1
0
mirror of https://github.com/vlang/v.git synced 2023-08-10 21:13:21 +03:00
v/examples/web_crawler/web_crawler.v

26 lines
1000 B
V
Raw Normal View History

import net.http
import net.html
fn main() {
// http.fetch() sends an HTTP request to the URL with the given method and configurations.
config := http.FetchConfig{
user_agent: 'Mozilla/5.0 (Windows NT 6.1; Win64; x64; rv:88.0) Gecko/20100101 Firefox/88.0'
}
2021-08-17 18:36:10 +03:00
resp := http.fetch(http.FetchConfig{ ...config, url: 'https://tuicool.com' }) or {
println('failed to fetch data from the server')
return
}
// html.parse() parses and returns the DOM from the given text.
mut doc := html.parse(resp.body)
// html.DocumentObjectModel.get_tag_by_attribute_value() retrieves all the tags in the document that has the given attribute name and value.
tags := doc.get_tag_by_attribute_value('class', 'list_article_item')
for tag in tags {
el := tag.children[1].children[0].children[0].children[0]
href := el.attributes['href'] or { panic('key not found') }
title := el.attributes['title'] or { panic('key not found') }
println('href: ${href}')
println('title: ${title}')
println('')
}
}