Merge pull request #15 from Quad9DNS/hashbrown-integration

Replace std HashMap with hashbrown for performance boost

Author: esensar

Cargo.lock

@@ -23,6 +23,7 @@ confusables = { version = "0.1.0" }
 criterion = { version = "0.5.1" }
 exitcode = { version = "1.1.2" }
 futures = { version = "0.3.31" }
+hashbrown = { version = "0.16.1" }
 lazy_static = { version = "1.5.0" }
 metrics = { version = "0.24.2" }
 metrics-exporter-prometheus = { version = "0.17.0" }

Cargo.toml

@@ -97,9 +97,10 @@ async-stream.workspace = true
 clap = { workspace = true, features = ["derive"] }
 exitcode.workspace = true
 futures.workspace = true
-metrics.workspace = true
+hashbrown.workspace = true
 metrics-exporter-prometheus.workspace = true
 metrics-util.workspace = true
+metrics.workspace = true
 rdkafka = { workspace = true, optional = true }
 serde.workspace = true
 serde_json.workspace = true
@@ -108,8 +109,8 @@ snafu.workspace = true
 sysinfo.workspace = true
 tokio = { workspace = true, features = ["full"] }
 tokio-stream = { workspace = true, features = ["io-util", "sync"] }
-tracing.workspace = true
 tracing-subscriber.workspace = true
+tracing.workspace = true
 walkdir.workspace = true
 
 [dev-dependencies]

bin/stringsimile-service/Cargo.toml

@@ -1,5 +1,5 @@
+use hashbrown::HashMap;
 use std::{
-    collections::HashMap,
     fs::File,
     io::{BufReader, Seek},
     panic,

bin/stringsimile-service/src/processor.rs

@@ -8,6 +8,7 @@ version.workspace = true
 [dependencies]
 stringsimile-matcher.workspace = true
 
+hashbrown.workspace = true
 serde = { workspace = true, features = ["derive"] }
 serde_json.workspace = true
 snafu.workspace = true

crates/stringsimile-config/Cargo.toml

@@ -24,6 +24,7 @@ rules-bitflip = ["dep:lazy_static"]
 confusables = { workspace = true, optional = true }
 lazy_static = { workspace = true, optional = true }
 metrics.workspace = true
+hashbrown.workspace = true
 rphonetic = { workspace = true, optional = true }
 serde = { workspace = true, features = ["derive"] }
 serde_json.workspace = true

crates/stringsimile-matcher/Cargo.toml

@@ -1,10 +1,7 @@
 //! Bitflip rule implementation
 
-use std::{
-    collections::{HashMap, HashSet},
-    fmt::Debug,
-    io::Error,
-};
+use hashbrown::{HashMap, HashSet};
+use std::{fmt::Debug, io::Error};
 
 use serde::{Deserialize, Serialize};
 
@@ -83,7 +80,7 @@ impl BitflipRule {
             .flat_map(move |(i, c)| {
                 let string = target_str.to_owned();
                 bitflips
-                    .get(&(c.try_into().unwrap()))
+                    .get::<u8>(&(c.try_into().unwrap()))
                     .into_iter()
                     .flatten()
                     .map(move |c| {

crates/stringsimile-matcher/src/rules/bitflip.rs

@@ -1,9 +1,7 @@
 //! Group of related rules
 
-use std::{
-    collections::{BTreeMap, HashMap},
-    ops::Deref,
-};
+use hashbrown::HashMap;
+use std::{collections::BTreeMap, ops::Deref};
 
 use metrics::{Counter, counter};
 use serde_json::{Map, Value};