xiaomi-ads: add stats.music.xiaomi.com (#3159)

data/xiaomi-ads

@@ -14,6 +14,7 @@ logupdate.avlyun.sec.miui.com @ads
 misc.in.duokanbox.com @ads
 sentry.d.mi.com @ads
 sentry.d.xiaomi.net @ads
+stats.music.xiaomi.com @ads
 tjqonline.cn @ads
 tracker.ai.xiaomi.com @ads
 tracking.miui.com @ads

main.go

@@ -29,29 +29,10 @@ const (
 	RuleTypeInclude    string = "include"
 )
 
-var (
-	TypeChecker  = regexp.MustCompile(`^(domain|full|keyword|regexp|include)$`)
-	ValueChecker = regexp.MustCompile(`^[a-z0-9!\.-]+$`)
-	AttrChecker  = regexp.MustCompile(`^[a-z0-9!-]+$`)
-)
-
-var (
-	refMap    = make(map[string]*List)
-	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 []string
+	Attrs []*router.Domain_Attribute
-}
-
-type Inclusion struct {
-	Source      string
-	MustAttrs   []string
-	BannedAttrs []string
 }
 
 type List struct {
@@ -60,108 +41,150 @@ type List struct {
 }
 
 type ParsedList struct {
-	Name       string
+	Name      string
-	Inclusions []Inclusion
+	Inclusion map[string]bool
-	Entry      []Entry
+	Entry     []Entry
 }
 
-func makeProtoList(listName string, entries []Entry) (*router.GeoSite, error) {
+func (l *ParsedList) toPlainText(listName string) error {
-	site := &router.GeoSite{
+	var entryBytes []byte
-		CountryCode: listName,
+	for _, entry := range l.Entry {
-	}
+		var attrString string
-	for _, entry := range entries {
+		if entry.Attrs != nil {
-		pdomain := &router.Domain{Value: entry.Value}
+			for _, attr := range entry.Attrs {
-		for _, attr := range entry.Attrs {
+				attrString += "@" + attr.GetKey() + ","
-			pdomain.Attribute = append(pdomain.Attribute, &router.Domain_Attribute{
+			}
-				Key:        attr,
+			attrString = strings.TrimRight(":"+attrString, ",")
-				TypedValue: &router.Domain_Attribute_BoolValue{BoolValue: true},
-			})
 		}
+		// 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) {
+	site := &router.GeoSite{
+		CountryCode: l.Name,
+	}
+	for _, entry := range l.Entry {
 		switch entry.Type {
 		case RuleTypeDomain:
-			pdomain.Type = router.Domain_RootDomain
+			site.Domain = append(site.Domain, &router.Domain{
+				Type:      router.Domain_RootDomain,
+				Value:     entry.Value,
+				Attribute: entry.Attrs,
+			})
+
 		case RuleTypeRegexp:
-			pdomain.Type = router.Domain_Regex
+			// 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:
-			pdomain.Type = router.Domain_Plain
+			site.Domain = append(site.Domain, &router.Domain{
+				Type:      router.Domain_Plain,
+				Value:     entry.Value,
+				Attribute: entry.Attrs,
+			})
+
 		case RuleTypeFullDomain:
-			pdomain.Type = router.Domain_Full
+			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)
 		}
-		site.Domain = append(site.Domain, pdomain)
 	}
 	return site, nil
 }
 
-func writePlainList(exportedName string) error {
+func exportPlainTextList(list []string, refName string, pl *ParsedList) {
-	targetList, exist := finalMap[strings.ToUpper(exportedName)]
+	for _, listName := range list {
-	if !exist || len(targetList) == 0 {
+		if strings.EqualFold(refName, listName) {
-		return fmt.Errorf("'%s' list does not exist or is empty.", exportedName)
+			if err := pl.toPlainText(strings.ToLower(refName)); err != nil {
-	}
+				fmt.Println("Failed:", err)
-	file, err := os.Create(filepath.Join(*outputDir, strings.ToLower(exportedName) + ".txt"))
+				continue
-	if err != nil {
+			}
-		return err
+			fmt.Printf("'%s' has been generated successfully.\n", listName)
-	}
-	defer file.Close()
-	w := bufio.NewWriter(file)
-	for _, entry := range targetList {
-		// Entry output format is: type:domain.tld:@attr1,@attr2
-		var attrString string
-		if entry.Attrs != nil {
-			attrString = ":@" + strings.Join(entry.Attrs, ",@")
 		}
-		fmt.Fprintln(w, entry.Type + ":" + entry.Value + attrString)
 	}
-	return w.Flush()
 }
 
-func parseEntry(line string) (Entry, error) {
+func removeComment(line string) string {
-	var entry Entry
+	idx := strings.Index(line, "#")
-	parts := strings.Fields(line)
+	if idx == -1 {
+		return line
+	}
+	return strings.TrimSpace(line[:idx])
+}
 
-	// Parse type and value
+func parseDomain(domain string, entry *Entry) error {
-	rawTypeVal := parts[0]
+	kv := strings.Split(domain, ":")
-	kv := strings.Split(rawTypeVal, ":")
 	if len(kv) == 1 {
-		entry.Type = RuleTypeDomain // Default type
+		entry.Type = RuleTypeDomain
-		entry.Value = strings.ToLower(rawTypeVal)
+		entry.Value = strings.ToLower(kv[0])
-	} else if len(kv) == 2 {
+		return nil
+	}
+
+	if len(kv) == 2 {
 		entry.Type = strings.ToLower(kv[0])
-		if entry.Type == RuleTypeRegexp {
+
+		if strings.EqualFold(entry.Type, RuleTypeRegexp) {
 			entry.Value = kv[1]
 		} else {
 			entry.Value = strings.ToLower(kv[1])
 		}
-	} else {
+
-		return entry, fmt.Errorf("invalid format: %s", line)
+		return nil
-	}
-	// Check type and value
-	if !TypeChecker.MatchString(entry.Type) {
-		return entry, fmt.Errorf("invalid type: %s", entry.Type)
-	}
-	if entry.Type == RuleTypeRegexp {
-		if _, err := regexp.Compile(entry.Value); err != nil {
-			return entry, fmt.Errorf("invalid regexp: %s", entry.Value)
-		}
-	} else if !ValueChecker.MatchString(entry.Value) {
-		return entry, fmt.Errorf("invalid value: %s", entry.Value)
 	}
 
-	// Parse/Check attributes
+	return fmt.Errorf("invalid format: %s", domain)
-	for _, part := range parts[1:] {
+}
-		if !strings.HasPrefix(part, "@") {
+
-			return entry, fmt.Errorf("invalid attribute: %s", part)
+func parseAttribute(attr string) (*router.Domain_Attribute, error) {
-		}
+	var attribute router.Domain_Attribute
-		attr := strings.ToLower(part[1:]) // Trim attribute prefix `@` character
+	if len(attr) == 0 || attr[0] != '@' {
-		if !AttrChecker.MatchString(attr) {
+		return &attribute, fmt.Errorf("invalid attribute: %s", attr)
-			return entry, fmt.Errorf("invalid attribute key: %s", attr)
+	}
+
+	attribute.Key = strings.ToLower(attr[1:]) // Trim attribute prefix `@` character
+	attribute.TypedValue = &router.Domain_Attribute_BoolValue{BoolValue: true}
+	return &attribute, nil
+}
+
+func parseEntry(line string) (Entry, error) {
+	line = strings.TrimSpace(line)
+	parts := strings.Split(line, " ")
+
+	var entry Entry
+	if len(parts) == 0 {
+		return entry, fmt.Errorf("empty entry")
+	}
+
+	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
 		}
 		entry.Attrs = append(entry.Attrs, attr)
 	}
-	// Sort attributes
-	sort.Slice(entry.Attrs, func(i, j int) bool {
-		return entry.Attrs[i] < entry.Attrs[j]
-	})
 
 	return entry, nil
 }
@@ -178,13 +201,9 @@ func Load(path string) (*List, error) {
 	}
 	scanner := bufio.NewScanner(file)
 	for scanner.Scan() {
-		line := scanner.Text()
+		line := strings.TrimSpace(scanner.Text())
-		// Remove comments
+		line = removeComment(line)
-		if idx := strings.Index(line, "#"); idx != -1 {
+		if len(line) == 0 {
-			line = line[:idx]
-		}
-		line = strings.TrimSpace(line)
-		if line == "" {
 			continue
 		}
 		entry, err := parseEntry(line)
@@ -197,80 +216,99 @@ func Load(path string) (*List, error) {
 	return list, nil
 }
 
-func ParseList(refList *List) error {
+func isMatchAttr(Attrs []*router.Domain_Attribute, includeKey string) bool {
-	//TODO: one Entry -> multiple ParsedLists
+	isMatch := false
-	pl := &ParsedList{Name: refList.Name}
+	mustMatch := true
-	for _, entry := range refList.Entry {
+	matchName := includeKey
-		if entry.Type == RuleTypeInclude {
+	if strings.HasPrefix(includeKey, "!") {
-			inc := Inclusion{Source: strings.ToUpper(entry.Value)}
+		isMatch = true
-			for _, attr := range entry.Attrs {
+		mustMatch = false
-				if strings.HasPrefix(attr, "-") {
+		matchName = strings.TrimLeft(includeKey, "!")
-					inc.BannedAttrs = append(inc.BannedAttrs, attr[1:]) // Trim attribute prefix `-` character
+	}
-				} else {
+
-					inc.MustAttrs = append(inc.MustAttrs, attr)
+	for _, Attr := range Attrs {
-				}
+		attrName := Attr.Key
+		if mustMatch {
+			if matchName == attrName {
+				isMatch = true
+				break
 			}
-			pl.Inclusions = append(pl.Inclusions, inc)
 		} else {
-			pl.Entry = append(pl.Entry, entry)
+			if matchName == attrName {
+				isMatch = false
+				break
+			}
 		}
 	}
-	plMap[refList.Name] = pl
+	return isMatch
-	return nil
 }
 
-func ResolveList(pl *ParsedList) error {
+func createIncludeAttrEntrys(list *List, matchAttr *router.Domain_Attribute) []Entry {
-	if _, pldone := finalMap[pl.Name]; pldone { return nil }
+	newEntryList := make([]Entry, 0, len(list.Entry))
-
+	matchName := matchAttr.Key
-	if cirIncMap[pl.Name] {
+	for _, entry := range list.Entry {
-		return fmt.Errorf("circular inclusion in: %s", pl.Name)
+		matched := isMatchAttr(entry.Attrs, matchName)
-	}
+		if matched {
-	cirIncMap[pl.Name] = true
+			newEntryList = append(newEntryList, entry)
-	defer delete(cirIncMap, pl.Name)
-
-	entry2String := func(e Entry) string { // Attributes already sorted
-		return e.Type + ":" + e.Value + "@" + strings.Join(e.Attrs, "@")
-	}
-	isMatchAttrFilters := func(entry Entry, incFilter Inclusion) bool {
-		if len(incFilter.MustAttrs) == 0 && len(incFilter.BannedAttrs) == 0 { return true }
-
-		attrMap := make(map[string]bool)
-		for _, attr := range entry.Attrs {
-			attrMap[attr] = true
-		}
-		for _, m := range incFilter.MustAttrs {
-			if !attrMap[m] { return false }
-		}
-		for _, b := range incFilter.BannedAttrs {
-			if attrMap[b] { return false }
-		}
-		return true
-	}
-
-	bscDupMap := make(map[string]bool) // Used for basic duplicates detection
-	var finalList []Entry
-	for _, dentry := range pl.Entry {
-		if dstring := entry2String(dentry); !bscDupMap[dstring] {
-			bscDupMap[dstring] = true
-			finalList = append(finalList, dentry)
 		}
 	}
+	return newEntryList
+}
 
-	for _, inc := range pl.Inclusions {
+func ParseList(list *List, ref map[string]*List) (*ParsedList, error) {
-		if err := ResolveList(plMap[inc.Source]); err != nil {
+	pl := &ParsedList{
-			return err
+		Name:      list.Name,
-		}
+		Inclusion: make(map[string]bool),
-		for _, ientry := range finalMap[inc.Source] {
+	}
-			if isMatchAttrFilters(ientry, inc) {
+	entryList := list.Entry
-				if istring := entry2String(ientry); !bscDupMap[istring] {
+	for {
-					bscDupMap[istring] = true
+		newEntryList := make([]Entry, 0, len(entryList))
-					finalList = append(finalList, ientry)
+		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...)
+						}
+					}
+				} 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...)
 				}
+				hasInclude = true
+			} else {
+				newEntryList = append(newEntryList, entry)
 			}
 		}
+		entryList = newEntryList
+		if !hasInclude {
+			break
+		}
 	}
-	finalMap[pl.Name] = finalList
+	pl.Entry = entryList
-	return nil
+
+	return pl, nil
 }
 
 func main() {
@@ -279,7 +317,7 @@ func main() {
 	dir := *dataPath
 	fmt.Println("Use domain lists in", dir)
 
-	// Generate refMap
+	ref := make(map[string]*List)
 	err := filepath.Walk(dir, func(path string, info os.FileInfo, err error) error {
 		if err != nil {
 			return err
@@ -291,7 +329,7 @@ func main() {
 		if err != nil {
 			return err
 		}
-		refMap[list.Name] = list
+		ref[list.Name] = list
 		return nil
 	})
 	if err != nil {
@@ -299,22 +337,6 @@ func main() {
 		os.Exit(1)
 	}
 
-	// Generate plMap
-	for _, refList := range refMap {
-		if err := ParseList(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 {
@@ -323,28 +345,43 @@ func main() {
 		}
 	}
 
-	// Export plaintext list
-	if *exportLists != "" {
-		exportedListSlice := strings.Split(*exportLists, ",")
-		for _, exportedList := range exportedListSlice {
-			if err := writePlainList(exportedList); err != nil {
-				fmt.Println("Failed to write list:", err)
-				continue
-			}
-			fmt.Printf("list: '%s' has been generated successfully.\n", exportedList)
-		}
-	}
-
-	// Generate dat file
 	protoList := new(router.GeoSiteList)
-	for siteName, siteEntries := range finalMap {
+	var existList []string
-		site, err := makeProtoList(siteName, siteEntries)
+	for refName, list := range ref {
+		pl, err := ParseList(list, ref)
+		if err != nil {
+			fmt.Println("Failed:", err)
+			os.Exit(1)
+		}
+		site, err := pl.toProto()
 		if err != nil {
 			fmt.Println("Failed:", 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
@@ -352,11 +389,11 @@ func main() {
 
 	protoBytes, err := proto.Marshal(protoList)
 	if err != nil {
-		fmt.Println("Failed to marshal:", err)
+		fmt.Println("Failed:", err)
 		os.Exit(1)
 	}
 	if err := os.WriteFile(filepath.Join(*outputDir, *outputName), protoBytes, 0644); err != nil {
-		fmt.Println("Failed to write output:", err)
+		fmt.Println("Failed:", err)
 		os.Exit(1)
 	} else {
 		fmt.Println(*outputName, "has been generated successfully.")