1 unstable release

0.1.16 Aug 25, 2022
0.1.15 Aug 24, 2022

#6 in #crawl

38 downloads per month

MIT/Apache

21KB
447 lines

gar-crawl

A high level HTML crawler with a concise builder.

The goal of this crate is to accomplish crawling and scraping tasks with minimal boilerplate. Default propagators are provided or you can make your own, and you can modify the Reqwest client used before building a crawler.

examples

Basic usage with default options ( crawl depth: 2, workers: 40, revisit: false )

Crawler::builder()
    .add_default_propagators()                         // crawl to href and src links
    .add_handler("*[href]", |args| {                   // add handler
        if let Some(href) = args.element.unwrap().value().attr("href") {
            println!("{href}");
        }
    })
    .build()?                                          // construct crawler
    .crawl("https://example.org")                      // begin crawl
    .await?;

Example with more features used

Crawler::builder()
    .add_default_propagators()                         // crawl to href and src links
    .revisit(true)                                     // default false
    .whitelist("https://example.org")                  // stay on this site
    .user_agent("Mozilla/5.0 (X11; Linux x86_64)...")  // set user agent
    .proxy("127.0.0.1:8080", "/path/to/cacert.der")?   // set up https proxy
    .add_handler("*[href]", |args| {                   // add handler
        if let Some(href) = args.element.unwrap().value().attr("href") {
            println!("{href}");
        }
    })
    .on_page(|args| {
        // do stuff with page
    })
    .depth(3)                                          // default 2
    .workers(100)                                      // default 40
    .timeout(5, 0)                                     // timeout requests after 5 seconds
    .build()?                                          // construct crawler
    .crawl("https://example.org")                      // begin crawl
    .await?;

See examples/ or gar-crawl-cli/ for more examples

Dependencies

~9–23MB
~353K SLoC