feat: add tldts dependency for improved hostname parsing (#5128)

packages/phishing/package.json

@@ -24,6 +24,7 @@
     "@polkadot/util": "^13.5.1",
     "@polkadot/util-crypto": "^13.5.1",
     "@polkadot/x-fetch": "^13.5.1",
+    "tldts": "^7.0.8",
     "tslib": "^2.8.1"
   },
   "devDependencies": {

packages/phishing/src/bundle.ts

@@ -3,6 +3,8 @@
 
 import type { AddressList, HostList } from './types.js';
 
+import { parse } from 'tldts';
+
 import { u8aEq } from '@polkadot/util';
 import { decodeAddress } from '@polkadot/util-crypto';
 
@@ -37,21 +39,26 @@ const cacheAddr: CacheAddrList = {
 const cacheHost: Record<string, CacheHostList> = {};
 
 function splitHostParts (host: string): string[] {
-  return host
-    // split domain
+  const parsed = parse(host, { allowPrivateDomains: true });
+
+  if (!parsed.hostname) {
+    return [];
+  }
+
+  return parsed.hostname
+    .replace(/\.$/, '') // remove trailing dot
     .split('.')
-    // reverse order
     .reverse();
 }
 
 function extractHostParts (host: string): string[] {
-  return splitHostParts(
-    host
-      // remove protocol
-      .replace(/https:\/\/|http:\/\/|wss:\/\/|ws:\/\//, '')
-      // get the domain-only part
-      .split('/')[0]
-  );
+  const parsed = parse(host, { allowPrivateDomains: true });
+
+  if (!parsed.hostname) {
+    return [];
+  }
+
+  return splitHostParts(parsed.hostname);
 }
 
 async function retrieveAddrCache (allowCached = true): Promise<CacheAddrList> {

yarn.lock

@@ -508,6 +508,7 @@ __metadata:
     "@polkadot/x-fetch": "npm:^13.5.1"
     "@types/js-yaml": "npm:^4.0.9"
     js-yaml: "npm:^4.1.0"
+    tldts: "npm:^7.0.8"
     tslib: "npm:^2.8.1"
   languageName: unknown
   linkType: soft
@@ -8985,6 +8986,24 @@ __metadata:
   languageName: node
   linkType: hard
 
+"tldts-core@npm:^7.0.8":
+  version: 7.0.8
+  resolution: "tldts-core@npm:7.0.8"
+  checksum: 10/a43d1a485ef460fd897bcfc618c4bc57e33f59a97526a08cffa8d8e3447abee82c9ec3bc652c2aebd8faf4a4f10cabd63211f9e955af4fca20258b2dba42bae9
+  languageName: node
+  linkType: hard
+
+"tldts@npm:^7.0.8":
+  version: 7.0.8
+  resolution: "tldts@npm:7.0.8"
+  dependencies:
+    tldts-core: "npm:^7.0.8"
+  bin:
+    tldts: bin/cli.js
+  checksum: 10/b3cf1c0d78c982e99618eccacd628e5af1b4ef3700d6fb182eba78eda5dbb944fb4a154be220cc60342ab2c52de52876ef632df00e8cb2d1ceef01c833ef9e7c
+  languageName: node
+  linkType: hard
+
 "tmp@npm:^0.0.33":
   version: 0.0.33
   resolution: "tmp@npm:0.0.33"