Merge sites according to PSL by default (#97)

Author: sheepzh

package.json

@@ -20,6 +20,7 @@
     "@types/generate-json-webpack-plugin": "^0.3.4",
     "@types/jest": "^27.5.0",
     "@types/node": "^17.0.31",
+    "@types/psl": "^1.1.0",
     "@types/webpack": "^5.28.0",
     "@types/webpack-bundle-analyzer": "^4.4.1",
     "babel-loader": "^8.2.5",
@@ -49,7 +50,8 @@
     "clipboardy": "^3.0.0",
     "echarts": "^5.3.2",
     "element-plus": "1.2.0-beta.6",
+    "psl": "^1.8.0",
     "vue": "^3.2.33",
     "vue-router": "^4.0.14"
   }
-}
+}

src/app/components/rule-merge/alert-info.ts

@@ -6,19 +6,38 @@
  */
 
 import { ElAlert } from "element-plus"
-import { t } from "@app/locale"
+import { t, tN } from "@app/locale"
 import { defineComponent, h } from "vue"
 import { MergeRuleMessage } from "@app/locale/components/merge-rule"
+import { PSL_HOMEPAGE } from "@util/constant/url"
 
-const liKeys: (keyof MergeRuleMessage)[] = ['infoAlert0', 'infoAlert1', 'infoAlert2', 'infoAlert3', 'infoAlert4', 'infoAlert5']
+const liKeys: (keyof MergeRuleMessage)[] = ['infoAlert0', 'infoAlert1', 'infoAlert2', 'infoAlert3', 'infoAlert4']
 const title = t(msg => msg.mergeRule.infoAlertTitle)
 
+const pslStyle: Partial<CSSStyleDeclaration> = {
+    fontSize: "var(--el-alert-description-font-size)",
+    color: "var(--el-color-info)",
+    marginLeft: "2px",
+    marginRight: "2px",
+}
+
+function renderPslLink() {
+    return h('a', {
+        href: PSL_HOMEPAGE,
+        style: pslStyle,
+        target: '_blank'
+    }, 'Public Suffix List')
+}
+
 const _default = defineComponent({
     name: "RuleMergeAlertInfo",
     render() {
         return h(ElAlert,
             { type: 'info', title },
-            () => liKeys.map(key => h('li', t(msg => msg.mergeRule[key])))
+            () => [
+                ...liKeys.map(key => h('li', t(msg => msg.mergeRule[key]))),
+                h('li', tN(msg => msg.mergeRule.infoAlert5, { psl: renderPslLink() }))
+            ]
         )
     }
 })

src/app/locale/components/merge-rule.ts

@@ -40,7 +40,7 @@ const _default: Messages<MergeRuleMessage> = {
         infoAlert2: '合并后域名可填具体的域名，或者填数字，或者不填',
         infoAlert3: '如果填数字，则表示合并后域名的级数。比如存在规则【 *.*.edu.cn >>> 3 】，那么 www.hust.edu.cn 将被合并至 hust.edu.cn',
         infoAlert4: '如果不填，则表示原域名不会被合并',
-        infoAlert5: '如果没有匹配任何规则，则默认合并至 2 级域名'
+        infoAlert5: '如果没有匹配任何规则，则默认会合并至 {psl} 的前一级'
     },
     zh_TW: {
         resultOfOrigin: '不合並',
@@ -57,7 +57,7 @@ const _default: Messages<MergeRuleMessage> = {
         infoAlert2: '合並後網域可填具體的網域，或者填數字，或者不填',
         infoAlert3: '如果填數字，則表示合並後網域的級數。比如存在規則【 *.*.edu.cn >>> 3 】，那麼 www.hust.edu.cn 將被合並至 hust.edu.cn',
         infoAlert4: '如果不填，則表示原網域不會被合並',
-        infoAlert5: '如果沒有匹配任何規則，則默認合並至 2 級網域'
+        infoAlert5: '如果沒有匹配任何規則，則默認會合並至 {psl} 的前一級',
     },
     en: {
         resultOfOrigin: 'Not Merge',
@@ -74,7 +74,7 @@ const _default: Messages<MergeRuleMessage> = {
         infoAlert2: 'The merged site can be filled with a specific site, number, or blank',
         infoAlert3: 'Number means the level of merged site. For example, there is a rule "*.*.edu.cn >>> 3", then "www.hust.edu.cn" will be merged to "hust.edu.cn"',
         infoAlert4: 'Blank means the origin site will not be merged',
-        infoAlert5: 'If no rule are matched, sites will be merged into the second-level domain name by default'
+        infoAlert5: 'If no rule is matched, it will default to the level before {psl}'
     },
     ja: {
         resultOfOrigin: '不合并',
@@ -91,7 +91,7 @@ const _default: Messages<MergeRuleMessage> = {
         infoAlert2: '統合されたドメイン名の後、特定のドメイン名を入力するか、番号を入力するか、空白のままにすることができます',
         infoAlert3: '数字を記入する場合は、ドメイン名のレベルが予約されていることを意味します。 たとえば、ルール [*.*.edu.cn >>> 3 ] がある場合、www.hust.edu.cn は hust.edu.cn にマージされます。',
         infoAlert4: '記入しない場合は、元のドメイン名が統合されないことを意味します',
-        infoAlert5: '一致するルールがない場合、デフォルトで第 2 レベル ドメイン名にマージされます'
+        infoAlert5: '一致するルールがない場合、デフォルトで {psl} より前のレベルになります'
     }
 }
 

src/service/components/host-merge-ruler.ts

@@ -7,6 +7,7 @@
 
 import HostMergeRuleItem from "@entity/dto/host-merge-rule-item"
 import { isIpAndPort } from "@util/pattern"
+import psl from "psl"
 
 /**
  * Ruler to merge host
@@ -74,11 +75,16 @@ export default class CustomizedHostMergeRuler implements IHostMergeRuler {
     merge(origin: string): string {
         // First check the static rules
         let merged = this.noRegMergeRules[origin]
-        // The check the regular rules
+        // Then check the regular rules
         let matchResult: undefined | RegRuleItem = undefined
         merged === undefined && (matchResult = this.regulars.find(item => item.reg.test(origin)))
         matchResult && (merged = matchResult.result)
-        return this.merge0(merged === undefined ? 1 : merged, origin)
+        if (merged === undefined) {
+            // No rule matched
+            return isIpAndPort(origin) ? origin : psl.get(origin)
+        } else {
+            return this.merge0(merged, origin)
+        }
     }
 
     private merge0(merged: string | number, origin: string): string {

src/util/constant/url.ts

@@ -88,4 +88,9 @@ export function iconUrlOfBrowser(protocol: string, host: string): string {
 /**
  * @since 0.2.9
  */
-export const MEAT_URL = 'https://dun.mianbaoduo.com/@yangyang'
+export const MEAT_URL = 'https://dun.mianbaoduo.com/@yangyang'
+
+/**
+ * @since 0.9.3
+ */
+export const PSL_HOMEPAGE = 'https://publicsuffix.org/'

test/service/host-merge-ruler.test.ts

@@ -21,9 +21,9 @@ test('specify merged url', () => {
     test1('www.baidu.com', targetUrl, 'www.baidu.com', targetUrl)
     test1('www.*.com', targetUrl, 'www.baidu.com', targetUrl)
     test1('www.*.*', targetUrl, 'www.baidu.123', targetUrl)
-    // Not match
+    // Not match, so merged with the public suffix
     test1('www.*.*', targetUrl, 'www1.baidu.123', 'baidu.123')
-    test1('www.baidu.com', targetUrl, 'www.baidu.com.hk', 'com.hk')
+    test1('www.baidu.com', targetUrl, 'www.baidu.com.hk', 'baidu.com.hk')
 })
 
 test('specify kept dots', () => {
@@ -38,6 +38,6 @@ test('specify kept dots', () => {
 test('specify nothing', () => {
     test1('*.*.edu.cn', '', 'www.hust.edu.cn', 'www.hust.edu.cn')
     test1('*.*.edu.cn', '', '123.hust.edu.cn', '123.hust.edu.cn')
-    // Not match
-    test1('*.*.edu.cn', '', 'hust.edu.cn', 'edu.cn')
+    // Not match, so merged with the public suffix
+    test1('*.*.edu.cn', '', 'xx.xx.hust.edu.cn', 'hust.edu.cn')
 })