hyperscan.proto

syntax = "proto3";

package envoy.extensions.matching.input_matchers.hyperscan.v3alpha;

import "udpa/annotations/status.proto";
import "validate/validate.proto";

option java_package = "io.envoyproxy.envoy.extensions.matching.input_matchers.hyperscan.v3alpha";
option java_outer_classname = "HyperscanProto";
option java_multiple_files = true;
option go_package = "github.com/envoyproxy/go-control-plane/contrib/envoy/extensions/matching/input_matchers/hyperscan/v3alpha";
option (udpa.annotations.file_status).package_version_status = ACTIVE;

// [#protodoc-title: Hyperscan matcher]
// Hyperscan :ref:`configuration overview <config_hyperscan>`.
// [#extension: envoy.matching.input_matchers.hyperscan]

// `Hyperscan <https://github.com/intel/hyperscan>`_ regex matcher. The matcher uses the Hyperscan
// engine which exploits x86 SIMD instructions to accelerate matching large numbers of regular
// expressions simultaneously across streams of data.
message Hyperscan {
  // [#next-free-field: 11]
  message Regex {
    // The regex expression.
    //
    // The expression must represent only the pattern to be matched, with no delimiters or flags.
    string regex = 1 [(validate.rules).string = {min_len: 1}];

    // The ID of the regex expression.
    //
    // This option is designed to be used on the sub-expressions in logical combinations.
    uint32 id = 2;

    // Matching will be performed case-insensitively.
    //
    // The expression may still use PCRE tokens (notably ``(?i)`` and ``(?-i)``) to switch
    // case-insensitive matching on and off.
    bool caseless = 3;

    // Matching a ``.`` will not exclude newlines.
    bool dot_all = 4;

    // ``^`` and ``$`` anchors match any newlines in data.
    bool multiline = 5;

    // Allow expressions which can match against an empty string.
    //
    // This option instructs the compiler to allow expressions that can match against empty buffers,
    // such as ``.?``, ``.*``, ``(a|)``. Since Hyperscan can return every possible match for an expression,
    // such expressions generally execute very slowly.
    bool allow_empty = 6;

    // Treat the pattern as a sequence of UTF-8 characters.
    bool utf8 = 7;

    // Use Unicode properties for character classes.
    //
    // This option instructs Hyperscan to use Unicode properties, rather than the default ASCII
    // interpretations, for character mnemonics like ``\w`` and ``\s`` as well as the POSIX character
    // classes. It is only meaningful in conjunction with ``utf8``.
    bool ucp = 8;

    // Logical combination.
    //
    // This option instructs Hyperscan to parse this expression as logical combination syntax.
    // Logical constraints consist of operands, operators and parentheses. The operands are
    // expression indices, and operators can be ``!``, ``&`` or ``|``.
    bool combination = 9;

    // Don’t do any match reporting.
    //
    // This option instructs Hyperscan to ignore match reporting for this expression. It is
    // designed to be used on the sub-expressions in logical combinations.
    bool quiet = 10;
  }

  // Specifies a set of regex expressions that the input should match on.
  repeated Regex regexes = 1 [(validate.rules).repeated = {min_items: 1}];
}