bing-image-scrapper 0.1.0

use regex::Regex;
use reqwest::header::{HeaderMap, USER_AGENT};

pub async fn scrape_bing_images(query: &str) -> Result<Vec<String>, Box<dyn std::error::Error>> {
    let url = format!("https://www.bing.com/images/search?q={}", query);
    let client = create_client()?;
    let html = client.get(url).send().await?.text().await?;
    let image_urls = extract_image_urls(&html)?;
    Ok(image_urls)
}

fn create_client() -> Result<reqwest::Client, Box<dyn std::error::Error>> {
    let mut headers = HeaderMap::new();
    headers.insert(
        USER_AGENT,
        "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36"
            .parse()?,
    );

    Ok(reqwest::Client::builder()
        .default_headers(headers)
        .build()?)
}

fn extract_image_urls(html: &str) -> Result<Vec<String>, Box<dyn std::error::Error>> {
    let re = Regex::new(r#"murl&quot;:&quot;(https?://[^&]+?)&quot;"#)?;
    let mut image_urls = Vec::new();

    for cap in re.captures_iter(html) {
        if let Some(murl) = cap.get(1) {
            image_urls.push(murl.as_str().to_string());
        }
    }

    Ok(image_urls)
}