feat(matcher): support google regex in gfwlist

README.md

@@ -41,6 +41,7 @@
 1. 默认配置（`ts-dns.toml`），开箱即用
   ```toml
   listen = ":53"
+  gfwlist = "gfwlist.txt"
   cnip = "cnip.txt"
 
   [groups]
@@ -49,7 +50,6 @@
 
     [groups.dirty]
     dns = [""]  # 省略
-    rules = ["google.com"]
   ```
 
 2. 指定hosts文件和自定义hosts
@@ -94,12 +94,6 @@
 * 默认使用ECS转发DNS请求
 * DNS并发响应
 
-## 依赖
-* [github.com/miekg/dns](https://github.com/miekg/dns)
-* [github.com/coreos/go-semver/semver](https://github.com/coreos/go-semver/semver)
-* [github.com/BurntSushi/toml](https://github.com/BurntSushi/toml)
-
 ## 特别鸣谢
-* [github.com/janeczku/go-ipset](https://github.com/janeczku/go-ipset)
 * [github.com/arloan/prdns](https://github.com/arloan/prdns)
 * [github.com/gfwlist/gfwlist](https://github.com/gfwlist/gfwlist)

go.mod

@@ -10,5 +10,6 @@ require (
 	github.com/coreos/go-semver v0.3.0 // indirect
 	github.com/janeczku/go-ipset v0.0.0-20170206212442-499ed3217c4b
 	github.com/miekg/dns v1.1.28
+	github.com/stretchr/testify v1.2.2
 	golang.org/x/net v0.0.0-20200301022130-244492dfa37a
 )

matcher/adblock.go

@@ -69,8 +69,14 @@ func NewABPByText(text string) (matcher *ABPlus) {
 	}
 	matcher = &ABPlus{isBlocked: map[string]bool{}}
 	for _, line := range strings.Split(text, "\n") {
-		if line == "" || line[0] == '!' || line[0] == '/' || line[0] == '[' {
-			continue // 忽略空行、注释行、path类规则、类型声明
+		if line == "" || line[0] == '!' || line[0] == '[' {
+			continue // 忽略空行、注释行、类型声明
+		} else if line[0] == '/' { // path类规则
+			if line[:13] == "/^https?:\\/\\/" && line[len(line)-5:] == "\\/.*/" { // google正则补丁
+				reg := regexp.MustCompile(line[13 : len(line)-5])
+				matcher.blockedRegs = append(matcher.blockedRegs, reg)
+			}
+			continue
 		}
 		line = strings.Replace(line, "%2F", "/", -1)
 

matcher/adblock_test.go

@@ -12,10 +12,12 @@ var text = `[AutoProxy]
 !comment
 |http://1.1.1.1
 unknown
+
 |https://*.youtube.com/path
-.google.com
+.abc.com
 @@||cip.cc
 @@||*.cn
+/^https?:\/\/([^\/]+\.)*google\.(ac|ad|ae|af|al|am|as|at|az|ba|be|bf|bg|bi|bj|bs|bt|by|ca|cat|cd|cf|cg|ch|ci|cl|cm|co.ao|co.bw|co.ck|co.cr|co.id|co.il|co.in|co.jp|co.ke|co.kr|co.ls|co.ma|com|com.af|com.ag|com.ai|com.ar|com.au|com.bd|com.bh|com.bn|com.bo|com.br|com.bz|com.co|com.cu|com.cy|com.do|com.ec|com.eg|com.et|com.fj|com.gh|com.gi|com.gt|com.hk|com.jm|com.kh|com.kw|com.lb|com.ly|com.mm|com.mt|com.mx|com.my|com.na|com.nf|com.ng|com.ni|com.np|com.om|com.pa|com.pe|com.pg|com.ph|com.pk|com.pr|com.py|com.qa|com.sa|com.sb|com.sg|com.sl|com.sv|com.tj|com.tr|com.tw|com.ua|com.uy|com.vc|com.vn|co.mz|co.nz|co.th|co.tz|co.ug|co.uk|co.uz|co.ve|co.vi|co.za|co.zm|co.zw|cv|cz|de|dj|dk|dm|dz|ee|es|eu|fi|fm|fr|ga|ge|gg|gl|gm|gp|gr|gy|hk|hn|hr|ht|hu|ie|im|iq|is|it|it.ao|je|jo|kg|ki|kz|la|li|lk|lt|lu|lv|md|me|mg|mk|ml|mn|ms|mu|mv|mw|mx|ne|nl|no|nr|nu|org|pl|pn|ps|pt|ro|rs|ru|rw|sc|se|sh|si|sk|sm|sn|so|sr|st|td|tg|tk|tl|tm|tn|to|tt|us|vg|vn|vu|ws)\/.*/
 `
 
 func TestNewChecker(t *testing.T) {
@@ -42,11 +44,11 @@ func TestNewChecker(t *testing.T) {
 	// 判断空串
 	matched, ok := matcher.Match("")
 	assert.Equal(t, ok, false)
-	// 规则.google.com不匹配google.com
-	matched, ok = matcher.Match("google.com")
+	// 规则.abc.com不匹配abc.com
+	matched, ok = matcher.Match("abc.com")
 	assert.Equal(t, ok, false)
-	// 但匹配test.google.com
-	matched, ok = matcher.Match("test.google.com")
+	// 但匹配test.abc.com
+	matched, ok = matcher.Match("test.abc.com")
 	assert.Equal(t, ok, true)
 	// 匹配白名单@@||cip.cc
 	matched, ok = matcher.Match("cip.cc.")
@@ -60,4 +62,8 @@ func TestNewChecker(t *testing.T) {
 	matched, ok = matcher.Match("www.youtube.com")
 	assert.Equal(t, ok, true)
 	assert.Equal(t, matched, true)
+	// 匹配google正则
+	matched, ok = matcher.Match("google.com")
+	assert.Equal(t, matched, true)
+	assert.Equal(t, ok, true)
 }

ts-dns-full.toml

@@ -26,7 +26,6 @@ max_ttl = 86400  # 最大ttl，单位为秒
   dot = ["1.0.0.1:[email protected]"]  # dns over tls服务器
   # 警告：如果本机的dns指向ts-dns自身，且DoH地址中的域名被归类到该组，则会出现递归解析的情况，此时需要在上面的hosts中指定对应IP
   doh = ["https://cloudflare-dns.com/dns-query"]  # dns over https服务器
-  rules = ["google.com"]  # 官方gfwlist里只有".google.com"规则，无法匹配"google.com"，所以手动加上
 
   # 警告：进程启动时会覆盖已有同名IPSet
   ipset = "blocked"  # 目标IPSet名称，该组所有域名的ipv4解析结果将加入到该IPSet中

ts-dns.toml

@@ -11,4 +11,3 @@ cnip = "cnip.txt"
 
   [groups.dirty]
   dns = ["208.67.222.222:5353", "176.103.130.130:5353"]
-  rules = ["google.com"]