Lib.rs

reqwest-scraper - 使用 reqwest 进行网页抓取集成

扩展 reqwest 以支持多种网页抓取方法。

特性

• 使用 JsonPath 在 json 响应中选择字段
• 使用 CSS 选择器 在 HTML 响应中选择元素
• 使用 xpath 表达式 评估 HTML 响应中的值
• 宏提取

入门指南

• 添加依赖

  reqwest = { version = "0.12", features = ["json"] }
  reqwest-scraper="0.3.2"
  

• 使用 ScraperResponse

  use reqwest_scraper::ScraperResponse;
  

JsonPath

• Json::选择<T:DeserializeOwned>(路径: &str) -> 结果<Vec<T>>
• Json::选择一个<T:DeserializeOwned>(路径: &str) -> 结果<T>
• Json::选择为字符串(路径: &str) -> 结果<String>

示例:

use reqwest_scraper::ScraperResponse;

pub async fn request() -> Result<()> {
    let json = reqwest::Client::builder()
        .build()?
        .get("https://api.github.com/search/repositories?q=rust")
        .header("User-Agent", "Rust Reqwest")
        .send()
        .await?
        .jsonpath()
        .await?;

    let total_count = json.select_as_str("$.total_count")?;
    let names: Vec<String> = json.select("$.items[*].full_name")?;

    println!("{}", total_count);
    println!("{}", names.join("\t"));

    Ok(())
}

CSS 选择器

• Html::选择(选择器: &str) -> 结果<Selectable>
• Selectable::迭代() ->实现迭代器<SelectItem>
• Selectable::第一个() -> Option<SelectItem>
• SelectItem::名称() -> &str
• SelectItem::id() -> Option<&str>
• SelectItem::有类(类: &str,大小写敏感:CaseSensitivity) -> bool
• SelectItem::类() ->Classes
• SelectItem::属性() ->Attrs
• SelectItem::属性(属性: &str) -> Option<&str>
• SelectItem::文本() ->String
• SelectItem::html() ->String
• SelectItem::内部_html() ->String
• SelectItem::子元素() ->实现迭代器<SelectItem>
• SelectItem::查找(选择器: &str) -> 结果<Selectable>

示例:

use reqwest_scraper::ScraperResponse;

async fn request() -> Result<()> {
    let html = reqwest::get("https://github.com/holmofy")
        .await?
        .css_selector()
        .await?;

    assert_eq!(
        html.select(".p-name")?.iter().nth(0).unwrap().text().trim(),
        "holmofy"
    );

    let select_result = html.select(".vcard-details > li.vcard-detail")?;

    for detail_item in select_result.iter() {
        println!("{}", detail_item.attr("aria-label").unwrap())
    }

    Ok(())
}

XPath

• XHtml::选择(xpath: &str) -> 结果<XPathResult>
• XPathResult::as_nodes() -> Vec<Node>
• XPathResult::as_strs() -> Vec<String>
• XPathResult::as_node() -> Option<Node>
• XPathResult::as_str() -> Option<String>
• Node::名称() ->String
• Node::id() -> Option<String>
• Node::类() -> HashSet<String>
• Node::属性(属性: &str) -> Option<String>
• Node::有属性(属性: &str) -> bool
• Node::文本() ->String
• TODO: Node::html() -> String
• TODO: Node::inner_html() -> String
• Node::子元素() -> Vec<Node>
• Node::findnodes(relative_xpath: &str) -> 结果<Vec<Node>>
• Node::findvalues(relative_xpath: &str) -> 结果<Vec<String>>
• Node::findnode(relative_xpath: &str) -> 结果<Option<Node>>
• Node::findvalue(relative_xpath: &str) -> 结果<Option<String>>

示例:

async fn request() -> Result<()> {
    let html = reqwest::get("https://github.com/holmofy")
        .await?
        .xpath()
        .await?;

    // simple extract element
    let name = html
        .select("//span[contains(@class,'p-name')]")?
        .as_node()
        .unwrap()
        .text();
    println!("{}", name);
    assert_eq!(name.trim(), "holmofy");

    // iterate elements
    let select_result = html
        .select("//ul[contains(@class,'vcard-details')]/li[contains(@class,'vcard-detail')]")?
        .as_nodes();

    println!("{}", select_result.len());

    for item in select_result.into_iter() {
        let attr = item.attr("aria-label").unwrap_or_else(|| "".into());
        println!("{}", attr);
        println!("{}", item.text());
    }

    // attribute extract
    let select_result = html
        .select("//ul[contains(@class,'vcard-details')]/li[contains(@class,'vcard-detail')]/@aria-label")?
        .as_strs();

    println!("{}", select_result.len());
    select_result.into_iter().for_each(|s| println!("{}", s));

    Ok(())
}

宏提取

使用 FromCssSelector & selector 将 html 元素提取到结构中

// define struct and derive the FromCssSelector trait
#[derive(Debug, FromCssSelector)]
#[selector(path = "#user-repositories-list > ul > li")]
struct Repo {
    #[selector(path = "a[itemprop~='name']", default = "<unname>", text)]
    name: String,

    #[selector(path = "span[itemprop~='programmingLanguage']", text)]
    program_lang: Option<String>,

    #[selector(path = "div.topics-row-container>a", text)]
    topics: Vec<String>,
}

// request
let html = reqwest::get("https://github.com/holmofy?tab=repositories")
    .await?
    .css_selector()
    .await?;

// Use the generated `from_html` method to extract data into the struct
let items = Repo::from_html(html)?;
items.iter().for_each(|item| println!("{:?}", item));

使用 FromXPath & xpath 将 html 元素提取到结构中

// define struct and derive the FromXPath trait
#[derive(Debug, FromXPath)]
#[xpath(path = "//div[@id='user-repositories-list']/ul/li")]
struct Repo {
    #[xpath(path = ".//a[contains(@itemprop,'name')]/text()", default = "<unname>")]
    name: String,

    #[xpath(path = ".//span[contains(@itemprop,'programmingLanguage')]/text()")]
    program_lang: Option<String>,

    #[xpath(path = ".//div[contains(@class,'topics-row-container')]/a/text()")]
    topics: Vec<String>,
}

let html = reqwest::get("https://github.com/holmofy?tab=repositories")
    .await?
    .xpath()
    .await?;

// Use the generated `from_xhtml` method to extract data into the struct
let items = Repo::from_xhtml(html)?;
items.iter().for_each(|item| println!("{:?}", item));

相关项目

• reqwest
• scraper
• nipper
• jsonpath_lib
• unhtml.rs
• xpath-scraper