package pages import ( "log" "net/http" "os" "strings" "time" "gitler.moe/suwako/gitlin/utils" "github.com/gocolly/colly" "github.com/gofiber/fiber/v2" ) type Dir struct { Readme string Username string Reponame string DirName string Branch string } type DirFiles struct { Name string Path string Commit string Date string Type string Branch string Username string Reponame string DirName string } func DirView(c *fiber.Ctx) error { var dirArray []Dir var dirFilesArray []DirFiles var readmeOutput string resp, statusErr := http.Get("https://github.com/" + c.Params("user") + "/" + c.Params("repo") + "/tree/" + c.Params("branch") + "/" + c.Params("+")) if statusErr != nil { log.Println(statusErr) } if resp.StatusCode == 404 { // I need a better way to do this return c.Status(404).Render("error", fiber.Map{ "title": "Error", "ver": utils.Ver, "ves": utils.Ves, "error": "Directory" + c.Params("+") + "not found", }) } // Scraping Scrape := Dir{} Scrape.Username = c.Params("user") Scrape.Reponame = c.Params("repo") Scrape.DirName = c.Params("+") Scrape.Branch = c.Params("branch") UserAgent, ok := os.LookupEnv("GITLIN_USER_AGENT") if !ok { UserAgent = "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/110.0.0.0 Safari/537.36" } sc := colly.NewCollector(colly.AllowedDomains("github.com"), colly.UserAgent(UserAgent)) sc.Limit(&colly.LimitRule{ DomainGlob: "github.githubassets.com/*", Delay: 15 * time.Second, RandomDelay: 15 * time.Second, }) sc.OnHTML("div#readme", func(e *colly.HTMLElement) { Scrape.Readme = e.ChildText("a[href='#readme']") }) sc.OnHTML("div#readme div.Box-body", func(e *colly.HTMLElement) { Content, _ := e.DOM.Html() readmeOutput = strings.Replace(strings.Replace(strings.Replace(strings.Replace(string(utils.UGCPolicy().SanitizeBytes([]byte(Content))), "https://github.com", "", -1), "user-content-", "", -1), "https://camo.githubusercontent.com", "/camo", -1), "https://raw.githubusercontent.com", "/raw", -1) }) sc.OnHTML("div.js-details-container div.Details-content--hidden-not-important", func(e *colly.HTMLElement) { e.ForEach("div.js-navigation-item", func(i int, el *colly.HTMLElement) { var FileType string if el.ChildAttr("div.flex-shrink-0 svg", "aria-label") == "Directory" { FileType = "dir" } else { FileType = "file" } tstring := el.ChildAttr("relative-time", "datetime") if tstring != "" { tstring = strings.ReplaceAll(tstring, "T", "、") tstring = strings.ReplaceAll(tstring, "Z", "") tstring = strings.Split(tstring, "+")[0] tstrings := strings.Split(tstring, "-") tstring = tstrings[0] + "-" + tstrings[1] + "-" + tstrings[2] } if el.ChildText("div.flex-auto span.d-block a.js-navigation-open") != "" { dirFilesArray = append(dirFilesArray, DirFiles{ Name: el.ChildText("div.flex-auto span.d-block a.js-navigation-open"), Path: el.ChildText("div.flex-auto span.d-block a.js-navigation-open"), Commit: el.ChildText("div.flex-auto span.d-block a.Link--secondary"), Date: tstring, Type: FileType, Username: Scrape.Username, Reponame: Scrape.Reponame, DirName: Scrape.DirName, Branch: Scrape.Branch, }) } }) }) sc.Visit("https://github.com/" + c.Params("user") + "/" + c.Params("repo") + "/tree/" + c.Params("branch") + "/" + c.Params("+")) // Add scrape-based info to dirArray dirArray = append(dirArray, Scrape) return c.Render("dir", fiber.Map{ "title": c.Params("+") + " フォルダ | " + c.Params("user") + "/" + c.Params("repo"), "dir": dirArray, "ver": utils.Ver, "ves": utils.Ves, "files": dirFilesArray, "readme": string(readmeOutput), }) }