Use published rather than vendored version of Vectorscan (#168)

* Delete vendored version of vectorscan, and instead, use the new `vectorscan-rs` and `vectorscan-rs-sys` crates
published on crates.io.
* Update CHANGELOG

CHANGELOG.md

@@ -5,6 +5,7 @@ All notable changes to the project will be documented in this file.
 
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/), and this project aspires to use [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+
 ## Unreleased
 
 ### Additions
@@ -31,6 +32,8 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 
 - The `rules list` command now includes additional fields when using JSON format ([#161](https://github.com/praetorian-inc/noseyparker/pull/161)).
 
+- The `vectorscan` and `vectorscan-sys` crates have been split off into a [separate project](https://github.com/bradlarsen/vectorscan-rs) with crates published on crates.io ([#168](https://github.com/praetorian-inc/noseyparker/pull/168)).
+
 
 ## [v0.17.0](https://github.com/praetorian-inc/noseyparker/releases/v0.17.0) (2024-03-05)
 

Cargo.toml

@@ -35,10 +35,10 @@ opt-level = 3
 [profile.dev.package.similar]
 opt-level = 3
 
-# `vectorscan` and `vectorscan-sys` benefit from being build in release mode,
+# `vectorscan-rs` and `vectorscan-rs-sys` benefit from being build in release mode,
 # even as a dev dependency: doing so saves about 6 seconds on each integration
 # test case!
-[profile.dev.package.vectorscan]
+[profile.dev.package.vectorscan-rs]
 opt-level = 3
-[profile.dev.package.vectorscan-sys]
+[profile.dev.package.vectorscan-rs-sys]
 opt-level = 3

crates/noseyparker-cli/Cargo.toml

@@ -33,11 +33,11 @@ disable_trace = ["log/release_max_level_debug", "tracing/release_max_level_debug
 
 # Specialize the build of Vectorscan for the microarchitecture of the build system's CPU.
 # This will result in binary that is not portable to other CPUs.
-vectorscan_cpu_native = ["vectorscan/cpu_native"]
+vectorscan_cpu_native = ["vectorscan-rs/cpu_native"]
 
 # Specialize the build of Vectorscan to use available SIMD instructions on the build system's CPU.
 # This will result in binary that is not portable to other CPUs.
-vectorscan_simd_specialization = ["vectorscan/simd_specialization"]
+vectorscan_simd_specialization = ["vectorscan-rs/simd_specialization"]
 
 # Enable all Vectorscan features that can improve speed but break binary portability.
 vectorscan_fast_nonportable = ["vectorscan_cpu_native", "vectorscan_simd_specialization"]
@@ -90,7 +90,7 @@ tracing = "0.1"
 tracing-log = "0.2"
 tracing-subscriber = { version = "0.3", features = ["tracing-log", "ansi", "env-filter", "smallvec", "fmt"], default_features = false }
 url = "2.3"
-vectorscan = { path = "../vectorscan" }
+vectorscan-rs = { version = "0.0.1" }
 
 [dev-dependencies]
 assert_cmd = { version = "2.0", features = ["color-auto"] }

crates/noseyparker-cli/src/cmd_rules/cmd_rules_check.rs

@@ -2,7 +2,7 @@ use anyhow::{bail, Context, Result};
 use regex::Regex;
 use std::collections::HashSet;
 use tracing::{debug_span, error, error_span, info, warn};
-use vectorscan::{BlockDatabase, Flag, Pattern, Scan};
+use vectorscan_rs::{BlockDatabase, Flag, Pattern, Scan};
 
 use noseyparker::rules_database::RulesDatabase;
 use noseyparker_rules::{Rule, RulesetSyntax};
@@ -189,7 +189,7 @@ fn hs_compile_pattern(pat: &str) -> Result<BlockDatabase> {
 // fn hs_compile_pattern_streaming(pat: &str) -> Result<StreamingDatabase> {
 //     let pattern = pattern!{pat};
 //     let mut pattern = pattern.left_most();
-//     pattern.som = Some(vectorscan::SomHorizon::Large);
+//     pattern.som = Some(vectorscan_rs::SomHorizon::Large);
 //     let db: StreamingDatabase = pattern.build()?;
 //     Ok(db)
 // }
@@ -270,7 +270,7 @@ fn check_rule(rule_num: usize, rule: &Rule) -> Result<CheckStats> {
             num_errors += 1;
         }
         Ok(db) => {
-            let mut scanner = vectorscan::BlockScanner::new(&db)?;
+            let mut scanner = vectorscan_rs::BlockScanner::new(&db)?;
 
             let mut num_succeeded = 0;
             let mut num_failed = 0;

crates/noseyparker/Cargo.toml

@@ -49,7 +49,7 @@ thiserror = "1"
 tokio = "1.23"
 tracing = "0.1"
 url = "2.3"
-vectorscan = { path = "../vectorscan" }
+vectorscan-rs = { version = "0.0.1" }
 
 [dev-dependencies]
 pretty_assertions = "1.3"

crates/noseyparker/src/matcher.rs

@@ -65,7 +65,7 @@ struct UserData {
 /// If doing multi-threaded scanning, use a separate `Matcher` for each thread.
 pub struct Matcher<'a> {
     /// A scratch buffer for Vectorscan
-    vs_scanner: vectorscan::BlockScanner<'a>,
+    vs_scanner: vectorscan_rs::BlockScanner<'a>,
 
     /// The rules database used for matching
     rules_db: &'a RulesDatabase,
@@ -114,7 +114,7 @@ impl<'a> Matcher<'a> {
             raw_matches_scratch,
             input_len: 0,
         };
-        let vs_scanner = vectorscan::BlockScanner::new(&rules_db.vsdb)?;
+        let vs_scanner = vectorscan_rs::BlockScanner::new(&rules_db.vsdb)?;
         Ok(Matcher {
             vs_scanner,
             rules_db,
@@ -141,7 +141,7 @@ impl<'a> Matcher<'a> {
                     start_idx,
                     end_idx: to,
                 });
-                vectorscan::Scan::Continue
+                vectorscan_rs::Scan::Continue
             })?;
         Ok(())
     }