Download videos by hashtag; get json data without video downloading; limit option

This commit is contained in:
alexpin 2020-02-25 00:56:19 +02:00
parent 1b3f985f42
commit f724f0f2a2
13 changed files with 308 additions and 165 deletions

View File

@ -22,6 +22,8 @@ Clone this repository and run `go build` to build the executable.
* `-batch-file` - File containing URLs/Usernames to download, one value per line. Lines starting with '#', are considered as comments and ignored.
* `-deadline` - Sets the timout for scraper logic in seconds (used as a workaround for context deadline exceeded error) (default 1500)
* `-quiet` - Supress output
* `-json` - Returns whole data, that was scraped from TikTok, in json
* `-limit` - Sets the max count of video that will be downloaded (default infinity)
## Acknowledgments
This software uses the **chromedp** for web scraping, it can be found here: https://github.com/chromedp/chromedp \

View File

@ -65,6 +65,7 @@ func runScrapeWithInfo(ctx context.Context, jsAction string, url string) string
// Navigate to user's page
chromedp.Navigate(url),
// Execute url grabber script
chromedp.WaitReady("video"),
chromedp.EvaluateAsDevTools(utils.ReadFileAsString("scraper.js"), &jsOutput),
chromedp.EvaluateAsDevTools(jsAction, &jsOutput),
)
@ -74,7 +75,7 @@ func runScrapeWithInfo(ctx context.Context, jsAction string, url string) string
err = chromedp.Run(ctx, chromedp.EvaluateAsDevTools("currentState.preloadCount.toString()", &jsOutput))
utils.CheckErr(err)
if jsOutput != "0" {
utils.Logf("\rPreloading... Currently loaded %s items.", jsOutput)
utils.Logf("\rPreloading... %s items have been founded.", jsOutput)
} else {
utils.Logf("\rPreloading...")
}

View File

@ -0,0 +1,19 @@
package client
import (
models "../models"
config "../models/config"
"fmt"
)
// GetUserUploads - Get all uploads marked with given hashtag
func GetHashtagUploads(hashtagURL string) []models.Upload {
jsMethod := fmt.Sprintf("bootstrapIteratingVideos(%d)", config.Config.Limit)
actionOutput := executeClientAction(hashtagURL, jsMethod)
return models.ParseUploads(actionOutput)
}
func GetHashtagUploadsJson(hashtagURL string) string {
jsMethod := fmt.Sprintf("bootstrapIteratingVideos(%d)", config.Config.Limit)
return executeClientAction(hashtagURL, jsMethod)
}

View File

@ -2,10 +2,18 @@ package client
import (
models "../models"
config "../models/config"
"fmt"
)
// GetMusicUploads - Get all uploads by given music
func GetMusicUploads(url string) []models.Upload {
actionOutput := executeClientAction(url, "bootstrapIteratingVideos()")
jsMethod := fmt.Sprintf("bootstrapIteratingVideos(%d)", config.Config.Limit)
actionOutput := executeClientAction(url, jsMethod)
return models.ParseUploads(actionOutput)
}
func GetMusicUploadsJson(url string) string {
jsMethod := fmt.Sprintf("bootstrapIteratingVideos(%d)", config.Config.Limit)
return executeClientAction(url, jsMethod)
}

View File

@ -2,10 +2,18 @@ package client
import (
models "../models"
config "../models/config"
"fmt"
)
// GetUserUploads - Get all uploads by user
func GetUserUploads(username string) []models.Upload {
actionOutput := executeClientAction(`https://www.tiktok.com/@`+username, "bootstrapIteratingVideos()")
jsMethod := fmt.Sprintf("bootstrapIteratingVideos(%d)", config.Config.Limit)
actionOutput := executeClientAction(`https://www.tiktok.com/@`+username, jsMethod)
return models.ParseUploads(actionOutput)
}
func GetUserUploadsJson(username string) string {
jsMethod := fmt.Sprintf("bootstrapIteratingVideos(%d)", config.Config.Limit)
return executeClientAction(`https://www.tiktok.com/@`+username, jsMethod)
}

View File

@ -15,6 +15,8 @@ var Config struct {
MetaData bool
Quiet bool
Deadline int
Limit int
JSONOnly bool
}
// GetConfig - Returns Config object
@ -25,6 +27,8 @@ func GetConfig() {
metadata := flag.Bool("metadata", false, "Write video metadata to a .json file")
quiet := flag.Bool("quiet", false, "Supress output")
deadline := flag.Int("deadline", 1500, "Sets the timout for scraper logic in seconds (used as a workaround for 'context deadline exceeded' error)")
limit := flag.Int("limit", 0, "Sets the videos count limit (useful when there too many videos from the user or by hashtag)")
jsonOnly := flag.Bool("json", false, "Just get JSON data from scraper (without video downloading)")
flag.Parse()
args := flag.Args()
@ -44,5 +48,10 @@ func GetConfig() {
Config.Debug = *debug
Config.MetaData = *metadata
Config.Quiet = *quiet
Config.Deadline = *deadline
if *jsonOnly {
Config.Quiet = true
}
Config.Deadline = *deadline
Config.Limit = *limit
Config.JSONOnly = *jsonOnly;
}

View File

@ -1,7 +1,7 @@
optStrings = {
selectors: {
feedLoading: 'div.tiktok-loading.feed-loading',
modalArrowLeft: 'div.video-card-modal > div > img.arrow-right',
modalArrowRight: 'div.video-card-modal > div > img.arrow-right',
modalClose: '.video-card-modal > div > div.close',
modalPlayer: 'div > div > main > div.video-card-modal > div > div.video-card-big > div.video-card-container > div > div > video',
modalShareInput: '.copy-link-container > input',
@ -30,6 +30,7 @@ optStrings = {
currentState = {
preloadCount: 0,
finished: false,
limit: 100
};
createVidUrlElement = function(outputObj) {
@ -37,7 +38,7 @@ createVidUrlElement = function(outputObj) {
urlSetElement.innerText = JSON.stringify(outputObj);
document.getElementsByTagName(optStrings.tags.resultParentTag)[0].appendChild(urlSetElement);
currentState.finished = true;
}
};
buldVidUrlArray = function(finishCallback) {
var feedItem = document.getElementsByClassName(optStrings.classes.feedVideoItem)[0];
@ -46,8 +47,14 @@ buldVidUrlArray = function(finishCallback) {
var videoArray = [];
var intervalID = window.setInterval(x => {
videoArray.push(getCurrentModalVideo());
var arrowRight = document.querySelectorAll(optStrings.selectors.modalArrowLeft)[0];
if(currentState.limit > 0) {
if (videoArray.length >= currentState.limit) {
window.clearInterval(intervalID);
document.querySelector(optStrings.selectors.modalClose).click();
finishCallback(videoArray);
}
}
var arrowRight = document.querySelectorAll(optStrings.selectors.modalArrowRight)[0];
if (arrowRight.classList.contains(optStrings.classes.modalCloseDisabled)) {
window.clearInterval(intervalID);
document.querySelector(optStrings.selectors.modalClose).click();
@ -78,7 +85,7 @@ getCurrentModalVideo = function() {
link: soundHref,
},
};
}
};
getCurrentVideo = function() {
var player = document.querySelector(optStrings.selectors.videoPlayer);
@ -100,13 +107,19 @@ getCurrentVideo = function() {
link: soundHref,
},
};
}
};
scrollWhileNew = function(finishCallback) {
var state = { count: 0 };
var intervalID = window.setInterval(x => {
var oldCount = state.count;
state.count = document.getElementsByClassName(optStrings.classes.feedVideoItem).length;
if(currentState.limit > 0) {
if (currentState.preloadCount >= currentState.limit || state.count >= currentState.limit) {
finishCallback(createVidUrlElement);
window.clearInterval(intervalID);
}
}
if (oldCount !== state.count) {
currentState.preloadCount = state.count;
window.scrollTo(0, document.body.scrollHeight);
@ -121,7 +134,8 @@ scrollWhileNew = function(finishCallback) {
}, 1000);
};
bootstrapIteratingVideos = function() {
bootstrapIteratingVideos = function(limit) {
currentState.limit = limit;
scrollWhileNew(buldVidUrlArray);
return 'bootstrapIteratingVideos';
};
@ -130,7 +144,7 @@ bootstrapGetCurrentVideo = function() {
var video = getCurrentVideo();
createVidUrlElement(video);
return 'bootstrapGetCurrentVideo';
}
};
init = () => {
const newProto = navigator.__proto__;

16
utils/getHashtag.go Normal file
View File

@ -0,0 +1,16 @@
package utils
import (
res "../resources"
"fmt"
"strings"
)
// GetHashtagFromURL - Get's tag name from passed url
func GetHashtagFromURL(str string) string {
if match := strings.Contains(str, "/tag/"); match {
return strings.Split(str, "/tag/")[1]
}
panic(fmt.Sprintf(res.ErrorCouldNotRecogniseURL, str))
}

View File

@ -0,0 +1,36 @@
package workflows
import (
client "../client"
config "../models/config"
utils "../utils"
"fmt"
"strings"
)
// CanUseDownloadHashtag - Test's if this workflow can be used for parameter
func CanUseDownloadHashtag(url string) bool {
match := strings.Contains(url, "/tag/")
return match
}
// DownloadHashtag - Download videos marked with given hashtag
func DownloadHashtag(url string) {
uploads := client.GetHashtagUploads(url)
uploadCount := len(uploads)
hashtag := utils.GetHashtagFromURL(url)
downloadDir := fmt.Sprintf("%s/%s", config.Config.OutputPath, hashtag)
utils.InitOutputDirectory(downloadDir)
for index, upload := range uploads {
downloadVideo(upload, downloadDir)
utils.Logf("\r[%d/%d] Downloaded", index+1, uploadCount)
}
utils.Log()
}
func GetHashtagJson(url string) {
uploads := client.GetHashtagUploads(url)
fmt.Printf("%s", uploads)
}

View File

@ -29,3 +29,8 @@ func DownloadMusic(url string) {
}
utils.Log()
}
func GetMusicJson(url string) {
uploads := client.GetMusicUploadsJson(url)
fmt.Printf("%s", uploads)
}

View File

@ -28,3 +28,8 @@ func DownloadUser(username string) {
}
utils.Log()
}
func GetUserVideosJson(username string) {
uploads := client.GetUserUploadsJson(username)
fmt.Printf("%s", uploads)
}

View File

@ -1,6 +1,7 @@
package workflows
import (
config "../models/config"
res "../resources"
utils "../utils"
)
@ -10,7 +11,11 @@ func StartWorkflowByParameter(url string) {
// Music
if CanUseDownloadMusic(url) {
if config.Config.JSONOnly {
GetMusicJson(url)
} else {
DownloadMusic(url)
}
return
}
@ -22,7 +27,22 @@ func StartWorkflowByParameter(url string) {
// Tiktok user
if CanUseDownloadUser(url) {
if config.Config.JSONOnly {
GetUserVideosJson(utils.GetUsernameFromString(url))
} else {
DownloadUser(utils.GetUsernameFromString(url))
}
return
}
// Tiktok hashtag
if CanUseDownloadHashtag(url) {
if config.Config.JSONOnly {
GetHashtagJson(url)
} else {
DownloadHashtag(url)
}
return
}