Skip to content

Commit

Permalink
scraper: Rewrite VirtualPorn using API as a Source (xbapps#1654)
Browse files Browse the repository at this point in the history
  • Loading branch information
@toshski
toshski authored Apr 2, 2024
1 parent 42a24eb commit 219f19a
Show file tree
Hide file tree
Showing 3 changed files with 208 additions and 107 deletions.
7 changes: 7 additions & 0 deletions pkg/migrations/migrations.go
View file
Original file line number Diff line number Diff line change
Expand Up @@ -1925,6 +1925,13 @@ func Migrate() {
return err
},
},
{
ID: "0077-Update-VirtualPorn-ids",
Migrate: func(tx *gorm.DB) error {
err := scrape.UpdateVirtualPornIds()
return err
},
},
})

if err := m.Migrate(); err != nil {
Expand Down
2 changes: 1 addition & 1 deletion pkg/models/model_external_reference.go
View file
Original file line number Diff line number Diff line change
Expand Up @@ -518,7 +518,7 @@ func (scrapeRules ActorScraperConfig) buildGenericActorScraperRules() {

siteDetails = GenericScraperRuleSet{}
siteDetails.Domain = "virtualporn.com"
siteDetails.SiteRules = append(siteDetails.SiteRules, GenericActorScraperRule{XbvrField: "image_url", Selector: `div.model__img-wrapper > img`, ResultType: "attr", Attribute: "src"})
siteDetails.SiteRules = append(siteDetails.SiteRules, GenericActorScraperRule{XbvrField: "image_url", Selector: `section[data-cy="actorProfilePicture"] img`, ResultType: "attr", Attribute: "src"})
scrapeRules.GenericActorScrapingConfig["bvr scrape"] = siteDetails

siteDetails = GenericScraperRuleSet{}
Expand Down
306 changes: 200 additions & 106 deletions pkg/scrape/virtualporn.go
View file
Original file line number Diff line number Diff line change
Expand Up @@ -2,150 +2,167 @@ package scrape

import (
"encoding/json"
"errors"
"regexp"
"strconv"
"strings"
"sync"

"github.com/gocolly/colly/v2"
"github.com/nleeper/goment"
"github.com/mozillazg/go-slugify"
"github.com/thoas/go-funk"
"github.com/tidwall/gjson"
"github.com/xbapps/xbvr/pkg/models"
)

func VirtualPorn(wg *sync.WaitGroup, updateSite bool, knownScenes []string, out chan<- models.ScrapedScene, singleSceneURL string, singeScrapeAdditionalInfo string, limitScraping bool) error {
// this scraper is non-standard in that it gathers info via an api rather than scraping html pages
defer wg.Done()
scraperID := "bvr"
siteID := "VirtualPorn"
logScrapeStart(scraperID, siteID)
nextApiUrl := ""

sceneCollector := createCollector("virtualporn.com")
siteCollector := createCollector("virtualporn.com")
pageCnt := 1

sceneCollector.OnHTML(`html`, func(e *colly.HTMLElement) {
sc := models.ScrapedScene{}
sc.ScraperID = scraperID
sc.SceneType = "VR"
sc.Studio = "BangBros"
sc.Site = siteID
sc.HomepageURL = strings.Split(e.Request.URL.String(), "?")[0]
sc.MembersUrl = "https://members.bangbros.com/product/655/movie/" + strings.Replace(strings.Split(e.Request.URL.String(), "/")[3], "video", "", 1)

// Title / Cover / ID / Filenames
e.ForEach(`dl8-video`, func(id int, e *colly.HTMLElement) {
sc.Title = strings.TrimSpace(e.Attr("title"))

tmpCover := e.Request.AbsoluteURL(e.Request.AbsoluteURL(e.Attr("poster")))
sc.Covers = append(sc.Covers, tmpCover)

tmp := strings.Split(tmpCover, "/")
sc.SceneID = strings.Replace(tmp[5], "bvr", "bvr-", 1)

e.ForEach(`source`, func(id int, e *colly.HTMLElement) {
tmpFile := strings.Split(e.Attr("src"), "/")
sc.Filenames = append(sc.Filenames, strings.Replace(tmpFile[len(tmpFile)-1], "trailer-", "", -1))
apiCollector := createCollector("site-api.project1service.com")
offset := 0

apiCollector.OnResponse(func(r *colly.Response) {
sceneListJson := gjson.ParseBytes(r.Body)

processScene := func(scene gjson.Result) {
sc := models.ScrapedScene{}
sc.ScraperID = scraperID
sc.SceneType = "VR"
sc.Studio = "BangBros"
sc.Site = siteID
id := strconv.Itoa(int(scene.Get("id").Int()))
sc.SceneID = "bvr-" + id

sc.Title = scene.Get("title").String()
sc.HomepageURL = "https://virtualporn.com/video/" + id + "/" + slugify.Slugify(strings.ReplaceAll(sc.Title, "'", ""))
sc.MembersUrl = "https://site-ma.virtualporn.com/scene/" + id + "/" + slugify.Slugify(strings.ReplaceAll(sc.Title, "'", ""))
sc.Synopsis = scene.Get("description").String()
dateParts := strings.Split(scene.Get("dateReleased").String(), "T")
sc.Released = dateParts[0]

scene.Get("images.poster").ForEach(func(key, imgGroup gjson.Result) bool {
if key.String() == "0" {
imgurl := imgGroup.Get("xl.urls.webp").String()
if imgurl != "" {
sc.Covers = append(sc.Covers, imgurl)
}

} else {
imgurl := imgGroup.Get("xl.urls.webp").String()
if imgurl != "" {
if len(sc.Covers) == 0 {
sc.Covers = append(sc.Covers, imgurl)
} else {
sc.Gallery = append(sc.Gallery, imgurl)
}
}
}
return true
})
})

file5kExists := false
for _, filename := range sc.Filenames {
if strings.Contains(filename, "5k") {
file5kExists = true
}
}
if !file5kExists {
sc.Filenames = append(sc.Filenames, strings.Replace(sc.SceneID, "bvr-", "bvr", -1)+"-5k.mp4")
}

// Gallery
e.ForEach(`div.player__thumbs img`, func(id int, e *colly.HTMLElement) {
sc.Gallery = append(sc.Gallery, e.Attr("src"))
})
// Cast
sc.ActorDetails = make(map[string]models.ActorDetails)
scene.Get("actors").ForEach(func(key, actor gjson.Result) bool {
name := actor.Get("name").String()
if actor.Get("gender").String() == "female" {
sc.Cast = append(sc.Cast, name)
}
sc.ActorDetails[actor.Get("name").String()] = models.ActorDetails{Source: scraperID + " scrape", ProfileUrl: "https://virtualporn.com/model/" + strconv.Itoa(int(actor.Get("id").Int())) + "/" + slugify.Slugify(name)}
return true
})

// trailer details
sc.TrailerType = "scrape_html"
params := models.TrailerScrape{SceneUrl: sc.HomepageURL, HtmlElement: "dl8-video source", ContentPath: "src", QualityPath: "quality"}
strParams, _ := json.Marshal(params)
sc.TrailerSrc = string(strParams)

// Cast
sc.ActorDetails = make(map[string]models.ActorDetails)
e.ForEach(`div.player__stats p.player__stats__cast a`, func(id int, e *colly.HTMLElement) {
if strings.TrimSpace(e.Text) != "" {
sc.Cast = append(sc.Cast, strings.TrimSpace(strings.ReplaceAll(e.Text, "!", "")))
sc.ActorDetails[strings.TrimSpace(strings.ReplaceAll(e.Text, "!", ""))] = models.ActorDetails{Source: scraperID + " scrape", ProfileUrl: e.Request.AbsoluteURL(e.Attr("href"))}
}
})
// Tags
scene.Get("tags").ForEach(func(key, tag gjson.Result) bool {
if tag.Get("isVisible").Bool() {
sc.Tags = append(sc.Tags, tag.Get("name").String())
}
return true
})

// Tags
e.ForEach(`div.video__tags__list a.tags`, func(id int, e *colly.HTMLElement) {
tag := strings.TrimSpace(e.Text)
if tag != "" {
sc.Tags = append(sc.Tags, strings.ToLower(tag))
}
})
// trailer & filename details
sc.TrailerType = "urls"
var trailers []models.VideoSource
scene.Get("children").ForEach(func(key, child gjson.Result) bool {
child.Get("videos.full.files").ForEach(func(key, file gjson.Result) bool {
quality := file.Get("format").String()
url := file.Get("urls.view").String()
filename := file.Get("urls.download").String()
if url != "" {
trailers = append(trailers, models.VideoSource{URL: url, Quality: quality})
}
pos := strings.Index(filename, "?filename=")
if pos != -1 {
sc.Filenames = append(sc.Filenames, filename[pos+10:])
}
return true
})
return true
})
trailerJson, _ := json.Marshal(models.VideoSourceResponse{VideoSources: trailers})
sc.TrailerSrc = string(trailerJson)

// Synposis
e.ForEach(`p.player__description`, func(id int, e *colly.HTMLElement) {
sc.Synopsis = strings.TrimSpace(e.Text)
})
out <- sc

// Release date / Duration
tmpDate, _ := goment.New(strings.TrimSpace(e.Request.Ctx.GetAny("date").(string)), "MMM DD, YYYY")
sc.Released = tmpDate.Format("YYYY-MM-DD")
tmpDuration, err := strconv.Atoi(strings.TrimSpace(strings.Replace(e.Request.Ctx.GetAny("dur").(string), "mins", "", -1)))
if err == nil {
sc.Duration = tmpDuration
}
total := int(sceneListJson.Get("meta.total").Int())
scenes := sceneListJson.Get("result")
if strings.Contains(r.Request.URL.RawQuery, "offset=") {
scenes.ForEach(func(key, scene gjson.Result) bool {
// check if we have the scene already
matches := funk.Filter(knownScenes, func(s string) bool {
return strings.Contains(s, scene.Get("id").String())
})
if funk.IsEmpty(matches) {
processScene(scene)
}
return true
})
} else {
processScene(scenes)
}

out <- sc
})

siteCollector.OnHTML(`body`, func(e *colly.HTMLElement) {
sceneCnt := 0
e.ForEach(`div.recommended__item`, func(id int, e *colly.HTMLElement) {
sceneCnt += 1
})

if sceneCnt > 0 {
pageCnt += 1
offset += 24
if offset < total {
if !limitScraping {
siteCollector.Visit("https://virtualporn.com/videos/" + strconv.Itoa(pageCnt))
apiCollector.Visit("https://site-api.project1service.com/v2/releases?type=scene&limit=24&offset=" + strconv.Itoa(offset))
}
}
})

siteCollector.OnHTML(`div.recommended__item`, func(e *colly.HTMLElement) {
sceneURL := e.Request.AbsoluteURL(e.ChildAttr(`a`, "href"))

// If scene exist in database, there's no need to scrape
if !funk.ContainsString(knownScenes, sceneURL) {

//Date & Duration from main index
ctx := colly.NewContext()
e.ForEach(`span.recommended__item__info__date`, func(id int, e *colly.HTMLElement) {
if id == 0 {
ctx.Put("date", strings.TrimSpace(e.Text))
}
})
e.ForEach(`span.recommended__item__time`, func(id int, e *colly.HTMLElement) {
if id == 0 {
ctx.Put("dur", strings.TrimSpace(e.Text))
}
siteCollector.OnHTML(`script`, func(e *colly.HTMLElement) {
// only interested in a script containg window\.__JUAN\.rawInstance
re := regexp.MustCompile(`window\.__JUAN\.rawInstance = (\{.*?\});`)
matches := re.FindStringSubmatch(e.Text)
if len(matches) > 1 {
instanceJson := gjson.ParseBytes([]byte(matches[1]))
token := instanceJson.Get("jwt").String()
// set up api requests to use the token in the Instance Header
apiCollector.OnRequest(func(r *colly.Request) {
r.Headers.Set("Instance", token)
})

sceneCollector.Request("GET", sceneURL, nil, ctx, nil)
apiCollector.Visit(nextApiUrl)
}
})

if singleSceneURL != "" {
ctx := colly.NewContext()
ctx.Put("dur", "")
ctx.Put("date", "")
urlParts := strings.Split(singleSceneURL, "/")
id := urlParts[len(urlParts)-2]
offset = 9999 // do read more pages, we only need 1
nextApiUrl = "https://site-api.project1service.com/v2/releases/" + id
siteCollector.Visit("https://virtualporn.com/videos")

sceneCollector.Request("GET", singleSceneURL, nil, ctx, nil)
} else {
siteCollector.Visit("https://virtualporn.com/videos/" + strconv.Itoa(pageCnt))
// call virtualporn.com, this is just to get the instance token to use the api for this session
nextApiUrl = "https://site-api.project1service.com/v2/releases?type=scene&limit=24&offset=" + strconv.Itoa(offset)
siteCollector.Visit("https://virtualporn.com/videos")
}

if updateSite {
Expand All @@ -158,3 +175,80 @@ func VirtualPorn(wg *sync.WaitGroup, updateSite bool, knownScenes []string, out
func init() {
registerScraper("bvr", "VirtualPorn", "https://images.cn77nd.com/members/bangbros/favicon/apple-icon-60x60.png", "virtualporn.com", VirtualPorn)
}

// one off conversion routine called by migrations.go
func UpdateVirtualPornIds() error {
collector := createCollector("virtualporn.com")
apiCollector := createCollector("site-api.project1service.com")
offset := 0
sceneCnt := 0

collector.OnHTML(`script`, func(e *colly.HTMLElement) {
// only interested in a script containg window\.__JUAN\.rawInstance
re := regexp.MustCompile(`window\.__JUAN\.rawInstance = (\{.*?\});`)
matches := re.FindStringSubmatch(e.Text)
if len(matches) > 1 {
instanceJson := gjson.ParseBytes([]byte(matches[1]))
token := instanceJson.Get("jwt").String()
// set up api requests to use the token in the Instance Header
apiCollector.OnRequest(func(r *colly.Request) {
r.Headers.Set("Instance", token)
})
apiCollector.Visit("https://site-api.project1service.com/v2/releases?type=scene&limit=100&offset=" + strconv.Itoa(offset))
}
})

apiCollector.OnResponse(func(r *colly.Response) {
db, _ := models.GetDB()
defer db.Close()

sceneListJson := gjson.ParseBytes(r.Body)
sceneCnt = int(sceneListJson.Get("meta.total").Int())
scenes := sceneListJson.Get("result")
scenes.ForEach(func(key, apiScene gjson.Result) bool {
id := strconv.Itoa(int(apiScene.Get("id").Int()))
title := apiScene.Get("title").String()
dateParts := strings.Split(apiScene.Get("dateReleased").String(), "T")
releasedDate := dateParts[0]
var scene models.Scene
scene.GetIfExist("bvr-" + id)
if scene.ID > 0 {
// get the next record, this one already matches the new id
return true
}
db.Where("scraper_id = ? and release_date_text = ?", "bvr", releasedDate).Find(&scene)
if scene.ID > 0 {
oldSceneId := scene.SceneID
log.Infof("Updating SceneId %s to %s ", oldSceneId, "bvr-"+id)
scene.LegacySceneID = scene.SceneID
scene.SceneID = "bvr-" + id
scene.SceneURL = "https://virtualporn.com/video/" + id + "/" + slugify.Slugify(strings.ReplaceAll(title, "'", ""))
scene.MemberURL = "https://site-ma.virtualporn.com/scene/" + id + "/" + slugify.Slugify(strings.ReplaceAll(title, "'", ""))

scene.Save()
result := db.Model(&models.Action{}).Where("scene_id = ?", oldSceneId).Update("scene_id", scene.SceneID)
if result.Error != nil {
log.Infof("Converting Actions for VirtualPorn Scene %s to %s failed, %s", oldSceneId, scene.SceneID, result.Error)
}
result = db.Model(&models.ExternalReferenceLink{}).Where("internal_table = 'scenes' and internal_name_id = ?", oldSceneId).Update("internal_name_id", scene.SceneID)
if result.Error != nil {
log.Infof("Converting External Reference Links for VirtualPorn Scene %s to %s failed, %s", oldSceneId, scene.SceneID, result.Error)
}
}
return true
})
offset += 100
if offset < sceneCnt {
apiCollector.Visit("https://site-api.project1service.com/v2/releases?type=scene&limit=24&offset=" + strconv.Itoa(offset))
}
})

collector.Visit("https://virtualporn.com/videos")

if sceneCnt > 0 {
return nil
} else {
return errors.New("No scenes updated")
}

}

0 comments on commit 219f19a

Please sign in to comment.