From 6ccb9ff14547c7c75bd0f49b6be8a733008ec4fa Mon Sep 17 00:00:00 2001
From: BROBIRD <7692707+BROBIRD@users.noreply.github.com>
Date: Tue, 2 Jun 2026 02:36:46 +0800
Subject: [PATCH] fix some bugs

---
 .github/workflows/update.yml |   5 +-
 scripts/gfwlist_parser.py    | 114 ++++++++++++++++++++++++++++++++++-
 2 files changed, 116 insertions(+), 3 deletions(-)

diff --git a/.github/workflows/update.yml b/.github/workflows/update.yml
index a8251cc..f428df7 100644
--- a/.github/workflows/update.yml
+++ b/.github/workflows/update.yml
@@ -82,8 +82,8 @@ jobs:
             return 1
           }
           
-          # 处理我们生成的分离文件
-          for file in Clash/Providers/*_domain.yaml Clash/Providers/*_ip.yaml; do
+          # 处理所有分离文件（包括生成的和已存在的）
+          for file in Clash/Providers/*_domain.yaml Clash/Providers/*_ip.yaml Clash/Providers/Ruleset/*_domain.yaml Clash/Providers/Ruleset/*_ip.yaml; do
             if [ -f "$file" ]; then
               filename=$(basename "$file")
               if [[ "$filename" == *"_domain.yaml" ]]; then
@@ -113,6 +113,7 @@ jobs:
           
           # 清理临时文件
           rm -f Clash/Providers/*_domain.yaml Clash/Providers/*_ip.yaml
+          rm -f Clash/Providers/Ruleset/*_domain.yaml Clash/Providers/Ruleset/*_ip.yaml
           rm -f Clash/Ruleset/*_domain.list Clash/Ruleset/*_ip.list
           # 清理 mihomo 二进制文件
           rm -f mihomo
diff --git a/scripts/gfwlist_parser.py b/scripts/gfwlist_parser.py
index 75407b6..7034abb 100644
--- a/scripts/gfwlist_parser.py
+++ b/scripts/gfwlist_parser.py
@@ -3,6 +3,7 @@ import base64
 import re
 import os
 import sys
+from glob import glob
 
 def is_ip(s):
     if not s:
@@ -187,6 +188,115 @@ def format_ip_cidr_acl_rules(ip_rules):
             rules.append(ip)
     return rules
 
+def parse_yaml_rule_line(line):
+    line = line.strip()
+    if not line:
+        return None, None
+    rule = None
+    if line.startswith('- '):
+        rule = line[2:]
+    elif line.startswith('  - '):
+        rule = line[4:]
+    else:
+        return None, None
+    if ',' in rule:
+        return rule.split(',', 1)
+    return rule, ''
+
+def parse_list_rule_line(line):
+    line = line.strip()
+    if not line or line.startswith('#'):
+        return None, None
+    if ',' in line:
+        return line.split(',', 1)
+    return line, ''
+
+def is_domain_rule_type(rule_type):
+    return rule_type in ('DOMAIN', 'DOMAIN-SUFFIX', 'DOMAIN-KEYWORD', 'DOMAIN-REGEX')
+
+def is_ip_rule_type(rule_type):
+    return rule_type in ('IP-CIDR', 'IP-CIDR6', 'IP-SUFFIX', 'IP-ASN')
+
+def split_existing_rule_files():
+    print("Splitting existing rule files for mrs conversion...")
+
+    for filepath in glob('Clash/Providers/*.yaml'):
+        if '_domain' in filepath or '_ip' in filepath:
+            continue
+        split_yaml_file(filepath)
+
+    for filepath in glob('Clash/Providers/Ruleset/*.yaml'):
+        if '_domain' in filepath or '_ip' in filepath:
+            continue
+        split_yaml_file(filepath)
+
+    for filepath in glob('Clash/Ruleset/*.list'):
+        if '_domain' in filepath or '_ip' in filepath:
+            continue
+        split_list_file(filepath)
+
+def split_yaml_file(filepath):
+    domains = []
+    ips = []
+    try:
+        with open(filepath, 'r', encoding='utf-8') as f:
+            for line in f:
+                rule_type, rule_value = parse_yaml_rule_line(line)
+                if not rule_type or not rule_value:
+                    continue
+                if is_domain_rule_type(rule_type):
+                    domains.append((rule_type, rule_value.strip()))
+                elif is_ip_rule_type(rule_type):
+                    ips.append((rule_type, rule_value.strip()))
+    except Exception as e:
+        print(f"Warning: Failed to read {filepath}: {e}")
+        return
+
+    base_name = os.path.splitext(filepath)[0]
+    if domains:
+        unique_domains = sorted(set(domains))
+        domain_yaml = "payload:\n"
+        for rule_type, d in unique_domains:
+            domain_yaml += f"  - {rule_type},{d}\n"
+        write_file(f"{base_name}_domain.yaml", domain_yaml)
+    if ips:
+        unique_ips = sorted(set(ips))
+        ip_yaml = "payload:\n"
+        for rule_type, ip in unique_ips:
+            ip_yaml += f"  - {rule_type},{ip}\n"
+        write_file(f"{base_name}_ip.yaml", ip_yaml)
+
+def split_list_file(filepath):
+    domains = []
+    ips = []
+    try:
+        with open(filepath, 'r', encoding='utf-8') as f:
+            for line in f:
+                rule_type, rule_value = parse_list_rule_line(line)
+                if not rule_type or not rule_value:
+                    continue
+                if is_domain_rule_type(rule_type):
+                    domains.append((rule_type, rule_value.strip()))
+                elif is_ip_rule_type(rule_type):
+                    ips.append((rule_type, rule_value.strip()))
+    except Exception as e:
+        print(f"Warning: Failed to read {filepath}: {e}")
+        return
+
+    base_name = os.path.splitext(filepath)[0]
+    if domains:
+        unique_domains = sorted(set(domains))
+        domain_text = ""
+        for rule_type, d in unique_domains:
+            domain_text += f"{rule_type},{d}\n"
+        write_file(f"{base_name}_domain.list", domain_text)
+    if ips:
+        unique_ips = sorted(set(ips))
+        ip_text = ""
+        for rule_type, ip in unique_ips:
+            ip_text += f"{rule_type},{ip}\n"
+        write_file(f"{base_name}_ip.list", ip_text)
+
 def generate_acl_file(domain_list, ip_list, filename, title="GFWList Rules"):
     header = f"""#**********************************************************************
 # {title}
@@ -336,5 +446,7 @@ def main():
     generate_clash_ruleset_list(domain_whitelist, ip_whitelist, 'Clash/Ruleset/UnBan.list', 'GFWList 白名单')
     print("Generated: Clash/Ruleset/UnBan.list")
 
+    split_existing_rule_files()
+
 if __name__ == "__main__":
-    main()
\ No newline at end of file
+    main()