Update category-ads (#3222)

* remove ads attr from openaicom.imgix.net

imgix.net is serving for pictures, not ads/tracking

* category-ads: include more ad domains

.github/workflows/build.yml

@@ -33,15 +33,17 @@ jobs:
           echo "TAG_NAME=$(date +%Y%m%d%H%M%S)" >> $GITHUB_ENV
         shell: bash
 
-      - name: Build dlc.dat file
+      - name: Build dlc.dat and plain lists
         run: |
           cd code || exit 1
           go run ./ --outputdir=../ --exportlists=category-ads-all,tld-cn,cn,tld-\!cn,geolocation-\!cn,apple,icloud
+          go run ./cmd/datdump/main.go --inputdata=../dlc.dat --outputdir=../ --exportlists=_all_
           cd ../ && rm -rf code
 
       - name: Generate dlc.dat sha256 hash
         run: |
           sha256sum dlc.dat > dlc.dat.sha256sum
+          sha256sum dlc.dat_plain.yml > dlc.dat_plain.yml.sha256sum
 
       - name: Generate Zip
         run: |
@@ -66,6 +68,6 @@ jobs:
 
       - name: Release and upload assets
         run: |
-          gh release create ${{ env.TAG_NAME }} --generate-notes --latest --title ${{ env.RELEASE_NAME }} ./dlc.dat ./dlc.dat.*
+          gh release create ${{ env.TAG_NAME }} --generate-notes --latest --title ${{ env.RELEASE_NAME }} ./dlc.dat ./dlc.dat.* ./dlc.dat_plain.yml ./dlc.dat_plain.yml.*
         env:
           GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}

.gitignore

@@ -8,4 +8,5 @@
 dlc.dat
 
 # Exported plaintext lists.
+/*.yml
 /*.txt

README.md

@@ -10,6 +10,14 @@ This project is not opinionated. In other words, it does NOT endorse, claim or i
 
 - **dlc.dat**：[https://github.com/v2fly/domain-list-community/releases/latest/download/dlc.dat](https://github.com/v2fly/domain-list-community/releases/latest/download/dlc.dat)
 - **dlc.dat.sha256sum**：[https://github.com/v2fly/domain-list-community/releases/latest/download/dlc.dat.sha256sum](https://github.com/v2fly/domain-list-community/releases/latest/download/dlc.dat.sha256sum)
+- **dlc.dat_plain.yml**：[https://github.com/v2fly/domain-list-community/releases/latest/download/dlc.dat_plain.yml](https://github.com/v2fly/domain-list-community/releases/latest/download/dlc.dat_plain.yml)
+- **dlc.dat_plain.yml.sha256sum**：[https://github.com/v2fly/domain-list-community/releases/latest/download/dlc.dat_plain.yml.sha256sum](https://github.com/v2fly/domain-list-community/releases/latest/download/dlc.dat_plain.yml.sha256sum)
+
+## Notice
+
+Rules with `@!cn` attribute has been cast out from cn lists. `geosite:geolocation-cn@!cn` is no longer available. Check [#390](https://github.com/v2fly/domain-list-community/issues/390), [#3119](https://github.com/v2fly/domain-list-community/pull/3119) and [#3198](https://github.com/v2fly/domain-list-community/pull/3198) for more information.
+
+Please report if you have any problems or questions.
 
 ## Usage example
 
@@ -85,38 +93,45 @@ All data are under `data` directory. Each file in the directory represents a sub
 # comments
 include:another-file
 domain:google.com @attr1 @attr2
+full:analytics.google.com @ads
 keyword:google
-regexp:www\.google\.com$
-full:www.google.com
+regexp:^odd[1-7]\.example\.org(\.[a-z]{2})?$
 ```
 
 **Syntax:**
 
+> [!NOTE]
+> Adding new `regexp` and `keyword` rules is discouraged because it is easy to use them incorrectly, and proxy software cannot efficiently match these types of rules.
+
+> [!NOTE]
 > The following types of rules are **NOT** fully compatible with the ones that defined by user in V2Ray config file. Do **Not** copy and paste directly.
 
 - Comment begins with `#`. It may begin anywhere in the file. The content in the line after `#` is treated as comment and ignored in production.
-- Inclusion begins with `include:`, followed by the file name of an existing file in the same directory.
 - Subdomain begins with `domain:`, followed by a valid domain name. The prefix `domain:` may be omitted.
-- Keyword begins with `keyword:`, followed by a string.
-- Regular expression begins with `regexp:`, followed by a valid regular expression (per Golang's standard).
 - Full domain begins with `full:`, followed by a complete and valid domain name.
-- Domains (including `domain`, `keyword`, `regexp` and `full`) may have one or more attributes. Each attribute begins with `@` and followed by the name of the attribute.
-
-> **Note:** Adding new `regexp` and `keyword` rules is discouraged because it is easy to use them incorrectly, and proxy software cannot efficiently match these types of rules.
+- Keyword begins with `keyword:`, followed by a substring of a valid domain name.
+- Regular expression begins with `regexp:`, followed by a valid regular expression (per Golang's standard).
+- Domain rules (including `domain`, `full`, `keyword`, and `regexp`) may have none, one or more attributes. Each attribute begins with `@` and followed by the name of the attribute. Attributes will remain available in final lists and `dlc.dat`.
+- Domain rules may have none, one or more affiliations, which additionally adds the domain rule into the affiliated target list. Each affiliation begins with `&` and followed by the name of the target list (nomatter whether the target has a dedicated file in data path). This is a method for data management, and will not remain in the final lists or `dlc.dat`.
+- Inclusion begins with `include:`, followed by the name of another valid domain list. A simple `include:listb` in file `lista` means adding all domain rules of `listb` into `lista`. Inclusions with attributes stands for selective inclusion. `include:listb @attr1 @-attr2` means only adding those domain rules *with* `@attr1` **and** *without* `@attr2`. This is a special type for data management, and will not remain in the final lists or `dlc.dat`.
 
 ## How it works
 
 The entire `data` directory will be built into an external `geosite` file for Project V. Each file in the directory represents a section in the generated file.
 
-To generate a section:
+**General steps:**
 
-1. Remove all the comments in the file.
-2. Replace `include:` lines with the actual content of the file.
-3. Omit all empty lines.
-4. Generate each `domain:` line into a [sub-domain routing rule](https://github.com/v2fly/v2ray-core/blob/master/app/router/routercommon/common.proto#L21).
-5. Generate each `full:` line into a [full domain routing rule](https://github.com/v2fly/v2ray-core/blob/master/app/router/routercommon/common.proto#L23).
-6. Generate each `keyword:` line into a [plain domain routing rule](https://github.com/v2fly/v2ray-core/blob/master/app/router/routercommon/common.proto#L17).
-7. Generate each `regexp:` line into a [regex domain routing rule](https://github.com/v2fly/v2ray-core/blob/master/app/router/routercommon/common.proto#L19).
+1. Read files in the data path (ignore all comments and empty lines).
+2. Parse and resolve source data, turn affiliations and inclusions into actual domain rules in proper lists.
+3. Deduplicate and sort rules in every list.
+4. Export desired plain text lists.
+5. Generate `dlc.dat`:
+   - turn each `domain:` line into a [sub-domain routing rule](https://github.com/v2fly/v2ray-core/blob/master/app/router/routercommon/common.proto#L21).
+   - turn each `full:` line into a [full domain routing rule](https://github.com/v2fly/v2ray-core/blob/master/app/router/routercommon/common.proto#L23).
+   - turn each `keyword:` line into a [plain domain routing rule](https://github.com/v2fly/v2ray-core/blob/master/app/router/routercommon/common.proto#L17).
+   - turn each `regexp:` line into a [regex domain routing rule](https://github.com/v2fly/v2ray-core/blob/master/app/router/routercommon/common.proto#L19).
+
+Read [main.go](./main.go) for details.
 
 ## How to organize domains
 
@@ -126,7 +141,7 @@ Theoretically any string can be used as the name, as long as it is a valid file
 
 ### Attributes
 
-Attribute is useful for sub-group of domains, especially for filtering purpose. For example, the list of `google` domains may contains its main domains, as well as domains that serve ads. The ads domains may be marked by attribute `@ads`, and can be used as `geosite:google@ads` in V2Ray routing.
+Attribute is useful for sub-group of domains, especially for filtering purpose. For example, the list of `google` may contains its main domains, as well as domains that serve ads. The ads domains may be marked by attribute `@ads`, and can be used as `geosite:google@ads` in V2Ray routing. Domains and services that originate from outside China mainland but have access point in China mainland, may be marked by attribute `@cn`.
 
 ## Contribution guideline
 

cmd/datdump/main.go

@@ -0,0 +1,164 @@
+package main
+
+import (
+	"bufio"
+	"flag"
+	"fmt"
+	"os"
+	"path/filepath"
+	"strings"
+
+	"github.com/v2fly/domain-list-community/internal/dlc"
+	router "github.com/v2fly/v2ray-core/v5/app/router/routercommon"
+	"google.golang.org/protobuf/proto"
+)
+
+var (
+	inputData   = flag.String("inputdata", "dlc.dat", "Name of the geosite dat file")
+	outputDir   = flag.String("outputdir", "./", "Directory to place all generated files")
+	exportLists = flag.String("exportlists", "", "Lists to be exported, separated by ',' (empty for _all_)")
+)
+
+type DomainRule struct {
+	Type  string
+	Value string
+	Attrs []string
+}
+
+type DomainList struct {
+	Name  string
+	Rules []DomainRule
+}
+
+func (d *DomainRule) domain2String() string {
+	dstring := d.Type + ":" + d.Value
+	if len(d.Attrs) != 0 {
+		dstring += ":@" + strings.Join(d.Attrs, ",@")
+	}
+	return dstring
+}
+
+func loadGeosite(path string) ([]DomainList, map[string]*DomainList, error) {
+	data, err := os.ReadFile(path)
+	if err != nil {
+		return nil, nil, fmt.Errorf("failed to read geosite file: %w", err)
+	}
+	vgeositeList := new(router.GeoSiteList)
+	if err := proto.Unmarshal(data, vgeositeList); err != nil {
+		return nil, nil, fmt.Errorf("failed to unmarshal: %w", err)
+	}
+	domainLists := make([]DomainList, len(vgeositeList.Entry))
+	domainListByName := make(map[string]*DomainList, len(vgeositeList.Entry))
+	for i, vsite := range vgeositeList.Entry {
+		rules := make([]DomainRule, 0, len(vsite.Domain))
+		for _, vdomain := range vsite.Domain {
+			rule := DomainRule{Value: vdomain.Value}
+			switch vdomain.Type {
+			case router.Domain_RootDomain:
+				rule.Type = dlc.RuleTypeDomain
+			case router.Domain_Regex:
+				rule.Type = dlc.RuleTypeRegexp
+			case router.Domain_Plain:
+				rule.Type = dlc.RuleTypeKeyword
+			case router.Domain_Full:
+				rule.Type = dlc.RuleTypeFullDomain
+			default:
+				return nil, nil, fmt.Errorf("invalid rule type: %+v", vdomain.Type)
+			}
+			for _, vattr := range vdomain.Attribute {
+				rule.Attrs = append(rule.Attrs, vattr.Key)
+			}
+			rules = append(rules, rule)
+		}
+		domainLists[i] = DomainList{
+			Name:  strings.ToUpper(vsite.CountryCode),
+			Rules: rules,
+		}
+		domainListByName[domainLists[i].Name] = &domainLists[i]
+	}
+	return domainLists, domainListByName, nil
+}
+
+func exportSite(name string, domainListByName map[string]*DomainList) error {
+	domainList, ok := domainListByName[strings.ToUpper(name)]
+	if !ok {
+		return fmt.Errorf("list '%s' does not exist", name)
+	}
+	if len(domainList.Rules) == 0 {
+		return fmt.Errorf("list '%s' is empty", name)
+	}
+	file, err := os.Create(filepath.Join(*outputDir, name+".yml"))
+	if err != nil {
+		return err
+	}
+	defer file.Close()
+	w := bufio.NewWriter(file)
+	fmt.Fprintf(w, "%s:\n", name)
+	for _, domain := range domainList.Rules {
+		fmt.Fprintf(w, "  - %q\n", domain.domain2String())
+	}
+	return w.Flush()
+}
+
+func exportAll(filename string, domainLists []DomainList) error {
+	file, err := os.Create(filepath.Join(*outputDir, filename))
+	if err != nil {
+		return err
+	}
+	defer file.Close()
+	w := bufio.NewWriter(file)
+	w.WriteString("lists:\n")
+	for _, domainList := range domainLists {
+		fmt.Fprintf(w, "  - name: %s\n", strings.ToLower(domainList.Name))
+		fmt.Fprintf(w, "    length: %d\n", len(domainList.Rules))
+		w.WriteString("    rules:\n")
+		for _, domain := range domainList.Rules {
+			fmt.Fprintf(w, "      - %q\n", domain.domain2String())
+		}
+	}
+	return w.Flush()
+}
+
+func main() {
+	flag.Parse()
+
+	// Create output directory if not exist
+	if _, err := os.Stat(*outputDir); os.IsNotExist(err) {
+		if mkErr := os.MkdirAll(*outputDir, 0755); mkErr != nil {
+			fmt.Println("Failed to create output directory:", mkErr)
+			os.Exit(1)
+		}
+	}
+
+	fmt.Printf("Loading %s...\n", *inputData)
+	domainLists, domainListByName, err := loadGeosite(*inputData)
+	if err != nil {
+		fmt.Println("Failed to loadGeosite:", err)
+		os.Exit(1)
+	}
+
+	var exportListSlice []string
+	for raw := range strings.SplitSeq(*exportLists, ",") {
+		if trimmed := strings.TrimSpace(raw); trimmed != "" {
+			exportListSlice = append(exportListSlice, trimmed)
+		}
+	}
+	if len(exportListSlice) == 0 {
+		exportListSlice = []string{"_all_"}
+	}
+
+	for _, eplistname := range exportListSlice {
+		if strings.EqualFold(eplistname, "_all_") {
+			if err := exportAll(filepath.Base(*inputData)+"_plain.yml", domainLists); err != nil {
+				fmt.Println("Failed to exportAll:", err)
+				continue
+			}
+		} else {
+			if err := exportSite(eplistname, domainListByName); err != nil {
+				fmt.Println("Failed to exportSite:", err)
+				continue
+			}
+		}
+		fmt.Printf("list: '%s' has been exported successfully.\n", eplistname)
+	}
+}

data/annas-archive

@@ -1,3 +1,5 @@
+annas-archive.in
 annas-archive.li
 annas-archive.org
+annas-archive.pm
 annas-archive.se

data/binance

@@ -28,6 +28,7 @@ binancezh.top
 
 # API
 binanceapi.com
+binanceru.net
 bnbstatic.com
 bntrace.com
 nftstatic.com

data/category-ads

@@ -14,12 +14,15 @@ include:bytedance-ads
 include:category-ads-ir
 include:cctv @ads
 include:clearbit-ads
+include:disney @ads
 include:dmm-ads
 include:duolingo-ads
 include:emogi-ads
 include:flurry-ads
+include:gamersky @ads
 include:google-ads
 include:growingio-ads
+include:hetzner @ads
 include:hiido-ads
 include:hotjar-ads
 include:hunantv-ads
@@ -37,14 +40,18 @@ include:netease-ads
 include:newrelic-ads
 include:ogury-ads
 include:ookla-speedtest-ads
+include:openai @ads
 include:openx-ads
 include:picacg @ads
+include:pikpak @ads
+include:pixiv @ads
 include:pocoiq-ads
 include:pubmatic-ads
 include:qihoo360-ads
+include:samsung @ads
 include:segment-ads
-include:sensorsdata-ads
 include:sina-ads
+include:snap @ads
 include:sohu-ads
 include:spotify-ads
 include:supersonic-ads
@@ -53,6 +60,7 @@ include:tappx-ads
 include:television-ads
 include:tencent-ads
 include:tendcloud @ads
+include:twitter @ads
 include:uberads-ads
 include:umeng-ads
 include:unity-ads
@@ -194,6 +202,9 @@ reachmax.cn
 # 热云数据
 reyun.com
 
+# 神策数据
+static.sensorsdata.cn
+
 # 诸葛io
 zhugeapi.com
 zhugeapi.net

data/category-ads-all

@@ -20,7 +20,6 @@ mgid.com @ads
 ns1p.net @ads
 pubmatic.com @ads
 sigmob.com @ads
-snapads.com @ads
 spotxchange.com @ads
 unimhk.com @ads
 upapi.net @ads

data/category-ai-!cn

@@ -7,10 +7,12 @@ include:elevenlabs
 include:google-deepmind
 include:groq
 include:huggingface
+include:liveperson
 include:openai
 include:perplexity
 include:poe
 include:xai
+include:youmind
 
 # CodeRabbit
 coderabbit.ai

data/category-bank-cn

@@ -1,9 +1,9 @@
-include:boc
-include:ccb
-include:citic
-include:cmb
+include:boc @-!cn
+include:ccb @-!cn
+include:citic @-!cn
+include:cmb @-!cn
 include:hsbc-cn
-include:icbc
+include:icbc @-!cn
 include:unionpay
 
 abchina.com

data/category-dev-cn

@@ -4,7 +4,7 @@ include:apipost
 include:baltamatica
 include:cnblogs
 include:csdn
-include:deepin
+include:deepin @-!cn
 include:gitee
 include:goproxy
 include:huawei-dev

data/category-emby

@@ -188,3 +188,6 @@ emby1.69yun69.com
 
 # 云梯
 yunti.online
+
+# 守候网络
+server2.cn2gias.uk

data/category-entertainment-cn

@@ -6,7 +6,7 @@ include:aamgame
 include:acfun
 include:acplay
 include:bestv
-include:bilibili
+include:bilibili @-!cn
 include:ciweimao
 include:dedao
 include:douyin
@@ -18,7 +18,7 @@ include:gamersky
 include:gitv
 include:hunantv
 include:huya
-include:iqiyi
+include:iqiyi @-!cn
 include:ku6
 include:kuaikan
 include:kuaishou

data/category-finance

@@ -12,6 +12,7 @@ include:itiger
 include:longbridge
 include:n26
 include:schwab
+include:standardchartered
 include:wise
 
 fxcorporate.com

data/category-games-!cn

@@ -5,6 +5,7 @@ include:curseforge
 include:cygames
 include:ea
 include:embark
+include:eneba
 include:epicgames
 include:escapefromtarkov
 include:faceit
@@ -24,6 +25,7 @@ include:pandanet
 include:pinkcore
 include:playstation
 include:projectsekai
+include:pubg
 include:purikonejp
 include:riot
 include:roblox

data/category-games-cn

@@ -9,13 +9,15 @@ include:mihoyo-cn
 include:tencent-games
 include:tiancity
 include:vrzwk
-include:xd
+include:xd @-!cn
 include:yokaverse
 
 # 北京奇客创想科技有限公司
 7k7k.com
 # 刀锋盒子 皖B2-20190103-4
 9xgame.com
+# 韵律谱面研究站 桂ICP备20001846号-3
+arcaea.cn
 # 《异象回声》游戏官网 沪ICP备2023010411号-1
 astral-vector.com
 # 九九互动 粤ICP备19068416号

data/category-hospital-cn

@@ -10,3 +10,6 @@ yctdyy.com
 
 # 南方医科大学深圳医院
 smuszh.com
+
+# 四川大学华西医院
+cd120.com

data/category-scholar-cn

@@ -52,6 +52,7 @@ hwshu.com # 瀚文民国书库
 hytung.cn # 瀚堂典藏古籍
 incopat.com # incoPat 专利数据库
 lawyee.org # 北大法意网 中国法律资料库
+libvideo.com # 知识视界 武汉缘来文化
 neohytung.com # 瀚堂近代报刊
 nmrdata.com # 微谱数据
 nssd.cn # 国家哲学社会科学学术期刊数据库
@@ -67,5 +68,3 @@ unihan.com.cn # 书同文
 wenxin-ge.com # 文心阁古籍全文数据库
 wind.com.cn # Wind 资讯金融
 yiigle.com # 中华医学期刊全文数据库
-
-full:www.libvideo.com # 武汉缘来文化-知识视界

data/category-social-media-cn

@@ -6,7 +6,7 @@ include:gracg
 include:hupu
 include:meipian
 include:okjike
-include:sina
+include:sina @-!cn
 include:xiaohongshu
 include:yy
 include:zhihu

data/eneba

@@ -0,0 +1,2 @@
+eneba.com
+eneba.games

data/epochmediagroup

@@ -46,16 +46,20 @@ ntd.com
 ntd.tv
 ntdca.com
 ntdimg.com
+ntdtv-dc.com
 ntdtv.ca
+ntdtv.co.il
 ntdtv.co.kr
 ntdtv.com
 ntdtv.com.tw
+ntdtv.fr
 ntdtv.jp
 ntdtv.kr
 ntdtv.org
 ntdtv.ru
-ntdtv-dc.com
+ntdtv.se
 ntdtvla.com
+ntdvideo.tw
 ntdvn.com
 persianepochtimes.com
 renminbao.com

data/geolocation-!cn

@@ -102,6 +102,8 @@ include:w3schools
 include:zotero
 
 chemequations.com # 线上化学方程式！
+geogebra.org
+wolframalpha.com
 
 # Entertainment & Games & Music & Podcasts & Videos
 include:category-entertainment
@@ -244,29 +246,31 @@ include:zeplin
 include:zoho
 include:zoom
 
-biliplus.com # BiliPlus
-
-# Graphing for Science and Engineering
-originlab.com
-
-# Online LaTeX Editor
-cloudlatex.io
-overleaf.com
-
-# Translator & Dictionary
-include:linguee
-
-collinsdictionary.com
-ldoceonline.com
-immersivetranslate.com # 沉浸式翻译 (国际版)
-
-# Aurora Open Source Software (https://gitlab.com/AuroraOSS)
+## Aurora Open Source Software (https://gitlab.com/AuroraOSS)
 auroraoss.com
-
-# CookiePro, provides cookies and tracking
+## BiliPlus
+biliplus.com
+## CataBoom
+cataboom.com
+## Consent Management Platforms / Cookie service
+consentpro.com
 cookiepro.com
 cookielaw.org
 onetrust.com
+osano.com
+usercentrics.eu
+## Greasy Fork
+greasyfork.org
+## Online LaTeX Editor
+cloudlatex.io
+overleaf.com
+## Translator & Dictionary
+include:linguee
+collinsdictionary.com
+ldoceonline.com
+immersivetranslate.com # 沉浸式翻译 (国际版)
+## OriginLab (Graphing for Science and Engineering)
+originlab.com
 
 # Software development
 include:category-dev
@@ -284,9 +288,6 @@ include:rarbg
 dmhy.org
 rutor.info
 
-# User scripts
-greasyfork.org
-
 # VPN services
 include:category-vpnservices
 
@@ -301,6 +302,7 @@ include:wikimedia
 
 atwiki.jp
 touhouwiki.net
+wiki.gg
 
 # Others
 include:avaxhome

data/geolocation-cn

@@ -26,6 +26,9 @@ include:getui
 include:jiguang
 include:umeng
 
+# 神策数据
+sensorsdata.cn
+
 # category-httpdns-cn is mainly for advertising purpose
 include:category-httpdns-cn
 
@@ -35,9 +38,9 @@ include:category-httpdns-cn
 # Bank & Finance & Insurance & Securities
 include:category-bank-cn
 include:category-securities-cn
-include:eastmoney
+include:eastmoney @-!cn
 include:everbright
-include:pingan
+include:pingan @-!cn
 include:taikang
 
 ## 航财通·校园付
@@ -89,11 +92,11 @@ pkoplink.com
 
 # E-commerce
 include:58tongcheng
-include:ctrip
+include:ctrip @-!cn
 include:dangdang
-include:dewu
+include:dewu @-!cn
 include:dongjiao
-include:jd
+include:jd @-!cn
 include:lianjia
 include:meituan
 include:miaomiaozhe
@@ -408,19 +411,19 @@ zhaopin.cn
 # Tech companies & Orginations
 include:aisiku # 北京艾斯酷科技有限公司
 include:akiko # 秋子酱科技
-include:alibaba
+include:alibaba @-!cn
 include:baidu
 include:beisen
 include:bluepoch
-include:bytedance
-include:didi
+include:bytedance @-!cn
+include:didi @-!cn
 include:dingdatech # 叮哒出行(杭州金通互联科技有限公司)
 include:dji
 include:gree
 include:haier
 include:hikvision
 include:honor
-include:huawei
+include:huawei @-!cn
 include:hupun # 杭州湖畔网络技术有限公司
 include:iflytek
 include:ishumei # 北京数美时代科技有限公司
@@ -432,14 +435,14 @@ include:meizu
 include:midea
 include:narwal # 云鲸科技
 include:netease
-include:oppo
+include:oppo @-!cn
 include:qihoo360
 include:sumkoo #北京尚古创新科技有限公司
 include:tcl
-include:tencent
+include:tencent @-!cn
 include:tongfang
-include:vivo
-include:xiaomi
+include:vivo @-!cn
+include:xiaomi @-!cn
 include:xunlei
 include:youquan # 祐全科技
 include:yuanbei # 上海圆贝信息科技有限公司
@@ -476,13 +479,20 @@ xsbapp.cn
 
 # Telecommunication
 include:chinabroadnet
-include:chinamobile
-include:chinatelecom
+include:chinamobile @-!cn
+include:chinatelecom @-!cn
 include:chinatower
-include:chinaunicom
+include:chinaunicom @-!cn
 
 # 在线工具
-include:ipip # IPIP ip地理位置数据库
+## IPIP ip地理位置数据库
+include:ipip @-!cn
+
+## 白描
+baimiao.tech
+baimiaoapp.com
+shinescan.tech
+uzero.cn
 
 chaziyu.com # 滇ICP备2024035496号
 fofa.info # Fofa网站测绘（华顺信安）

data/huawei-dev

@@ -1,23 +1,23 @@
-bisheng.cn @cn
-bishengcompiler.cn @cn
-devui.design @cn
-gneec.com @cn
-gneec.com.cn @cn
-gneec3.com @cn
-gneec4.com @cn
-gneec7.com @cn
-harmonyos.com @cn
-hiascend.cn @cn
-hiascend.com @cn
-hiclc.com @cn
-hikunpeng.cn @cn
-hikunpeng.com @cn
-hikunpeng.com.cn @cn
-hikunpeng.net @cn
-hisilicon.com @cn
-hisilicon.com.cn @cn
-huaweiapaas.com @cn
-mindspore.cn @cn
-owsgo.com @cn
-teleows.com @cn
-saasops.tech @cn
+bisheng.cn
+bishengcompiler.cn
+devui.design
+gneec.com
+gneec.com.cn
+gneec3.com
+gneec4.com
+gneec7.com
+harmonyos.com
+hiascend.cn
+hiascend.com
+hiclc.com
+hikunpeng.cn
+hikunpeng.com
+hikunpeng.com.cn
+hikunpeng.net
+hisilicon.com
+hisilicon.com.cn
+huaweiapaas.com
+mindspore.cn
+owsgo.com
+saasops.tech
+teleows.com

data/kechuang

@@ -1,2 +1,2 @@
 kechuang.org
-full:kc.kexinshe.com @cn
+kexinshe.com

data/kurogames

@@ -1,4 +1,4 @@
-kurogames.com @cn
+kurogames.com
 
 # Wuthering Waves
-aki-game.com @cn
+aki-game.com

data/liveperson

@@ -0,0 +1,2 @@
+liveperson.net
+lpsnmedia.net

data/okx

@@ -1,8 +1,9 @@
 okex.com
-okx.com
 okx-dns.com
 okx-dns1.com
 okx-dns2.com
+okx.cab
+okx.com
 
 # OKC Browser
 oklink.com @cn

data/openai

@@ -1,6 +1,7 @@
 # Main domain
-chatgpt.com
 chat.com
+chatgpt.com
+crixet.com
 oaistatic.com
 oaiusercontent.com
 openai.com
@@ -10,13 +11,13 @@ sora.com
 openai.com.cdn.cloudflare.net
 full:openaiapi-site.azureedge.net
 full:openaicom-api-bdcpf8c6d2e9atf6.z01.azurefd.net
+full:openaicom.imgix.net
 full:openaicomproductionae4b.blob.core.windows.net
 full:production-openaicom-storage.azureedge.net
 regexp:^chatgpt-async-webps-prod-\S+-\d+\.webpubsub\.azure\.com$
 
 # tracking
 full:o33249.ingest.sentry.io @ads
-full:openaicom.imgix.net @ads
 full:browser-intake-datadoghq.com @ads
 
 # Advanced Voice

data/pubg

@@ -0,0 +1,5 @@
+kraftonde.com
+playbattlegrounds.com
+pubg.com
+
+full:pubg1.battleye.com

data/samsung

@@ -8,6 +8,7 @@ galaxyappstore.com
 galaxymobile.jp
 game-platform.net
 knoxemm.com
+ospserver.net
 samsung.com
 samsungads.com @ads
 samsungapps.com

data/sensorsdata-ads

@@ -1 +0,0 @@
-static.sensorsdata.cn @ads

data/standardchartered

@@ -0,0 +1,4 @@
+sc.com
+standardchartered.com
+full:standchartbank.sc.omtrdc.net
+full:standchartbank.tt.omtrdc.net

data/tencent-dev

@@ -11,4 +11,4 @@ tencentcloud.com
 tjstats.com
 wegamedeveloper.com
 weixinbridge.com
-weui.io @cn
+weui.io

data/terabox

@@ -1,2 +1,7 @@
+1024terabox.com
+bestclouddrive.com
+freeterabox.com
+nephobox.com
 terabox.com
+terabox1024.com
 teraboxcdn.com

data/xiaohongshu

@@ -1,5 +1,6 @@
 include:askdiandian
 
+rednotecdn.com
 xhscdn.com
 xhscdn.net
 xhslink.com

data/xiaomi

@@ -1,5 +1,6 @@
 include:xiaomi-ads
 include:xiaomi-ai
+include:xiaomi-iot
 
 mgslb.com
 mi-idc.com

data/xiaomi-iot

@@ -0,0 +1,5 @@
+# Xiaomi IoT Services
+account.xiaomi.com
+cn-ha.mqtt.io.mi.com
+ha.api.io.mi.com
+miot-spec.org

data/xvideos

@@ -1,2 +1,5 @@
+xv-ru.com
+xvideos-ar.com
 xvideos-cdn.com
+xvideos-india.com
 xvideos.com

data/youmind

@@ -0,0 +1,4 @@
+# Youmind
+youmind.ai
+youmind.com
+youmind.site

internal/dlc/dlc.go

@@ -0,0 +1,9 @@
+package dlc
+
+const (
+	RuleTypeDomain     string = "domain"
+	RuleTypeFullDomain string = "full"
+	RuleTypeKeyword    string = "keyword"
+	RuleTypeRegexp     string = "regexp"
+	RuleTypeInclude    string = "include"
+)

main.go

@@ -7,9 +7,10 @@ import (
 	"os"
 	"path/filepath"
 	"regexp"
-	"sort"
+	"slices"
 	"strings"
 
+	"github.com/v2fly/domain-list-community/internal/dlc"
 	router "github.com/v2fly/v2ray-core/v5/app/router/routercommon"
 	"google.golang.org/protobuf/proto"
 )
@@ -21,294 +22,361 @@ var (
 	exportLists = flag.String("exportlists", "", "Lists to be flattened and exported in plaintext format, separated by ',' comma")
 )
 
-const (
-	RuleTypeDomain     string = "domain"
-	RuleTypeFullDomain string = "full"
-	RuleTypeKeyword    string = "keyword"
-	RuleTypeRegexp     string = "regexp"
-	RuleTypeInclude    string = "include"
+var (
+	refMap    = make(map[string][]*Entry)
+	plMap     = make(map[string]*ParsedList)
+	finalMap  = make(map[string][]*Entry)
+	cirIncMap = make(map[string]bool) // Used for circular inclusion detection
 )
 
 type Entry struct {
 	Type  string
 	Value string
-	Attrs []*router.Domain_Attribute
+	Attrs []string
+	Plain string
+	Affs  []string
 }
 
-type List struct {
-	Name  string
-	Entry []Entry
+type Inclusion struct {
+	Source    string
+	MustAttrs []string
+	BanAttrs  []string
 }
 
 type ParsedList struct {
-	Name      string
-	Inclusion map[string]bool
-	Entry     []Entry
+	Name       string
+	Inclusions []*Inclusion
+	Entries    []*Entry
 }
 
-func (l *ParsedList) toPlainText(listName string) error {
-	var entryBytes []byte
-	for _, entry := range l.Entry {
-		var attrString string
-		if entry.Attrs != nil {
-			for _, attr := range entry.Attrs {
-				attrString += "@" + attr.GetKey() + ","
-			}
-			attrString = strings.TrimRight(":"+attrString, ",")
-		}
-		// Entry output format is: type:domain.tld:@attr1,@attr2
-		entryBytes = append(entryBytes, []byte(entry.Type+":"+entry.Value+attrString+"\n")...)
-	}
-	if err := os.WriteFile(filepath.Join(*outputDir, listName+".txt"), entryBytes, 0644); err != nil {
-		return err
-	}
-	return nil
-}
-
-func (l *ParsedList) toProto() (*router.GeoSite, error) {
+func makeProtoList(listName string, entries []*Entry) (*router.GeoSite, error) {
 	site := &router.GeoSite{
-		CountryCode: l.Name,
+		CountryCode: listName,
+		Domain:      make([]*router.Domain, 0, len(entries)),
 	}
-	for _, entry := range l.Entry {
-		switch entry.Type {
-		case RuleTypeDomain:
-			site.Domain = append(site.Domain, &router.Domain{
-				Type:      router.Domain_RootDomain,
-				Value:     entry.Value,
-				Attribute: entry.Attrs,
+	for _, entry := range entries {
+		pdomain := &router.Domain{Value: entry.Value}
+		for _, attr := range entry.Attrs {
+			pdomain.Attribute = append(pdomain.Attribute, &router.Domain_Attribute{
+				Key:        attr,
+				TypedValue: &router.Domain_Attribute_BoolValue{BoolValue: true},
 			})
-
-		case RuleTypeRegexp:
-			// check regexp validity to avoid runtime error
-			_, err := regexp.Compile(entry.Value)
-			if err != nil {
-				return nil, fmt.Errorf("invalid regexp in list %s: %s", l.Name, entry.Value)
-			}
-			site.Domain = append(site.Domain, &router.Domain{
-				Type:      router.Domain_Regex,
-				Value:     entry.Value,
-				Attribute: entry.Attrs,
-			})
-
-		case RuleTypeKeyword:
-			site.Domain = append(site.Domain, &router.Domain{
-				Type:      router.Domain_Plain,
-				Value:     entry.Value,
-				Attribute: entry.Attrs,
-			})
-
-		case RuleTypeFullDomain:
-			site.Domain = append(site.Domain, &router.Domain{
-				Type:      router.Domain_Full,
-				Value:     entry.Value,
-				Attribute: entry.Attrs,
-			})
-
-		default:
-			return nil, fmt.Errorf("unknown domain type: %s", entry.Type)
 		}
+
+		switch entry.Type {
+		case dlc.RuleTypeDomain:
+			pdomain.Type = router.Domain_RootDomain
+		case dlc.RuleTypeRegexp:
+			pdomain.Type = router.Domain_Regex
+		case dlc.RuleTypeKeyword:
+			pdomain.Type = router.Domain_Plain
+		case dlc.RuleTypeFullDomain:
+			pdomain.Type = router.Domain_Full
+		}
+		site.Domain = append(site.Domain, pdomain)
 	}
 	return site, nil
 }
 
-func exportPlainTextList(list []string, refName string, pl *ParsedList) {
-	for _, listName := range list {
-		if strings.EqualFold(refName, listName) {
-			if err := pl.toPlainText(strings.ToLower(refName)); err != nil {
-				fmt.Println("Failed:", err)
-				continue
-			}
-			fmt.Printf("'%s' has been generated successfully.\n", listName)
-		}
+func writePlainList(exportedName string) error {
+	targetList, exist := finalMap[strings.ToUpper(exportedName)]
+	if !exist || len(targetList) == 0 {
+		return fmt.Errorf("list %q does not exist or is empty.", exportedName)
 	}
-}
-
-func removeComment(line string) string {
-	idx := strings.Index(line, "#")
-	if idx == -1 {
-		return line
+	file, err := os.Create(filepath.Join(*outputDir, strings.ToLower(exportedName)+".txt"))
+	if err != nil {
+		return err
 	}
-	return strings.TrimSpace(line[:idx])
-}
-
-func parseDomain(domain string, entry *Entry) error {
-	kv := strings.Split(domain, ":")
-	if len(kv) == 1 {
-		entry.Type = RuleTypeDomain
-		entry.Value = strings.ToLower(kv[0])
-		return nil
+	defer file.Close()
+	w := bufio.NewWriter(file)
+	for _, entry := range targetList {
+		fmt.Fprintln(w, entry.Plain)
 	}
-
-	if len(kv) == 2 {
-		entry.Type = strings.ToLower(kv[0])
-
-		if strings.EqualFold(entry.Type, RuleTypeRegexp) {
-			entry.Value = kv[1]
-		} else {
-			entry.Value = strings.ToLower(kv[1])
-		}
-
-		return nil
-	}
-
-	return fmt.Errorf("invalid format: %s", domain)
-}
-
-func parseAttribute(attr string) (*router.Domain_Attribute, error) {
-	var attribute router.Domain_Attribute
-	if len(attr) == 0 || attr[0] != '@' {
-		return &attribute, fmt.Errorf("invalid attribute: %s", attr)
-	}
-
-	attribute.Key = strings.ToLower(attr[1:]) // Trim attribute prefix `@` character
-	attribute.TypedValue = &router.Domain_Attribute_BoolValue{BoolValue: true}
-	return &attribute, nil
+	return w.Flush()
 }
 
 func parseEntry(line string) (Entry, error) {
-	line = strings.TrimSpace(line)
-	parts := strings.Split(line, " ")
-
 	var entry Entry
+	parts := strings.Fields(line)
 	if len(parts) == 0 {
-		return entry, fmt.Errorf("empty entry")
+		return entry, fmt.Errorf("empty line: %q", line)
 	}
 
-	if err := parseDomain(parts[0], &entry); err != nil {
-		return entry, err
-	}
-
-	for i := 1; i < len(parts); i++ {
-		attr, err := parseAttribute(parts[i])
-		if err != nil {
-			return entry, err
+	// Parse type and value
+	v := parts[0]
+	colonIndex := strings.Index(v, ":")
+	if colonIndex == -1 {
+		entry.Type = dlc.RuleTypeDomain // Default type
+		entry.Value = strings.ToLower(v)
+		if !validateDomainChars(entry.Value) {
+			return entry, fmt.Errorf("invalid domain: %q", entry.Value)
 		}
-		entry.Attrs = append(entry.Attrs, attr)
+	} else {
+		typ := strings.ToLower(v[:colonIndex])
+		val := v[colonIndex+1:]
+		switch typ {
+		case dlc.RuleTypeRegexp:
+			if _, err := regexp.Compile(val); err != nil {
+				return entry, fmt.Errorf("invalid regexp %q: %w", val, err)
+			}
+			entry.Type = dlc.RuleTypeRegexp
+			entry.Value = val
+		case dlc.RuleTypeInclude:
+			entry.Type = dlc.RuleTypeInclude
+			entry.Value = strings.ToUpper(val)
+			if !validateSiteName(entry.Value) {
+				return entry, fmt.Errorf("invalid include list name: %q", entry.Value)
+			}
+		case dlc.RuleTypeDomain, dlc.RuleTypeFullDomain, dlc.RuleTypeKeyword:
+			entry.Type = typ
+			entry.Value = strings.ToLower(val)
+			if !validateDomainChars(entry.Value) {
+				return entry, fmt.Errorf("invalid domain: %q", entry.Value)
+			}
+		default:
+			return entry, fmt.Errorf("invalid type: %q", typ)
+		}
+	}
+
+	// Parse/Check attributes and affiliations
+	for _, part := range parts[1:] {
+		if strings.HasPrefix(part, "@") {
+			attr := strings.ToLower(part[1:]) // Trim attribute prefix `@` character
+			if !validateAttrChars(attr) {
+				return entry, fmt.Errorf("invalid attribute: %q", attr)
+			}
+			entry.Attrs = append(entry.Attrs, attr)
+		} else if strings.HasPrefix(part, "&") {
+			aff := strings.ToUpper(part[1:]) // Trim affiliation prefix `&` character
+			if !validateSiteName(aff) {
+				return entry, fmt.Errorf("invalid affiliation: %q", aff)
+			}
+			entry.Affs = append(entry.Affs, aff)
+		} else {
+			return entry, fmt.Errorf("invalid attribute/affiliation: %q", part)
+		}
+	}
+	// Sort attributes
+	slices.Sort(entry.Attrs)
+	// Formated plain entry: type:domain.tld:@attr1,@attr2
+	entry.Plain = entry.Type + ":" + entry.Value
+	if len(entry.Attrs) != 0 {
+		entry.Plain = entry.Plain + ":@" + strings.Join(entry.Attrs, ",@")
 	}
 
 	return entry, nil
 }
 
-func Load(path string) (*List, error) {
+func validateDomainChars(domain string) bool {
+	for i := range domain {
+		c := domain[i]
+		if (c >= 'a' && c <= 'z') || (c >= '0' && c <= '9') || c == '.' || c == '-' {
+			continue
+		}
+		return false
+	}
+	return true
+}
+
+func validateAttrChars(attr string) bool {
+	for i := range attr {
+		c := attr[i]
+		if (c >= 'a' && c <= 'z') || (c >= '0' && c <= '9') || c == '!' || c == '-' {
+			continue
+		}
+		return false
+	}
+	return true
+}
+
+func validateSiteName(name string) bool {
+	for i := range name {
+		c := name[i]
+		if (c >= 'A' && c <= 'Z') || (c >= '0' && c <= '9') || c == '!' || c == '-' {
+			continue
+		}
+		return false
+	}
+	return true
+}
+
+func loadData(path string) error {
 	file, err := os.Open(path)
 	if err != nil {
-		return nil, err
+		return err
 	}
 	defer file.Close()
 
-	list := &List{
-		Name: strings.ToUpper(filepath.Base(path)),
+	listName := strings.ToUpper(filepath.Base(path))
+	if !validateSiteName(listName) {
+		return fmt.Errorf("invalid list name: %s", listName)
 	}
 	scanner := bufio.NewScanner(file)
+	lineIdx := 0
 	for scanner.Scan() {
-		line := strings.TrimSpace(scanner.Text())
-		line = removeComment(line)
-		if len(line) == 0 {
+		line := scanner.Text()
+		lineIdx++
+		// Remove comments
+		if idx := strings.Index(line, "#"); idx != -1 {
+			line = line[:idx]
+		}
+		line = strings.TrimSpace(line)
+		if line == "" {
 			continue
 		}
 		entry, err := parseEntry(line)
 		if err != nil {
-			return nil, err
+			return fmt.Errorf("error in %s at line %d: %v", path, lineIdx, err)
 		}
-		list.Entry = append(list.Entry, entry)
+		refMap[listName] = append(refMap[listName], &entry)
 	}
-
-	return list, nil
+	return nil
 }
 
-func isMatchAttr(Attrs []*router.Domain_Attribute, includeKey string) bool {
-	isMatch := false
-	mustMatch := true
-	matchName := includeKey
-	if strings.HasPrefix(includeKey, "!") {
-		isMatch = true
-		mustMatch = false
-		matchName = strings.TrimLeft(includeKey, "!")
+func parseList(refName string, refList []*Entry) error {
+	pl, _ := plMap[refName]
+	if pl == nil {
+		pl = &ParsedList{Name: refName}
+		plMap[refName] = pl
 	}
-
-	for _, Attr := range Attrs {
-		attrName := Attr.Key
-		if mustMatch {
-			if matchName == attrName {
-				isMatch = true
-				break
+	for _, entry := range refList {
+		if entry.Type == dlc.RuleTypeInclude {
+			if len(entry.Affs) != 0 {
+				return fmt.Errorf("affiliation is not allowed for include:%s", entry.Value)
 			}
-		} else {
-			if matchName == attrName {
-				isMatch = false
-				break
-			}
-		}
-	}
-	return isMatch
-}
-
-func createIncludeAttrEntrys(list *List, matchAttr *router.Domain_Attribute) []Entry {
-	newEntryList := make([]Entry, 0, len(list.Entry))
-	matchName := matchAttr.Key
-	for _, entry := range list.Entry {
-		matched := isMatchAttr(entry.Attrs, matchName)
-		if matched {
-			newEntryList = append(newEntryList, entry)
-		}
-	}
-	return newEntryList
-}
-
-func ParseList(list *List, ref map[string]*List) (*ParsedList, error) {
-	pl := &ParsedList{
-		Name:      list.Name,
-		Inclusion: make(map[string]bool),
-	}
-	entryList := list.Entry
-	for {
-		newEntryList := make([]Entry, 0, len(entryList))
-		hasInclude := false
-		for _, entry := range entryList {
-			if entry.Type == RuleTypeInclude {
-				refName := strings.ToUpper(entry.Value)
-				if entry.Attrs != nil {
-					for _, attr := range entry.Attrs {
-						InclusionName := strings.ToUpper(refName + "@" + attr.Key)
-						if pl.Inclusion[InclusionName] {
-							continue
-						}
-						pl.Inclusion[InclusionName] = true
-
-						refList := ref[refName]
-						if refList == nil {
-							return nil, fmt.Errorf("list not found: %s", entry.Value)
-						}
-						attrEntrys := createIncludeAttrEntrys(refList, attr)
-						if len(attrEntrys) != 0 {
-							newEntryList = append(newEntryList, attrEntrys...)
-						}
-					}
+			inc := &Inclusion{Source: entry.Value}
+			for _, attr := range entry.Attrs {
+				if strings.HasPrefix(attr, "-") {
+					inc.BanAttrs = append(inc.BanAttrs, attr[1:]) // Trim attribute prefix `-` character
 				} else {
-					InclusionName := refName
-					if pl.Inclusion[InclusionName] {
-						continue
-					}
-					pl.Inclusion[InclusionName] = true
-					refList := ref[refName]
-					if refList == nil {
-						return nil, fmt.Errorf("list not found: %s", entry.Value)
-					}
-					newEntryList = append(newEntryList, refList.Entry...)
+					inc.MustAttrs = append(inc.MustAttrs, attr)
 				}
-				hasInclude = true
-			} else {
-				newEntryList = append(newEntryList, entry)
 			}
-		}
-		entryList = newEntryList
-		if !hasInclude {
-			break
+			pl.Inclusions = append(pl.Inclusions, inc)
+		} else {
+			for _, aff := range entry.Affs {
+				apl, _ := plMap[aff]
+				if apl == nil {
+					apl = &ParsedList{Name: aff}
+					plMap[aff] = apl
+				}
+				apl.Entries = append(apl.Entries, entry)
+			}
+			pl.Entries = append(pl.Entries, entry)
 		}
 	}
-	pl.Entry = entryList
+	return nil
+}
 
-	return pl, nil
+func polishList(roughMap *map[string]*Entry) []*Entry {
+	finalList := make([]*Entry, 0, len(*roughMap))
+	queuingList := make([]*Entry, 0, len(*roughMap)) // Domain/full entries without attr
+	domainsMap := make(map[string]bool)
+	for _, entry := range *roughMap {
+		switch entry.Type { // Bypass regexp, keyword and "full/domain with attr"
+		case dlc.RuleTypeRegexp:
+			finalList = append(finalList, entry)
+		case dlc.RuleTypeKeyword:
+			finalList = append(finalList, entry)
+		case dlc.RuleTypeDomain:
+			domainsMap[entry.Value] = true
+			if len(entry.Attrs) != 0 {
+				finalList = append(finalList, entry)
+			} else {
+				queuingList = append(queuingList, entry)
+			}
+		case dlc.RuleTypeFullDomain:
+			if len(entry.Attrs) != 0 {
+				finalList = append(finalList, entry)
+			} else {
+				queuingList = append(queuingList, entry)
+			}
+		}
+	}
+	// Remove redundant subdomains for full/domain without attr
+	for _, qentry := range queuingList {
+		isRedundant := false
+		pd := qentry.Value // To be parent domain
+		if qentry.Type == dlc.RuleTypeFullDomain {
+			pd = "." + pd // So that `domain:example.org` overrides `full:example.org`
+		}
+		for {
+			idx := strings.Index(pd, ".")
+			if idx == -1 {
+				break
+			}
+			pd = pd[idx+1:] // Go for next parent
+			if !strings.Contains(pd, ".") {
+				break
+			} // Not allow tld to be a parent
+			if domainsMap[pd] {
+				isRedundant = true
+				break
+			}
+		}
+		if !isRedundant {
+			finalList = append(finalList, qentry)
+		}
+	}
+	// Sort final entries
+	slices.SortFunc(finalList, func(a, b *Entry) int {
+		return strings.Compare(a.Plain, b.Plain)
+	})
+	return finalList
+}
+
+func resolveList(pl *ParsedList) error {
+	if _, pldone := finalMap[pl.Name]; pldone {
+		return nil
+	}
+
+	if cirIncMap[pl.Name] {
+		return fmt.Errorf("circular inclusion in: %s", pl.Name)
+	}
+	cirIncMap[pl.Name] = true
+	defer delete(cirIncMap, pl.Name)
+
+	isMatchAttrFilters := func(entry *Entry, incFilter *Inclusion) bool {
+		if len(incFilter.MustAttrs) == 0 && len(incFilter.BanAttrs) == 0 {
+			return true
+		}
+		if len(entry.Attrs) == 0 {
+			return len(incFilter.MustAttrs) == 0
+		}
+
+		for _, m := range incFilter.MustAttrs {
+			if !slices.Contains(entry.Attrs, m) {
+				return false
+			}
+		}
+		for _, b := range incFilter.BanAttrs {
+			if slices.Contains(entry.Attrs, b) {
+				return false
+			}
+		}
+		return true
+	}
+
+	roughMap := make(map[string]*Entry) // Avoid basic duplicates
+	for _, dentry := range pl.Entries { // Add direct entries
+		roughMap[dentry.Plain] = dentry
+	}
+	for _, inc := range pl.Inclusions {
+		incPl, exist := plMap[inc.Source]
+		if !exist {
+			return fmt.Errorf("list %q includes a non-existent list: %q", pl.Name, inc.Source)
+		}
+		if err := resolveList(incPl); err != nil {
+			return err
+		}
+		for _, ientry := range finalMap[inc.Source] {
+			if isMatchAttrFilters(ientry, inc) { // Add included entries
+				roughMap[ientry.Plain] = ientry
+			}
+		}
+	}
+	finalMap[pl.Name] = polishList(&roughMap)
+	return nil
 }
 
 func main() {
@@ -317,7 +385,7 @@ func main() {
 	dir := *dataPath
 	fmt.Println("Use domain lists in", dir)
 
-	ref := make(map[string]*List)
+	// Generate refMap
 	err := filepath.Walk(dir, func(path string, info os.FileInfo, err error) error {
 		if err != nil {
 			return err
@@ -325,75 +393,77 @@ func main() {
 		if info.IsDir() {
 			return nil
 		}
-		list, err := Load(path)
-		if err != nil {
+		if err := loadData(path); err != nil {
 			return err
 		}
-		ref[list.Name] = list
 		return nil
 	})
 	if err != nil {
-		fmt.Println("Failed:", err)
+		fmt.Println("Failed to loadData:", err)
 		os.Exit(1)
 	}
 
+	// Generate plMap
+	for refName, refList := range refMap {
+		if err := parseList(refName, refList); err != nil {
+			fmt.Println("Failed to parseList:", err)
+			os.Exit(1)
+		}
+	}
+
+	// Generate finalMap
+	for _, pl := range plMap {
+		if err := resolveList(pl); err != nil {
+			fmt.Println("Failed to resolveList:", err)
+			os.Exit(1)
+		}
+	}
+
 	// Create output directory if not exist
 	if _, err := os.Stat(*outputDir); os.IsNotExist(err) {
 		if mkErr := os.MkdirAll(*outputDir, 0755); mkErr != nil {
-			fmt.Println("Failed:", mkErr)
+			fmt.Println("Failed to create output directory:", mkErr)
 			os.Exit(1)
 		}
 	}
 
-	protoList := new(router.GeoSiteList)
-	var existList []string
-	for refName, list := range ref {
-		pl, err := ParseList(list, ref)
-		if err != nil {
-			fmt.Println("Failed:", err)
-			os.Exit(1)
+	// Export plaintext list
+	var exportListSlice []string
+	for raw := range strings.SplitSeq(*exportLists, ",") {
+		if trimmed := strings.TrimSpace(raw); trimmed != "" {
+			exportListSlice = append(exportListSlice, trimmed)
 		}
-		site, err := pl.toProto()
+	}
+	for _, exportList := range exportListSlice {
+		if err := writePlainList(exportList); err != nil {
+			fmt.Println("Failed to write list:", err)
+			continue
+		}
+		fmt.Printf("list %q has been generated successfully.\n", exportList)
+	}
+
+	// Generate dat file
+	protoList := new(router.GeoSiteList)
+	for siteName, siteEntries := range finalMap {
+		site, err := makeProtoList(siteName, siteEntries)
 		if err != nil {
-			fmt.Println("Failed:", err)
+			fmt.Println("Failed to makeProtoList:", err)
 			os.Exit(1)
 		}
 		protoList.Entry = append(protoList.Entry, site)
-
-		// Flatten and export plaintext list
-		if *exportLists != "" {
-			if existList != nil {
-				exportPlainTextList(existList, refName, pl)
-			} else {
-				exportedListSlice := strings.Split(*exportLists, ",")
-				for _, exportedListName := range exportedListSlice {
-					fileName := filepath.Join(dir, exportedListName)
-					_, err := os.Stat(fileName)
-					if err == nil || os.IsExist(err) {
-						existList = append(existList, exportedListName)
-					} else {
-						fmt.Printf("'%s' list does not exist in '%s' directory.\n", exportedListName, dir)
-					}
-				}
-				if existList != nil {
-					exportPlainTextList(existList, refName, pl)
-				}
-			}
-		}
 	}
-
 	// Sort protoList so the marshaled list is reproducible
-	sort.SliceStable(protoList.Entry, func(i, j int) bool {
-		return protoList.Entry[i].CountryCode < protoList.Entry[j].CountryCode
+	slices.SortFunc(protoList.Entry, func(a, b *router.GeoSite) int {
+		return strings.Compare(a.CountryCode, b.CountryCode)
 	})
 
 	protoBytes, err := proto.Marshal(protoList)
 	if err != nil {
-		fmt.Println("Failed:", err)
+		fmt.Println("Failed to marshal:", err)
 		os.Exit(1)
 	}
 	if err := os.WriteFile(filepath.Join(*outputDir, *outputName), protoBytes, 0644); err != nil {
-		fmt.Println("Failed:", err)
+		fmt.Println("Failed to write output:", err)
 		os.Exit(1)
 	} else {
 		fmt.Println(*outputName, "has been generated successfully.")