Crate web_parser

   

Web Parser + Search

This web page parser library allows asynchronous fetching and extracting of data from web pages in multiple formats.

• Asynchronous web search using the search engines [Google, Bing, Duck, Ecosia, Yahoo, Wiki] with domain blacklisting (feature search).
• You can also create a custom search engine by using the SearchEngine trait (feature search).
• Reading an HTML document from a URL with a randomized user-agent (User::random()).
• Selecting elements by CSS selectors and retrieving their attributes and content.
• Fetching the full page as plain text.
• Fetching and parsing page content as JSON with serde_json support.

This tool is well-suited for web scraping and data extraction tasks, offering flexible parsing of HTML, plain text, and JSON to enable comprehensive data gathering from various web sources.

Examples:

Web Search (feature: ‘search’):

Requires the chromedriver tool installed!

use web_parser::prelude::*;

#[tokio::main]
async fn main() -> Result<()> {
    // select search engine:
    let mut engine = SearchEngine::<Duck>::new(
        Some("bin/chromedriver/chromedriver.exe"),  // path to chromedriver (None = to use global PATH)
        Some(macron::path!("$/WebSearch/Profile1").to_str().unwrap()),  // path to save Chrome session
        false,  // run in headless mode (without browser interface)
    ).await?;

    println!("Searching results..");

    // send search query:
    let results = engine.search(
        "Rust (programming language)",  // search query
        &["support.google.com", "youtube.com"],  // black list
        1000  // sleep in millis
    ).await;

    // handle search results:
    match results {
        Ok(cites) => {
            println!("Reading result pages..");

            let contents = cites.read(5, &[
                "header", "footer", "style", "script", "noscript",
                "iframe", "button", "img", "svg"
            ]).await?;

            println!("Results: {contents:#?}");
        }
        Err(e) => eprintln!("Search error: {e}")
    }

    // stop search engine:
    engine.stop().await?;
    Ok(())
}

Web Parsing:

use web_parser::prelude::*;

#[tokio::main]
async fn main() -> Result<()> {
    // _____ READ PAGE AS HTML DOCUMENT: _____
    
    // read website page:
    let mut doc = Document::read("https://example.com/", User::random()).await?;

    // select title:
    let title = doc.select("h1")?.expect("No elements found");
    println!("Title: '{}'", title.text());

    // select descriptions:
    let mut descrs = doc.select_all("p")?.expect("No elements found");
    
    while let Some(descr) = descrs.next() {
        println!("Description: '{}'", descr.text())
    }


    // _____ READ PAGE AS PLAIN TEXT: _______

    let text: String = Document::text("https://example.com/", User::random()).await?;
    println!("Text: {text}");


    // _____ READ PAGE AS JSON: ______________

    let json: serde_json::Value = Document::json("https://example.com/", User::random()).await?.expect("Failed to parse JSON");
    println!("Json: {json}");
    

    Ok(())
}

Licensing:

Distributed under the MIT license.

Feedback:

You can find me here, also see my channel. I welcome your suggestions and feedback!

Copyright (c) 2025 Bulat Sh. (fuderis)

Re-exports

pub use error::Result;

pub use error::Error;

pub use document::User;

pub use document::Document;

pub use document::Node;

pub use document::Nodes;

Modules

document

error

prelude