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
use crate::{
client::{prepare_command, PreparedCommand},
resp::{
cmd, deserialize_byte_buf, CollectionResponse, CommandArgs, SingleArg, SingleArgCollection,
ToArgs, Value,
},
};
use serde::Deserialize;
use std::collections::HashMap;
/// A group of Redis commands related to [`Cuckoo filters`](https://redis.io/docs/stack/bloom/)
///
/// # See Also
/// [Cuckoo Filter Commands](https://redis.io/commands/?group=cf)
pub trait CuckooCommands<'a> {
/// Adds an item to the cuckoo filter, creating the filter if it does not exist.
///
/// Cuckoo filters can contain the same item multiple times, and consider each insert as separate.
/// You can use [`cf_addnx`](CuckooCommands::cf_addnx) to only add the item if it does not exist yet.
/// Keep in mind that deleting an element inserted using [`cf_addnx`](CuckooCommands::cf_addnx) may cause false-negative errors.
///
/// # Arguments
/// * `key` - The name of the filter
/// * `item` - The item to add
///
/// # See Also
/// * [<https://redis.io/commands/cf.add/>](https://redis.io/commands/cf.add/)
#[must_use]
fn cf_add(self, key: impl SingleArg, item: impl SingleArg) -> PreparedCommand<'a, Self, ()>
where
Self: Sized,
{
prepare_command(self, cmd("CF.ADD").arg(key).arg(item))
}
/// Adds an item to a cuckoo filter if the item did not exist previously.
///
/// See documentation on [`cf_add`](CuckooCommands::cf_add) for more information on this command.
///
/// This command is equivalent to a [`cf_exists`](CuckooCommands::cf_exists) + [`cf_add`](CuckooCommands::cf_add) command.
/// It does not insert an element into the filter if its fingerprint already exists in order to use the available capacity more efficiently.
/// However, deleting elements can introduce `false negative` error rate!
///
/// Note that this command is slower than [`cf_add`](CuckooCommands::cf_add) because it first checks whether the item exists.
///
/// # Arguments
/// * `key` - The name of the filter
/// * `item` - The item to add
///
/// # Return
/// * `true` - if the item did not exist in the filter.
/// * `false` - if the item already existed.
///
/// # See Also
/// * [<https://redis.io/commands/cf.addnx/>](https://redis.io/commands/cf.addnx/)
#[must_use]
fn cf_addnx(self, key: impl SingleArg, item: impl SingleArg) -> PreparedCommand<'a, Self, bool>
where
Self: Sized,
{
prepare_command(self, cmd("CF.ADDNX").arg(key).arg(item))
}
/// Returns the number of times an item may be in the filter.
///
/// Because this is a probabilistic data structure, this may not necessarily be accurate.
///
/// If you just want to know if an item exists in the filter,
/// use [`cf_exists`](CuckooCommands::cf_exists) because it is more efficient for that purpose.
///
/// # Arguments
/// * `key` - The name of the filter
/// * `item` - The item to count
///
/// # Return
/// the count of possible matching copies of the item in the filter.
///
/// # See Also
/// * [<https://redis.io/commands/cf.count/>](https://redis.io/commands/cf.count/)
#[must_use]
fn cf_count(self, key: impl SingleArg, item: impl SingleArg) -> PreparedCommand<'a, Self, usize>
where
Self: Sized,
{
prepare_command(self, cmd("CF.COUNT").arg(key).arg(item))
}
/// Deletes an item once from the filter.
///
/// If the item exists only once, it will be removed from the filter.
/// If the item was added multiple times, it will still be present.
///
/// # Danger !
/// Deleting elements that are not in the filter may delete a different item, resulting in false negatives!
///
/// # Arguments
/// * `key` - The name of the filter
/// * `item` - The item to delete from the filter
///
/// # Complexity
/// O(n), where n is the number of `sub-filters`. Both alternative locations are checked on all `sub-filters`.
///
/// # Return
/// * `true` - the item has been deleted from the filter.
/// * `false` - if the item was not found.
///
/// # See Also
/// * [<https://redis.io/commands/cf.del/>](https://redis.io/commands/cf.del/)
#[must_use]
fn cf_del(self, key: impl SingleArg, item: impl SingleArg) -> PreparedCommand<'a, Self, bool>
where
Self: Sized,
{
prepare_command(self, cmd("CF.DEL").arg(key).arg(item))
}
/// Check if an `item` exists in a Cuckoo Filter `key`
///
/// # Arguments
/// * `key` - The name of the filter
/// * `item` - The item to check for
///
/// # Return
/// * `true` - the item may exist in the filter
/// * `false` - if the item does not exist in the filter.
///
/// # See Also
/// * [<https://redis.io/commands/cf.exists/>](https://redis.io/commands/cf.exists/)
#[must_use]
fn cf_exists(self, key: impl SingleArg, item: impl SingleArg) -> PreparedCommand<'a, Self, bool>
where
Self: Sized,
{
prepare_command(self, cmd("CF.EXISTS").arg(key).arg(item))
}
/// Return information about `key`
///
/// # Arguments
/// * `key` - Name of the key to get info about
///
/// # Return
/// An instance of [`CfInfoResult`](CfInfoResult)
///
/// # See Also
/// * [<https://redis.io/commands/cf.info/>](https://redis.io/commands/cf.info/)
#[must_use]
fn cf_info(self, key: impl SingleArg) -> PreparedCommand<'a, Self, CfInfoResult>
where
Self: Sized,
{
prepare_command(self, cmd("CF.INFO").arg(key))
}
/// Adds one or more items to a cuckoo filter, allowing the filter to be created with a custom capacity if it does not exist yet.
///
/// These commands offers more flexibility over the [`cf_add`](CuckooCommands::cf_add) command, at the cost of more verbosity.
///
/// # Arguments
/// * `key` - The name of the filter
/// * `options` - see [`CfInsertOptions`](CfInsertOptions)
/// * `items` - One or more items to add.
///
/// # See Also
/// * [<https://redis.io/commands/cf.insert/>](https://redis.io/commands/cf.insert/)
#[must_use]
fn cf_insert<I: SingleArg>(
self,
key: impl SingleArg,
options: CfInsertOptions,
item: impl SingleArgCollection<I>,
) -> PreparedCommand<'a, Self, Vec<usize>>
where
Self: Sized,
{
prepare_command(
self,
cmd("CF.INSERT")
.arg(key)
.arg(options)
.arg("ITEMS")
.arg(item),
)
}
/// Adds one or more items to a cuckoo filter, allowing the filter to be created with a custom capacity if it does not exist yet.
///
/// This command is equivalent to a [`cf_exists`](CuckooCommands::cf_exists) + [`cf_add`](CuckooCommands::cf_add) command.
/// It does not insert an element into the filter if its fingerprint already exists and therefore better utilizes the available capacity.
/// However, if you delete elements it might introduce `false negative` error rate!
///
/// These commands offers more flexibility over the [`cf_add`](CuckooCommands::cf_add) and [`cf_addnx`](CuckooCommands::cf_addnx) commands,
/// at the cost of more verbosity.
///
/// # Complexity
/// `O(n + i)`, where n is the number of `sub-filters` and `i` is `maxIterations`.
/// Adding items requires up to 2 memory accesses per `sub-filter`.
/// But as the filter fills up, both locations for an item might be full.
/// The filter attempts to `Cuckoo` swap items up to maxIterations times.
///
/// # Arguments
/// * `key` - The name of the filter
/// * `options` - see [`CfInsertOptions`](CfInsertOptions)
/// * `items` - One or more items to add.
///
/// # Return
/// A collection of integers corresponding to the items specified. Possible values for each element are:
/// * `>0` - if the item was successfully inserted
/// * `0` - if the item already existed and [`cf_insertnx`](CuckooCommands::cf_insertnx) is used.
/// * `<0` - if an error occurred
///
/// # See Also
/// * [<https://redis.io/commands/cf.insert/>](https://redis.io/commands/cf.insert/)
#[must_use]
fn cf_insertnx<I: SingleArg, R: CollectionResponse<i64>>(
self,
key: impl SingleArg,
options: CfInsertOptions,
item: impl SingleArgCollection<I>,
) -> PreparedCommand<'a, Self, R>
where
Self: Sized,
{
prepare_command(
self,
cmd("CF.INSERTNX")
.arg(key)
.arg(options)
.arg("ITEMS")
.arg(item),
)
}
/// Restores a filter previously saved using [`cf_scandump`](CuckooCommands::cf_scandump).
///
/// See the [`cf_scandump`](CuckooCommands::cf_scandump) command for example usage.
///
/// This command overwrites any bloom filter stored under `key`.
/// Make sure that the bloom filter is not be changed between invocations.
///
/// # Arguments
/// * `key` - Name of the key to restore
/// * `iterator` - Iterator value associated with `data` (returned by [`cf_scandump`](CuckooCommands::cf_scandump))
/// * `data` - Current data chunk (returned by [`cf_scandump`](CuckooCommands::cf_scandump))
///
/// # See Also
/// [<https://redis.io/commands/cf.loadchunk/>](https://redis.io/commands/cf.loadchunk/)
#[must_use]
fn cf_loadchunk(
self,
key: impl SingleArg,
iterator: i64,
data: impl SingleArg,
) -> PreparedCommand<'a, Self, ()>
where
Self: Sized,
{
prepare_command(self, cmd("CF.LOADCHUNK").arg(key).arg(iterator).arg(data))
}
/// Check if one or more `items` exists in a Cuckoo Filter `key`
///
/// # Arguments
/// * `key` - The name of the filter
/// * `items` - One or more items to check for
///
/// # Return
/// Collection reply of boolean - for each item where `true` value means the corresponding item
/// may exist in the filter, and a `false` value means it does not exist in the filter.
///
/// # See Also
/// [<https://redis.io/commands/cf.mexists/>](https://redis.io/commands/cf.mexists/)
#[must_use]
fn cf_mexists<I: SingleArg, R: CollectionResponse<bool>>(
self,
key: impl SingleArg,
items: impl SingleArgCollection<I>,
) -> PreparedCommand<'a, Self, R>
where
Self: Sized,
{
prepare_command(self, cmd("CF.MEXISTS").arg(key).arg(items))
}
/// Create a Cuckoo Filter as `key` with a single sub-filter for the initial amount of `capacity` for items.
/// Because of how Cuckoo Filters work, the filter is likely to declare itself full before `capacity` is reached
/// and therefore fill rate will likely never reach 100%.
/// The fill rate can be improved by using a larger `bucketsize` at the cost of a higher error rate.
/// When the filter self-declare itself `full`, it will auto-expand by generating additional sub-filters at the cost of reduced performance and increased error rate.
/// The new sub-filter is created with size of the previous sub-filter multiplied by `expansion`.
/// Like bucket size, additional sub-filters grow the error rate linearly.
/// The size of the new sub-filter is the size of the last sub-filter multiplied by `expansion`.
///
/// The minimal false positive error rate is 2/255 ≈ 0.78% when bucket size of 1 is used.
/// Larger buckets increase the error rate linearly (for example, a bucket size of 3 yields a 2.35% error rate) but improve the fill rate of the filter.
///
/// `maxiterations` dictates the number of attempts to find a slot for the incoming fingerprint.
/// Once the filter gets full, high `maxIterations` value will slow down insertions.
///
/// Unused capacity in prior sub-filters is automatically used when possible. The filter can grow up to 32 times.
///
/// # Arguments
/// * `key` - The key under which the filter is found
/// * `capacity` - Estimated capacity for the filter.
/// Capacity is rounded to the next 2^n number.
/// The filter will likely not fill up to 100% of it's capacity.
/// Make sure to reserve extra capacity if you want to avoid expansions.
/// * `options` - See [`CfReserveOptions`](CfReserveOptions)
///
/// # See Also
/// [<https://redis.io/commands/cf.reserve/>](https://redis.io/commands/cf.reserve/)
#[must_use]
fn cf_reserve(
self,
key: impl SingleArg,
capacity: usize,
options: CfReserveOptions,
) -> PreparedCommand<'a, Self, ()>
where
Self: Sized,
{
prepare_command(self, cmd("CF.RESERVE").arg(key).arg(capacity).arg(options))
}
/// Begins an incremental save of the cuckoo filter.
/// This is useful for large cuckoo filters which cannot fit into the normal [`dump`](crate::commands::GenericCommands::dump)
/// and [`restore`](crate::commands::GenericCommands::restore) model.
///
/// # Arguments
/// * `key` - Name of the filter
/// * `iterator` - Iterator value; either 0 or the iterator from a previous invocation of this command.\
/// The first time this command is called, the value of `iterator` should be 0.
///
/// # Return
/// This command returns successive `(iterator, data)` pairs until `(0, vec![])` to indicate completion.
///
/// # See Also
/// [<https://redis.io/commands/cf.scandump/>](https://redis.io/commands/cf.scandump/)
#[must_use]
fn cf_scandump(
self,
key: impl SingleArg,
iterator: i64,
) -> PreparedCommand<'a, Self, CfScanDumpResult>
where
Self: Sized,
{
prepare_command(self, cmd("CF.SCANDUMP").arg(key).arg(iterator))
}
}
/// Result for the [`cf_info`](CuckooCommands::cf_info) command.
#[derive(Debug, Deserialize)]
pub struct CfInfoResult {
/// Size
#[serde(rename = "Size")]
pub size: usize,
/// Number of buckets
#[serde(rename = "Number of buckets")]
pub num_buckets: usize,
/// Number of filters
#[serde(rename = "Number of filters")]
pub num_filters: usize,
/// Number of items inserted
#[serde(rename = "Number of items inserted")]
pub num_items_inserted: usize,
/// Number of items deleted
#[serde(rename = "Number of items deleted")]
pub num_items_deleted: usize,
/// Bucket size
#[serde(rename = "Bucket size")]
pub bucket_size: usize,
/// Expansion rate
#[serde(rename = "Expansion rate")]
pub expansion_rate: usize,
/// Max iteration
#[serde(rename = "Max iterations")]
pub max_iteration: usize,
/// Additional information
#[serde(flatten)]
pub additional_info: HashMap<String, Value>,
}
/// Options for the [`cf_insert`](CuckooCommands::cf_insert) command.
#[derive(Default)]
pub struct CfInsertOptions {
command_args: CommandArgs,
}
impl CfInsertOptions {
/// Specifies the desired capacity of the new filter, if this filter does not exist yet.
///
/// If the filter already exists, then this parameter is ignored.
/// If the filter does not exist yet and this parameter is not specified,
/// then the filter is created with the module-level default capacity which is `1024`.
/// See [`cf_reserve`](CuckooCommands::cf_reserve) for more information on cuckoo filter capacities.
#[must_use]
pub fn capacity(mut self, capacity: usize) -> Self {
Self {
command_args: self.command_args.arg("CAPACITY").arg(capacity).build(),
}
}
/// If specified, prevents automatic filter creation if the filter does not exist.
///
/// Instead, an error is returned if the filter does not already exist.
/// This option is mutually exclusive with [`capacity`](CfInsertOptions::capacity).
#[must_use]
pub fn nocreate(mut self) -> Self {
Self {
command_args: self.command_args.arg("NOCREATE").build(),
}
}
}
impl ToArgs for CfInsertOptions {
fn write_args(&self, args: &mut CommandArgs) {
args.arg(&self.command_args);
}
}
/// Options for the [`cf_reserve`](CuckooCommands::cf_reserve) command.
#[derive(Default)]
pub struct CfReserveOptions {
command_args: CommandArgs,
}
impl CfReserveOptions {
/// Number of items in each bucket.
///
/// A higher bucket size value improves the fill rate but also causes a higher error rate and slightly slower performance.
/// The default value is 2.
#[must_use]
pub fn bucketsize(mut self, bucketsize: usize) -> Self {
Self {
command_args: self.command_args.arg("BUCKETSIZE").arg(bucketsize).build(),
}
}
/// Number of attempts to swap items between buckets before declaring filter as full and creating an additional filter.
///
/// A low value is better for performance and a higher number is better for filter fill rate.
/// The default value is 20.
pub fn maxiterations(mut self, maxiterations: usize) -> Self {
Self {
command_args: self.command_args.arg("MAXITERATIONS").arg(maxiterations).build(),
}
}
/// When a new filter is created, its size is the size of the current filter multiplied by `expansion`.
/// Expansion is rounded to the next `2^n` number.
/// The default value is 1.
#[must_use]
pub fn expansion(mut self, expansion: usize) -> Self {
Self {
command_args: self.command_args.arg("EXPANSION").arg(expansion).build(),
}
}
}
impl ToArgs for CfReserveOptions {
fn write_args(&self, args: &mut CommandArgs) {
args.arg(&self.command_args);
}
}
/// Result for the [`cf_scandump`](CuckooCommands::cf_scandump) command.
#[derive(Debug, Deserialize)]
pub struct CfScanDumpResult {
pub iterator: i64,
#[serde(deserialize_with = "deserialize_byte_buf")]
pub data: Vec<u8>,
}