Created
April 4, 2020 03:02
-
-
Save ronanyeah/c515615ddbf81c70aca17fceb3ca4d06 to your computer and use it in GitHub Desktop.
YouTube free movies scraper
This file contains 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
use scraper::{Html, Selector}; | |
use std::fs::File; | |
use std::io::prelude::*; | |
const USER_AGENT: &str = "Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/80.0.3987.163 Safari/537.36"; | |
const URL: &str = "https://www.youtube.com/feed/storefront?bp=kgEmCGQSIlBMSFBUeFR4dEMwaWJWWnJUMl9XS1dVbDJTQXhzS3VLd3iiBQIoAg%3D%3D"; | |
#[derive(serde::Deserialize, serde::Serialize)] | |
struct Movie { | |
id: String, | |
img: String, | |
title: String, | |
} | |
async fn fetch(client: reqwest::Client) -> Result<String, reqwest::Error> { | |
let res = client | |
.get(URL) | |
.header("User-Agent", USER_AGENT) | |
.send() | |
.await; | |
match res { | |
Err(err) => Err(err), | |
Ok(data) => data.text().await, | |
} | |
} | |
#[tokio::main] | |
async fn main() -> Result<(), Box<dyn std::error::Error>> { | |
let client = reqwest::Client::new(); | |
let txt = fetch(client).await?; | |
let selector = Selector::parse("script").expect("bad selector"); | |
let document = Html::parse_document(&txt); | |
let sel = document.select(&selector); | |
let res = sel | |
.map(|element| element.text().collect::<Vec<_>>()) | |
.find(|text| text.iter().any(|x| x.contains(&"ytInitialData"))) | |
.expect("no yt"); | |
let mut ok = res | |
.get(0) | |
.expect("bad index") | |
.split("\n") | |
.collect::<Vec<_>>() | |
.iter() | |
.filter(|x| !x.is_empty()) | |
.collect::<Vec<_>>() | |
.get(0) | |
.expect("bad index 2") | |
.chars() | |
.skip(30) | |
.collect::<String>(); | |
ok.pop(); | |
let json: serde_json::Value = serde_json::from_str(&ok)?; | |
let inner = &json.as_object().expect("1")["contents"] | |
.as_object() | |
.expect("2")["twoColumnBrowseResultsRenderer"] | |
.as_object() | |
.expect("3")["tabs"][0] | |
.as_object() | |
.expect("4")["tabRenderer"] | |
.as_object() | |
.expect("5")["content"] | |
.as_object() | |
.expect("6")["sectionListRenderer"] | |
.as_object() | |
.expect("7")["contents"][0] | |
.as_object() | |
.expect("6")["itemSectionRenderer"] | |
.as_object() | |
.expect("7")["contents"][0] | |
.as_object() | |
.expect("8")["shelfRenderer"] | |
.as_object() | |
.expect("9")["content"] | |
.as_object() | |
.expect("10")["gridRenderer"] | |
.as_object() | |
.expect("11")["items"] | |
.as_array() | |
.unwrap() | |
.iter() | |
.map(|item| { | |
let obj = item.as_object().unwrap()["gridMovieRenderer"] | |
.as_object() | |
.unwrap(); | |
Movie { | |
id: obj["videoId"].as_str().unwrap().to_string(), | |
img: obj["thumbnail"].as_object().unwrap()["thumbnails"][0] | |
.as_object() | |
.unwrap()["url"] | |
.as_str() | |
.unwrap() | |
.to_string(), | |
title: obj["title"].as_object().unwrap()["runs"][0] | |
.as_object() | |
.unwrap()["text"] | |
.as_str() | |
.unwrap() | |
.to_string(), | |
} | |
}) | |
.collect::<Vec<_>>(); | |
let mut file = File::create("./res.json")?; | |
file.write_all(serde_json::to_string(inner)?.as_bytes())?; | |
Ok(()) | |
} |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment