redis_handle,
            #[cfg(feature = "critical")]
            _critical_handle: critical_handle,
            #[cfg(feature = "metrics-prometheus")]
            _metrics_handle: metrics_handle,
        }
    }

    /// Enqueue a remove-command for the given type and id.
    pub fn remove<T: ImcCacheable>(&self, id: &T::Id) {
        let type_id = TypeId::of::<T>();
        let id_hash = hash_value(id);
        let _ = self.tx.send(CacheCmd::Remove(type_id, id_hash));
    }

    /// Enqueue a clear-command for the given type.
    pub fn clear<T: ImcCacheable>(&self) {
        let type_id = TypeId::of::<T>();
        let _ = self.tx.send(CacheCmd::Clear(type_id));
    }

    /// Enqueue a clear-all command (evicts every namespace).
    pub fn clear_all(&self) {
        let _ = self.tx.send(CacheCmd::ClearAll);
    }
}

impl Drop for CacheWorker {
    fn drop(&mut self) {
        *WORKER_TX.lock().unwrap_or_else(|e| e.into_inner()) = None;
        let _ = self.tx.send(CacheCmd::Shutdown);
        crate::log_event!(INFO, crate::log::WORKER, crate::log::STOP);
    }
}

// ---------------------------------------------------------------------------
// Background loop
// ---------------------------------------------------------------------------

pub(crate) fn worker_loop(rx: Receiver<CacheCmd>, config: WorkerConfig) {
    let sweep_interval = config.sweep_interval;

    loop {
        match rx.recv_timeout(sweep_interval) {
            Ok(cmd) => match cmd {
                CacheCmd::Remove(type_id, id_hash) => {
                    let mut stores = global().stores.write().unwrap();
                    if let Some(cache) = stores.get_mut(&type_id) {
                        cache.remove_data(id_hash);
                        crate::log_event!(DEBUG, crate::log::WORKER, crate::log::REMOVE,
                            type_id = ?type_id, id_hash = id_hash);
                    }
                }
                CacheCmd::Clear(type_id) => {
                    let mut stores = global().stores.write().unwrap();
                    if let Some(cache) = stores.get_mut(&type_id) {
                        cache.clear();
                        crate::log_event!(INFO, crate::log::WORKER, crate::log::CLEAR,
                            type_id = ?type_id);
                    }
                }
                CacheCmd::ClearAll => {
                    let mut stores = global().stores.write().unwrap();
                    stores.clear();
                    crate::log_event!(INFO, crate::log::WORKER, crate::log::CLEAR, all = true);
                }
                CacheCmd::Shutdown => {
                    crate::log_event!(INFO, crate::log::WORKER, crate::log::STOP, reason = "shutdown");
                    return;
                }
            },
            Err(mpsc::RecvTimeoutError::Timeout) => {}
            Err(mpsc::RecvTimeoutError::Disconnected) => return,
        }

        sweep_all();
    }
}

pub(crate) fn sweep_all() {
    let type_ids: Vec<TypeId> = {
        let stores = global().stores.read().unwrap();
        stores.keys().copied().collect()
    };

    for type_id in type_ids {
        let mut stores = global().stores.write().unwrap();
        if let Some(cache) = stores.get_mut(&type_id) {
            cache.remove_expired();
            cache.evict_to_max_size();
            crate::metrics::set_entries(cache.len());
        }
    }
}