feat!: added regex-lite feature

LICENSE-3rdparty.csv

@@ -335,6 +335,7 @@ ref-cast,https://github.com/dtolnay/ref-cast,MIT OR Apache-2.0,David Tolnay <dto
 ref-cast-impl,https://github.com/dtolnay/ref-cast,MIT OR Apache-2.0,David Tolnay <dtolnay@gmail.com>
 regex,https://github.com/rust-lang/regex,MIT OR Apache-2.0,"The Rust Project Developers, Andrew Gallant <jamslam@gmail.com>"
 regex-automata,https://github.com/rust-lang/regex/tree/master/regex-automata,MIT OR Apache-2.0,"The Rust Project Developers, Andrew Gallant <jamslam@gmail.com>"
+regex-lite,https://github.com/rust-lang/regex,MIT OR Apache-2.0,"The Rust Project Developers, Andrew Gallant <jamslam@gmail.com>"
 regex-syntax,https://github.com/rust-lang/regex/tree/master/regex-syntax,MIT OR Apache-2.0,"The Rust Project Developers, Andrew Gallant <jamslam@gmail.com>"
 reqwest,https://github.com/seanmonstar/reqwest,MIT OR Apache-2.0,Sean McArthur <sean@seanmonstar.com>
 resolv-conf,https://github.com/hickory-dns/resolv-conf,MIT OR Apache-2.0,The resolv-conf Authors

bin_tests/Cargo.toml

@@ -25,11 +25,13 @@ errno = "0.3"
 nix = { version = "0.29", features = ["signal", "socket"] }
 hex = "0.4"
 os_info = "3.14.0"
-regex = "1.0"
 
 [dev-dependencies]
 serial_test = "3.2"
 
+[features]
+regex-lite = ["libdd-common/regex-lite"]
+
 [lib]
 bench = false
 

bin_tests/tests/crashtracker_bin_test.rs

@@ -738,7 +738,7 @@ fn test_panic_hook_mode(mode: &str, expected_category: &str, expected_panic_mess
 
         // Check for location format (file:line:column) - always present in Debug builds
         // Location should end with pattern like " (path/file.rs:123:45)"
-        let location_regex = regex::Regex::new(r" \(.+?:\d+:\d+\)$").unwrap();
+        let location_regex = libdd_common::regex_engine::Regex::new(r" \(.+?:\d+:\d+\)$").unwrap();
         assert!(
             location_regex.is_match(message),
             "Expected panic message to end with location ' (file:line:column)', got: {}",

builder/Cargo.toml

@@ -29,6 +29,7 @@ library-config = []
 log = []
 ddsketch = []
 ffe = []
+regex-lite = ["libdd-common/regex-lite"]
 
 [lib]
 bench = false
@@ -44,7 +45,7 @@ tar = "0.4.45"
 tools = { path = "../tools" }
 toml = "0.8.19"
 serde = "1.0.209"
-regex = "1.10"
+libdd-common = { path = "../libdd-common", default-features = false }
 
 [[bin]]
 name = "release"

builder/src/utils.rs

@@ -2,7 +2,7 @@
 // SPDX-License-Identifier: Apache-2.0
 
 use anyhow::{anyhow, Result};
-use regex::Regex;
+use libdd_common::regex_engine::{Captures, Regex};
 use std::fs::{self, OpenOptions};
 use std::io::Write;
 use std::path::{Path, PathBuf};
@@ -37,7 +37,7 @@ pub(crate) fn adjust_extern_symbols(
     let re = Regex::new(r#"(?m)^(\s*)extern\s+(.+;)$"#).unwrap();
 
     // Replace function using captures
-    let new_content = re.replace_all(&content, |caps: &regex::Captures| {
+    let new_content = re.replace_all(&content, |caps: &Captures| {
         let full_match = caps.get(0).unwrap().as_str();
         let indent = &caps[1];
         let declaration = &caps[2];

datadog-ffe/Cargo.toml

@@ -18,7 +18,7 @@ chrono = { version = "0.4.38", default-features = false, features = ["now", "ser
 derive_more = { version = "2.0.0", default-features = false, features = ["from", "into"] }
 log = { version = "0.4.21", default-features = false, features = ["kv", "kv_serde"] }
 md5 = { version = "0.7.0", default-features = false }
-regex = "1.10.4"
+libdd-common = { version = "4.0.0", path = "../libdd-common", default-features = false, features = ["require-regex-full"] }
 semver = "1.0"
 serde-bool = { version = "0.1.3", default-features = false }
 serde_with = { version = "3.11.0", default-features = false, features = ["base64", "hex", "macros"] }

datadog-ffe/src/rules_based/ufc/models.rs

@@ -3,7 +3,7 @@
 
 use std::{collections::HashMap, sync::Arc};
 
-use regex::Regex;
+use libdd_common::regex_engine::Regex;
 use serde::{Deserialize, Serialize};
 
 use crate::rules_based::{EvaluationError, FlagType, Str, Timestamp};

datadog-live-debugger-ffi/Cargo.toml

@@ -25,6 +25,7 @@ log = "0.4.21"
 [features]
 default = ["cbindgen"]
 cbindgen = ["build_common/cbindgen", "libdd-common-ffi/cbindgen"]
+regex-lite = ["datadog-live-debugger/regex-lite"]
 
 [build-dependencies]
 build_common = { path = "../build-common" }

datadog-live-debugger/Cargo.toml

@@ -13,16 +13,17 @@ http-body-util = "0.1"
 "http" = "1"
 bytes =  "1.11.1"
 
-regex = "1.9.3"
 percent-encoding = "2.1"
 serde = { version = "1.0", features = ["derive"] }
 serde_json = "1.0"
 sys-info = { version = "0.9.0" }
 uuid = { version = "1.0", features = ["v4"] }
-regex-automata = "0.4.5"
 smallvec = "1.13.2"
 constcat = "0.4.1"
 tokio = "1.36.0"
 
+[features]
+regex-lite = ["libdd-common/regex-lite"]
+
 [lib]
 bench = false

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 libdd_common::regex_engine::Regex;
 use std::borrow::Cow;
 use std::cmp::min;
 use std::fmt::{Display, Formatter};

datadog-live-debugger/src/redacted_names.rs

@@ -3,7 +3,7 @@
 
 #![allow(invalid_reference_casting)]
 
-use regex_automata::dfa::regex::Regex;
+use libdd_common::regex_engine::Regex;
 use std::collections::HashSet;
 use std::sync::atomic::{AtomicBool, Ordering};
 use std::sync::LazyLock;
@@ -164,7 +164,7 @@ 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(libdd_common::regex_engine::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());
@@ -211,7 +211,7 @@ pub fn is_redacted_type<I: AsRef<[u8]>>(name: I) -> bool {
     if REDACTED_TYPES.contains(name) {
         true
     } else if !REDACTED_WILDCARD_TYPES_PATTERN.is_empty() {
-        REDACTED_TYPES_REGEX.is_match(name)
+        std::str::from_utf8(name).is_ok_and(|s| REDACTED_TYPES_REGEX.is_match(s))
     } else {
         false
     }

datadog-remote-config/Cargo.toml

@@ -22,6 +22,7 @@ client = [
     "tracing"
 ]
 live-debugger = ["datadog-live-debugger"]
+regex-lite = ["libdd-common/regex-lite"]
 ffe = ["datadog-ffe"]
 test = ["hyper/server", "hyper-util"]
 
@@ -46,7 +47,6 @@ tracing = { version = "0.1", default-features = false, optional = true }
 serde = "1.0"
 serde_json = { version = "1.0", features = ["raw_value"] }
 serde_with = "3"
-regex = "1.0"
 
 # Test feature
 hyper-util = { workspace = true, features = ["service"], optional = true }

datadog-remote-config/src/config/agent_task.rs

@@ -6,7 +6,7 @@ use serde::Deserialize;
 #[cfg(feature = "test")]
 use serde::Serialize;
 
-use regex::Regex;
+use libdd_common::regex_engine::Regex;
 use serde::de::{self, Deserializer};
 
 fn deserialize_case_id<'de, D>(deserializer: D) -> Result<String, D::Error>

libdd-common-ffi/Cargo.toml

@@ -15,6 +15,7 @@ bench =false
 [features]
 default = ["cbindgen"]
 cbindgen = ["build_common/cbindgen"]
+regex-lite = ["libdd-common/regex-lite"]
 
 [build-dependencies]
 build_common = { path = "../build-common" }

libdd-common/Cargo.toml

@@ -29,6 +29,7 @@ bytes = { version = "1.11.1" }
 pin-project = "1"
 rand = { version = "0.8", optional = true }
 regex = "1.5"
+regex-lite = { version = "0.1", optional = true }
 # Use hickory-dns instead of the default system DNS resolver to avoid fork safety issues.
 # The default resolver can hold locks or other global state that can cause deadlocks
 # or corruption when the process forks (e.g., in PHP-FPM or other forking environments).
@@ -98,6 +99,11 @@ use_webpki_roots = ["hyper-rustls/webpki-roots"]
 # Enable this feature to enable stubbing of cgroup
 # php directly import this crate and uses functions gated by this feature for their test
 cgroup_testing = []
+# Use regex-lite instead of regex for smaller binary size
+regex-lite = ["dep:regex-lite"]
+# Consumers that handle user-provided regexes can enable this feature to force
+# the full `regex` crate, overriding `regex-lite` when both are active.
+require-regex-full = []
 # FIPS mode uses the FIPS-compliant cryptographic provider (Unix only)
 fips = ["tls-core", "hyper-rustls/fips"]
 # Enable reqwest client builder support with file dump debugging

libdd-common/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 crate::regex_engine::Regex;
 #[cfg(target_os = "linux")]
 use std::collections::HashMap;
 use std::sync::LazyLock;

libdd-common/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 crate::regex_engine::Regex;
 use std::fs::File;
 use std::io::{BufRead, BufReader};
 use std::path::Path;

libdd-common/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 crate::regex_engine::Regex;
 
     static IN_REGEX: LazyLock<Regex> = LazyLock::new(|| Regex::new(r"in-\d+").unwrap());
     static CI_REGEX: LazyLock<Regex> = LazyLock::new(|| {

libdd-common/src/lib.rs

@@ -21,6 +21,7 @@ pub mod connector;
 #[cfg(feature = "reqwest")]
 pub mod dump_server;
 pub mod entity_id;
+pub mod regex_engine;
 #[macro_use]
 pub mod cstr;
 #[cfg(feature = "bench-utils")]

libdd-common/src/regex_engine.rs

@@ -0,0 +1,19 @@
+// Copyright 2021-Present Datadog, Inc. https://www.datadoghq.com/
+// SPDX-License-Identifier: Apache-2.0
+
+//! Workspace-wide regex engine re-exports.
+//!
+//! By default this module re-exports from the full [`regex`] crate.
+//! Enable the **`regex-lite`** feature to switch to [`regex_lite`] instead,
+//! which trades advanced features (Unicode classes, look-around, etc.) for
+//! smaller binary size and faster compile times.
+//!
+//! The **`require-regex-full`** feature forces the full `regex` crate even
+//! when `regex-lite` is enabled, for consumers that evaluate user-provided
+//! regexes requiring Unicode character class support.
+
+#[cfg(all(feature = "regex-lite", not(feature = "require-regex-full")))]
+pub use regex_lite::{escape, Captures, Regex, RegexBuilder, Replacer};
+
+#[cfg(not(all(feature = "regex-lite", not(feature = "require-regex-full"))))]
+pub use regex::{escape, Captures, Regex, RegexBuilder, Replacer};

libdd-crashtracker-ffi/Cargo.toml

@@ -22,6 +22,7 @@ required-features = ["collector_windows"]
 [features]
 default = ["cbindgen", "collector", "demangler", "receiver"]
 cbindgen = ["build_common/cbindgen"]
+regex-lite = ["libdd-common-ffi/regex-lite"]
 # Enables the in-process collection of crash-info
 collector = []
 collector_windows = []

libdd-data-pipeline-ffi/Cargo.toml

@@ -19,6 +19,7 @@ bench = false
 default = ["cbindgen", "catch_panic"]
 catch_panic = []
 cbindgen = ["build_common/cbindgen", "libdd-common-ffi/cbindgen"]
+regex-lite = ["libdd-data-pipeline/regex-lite"]
 
 [build-dependencies]
 build_common = { path = "../build-common" }

libdd-data-pipeline/Cargo.toml

@@ -64,14 +64,14 @@ path = "benches/trace_buffer.rs"
 libdd-capabilities-impl = { version = "1.0.0", path = "../libdd-capabilities-impl" }
 libdd-log = { path = "../libdd-log" }
 libdd-shared-runtime = { version = "0.1.0", path = "../libdd-shared-runtime" }
+regex = "1.5"
 clap = { version = "4.0", features = ["derive"] }
 criterion = "0.5.1"
 libdd-trace-utils = { path = "../libdd-trace-utils", features = [
     "test-utils",
 ] }
 httpmock = "0.8.0-alpha.1"
 rand = "0.8.5"
-regex = "1.5"
 tempfile = "3.3.0"
 tokio = { version = "1.23", features = [
     "rt",
@@ -90,3 +90,4 @@ https = [
     "libdd-dogstatsd-client/https",
 ]
 test-utils = []
+regex-lite = ["libdd-common/regex-lite"]

libdd-data-pipeline/src/telemetry/mod.rs

@@ -330,6 +330,9 @@ mod tests {
     use libdd_capabilities::HttpError;
     use libdd_shared_runtime::{SharedRuntime, WorkerHandle};
     use libdd_trace_utils::test_utils::poll_for_mock_hits;
+    // Use `regex::Regex` directly here because `httpmock`'s `body_matches`
+    // requires `Into<HttpMockRegex>`, which is only implemented for
+    // `regex::Regex`, not `regex_lite::Regex`.
     use regex::Regex;
     use tokio::time::sleep;
 

libdd-ddsketch-ffi/Cargo.toml

@@ -16,6 +16,7 @@ bench = false
 [features]
 default = ["cbindgen"]
 cbindgen = ["build_common/cbindgen", "libdd-common-ffi/cbindgen"]
+regex-lite = ["libdd-common-ffi/regex-lite"]
 
 [build-dependencies]
 build_common = { path = "../build-common" }