Created
October 5, 2018 09:32
-
-
Save hjhee/13bed41e342e66f236f9a6f390491033 to your computer and use it in GitHub Desktop.
A Image Crawler for Steam
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
package main | |
import ( | |
"bytes" | |
"fmt" | |
"io/ioutil" | |
"net/http" | |
"net/http/httputil" | |
"net/url" | |
"os" | |
"regexp" | |
"github.com/PuerkitoBio/goquery" | |
) | |
var demoURL = `https://steamcommunity.com/app/49520/homecontent/?userreviewsoffset=0&p=10&workshopitemspage=10&readytouseitemspage=10&mtxitemspage=10&itemspage=10&screenshotspage=10&videospage=10&artpage=10&allguidepage=10&webguidepage=10&integratedguidepage=10&discussionspage=10&numperpage=10&browsefilter=toprated&browsefilter=toprated&appid=49520&appHubSubSection=4&appHubSubSection=4&l=schinese&filterLanguage=default&searchText=moxxi` | |
var parameters = []string{"p", | |
"workshopitemspage", | |
"readytouseitemspage", | |
"mtxitemspage", | |
"itemspage", | |
"screenshotspage", | |
"videospage", | |
"artpage", | |
"allguidepage", | |
"webguidepage", | |
"integratedguidepage", | |
"discussionspage", | |
} | |
var re *regexp.Regexp | |
func parse(res *http.Response) error { | |
defer res.Body.Close() | |
body, err := ioutil.ReadAll(res.Body) | |
if err != nil { | |
return err | |
} | |
if len(body) == 0 { | |
fmt.Printf("empty body") | |
dump, err := httputil.DumpResponse(res, false) | |
if err == nil { | |
fmt.Printf("dump:\n%q\n", dump) | |
} | |
return fmt.Errorf("ContentLength is 0") | |
} | |
doc, err := goquery.NewDocumentFromReader(bytes.NewReader(body)) | |
if err != nil { | |
return err | |
} | |
doc.Find(".apphub_CardContentMain img").Each(func(i int, s *goquery.Selection) { | |
src, e := s.Attr("src") | |
if e == false { | |
fmt.Printf("error finding image #%d: attribute \".apphub_CardContentMain img\" does not exists", i) | |
return | |
} | |
q, err := url.Parse(src) | |
if err != nil { | |
fmt.Printf("error parsing url (%s)#%d: %s", src, i, err.Error()) | |
return | |
} | |
q.RawQuery = "" | |
fmt.Printf("img#%d: %s\n", i, q) | |
matches := re.FindStringSubmatch(q.String()) | |
filename := "img/" + matches[1] | |
r, _ := http.Get(q.String()) | |
b, _ := ioutil.ReadAll(r.Body) | |
r.Body.Close() | |
switch r.Header.Get("Content-Type") { | |
case "image/jpeg": | |
filename = filename + ".jpg" | |
case "image/png": | |
filename = filename + ".png" | |
case "image/gif": | |
filename = filename + ".gif" | |
} | |
fmt.Printf("img#%d: writing %s\n", i, filename) | |
f, _ := os.Create(filename) | |
defer f.Close() | |
f.Write(b) | |
f.Sync() | |
}) | |
return nil | |
} | |
func main() { | |
re = regexp.MustCompile(`\/(\w+)\/?$`) | |
u, _ := url.Parse(demoURL) | |
q := u.Query() | |
for i := 1; true; i++ { | |
p := fmt.Sprintf("%d", i) | |
for _, k := range parameters { | |
q.Set(k, p) | |
} | |
u.RawQuery = q.Encode() | |
fmt.Printf("#%d url: %s\n", i, u) | |
res, _ := http.Get(u.String()) | |
err := parse(res) | |
if err != nil { | |
break | |
} | |
} | |
} |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment