regex: add regex engine factory and config option (#21633)

Add support for injecting regex engine with its options on startup in a bootstrap option. These are some API, runtime and implementation changes.

* google_re2 in safe_regex will be deprecated and no longer be required. Regex::parseRegex will choose the registered regex engine to parse expressions into matchers.
* A new bootstrap option default_regex_engine will be introduced for regex engine selection.
* For compatibility, GoogleRE2 will be chosen as default regex engine if no regex engine is designated.

Signed-off-by: Xie Zhihao <zhihao.xie@intel.com>

Mirrored from https://github.com/envoyproxy/envoy @ 0a92cc86e7f8b88d5af0eb2e27b5c7ef64719e56

BUILD

@@ -243,6 +243,7 @@ proto_library(
         "//envoy/extensions/quic/proof_source/v3:pkg",
         "//envoy/extensions/rate_limit_descriptors/expr/v3:pkg",
         "//envoy/extensions/rbac/matchers/upstream_ip_port/v3:pkg",
+        "//envoy/extensions/regex_engines/v3:pkg",
         "//envoy/extensions/request_id/uuid/v3:pkg",
         "//envoy/extensions/resource_monitors/fixed_heap/v3:pkg",
         "//envoy/extensions/resource_monitors/injected_resource/v3:pkg",

envoy/config/bootstrap/v3/bootstrap.proto

@@ -41,7 +41,7 @@ option (udpa.annotations.file_status).package_version_status = ACTIVE;
 // <config_overview_bootstrap>` for more detail.
 
 // Bootstrap :ref:`configuration overview <config_overview_bootstrap>`.
-// [#next-free-field: 34]
+// [#next-free-field: 35]
 message Bootstrap {
   option (udpa.annotations.versioning).previous_message_type =
       "envoy.config.bootstrap.v2.Bootstrap";
@@ -330,6 +330,10 @@ message Bootstrap {
   // Optional path to a file with performance tracing data created by "Perfetto" SDK in binary
   // ProtoBuf format. The default value is "envoy.pftrace".
   string perf_tracing_file_path = 33;
+
+  // Optional overriding of default regex engine.
+  // If the value is not specified, Google RE2 will be used by default.
+  core.v3.TypedExtensionConfig default_regex_engine = 34;
 }
 
 // Administration interface :ref:`operations documentation

envoy/extensions/regex_engines/v3/BUILD

@@ -0,0 +1,9 @@
+# DO NOT EDIT. This file is generated by tools/proto_format/proto_sync.py.
+
+load("@envoy_api//bazel:api_build_system.bzl", "api_proto_package")
+
+licenses(["notice"])  # Apache 2
+
+api_proto_package(
+    deps = ["@com_github_cncf_udpa//udpa/annotations:pkg"],
+)

envoy/extensions/regex_engines/v3/google_re2.proto

@@ -0,0 +1,23 @@
+syntax = "proto3";
+
+package envoy.extensions.regex_engines.v3;
+
+import "udpa/annotations/status.proto";
+
+option java_package = "io.envoyproxy.envoy.extensions.regex_engines.v3";
+option java_outer_classname = "GoogleRe2Proto";
+option java_multiple_files = true;
+option go_package = "github.com/envoyproxy/go-control-plane/envoy/extensions/regex_engines/v3;regex_enginesv3";
+option (udpa.annotations.file_status).package_version_status = ACTIVE;
+
+// [#protodoc-title: Google RE2]
+
+// Google's `RE2 <https://github.com/google/re2>`_ regex engine. The regex string must adhere to
+// the documented `syntax <https://github.com/google/re2/wiki/Syntax>`_. The engine is designed
+// to complete execution in linear time as well as limit the amount of memory used.
+//
+// Envoy emits two stats for tracking the program size of regexes: the histogram `re2.program_size`,
+// which records the program size, and the counter `re2.exceeded_warn_level`, which is incremented
+// each time the program size exceeds the warn level threshold.
+message GoogleRE2 {
+}

envoy/type/matcher/v3/regex.proto

@@ -56,10 +56,12 @@ message RegexMatcher {
   }
 
   oneof engine_type {
-    option (validate.required) = true;
-
     // Google's RE2 regex engine.
-    GoogleRE2 google_re2 = 1 [(validate.rules).message = {required: true}];
+    GoogleRE2 google_re2 = 1 [
+      deprecated = true,
+      (validate.rules).message = {required: true},
+      (envoy.annotations.deprecated_at_minor_version) = "3.0"
+    ];
   }
 
   // The regex match string. The string must be supported by the configured engine. The regex is matched

versioning/BUILD

@@ -184,6 +184,7 @@ proto_library(
         "//envoy/extensions/quic/proof_source/v3:pkg",
         "//envoy/extensions/rate_limit_descriptors/expr/v3:pkg",
         "//envoy/extensions/rbac/matchers/upstream_ip_port/v3:pkg",
+        "//envoy/extensions/regex_engines/v3:pkg",
         "//envoy/extensions/request_id/uuid/v3:pkg",
         "//envoy/extensions/resource_monitors/fixed_heap/v3:pkg",
         "//envoy/extensions/resource_monitors/injected_resource/v3:pkg",