Migrate from regex to regex-lite

Cargo.toml

@@ -46,6 +46,10 @@ members = [
   "datadog-log-ffi",
 ]
 
+exclude = [
+  "regex-experiments"
+]
+
 # https://doc.rust-lang.org/cargo/reference/resolver.html
 resolver = "2"
 

datadog-live-debugger/Cargo.toml

@@ -10,7 +10,7 @@ ddcommon = { path = "../ddcommon" }
 hyper = { workspace = true}
 http-body-util = "0.1"
 
-regex = "1.9.3"
+regex-lite = "^0.1"
 percent-encoding = "2.1"
 serde = { version = "1.0", features = ["derive"] }
 serde_json = "1.0"

datadog-live-debugger/src/expr_eval.rs

@@ -6,7 +6,7 @@ use crate::expr_defs::{
     BinaryComparison, CollectionMatch, CollectionSource, Condition, DslPart, NumberSource,
     Reference, StringComparison, StringSource, Value,
 };
-use regex::Regex;
+use regex_lite::Regex;
 use std::borrow::Cow;
 use std::cmp::min;
 use std::fmt::{Display, Formatter};

datadog-live-debugger/src/redacted_names.rs

@@ -151,7 +151,7 @@ pub unsafe fn add_redacted_type<I: AsRef<[u8]>>(name: I) {
             regex_str.push('|')
         }
         let name = String::from_utf8_lossy(name);
-        regex_str.push_str(regex::escape(&name[..name.len() - 1]).as_str());
+        regex_str.push_str(regex_lite::escape(&name[..name.len() - 1]).as_str());
         regex_str.push_str(".*");
     } else {
         let added_types = &mut (*(&*ADDED_REDACTED_TYPES as *const Vec<Vec<u8>>).cast_mut());

datadog-trace-obfuscation/Cargo.toml

@@ -9,7 +9,7 @@ license.workspace = true
 
 [dependencies]
 anyhow = "1.0"
-regex = "1"
+regex-lite = "^0.1"
 serde = { version = "1.0.145", features = ["derive"] }
 serde_json = "1.0"
 url = "2.4.0"

datadog-trace-obfuscation/src/ip_address.rs

@@ -1,7 +1,7 @@
 // Copyright 2024-Present Datadog, Inc. https://www.datadoghq.com/
 // SPDX-License-Identifier: Apache-2.0
 
-use regex::Regex;
+use regex_lite::Regex;
 use std::{borrow::Cow, collections::HashSet, net::Ipv6Addr, sync::LazyLock};
 
 const ALLOWED_IP_ADDRESSES: [&str; 4] = [

datadog-trace-obfuscation/src/replacer.rs

@@ -2,7 +2,7 @@
 // SPDX-License-Identifier: Apache-2.0
 
 use datadog_trace_protobuf::pb;
-use regex::Regex;
+use regex_lite::Regex;
 use serde::Deserialize;
 
 #[derive(Deserialize)]
@@ -27,7 +27,7 @@ pub struct ReplaceRule {
     name: String,
 
     // re holds the regex pattern for matching.
-    re: regex::Regex,
+    re: regex_lite::Regex,
 
     // repl specifies the replacement string to be used when Pattern matches.
     repl: String,
@@ -97,7 +97,7 @@ pub fn parse_rules_from_string(
                 anyhow::bail!("Obfuscator Error: Error while parsing rule: {}", err)
             }
         };
-        let no_expansion = regex::Replacer::no_expansion(&mut &raw_rule.repl).is_some();
+        let no_expansion = regex_lite::Replacer::no_expansion(&mut &raw_rule.repl).is_some();
         vec.push(ReplaceRule {
             name: raw_rule.name,
             re: compiled_regex,
@@ -155,7 +155,7 @@ fn replace_all(
             #[allow(clippy::unwrap_used)]
             let m = cap.get(0).unwrap();
             scratch_space.push_str(&haystack[last_match..m.start()]);
-            regex::Replacer::replace_append(&mut replace, &cap, scratch_space);
+            regex_lite::Replacer::replace_append(&mut replace, &cap, scratch_space);
             last_match = m.end();
         }
         scratch_space.push_str(&haystack[last_match..]);
@@ -286,13 +286,13 @@ mod tests {
     fn test_replace_rule_eq() {
         let rule1 = replacer::ReplaceRule {
             name: "http.url".to_string(),
-            re: regex::Regex::new("(token/)([^/]*)").unwrap(),
+            re: regex_lite::Regex::new("(token/)([^/]*)").unwrap(),
             repl: "${1}?".to_string(),
             no_expansion: false,
         };
         let rule2 = replacer::ReplaceRule {
             name: "http.url".to_string(),
-            re: regex::Regex::new("(token/)([^/]*)").unwrap(),
+            re: regex_lite::Regex::new("(token/)([^/]*)").unwrap(),
             repl: "${1}?".to_string(),
             no_expansion: false,
         };
@@ -304,13 +304,13 @@ mod tests {
     fn test_replace_rule_neq() {
         let rule1 = replacer::ReplaceRule {
             name: "http.url".to_string(),
-            re: regex::Regex::new("(token/)([^/]*)").unwrap(),
+            re: regex_lite::Regex::new("(token/)([^/]*)").unwrap(),
             repl: "${1}?".to_string(),
             no_expansion: false,
         };
         let rule2 = replacer::ReplaceRule {
             name: "http.url".to_string(),
-            re: regex::Regex::new("(broken/)([^/]*)").unwrap(),
+            re: regex_lite::Regex::new("(broken/)([^/]*)").unwrap(),
             repl: "${1}?".to_string(),
             no_expansion: false,
         };

ddcommon/Cargo.toml

@@ -26,7 +26,7 @@ http-body-util = "0.1"
 tower-service = "0.3"
 cc = "1.1.31"
 pin-project = "1"
-regex = "1.5"
+regex-lite = "^0.1"
 rustls = { version = "0.23", default-features = false, optional = true }
 rustls-native-certs = { version = "0.8.1", optional = true }
 thiserror = "1.0"

ddcommon/src/azure_app_services.rs

@@ -1,7 +1,7 @@
 // Copyright 2021-Present Datadog, Inc. https://www.datadoghq.com/
 // SPDX-License-Identifier: Apache-2.0
 
-use regex::Regex;
+use regex_lite::Regex;
 use std::env;
 use std::sync::LazyLock;
 

ddcommon/src/entity_id/unix/container_id.rs

@@ -3,7 +3,7 @@
 
 //! This module provides functions to parse the container id from the cgroup file
 use super::CgroupFileParsingError;
-use regex::Regex;
+use regex_lite::Regex;
 use std::fs::File;
 use std::io::{BufRead, BufReader};
 use std::path::Path;

ddcommon/src/entity_id/unix/mod.rs

@@ -104,7 +104,7 @@ pub static ENTITY_ID: LazyLock<Option<&'static str>> = LazyLock::new(|| {
 #[cfg(test)]
 mod tests {
     use super::*;
-    use regex::Regex;
+    use regex_lite::Regex;
 
     static IN_REGEX: LazyLock<Regex> = LazyLock::new(|| Regex::new(r"in-\d+").unwrap());
     static CI_REGEX: LazyLock<Regex> = LazyLock::new(|| {

regex-experiments/.gitignore

@@ -0,0 +1 @@
+/target

regex-experiments/Cargo.toml

@@ -0,0 +1,8 @@
+[package]
+name = "cgroup-parse-bench"
+version = "0.1.0"
+edition = "2021"
+
+[dependencies]
+regex = "1"
+regex-lite = "0.1"