there we go

This commit is contained in:
TJ Horner
2019-03-13 12:16:40 -07:00
parent 2b8ca2a74d
commit 1c5856a243
5 changed files with 12 additions and 10 deletions

109
e621/download.go Normal file
View File

@ -0,0 +1,109 @@
package e621
import (
"fmt"
"io/ioutil"
"path"
"strconv"
"strings"
"sync"
"time"
"github.com/dustin/go-humanize"
)
// BeginDownload takes a slice of posts, a directory to save them in, and a
// number of concurrent workers to make. It blocks until all the post have
// been processed. It returns the number of successes, failures, and the total
// amount of posts.
func BeginDownload(posts *[]Post, saveDirectory *string, maxConcurrents *int) (*int, *int, *int) {
var wg sync.WaitGroup
var completed int
var successes int
var failures int
total := len(*posts)
// Distribute the posts based on the number of workers
ppw := len(*posts) / *maxConcurrents // ppw: posts per worker
mod := len(*posts) % *maxConcurrents // mod: remainder of posts
for i := 0; i < *maxConcurrents; i++ {
postsLower := i * ppw
postsUpper := i*ppw + ppw
if i == *maxConcurrents-1 {
// Give the last worker the remaining posts
// TODO: compensate it for labor
postsUpper += mod
}
wg.Add(1)
go work(i+1, (*posts)[postsLower:postsUpper], *saveDirectory, &completed, &successes, &failures, &total, &wg)
// Spawn workers with a little bit of a delay so as to not DDOS e621
// but also make the initial numbers show up correctly
time.Sleep(50 * time.Millisecond)
}
wg.Wait()
return &successes, &failures, &total
}
func work(wn int, posts []Post, directory string, completed *int, successes *int, failures *int, total *int, wg *sync.WaitGroup) {
defer wg.Done()
for _, post := range posts {
*completed++
fmt.Printf(
"[%d/%d] [w%d] Downloading post %d (%s) -> %s...\n",
*completed,
*total,
wn,
post.ID,
humanize.Bytes(uint64(post.FileSize)),
getSavePath(&post, &directory),
)
err := downloadPost(&post, directory)
if err != nil {
fmt.Printf("[w%d] Failed to download post %d: %v\n", wn, post.ID, err)
*failures++
} else {
*successes++
}
}
}
func getSavePath(post *Post, directory *string) string {
pathSliced := strings.Split(post.FileURL, ".")
extension := pathSliced[len(pathSliced)-1]
savePath := path.Join(*directory, strconv.Itoa(post.ID)+"."+extension)
return savePath
}
func downloadPost(post *Post, directory string) error {
savePath := getSavePath(post, &directory)
resp, err := HTTPGet(post.FileURL)
if err != nil {
return err
}
defer resp.Body.Close()
body, err := ioutil.ReadAll(resp.Body)
if err != nil {
return err
}
err = ioutil.WriteFile(savePath, body, 0755)
if err != nil {
return err
}
return nil
}

87
e621/e621.go Normal file
View File

@ -0,0 +1,87 @@
package e621
import (
"encoding/json"
"io/ioutil"
"net/http"
"strconv"
)
// Post represents an e621 post object returned by the e621 API.
type Post struct {
ID int `json:"id"`
Tags string `json:"tags"`
LockedTags bool `json:"locked_tags"`
Description string `json:"description"`
CreatedAt SerializedDate `json:"created_at"`
CreatorID int `json:"creator_id"`
Author string `json:"author"`
Change int `json:"change"`
Source string `json:"source"`
Score int `json:"score"`
FavoritesCount int `json:"fav_count"`
MD5Hash string `json:"md5"`
FileSize int `json:"file_size"`
FileURL string `json:"file_url"`
FileExt string `json:"file_ext"`
PreviewURL string `json:"preview_url"`
PreviewHeight int `json:"preview_height"`
PreviewWidth int `json:"preview_width"`
Rating string `json:"rating"`
Status string `json:"status"`
Width int `json:"width"`
Height int `json:"height"`
HasComments bool `json:"has_comments"`
HasNotes bool `json:"has_notes"`
HasChildren bool `json:"has_children"`
Children string `json:"children"`
ParentID int `json:"parent_id"`
Artist []string `json:"artist"`
Sources []string `json:"sources"`
}
// SerializedDate represents a serialized date passed via JSON
type SerializedDate struct {
JSONClass string `json:"json_class"`
Seconds int `json:"s"`
Nanoseconds int `json:"n"`
}
// GetPostsForTags gets a list of e621 Posts
func GetPostsForTags(tags string, limit int, sfw bool) ([]Post, error) {
client := &http.Client{}
var domain string
if sfw {
domain = "e926.net"
} else {
domain = "e621.net"
}
req, _ := http.NewRequest("GET", "https://"+domain+"/post/index.json", nil)
req.Header.Set("User-Agent", "e6dl: go edition (@tjhorner on Telegram)")
qs := req.URL.Query()
qs.Add("tags", tags)
qs.Add("limit", strconv.Itoa(limit))
req.URL.RawQuery = qs.Encode()
resp, err := client.Do(req)
if err != nil {
return nil, err
}
defer resp.Body.Close()
body, err := ioutil.ReadAll(resp.Body)
if err != nil {
return nil, err
}
var posts []Post
json.Unmarshal(body, &posts)
return posts, nil
}

16
e621/util.go Normal file
View File

@ -0,0 +1,16 @@
package e621
import (
"net/http"
)
// HTTPGet is a helper function that automatically adds the
// tool's UA to an HTTP GET request
func HTTPGet(url string) (*http.Response, error) {
client := &http.Client{}
req, _ := http.NewRequest("GET", url, nil)
req.Header.Set("User-Agent", "e6dl: go edition (@tjhorner on Telegram)")
return client.Do(req)
}