Use go-re2 (envoyproxy#116)

arminabf · Dec 22, 2022 · 219bb09 · 219bb09
1 parent ed57eba
commit 219bb09
Show file tree

Hide file tree

Showing 11 changed files with 26 additions and 196 deletions.
diff --git a/buildtools/re2/Dockerfile b/buildtools/re2/Dockerfile
diff --git a/buildtools/re2/cre2.patch b/buildtools/re2/cre2.patch
diff --git a/go.mod b/go.mod
@@ -7,6 +7,7 @@ require (
 	github.com/stretchr/testify v1.8.0
 	github.com/tetratelabs/proxy-wasm-go-sdk v0.20.1-0.20221031045735-89d180d022a5
 	github.com/tidwall/gjson v1.14.3
+	github.com/wasilibs/go-re2 v0.0.0-20221219074959-3ec67f9038f0
 )
 
 require (
@@ -16,7 +17,7 @@ require (
 	github.com/magefile/mage v1.14.0 // indirect
 	github.com/petar-dambovaliev/aho-corasick v0.0.0-20211021192214-5ab2d9280aa9 // indirect
 	github.com/pmezard/go-difflib v1.0.0 // indirect
-	github.com/tetratelabs/wazero v1.0.0-pre.3 // indirect
+	github.com/tetratelabs/wazero v1.0.0-pre.4.0.20221213074253-2e13f57f56a1 // indirect
 	github.com/tidwall/match v1.1.1 // indirect
 	github.com/tidwall/pretty v1.2.1 // indirect
 	golang.org/x/net v0.1.0 // indirect

diff --git a/go.sum b/go.sum
@@ -25,15 +25,17 @@ github.com/stretchr/testify v1.8.0 h1:pSgiaMZlXftHpm5L7V1+rVB+AZJydKsMxsQBIJw4PK
 github.com/stretchr/testify v1.8.0/go.mod h1:yNjHg4UonilssWZ8iaSj1OCr/vHnekPRkoO+kdMU+MU=
 github.com/tetratelabs/proxy-wasm-go-sdk v0.20.1-0.20221031045735-89d180d022a5 h1:gbsZkzeu+H7oX9xJA97eIuNHCuXfppuJh32mX2Cpeqc=
 github.com/tetratelabs/proxy-wasm-go-sdk v0.20.1-0.20221031045735-89d180d022a5/go.mod h1:A0osZ5uU1yRt5ZOdRRzIHxJZf8xzsxvEkeL8Ae698+s=
-github.com/tetratelabs/wazero v1.0.0-pre.3 h1:Z5fbogMUGcERzaQb9mQU8+yJSy0bVvv2ce3dfR4wcZg=
-github.com/tetratelabs/wazero v1.0.0-pre.3/go.mod h1:M8UDNECGm/HVjOfq0EOe4QfCY9Les1eq54IChMLETbc=
+github.com/tetratelabs/wazero v1.0.0-pre.4.0.20221213074253-2e13f57f56a1 h1:L+/AG1GzZc8u7tIl7ijAl508T/FHu9esMf+E3hZ1JVA=
+github.com/tetratelabs/wazero v1.0.0-pre.4.0.20221213074253-2e13f57f56a1/go.mod h1:u8wrFmpdrykiFK0DFPiFm5a4+0RzsdmXYVtijBKqUVo=
 github.com/tidwall/gjson v1.14.3 h1:9jvXn7olKEHU1S9vwoMGliaT8jq1vJ7IH/n9zD9Dnlw=
 github.com/tidwall/gjson v1.14.3/go.mod h1:/wbyibRr2FHMks5tjHJ5F8dMZh3AcwJEMf5vlfC0lxk=
 github.com/tidwall/match v1.1.1 h1:+Ho715JplO36QYgwN9PGYNhgZvoUSc9X2c80KVTi+GA=
 github.com/tidwall/match v1.1.1/go.mod h1:eRSPERbgtNPcGhD8UCthc6PmLEQXEWd3PRB5JTxsfmM=
 github.com/tidwall/pretty v1.2.0/go.mod h1:ITEVvHYasfjBbM0u2Pg8T2nJnzm8xPwvNhhsoaGGjNU=
 github.com/tidwall/pretty v1.2.1 h1:qjsOFOWWQl+N3RsoF5/ssm1pHmJJwhjlSbZ51I6wMl4=
 github.com/tidwall/pretty v1.2.1/go.mod h1:ITEVvHYasfjBbM0u2Pg8T2nJnzm8xPwvNhhsoaGGjNU=
+github.com/wasilibs/go-re2 v0.0.0-20221219074959-3ec67f9038f0 h1:+dy0jRJ7Y0sMNJPUkTeZ8qC9qc9tNWJ/Noha+L6w2ZE=
+github.com/wasilibs/go-re2 v0.0.0-20221219074959-3ec67f9038f0/go.mod h1:9YbcVrlaRryN9yCvk1fAjJTn5MLKPEd9/LnCJPkGWxY=
 golang.org/x/mod v0.6.0-dev.0.20220419223038-86c51ed26bb4 h1:6zppjxzCulZykYSLyVDYbneBfbaBIQPYMevg0bEwv2s=
 golang.org/x/net v0.1.0 h1:hZ/3BUoy5aId7sCpA/Tc5lt8DkFgdVS2onTpJsZ/fl0=
 golang.org/x/net v0.1.0/go.mod h1:Cx3nUiGt4eDBEyega/BKRp+/AlGL8hYe7U9odMt2Cco=

diff --git a/init_tinygo.go b/init_tinygo.go
@@ -7,5 +7,5 @@ package main
 
 import _ "github.com/corazawaf/coraza-proxy-wasm/internal/gc"
 
-// #cgo LDFLAGS: lib/libinjection.a lib/libre2.a lib/libcre2.a lib/libc++.a lib/libc++abi.a lib/libaho_corasick.a lib/libmimalloc.a lib/libgc.a
+// #cgo LDFLAGS: lib/libinjection.a lib/libaho_corasick.a lib/libmimalloc.a lib/libgc.a
 import "C"
diff --git a/internal/operators/rx.go b/internal/operators/rx.go
@@ -6,44 +6,41 @@
 package operators
 
 import (
-	"fmt"
-
 	"github.com/corazawaf/coraza/v3/rules"
-
-	"github.com/corazawaf/coraza-proxy-wasm/internal/re2"
+	re2 "github.com/wasilibs/go-re2"
 )
 
 type rx struct {
-	re    re2.RegExp
-	debug bool
+	re *re2.Regexp
 }
 
 var _ rules.Operator = (*rx)(nil)
 
 func newRX(options rules.OperatorOptions) (rules.Operator, error) {
-	o := &rx{}
 	data := options.Arguments
 
-	if data == `(?:\$(?:\((?:\(.*\)|.*)\)|\{.*})|\/\w*\[!?.+\]|[<>]\(.*\))` {
-		o.debug = true
-		fmt.Println("enabling rx debug!")
-	}
-
 	re, err := re2.Compile(data)
 	if err != nil {
 		return nil, err
 	}
-
-	o.re = re
-	return o, err
+	return &rx{re: re}, nil
 }
 
 func (o *rx) Evaluate(tx rules.TransactionState, value string) bool {
-	res := o.re.FindStringSubmatch8(value, func(i int, match string) {
-		tx.CaptureField(i, match)
-	})
-	if o.debug {
-		fmt.Println(res)
+
+	if tx.Capturing() {
+		match := o.re.FindStringSubmatch(value)
+		if len(match) == 0 {
+			return false
+		}
+		for i, c := range match {
+			if i == 9 {
+				return true
+			}
+			tx.CaptureField(i, c)
+		}
+		return true
+	} else {
+		return o.re.MatchString(value)
 	}
-	return res
 }
diff --git a/internal/re2/re2.go b/internal/re2/re2.go
diff --git a/lib/libc++.a b/lib/libc++.a
diff --git a/lib/libc++abi.a b/lib/libc++abi.a
diff --git a/lib/libcre2.a b/lib/libcre2.a
diff --git a/magefiles/magefile.go b/magefiles/magefile.go
@@ -203,7 +203,7 @@ tinygo build -gc=custom -opt=2 -o %s -scheduler=none -target=wasi %s`, filepath.
 
 // UpdateLibs updates the C++ filter dependencies.
 func UpdateLibs() error {
-	libs := []string{"aho-corasick", "bdwgc", "libinjection", "mimalloc", "re2"}
+	libs := []string{"aho-corasick", "bdwgc", "libinjection", "mimalloc"}
 	for _, lib := range libs {
 		if err := sh.RunV("docker", "build", "-t", "ghcr.io/corazawaf/coraza-proxy-wasm/buildtools-"+lib, filepath.Join("buildtools", lib)); err != nil {
 			return err