From cf08e0d8407b12b05568c8e3378cc832d9976e81 Mon Sep 17 00:00:00 2001
From: iBug <git@ibugone.com>
Date: Mon, 27 Jul 2020 03:00:37 +0800
Subject: [PATCH] Add domain checking and pattern matching

---
 build.py | 17 +++++++++++++++++
 code.js  | 52 +++++++++++++++++++++++++++++++++++++++++++++++++---
 2 files changed, 66 insertions(+), 3 deletions(-)

diff --git a/build.py b/build.py
index 5092b82..82f3fc8 100755
--- a/build.py
+++ b/build.py
@@ -6,6 +6,8 @@ import ipaddress
 import requests
 from requests.exceptions import RequestException, HTTPError
 
+import gfwlist
+
 
 SOURCES = {
     'ipdeny.com': 'http://www.ipdeny.com/ipblocks/data/aggregated/cn-aggregated.zone',
@@ -13,6 +15,9 @@ SOURCES = {
 }
 OUT_DIR = "dist"
 
+# Stub content to disable GFWList check
+GFWLIST_STUB = "var DOMAINS = {};\nvar BLACKPAT = [];\nvar WHITEPAT = [];\n"
+
 
 def fetch_and_convert(src):
     response = requests.get(src)
@@ -36,6 +41,9 @@ def main():
         code = f.read()
     code = code.replace("@@TIME@@", now.isoformat()[:-7])
 
+    gfwlist_part = gfwlist.generate_pac_partial()
+    gfwlist_stub = GFWLIST_STUB
+
     os.makedirs(OUT_DIR, mode=0o755, exist_ok=True)
     for key in SOURCES:
         print(f"Generating PAC script from source {key}")
@@ -45,10 +53,19 @@ def main():
             continue
         except HTTPError:
             continue
+
         filename = f"pac-{key}.txt"
+        filename_gfwlist = f"pac-gfwlist-{key}.txt"
         with open(os.path.join(OUT_DIR, filename), "w") as f:
             f.write(code)
             f.write(data)
+            f.write("\n")
+            f.write(gfwlist_stub)
+        with open(os.path.join(OUT_DIR, filename), "w") as f:
+            f.write(code)
+            f.write(data)
+            f.write("\n")
+            f.write(gfwlist_part)
 
 
 if __name__ == '__main__':
diff --git a/code.js b/code.js
index 00bd6a1..36c4a62 100644
--- a/code.js
+++ b/code.js
@@ -1,6 +1,9 @@
 // Author: iBug <ibugone.com>
 // Time: @@TIME@@
 
+var proxy = __PROXY__;
+var direct = "DIRECT";
+
 function belongsToSubnet(host, list) {
   var ip = host.split(".").map(Number);
   ip = 0x1000000 * ip[0] + 0x10000 * ip[1] + 0x100 * ip[2] + ip[3];
@@ -23,6 +26,38 @@ function belongsToSubnet(host, list) {
   return (masked ^ list[x][0]) == 0;
 }
 
+function hasMatchedPattern(text, patterns) {
+  for (var i = 0; i < patterns.length; i++) {
+    if (shExpMatch(text, patterns[i])
+      return true;
+  }
+  return false;
+}
+
+function checkDomainType(host) {
+  // Check if a domain is blacklisted or whitelisted
+  var segments = host.split(".").reverse();
+  var ptr = DOMAINS;
+  var type = DOMAINS["@"];
+  for (var i = 0; i < segments.length; i++) {
+    var segment = segments[i];
+    ptr = ptr[segment];
+    if (ptr === undefined)
+      break;
+    if (ptr["@"] !== undefined)
+      type = ptr["@"];
+  }
+  return type;
+}
+
+function hasWhitelistedPattern(url) {
+  return hasMatchedPattern(url, WHITEPAT);
+}
+
+function hasBlacklistedPattern(url) {
+  return hasMatchedPattern(url, BLACKPAT);
+}
+
 function isChina(host) {
   return belongsToSubnet(host, CHINA);
 }
@@ -31,10 +66,21 @@ function isLan(host) {
   return belongsToSubnet(host, LAN);
 }
 
-var proxy = "__PROXY__";
-var direct = "DIRECT";
-
 function FindProxyForURL(url, host) {
+  if (hasWhitelistedPattern(url)) {
+    return direct;
+  }
+  if (hasBlacklistedPattern(url)) {
+    return proxy;
+  }
+  var domainType = checkDomainType(host);
+  if (domainType === 0) {
+    return proxy;
+  } else if (domainType === 1) {
+    return direct;
+  }
+
+  // Fallback to IP whitelist
   var remote = dnsResolve(host);
   if (!remote || remote.indexOf(":") !== -1) {
     // resolution failed or is IPv6 addr