417 lines
12 KiB
Go
417 lines
12 KiB
Go
package sitemap
|
|
|
|
import (
|
|
"context"
|
|
"strings"
|
|
"sync"
|
|
"time"
|
|
"videoweb/pub/models/homeconfigmod"
|
|
"videoweb/pub/models/playletchaptermod"
|
|
"videoweb/pub/models/playletmod"
|
|
"videoweb/pub/models/playlettagmod"
|
|
"videoweb/pub/models/playletthememod"
|
|
"videoweb/pub/seo/h5path"
|
|
|
|
"go.mongodb.org/mongo-driver/bson"
|
|
"go.mongodb.org/mongo-driver/mongo/options"
|
|
)
|
|
|
|
type playEpisode struct {
|
|
Episode int64
|
|
LastMod time.Time
|
|
}
|
|
|
|
type idWithTime struct {
|
|
Id int64
|
|
Slug string
|
|
LastMod time.Time
|
|
IsOriginal bool
|
|
Episodes []playEpisode
|
|
}
|
|
|
|
type staticPageEntry struct {
|
|
path string
|
|
changeFreq string
|
|
priority float32
|
|
}
|
|
|
|
func buildStaticURLs(webAddr string, entries []staticPageEntry) []URL {
|
|
urls := make([]URL, 0, len(entries))
|
|
now := time.Now().Format(time.DateOnly)
|
|
for _, entry := range entries {
|
|
urls = append(urls, URL{
|
|
Loc: h5path.SitemapAbs(webAddr, entry.path),
|
|
LastMod: now,
|
|
ChangeFreq: entry.changeFreq,
|
|
Priority: entry.priority,
|
|
})
|
|
}
|
|
return urls
|
|
}
|
|
|
|
func entityPath(prefix, slug string, id int64) string {
|
|
return h5path.Entity(prefix, slug, id)
|
|
}
|
|
|
|
func siteStaticPages() []staticPageEntry {
|
|
return []staticPageEntry{
|
|
{path: "/", changeFreq: "daily", priority: 1.0},
|
|
{path: "/shorts", changeFreq: "daily", priority: 0.9},
|
|
{path: "/shorts/popular", changeFreq: "daily", priority: 0.7},
|
|
{path: "/shorts/original", changeFreq: "daily", priority: 0.7},
|
|
{path: "/shorts/random", changeFreq: "daily", priority: 0.7},
|
|
{path: "/series", changeFreq: "daily", priority: 0.9},
|
|
{path: "/series/popular", changeFreq: "daily", priority: 0.7},
|
|
{path: "/series/original", changeFreq: "daily", priority: 0.7},
|
|
{path: "/series/random", changeFreq: "daily", priority: 0.7},
|
|
{path: "/topics", changeFreq: "daily", priority: 0.8},
|
|
{path: "/rank", changeFreq: "daily", priority: 0.8},
|
|
{path: "/rank/new", changeFreq: "daily", priority: 0.7},
|
|
{path: "/rank/binge", changeFreq: "daily", priority: 0.7},
|
|
{path: "/home-road", changeFreq: "weekly", priority: 0.6},
|
|
{path: "/apps", changeFreq: "weekly", priority: 0.6},
|
|
{path: "/nav", changeFreq: "weekly", priority: 0.5},
|
|
{path: "/info/about-us", changeFreq: "yearly", priority: 0.3},
|
|
{path: "/info/user-agreement", changeFreq: "yearly", priority: 0.3},
|
|
{path: "/info/privacy-policy", changeFreq: "yearly", priority: 0.3},
|
|
{path: "/info/help-center", changeFreq: "yearly", priority: 0.3},
|
|
{path: "/info/2257", changeFreq: "yearly", priority: 0.3},
|
|
{path: "/info/dmca", changeFreq: "yearly", priority: 0.3},
|
|
}
|
|
}
|
|
|
|
// getURL 并发拉取短剧数据并写入 sitemap 分片。
|
|
// 路径对齐 https://91dj-h5.80sygtm.com/ 上有独立 canonical、且未被 noindex 的页面。
|
|
func getURL(sg *SitemapGenerator) {
|
|
webAddr := strings.TrimSpace(sg.WebAddr)
|
|
if webAddr == "" {
|
|
return
|
|
}
|
|
|
|
ctx := context.Background()
|
|
var (
|
|
wg sync.WaitGroup
|
|
listed listedPlaylets
|
|
themeItems []idWithTime
|
|
tagItems []idWithTime
|
|
homeConfigs []homeConfigItem
|
|
)
|
|
|
|
wg.Add(4)
|
|
go func() {
|
|
defer wg.Done()
|
|
listed = getPlayletList(ctx)
|
|
}()
|
|
go func() {
|
|
defer wg.Done()
|
|
themeItems = getThemeList(ctx)
|
|
}()
|
|
go func() {
|
|
defer wg.Done()
|
|
tagItems = getTagList(ctx)
|
|
}()
|
|
go func() {
|
|
defer wg.Done()
|
|
homeConfigs = getHomeConfigList(ctx)
|
|
}()
|
|
wg.Wait()
|
|
|
|
// 题材/标签没有上架短剧时不写入，避免空列表页被收录。
|
|
themeItems = keepWithListedPlaylets(themeItems, listed.themeIDs)
|
|
tagItems = keepWithListedPlaylets(tagItems, listed.tagIDs)
|
|
|
|
pageURLs := buildStaticURLs(webAddr, siteStaticPages())
|
|
for _, item := range themeItems {
|
|
pageURLs = append(pageURLs, newURL(webAddr, h5path.TopicDetail(item.Slug, item.Id), item.LastMod, "daily", 0.8))
|
|
}
|
|
for _, item := range tagItems {
|
|
if path := h5path.Tag(item.Slug, item.Id); path != "" {
|
|
pageURLs = append(pageURLs, newURL(webAddr, path, item.LastMod, "weekly", 0.6))
|
|
}
|
|
}
|
|
for _, item := range homeConfigs {
|
|
path := ""
|
|
switch item.Type {
|
|
case homeconfigmod.TypePlaylet:
|
|
path = h5path.MoreList(item.Slug, item.Id)
|
|
}
|
|
if path != "" {
|
|
pageURLs = append(pageURLs, newURL(webAddr, path, item.LastMod, "daily", 0.7))
|
|
}
|
|
}
|
|
if len(pageURLs) > 0 {
|
|
_ = sg.writeSitemapFile("sitemap-pages.xml", pageURLs)
|
|
}
|
|
|
|
// 详情页 /short-drama/{slug} 与播放页 /short-drama/{slug}/play/{集数} 都有独立 canonical。
|
|
// 原创且该集没有水印成片时不提交播放页，避免空壳播放地址被收录。
|
|
populatePlayEpisodes(ctx, listed.items)
|
|
if urls := buildPlayletPathURLs(webAddr, listed.items, "weekly", 0.8); len(urls) > 0 {
|
|
_ = sg.writeSitemapFiles("sitemap-playletdetail", urls)
|
|
}
|
|
if urls := buildPlayletEpisodeURLs(webAddr, listed.items, "weekly", 0.6); len(urls) > 0 {
|
|
_ = sg.writeSitemapFiles("sitemap-playletplay", urls)
|
|
}
|
|
_ = sg.writeSitemapFiles("sitemap-seriesplay", nil)
|
|
}
|
|
|
|
type listedPlaylets struct {
|
|
items []idWithTime
|
|
themeIDs map[int64]struct{}
|
|
tagIDs map[int64]struct{}
|
|
}
|
|
|
|
type homeConfigItem struct {
|
|
Id int64
|
|
Slug string
|
|
Type int32
|
|
LastMod time.Time
|
|
}
|
|
|
|
func newURL(webAddr, path string, lastMod time.Time, changeFreq string, priority float32) URL {
|
|
mod := formatLastMod(lastMod)
|
|
if mod == "" {
|
|
mod = time.Now().Format(time.DateOnly)
|
|
}
|
|
return URL{
|
|
Loc: h5path.SitemapAbs(webAddr, path),
|
|
LastMod: mod,
|
|
ChangeFreq: changeFreq,
|
|
Priority: priority,
|
|
}
|
|
}
|
|
|
|
func keepWithListedPlaylets(items []idWithTime, listed map[int64]struct{}) []idWithTime {
|
|
if len(items) == 0 || len(listed) == 0 {
|
|
return nil
|
|
}
|
|
out := make([]idWithTime, 0, len(items))
|
|
for _, item := range items {
|
|
if _, ok := listed[item.Id]; ok {
|
|
out = append(out, item)
|
|
}
|
|
}
|
|
return out
|
|
}
|
|
|
|
func addListedIDs(dst map[int64]struct{}, ids []int64) {
|
|
for _, id := range ids {
|
|
if id > 0 {
|
|
dst[id] = struct{}{}
|
|
}
|
|
}
|
|
}
|
|
|
|
func buildPlayletPathURLs(webAddr string, items []idWithTime, changeFreq string, priority float32) []URL {
|
|
urls := make([]URL, 0, len(items))
|
|
for _, item := range items {
|
|
path := h5path.Playlet(item.Slug, item.Id)
|
|
if path == "" {
|
|
continue
|
|
}
|
|
urls = append(urls, newURL(webAddr, path, item.LastMod, changeFreq, priority))
|
|
}
|
|
return urls
|
|
}
|
|
|
|
func buildPlayletEpisodeURLs(webAddr string, items []idWithTime, changeFreq string, priority float32) []URL {
|
|
urls := make([]URL, 0)
|
|
for _, item := range items {
|
|
for _, episode := range item.Episodes {
|
|
path := h5path.PlayletPlay(item.Slug, item.Id, episode.Episode)
|
|
if path == "" {
|
|
continue
|
|
}
|
|
lastMod := episode.LastMod
|
|
if lastMod.IsZero() {
|
|
lastMod = item.LastMod
|
|
}
|
|
urls = append(urls, newURL(webAddr, path, lastMod, changeFreq, priority))
|
|
}
|
|
}
|
|
return urls
|
|
}
|
|
|
|
func skipOriginalChapterWithoutWatermark(isOriginal bool, watermarkURL string) bool {
|
|
return isOriginal && strings.TrimSpace(watermarkURL) == ""
|
|
}
|
|
|
|
// populatePlayEpisodes 收集上架剧集的播放页。原创且没有水印成片的集不提交。
|
|
func populatePlayEpisodes(ctx context.Context, items []idWithTime) {
|
|
for start := 0; start < len(items); start += 500 {
|
|
end := start + 500
|
|
if end > len(items) {
|
|
end = len(items)
|
|
}
|
|
ids := make([]int64, 0, end-start)
|
|
byID := make(map[int64]*idWithTime, end-start)
|
|
for i := start; i < end; i++ {
|
|
if items[i].Id <= 0 {
|
|
continue
|
|
}
|
|
ids = append(ids, items[i].Id)
|
|
byID[items[i].Id] = &items[i]
|
|
}
|
|
if len(ids) == 0 {
|
|
continue
|
|
}
|
|
filter := bson.M{
|
|
playletchaptermod.FN.PlayletChapter.PlayletId: bson.M{"$in": ids},
|
|
}
|
|
for key, value := range playletchaptermod.VisibleStatusFilter() {
|
|
filter[key] = value
|
|
}
|
|
chapters, err := playletchaptermod.FindManyPlayletChapter(ctx, filter, options.Find().
|
|
SetSort(bson.D{
|
|
{Key: playletchaptermod.FN.PlayletChapter.CurrentEpisode, Value: 1},
|
|
{Key: playletchaptermod.FN.PlayletChapter.Id, Value: 1},
|
|
}).
|
|
SetProjection(bson.M{
|
|
playletchaptermod.FN.PlayletChapter.Id: 1,
|
|
playletchaptermod.FN.PlayletChapter.PlayletId: 1,
|
|
playletchaptermod.FN.PlayletChapter.CurrentEpisode: 1,
|
|
playletchaptermod.FN.PlayletChapter.WatermarkVideoUrl: 1,
|
|
playletchaptermod.FN.PlayletChapter.UpdatedAt: 1,
|
|
}))
|
|
if err != nil {
|
|
return
|
|
}
|
|
seen := make(map[int64]map[int64]struct{}, len(ids))
|
|
for _, ch := range chapters {
|
|
if ch == nil || ch.PlayletId <= 0 || ch.CurrentEpisode <= 0 {
|
|
continue
|
|
}
|
|
item := byID[ch.PlayletId]
|
|
if item == nil || skipOriginalChapterWithoutWatermark(item.IsOriginal, ch.WatermarkVideoUrl) {
|
|
continue
|
|
}
|
|
if seen[ch.PlayletId] == nil {
|
|
seen[ch.PlayletId] = make(map[int64]struct{})
|
|
}
|
|
if _, ok := seen[ch.PlayletId][ch.CurrentEpisode]; ok {
|
|
continue
|
|
}
|
|
seen[ch.PlayletId][ch.CurrentEpisode] = struct{}{}
|
|
item.Episodes = append(item.Episodes, playEpisode{Episode: ch.CurrentEpisode, LastMod: ch.UpdatedAt})
|
|
}
|
|
}
|
|
}
|
|
|
|
func formatLastMod(t time.Time) string {
|
|
if t.IsZero() {
|
|
return ""
|
|
}
|
|
return t.Format(time.DateOnly)
|
|
}
|
|
|
|
func getPlayletList(ctx context.Context) listedPlaylets {
|
|
result := listedPlaylets{
|
|
items: make([]idWithTime, 0),
|
|
themeIDs: make(map[int64]struct{}),
|
|
tagIDs: make(map[int64]struct{}),
|
|
}
|
|
filter := bson.M{
|
|
playletmod.FN.Playlet.Status: playletmod.StatusUp,
|
|
}
|
|
pageSize := int64(1000)
|
|
for i := int64(0); i < 500; i++ {
|
|
skip := i * pageSize
|
|
list, _ := playletmod.FindManyPlaylet(ctx, filter, options.Find().
|
|
SetSort(bson.M{playletmod.FN.Playlet.Id: -1}).
|
|
SetSkip(skip).
|
|
SetLimit(pageSize).
|
|
SetProjection(bson.M{
|
|
playletmod.FN.Playlet.Id: 1,
|
|
playletmod.FN.Playlet.Slug: 1,
|
|
playletmod.FN.Playlet.IsOriginal: 1,
|
|
playletmod.FN.Playlet.UpdatedAt: 1,
|
|
playletmod.FN.Playlet.OnlineAt: 1,
|
|
playletmod.FN.Playlet.ThemeIds: 1,
|
|
playletmod.FN.Playlet.TagIds: 1,
|
|
}))
|
|
for _, info := range list {
|
|
if info == nil || info.Id <= 0 {
|
|
continue
|
|
}
|
|
lastMod := info.UpdatedAt
|
|
if lastMod.IsZero() {
|
|
lastMod = info.OnlineAt
|
|
}
|
|
result.items = append(result.items, idWithTime{Id: info.Id, Slug: info.Slug, LastMod: lastMod, IsOriginal: info.IsOriginal})
|
|
addListedIDs(result.themeIDs, info.ThemeIds)
|
|
addListedIDs(result.tagIDs, info.TagIds)
|
|
}
|
|
if len(list) < int(pageSize) {
|
|
break
|
|
}
|
|
}
|
|
return result
|
|
}
|
|
|
|
func getThemeList(ctx context.Context) []idWithTime {
|
|
filter := bson.M{
|
|
playletthememod.FN.PlayletTheme.Status: true,
|
|
}
|
|
pageSize := int64(1000)
|
|
items := make([]idWithTime, 0)
|
|
for i := int64(0); i < 100; i++ {
|
|
skip := i * pageSize
|
|
list, _ := playletthememod.FindManyPlayletTheme(ctx, filter, options.Find().
|
|
SetSort(bson.M{playletthememod.FN.PlayletTheme.Rank: -1}).
|
|
SetSkip(skip).
|
|
SetLimit(pageSize))
|
|
for _, info := range list {
|
|
if info == nil || info.Id <= 0 || strings.TrimSpace(info.Name) == "" {
|
|
continue
|
|
}
|
|
items = append(items, idWithTime{Id: info.Id, Slug: info.Slug, LastMod: info.UpdatedAt})
|
|
}
|
|
if len(list) < int(pageSize) {
|
|
break
|
|
}
|
|
}
|
|
return items
|
|
}
|
|
|
|
func getTagList(ctx context.Context) []idWithTime {
|
|
filter := bson.M{
|
|
playlettagmod.FN.PlayletTag.Status: true,
|
|
}
|
|
pageSize := int64(1000)
|
|
items := make([]idWithTime, 0)
|
|
for i := int64(0); i < 100; i++ {
|
|
skip := i * pageSize
|
|
list, _ := playlettagmod.FindManyPlayletTag(ctx, filter, options.Find().
|
|
SetSort(bson.M{playlettagmod.FN.PlayletTag.Rank: -1}).
|
|
SetSkip(skip).
|
|
SetLimit(pageSize))
|
|
for _, info := range list {
|
|
if info == nil || info.Id <= 0 || strings.TrimSpace(info.Name) == "" {
|
|
continue
|
|
}
|
|
items = append(items, idWithTime{Id: info.Id, Slug: info.Slug, LastMod: info.UpdatedAt})
|
|
}
|
|
if len(list) < int(pageSize) {
|
|
break
|
|
}
|
|
}
|
|
return items
|
|
}
|
|
|
|
func getHomeConfigList(ctx context.Context) []homeConfigItem {
|
|
list, _ := homeconfigmod.FindManyHomeConfig(ctx, bson.M{
|
|
homeconfigmod.FN.HomeConfig.Status: true,
|
|
homeconfigmod.FN.HomeConfig.Type: bson.M{
|
|
"$in": []int32{homeconfigmod.TypePlaylet},
|
|
},
|
|
}, options.Find().SetSort(bson.M{homeconfigmod.FN.HomeConfig.Rank: -1}))
|
|
items := make([]homeConfigItem, 0, len(list))
|
|
for _, info := range list {
|
|
if info == nil || info.Id <= 0 {
|
|
continue
|
|
}
|
|
items = append(items, homeConfigItem{Id: info.Id, Slug: info.Slug, Type: info.Type, LastMod: info.UpdatedAt})
|
|
}
|
|
return items
|
|
}
|