pcgamedb/crawler/steamrip.go
2024-12-22 00:37:00 +08:00

143 lines
3.7 KiB
Go

package crawler
import (
"bytes"
"errors"
"fmt"
"regexp"
"strings"
"pcgamedb/constant"
"pcgamedb/db"
"pcgamedb/model"
"pcgamedb/utils"
"github.com/PuerkitoBio/goquery"
"go.uber.org/zap"
)
type SteamRIPCrawler struct {
logger *zap.Logger
}
func NewSteamRIPCrawler(logger *zap.Logger) *SteamRIPCrawler {
return &SteamRIPCrawler{
logger: logger,
}
}
func (c *SteamRIPCrawler) Name() string {
return "SteamRIPCrawler"
}
func (c *SteamRIPCrawler) CrawlByUrl(URL string) (*model.GameItem, error) {
resp, err := utils.Request().Get(URL)
if err != nil {
return nil, err
}
doc, err := goquery.NewDocumentFromReader(bytes.NewReader(resp.Body()))
if err != nil {
return nil, err
}
item, err := db.GetGameItemByUrl(URL)
if err != nil {
return nil, err
}
item.RawName = strings.TrimSpace(doc.Find(".entry-title").First().Text())
item.Name = SteamRIPFormatter(item.RawName)
item.Url = URL
item.Author = "SteamRIP"
item.Platform = "windows"
sizeRegex := regexp.MustCompile(`(?i)<li><strong>Game Size:\s?</strong>(.*?)</li>`)
sizeRegexRes := sizeRegex.FindStringSubmatch(string(resp.Body()))
if len(sizeRegexRes) != 0 {
item.Size = strings.TrimSpace(sizeRegexRes[1])
} else {
item.Size = "unknown"
}
megadbRegex := regexp.MustCompile(`(?i)(?:https?:)?(//megadb\.net/[^"]+)`)
megadbRegexRes := megadbRegex.FindStringSubmatch(string(resp.Body()))
if len(megadbRegexRes) != 0 {
item.Download = fmt.Sprintf("https:%s", megadbRegexRes[1])
}
if item.Download == "" {
gofileRegex := regexp.MustCompile(`(?i)(?:https?:)?(//gofile\.io/d/[^"]+)`)
gofileRegexRes := gofileRegex.FindStringSubmatch(string(resp.Body()))
if len(gofileRegexRes) != 0 {
item.Download = fmt.Sprintf("https:%s", gofileRegexRes[1])
}
}
if item.Download == "" {
filecryptRegex := regexp.MustCompile(`(?i)(?:https?:)?(//filecrypt\.co/Container/[^"]+)`)
filecryptRegexRes := filecryptRegex.FindStringSubmatch(string(resp.Body()))
if len(filecryptRegexRes) != 0 {
item.Download = fmt.Sprintf("https:%s", filecryptRegexRes[1])
}
}
if item.Download == "" {
return nil, errors.New("failed to find download link")
}
return item, nil
}
func (c *SteamRIPCrawler) Crawl(num int) ([]*model.GameItem, error) {
count := 0
resp, err := utils.Request().Get(constant.SteamRIPGameListURL)
if err != nil {
return nil, err
}
doc, err := goquery.NewDocumentFromReader(bytes.NewReader(resp.Body()))
if err != nil {
return nil, err
}
var items []*model.GameItem
var urls []string
var updateFlags []string // title
doc.Find(".az-list-item>a").Each(func(i int, s *goquery.Selection) {
u, exist := s.Attr("href")
if !exist {
return
}
urls = append(urls, fmt.Sprintf("%s%s", constant.SteamRIPBaseURL, u))
updateFlags = append(updateFlags, s.Text())
})
for i, u := range urls {
if count == num {
break
}
if db.IsSteamRIPCrawled(updateFlags[i]) {
continue
}
c.logger.Info("Crawling", zap.String("URL", u))
item, err := c.CrawlByUrl(u)
if err != nil {
c.logger.Warn("Failed to crawl", zap.Error(err), zap.String("URL", u))
continue
}
item.UpdateFlag = updateFlags[i]
if err := db.SaveGameItem(item); err != nil {
c.logger.Warn("Failed to save item", zap.Error(err))
continue
}
items = append(items, item)
count++
if err := OrganizeGameItem(item); err != nil {
c.logger.Warn("Failed to organize", zap.Error(err), zap.String("URL", u))
continue
}
}
return items, nil
}
func (c *SteamRIPCrawler) CrawlAll() ([]*model.GameItem, error) {
return c.Crawl(-1)
}
func SteamRIPFormatter(name string) string {
name = regexp.MustCompile(`\([^\)]+\)`).ReplaceAllString(name, "")
name = strings.Replace(name, "Free Download", "", -1)
name = strings.TrimSpace(name)
return name
}