# frozen_string_literal: true module Docs  class Rust < UrlScraper    self.type = 'rust'    self.release = '1.88.0'    self.base_url = 'https://doc.rust-lang.org/'    self.root_path = 'book/index.html'    self.initial_paths = %w(      reference/introduction.html      std/index.html      error-index.html)    self.links = {      home: 'https://www.rust-lang.org/',      code: 'https://github.com/rust-lang/rust'    }     html_filters.push 'rust/entries', 'rust/clean_html'     options[:only_patterns] = [      /\Abook\//,      /\Areference\//,      /\Acollections\//,      /\Astd\// ]     options[:skip] = %w(book/README.html book/ffi.html)    options[:skip_patterns] = [/(?<!\.html)\z/, /\/print\.html/, /\Abook\/second-edition\//]     options[:fix_urls] = ->(url) do      url.sub! %r{(#{Rust.base_url}.+/)\z}, '\1index.html'      url.sub! "#{Rust.base_url}nightly/", Rust.base_url      url.sub! '/unicode/u_str', '/unicode/str/'      url.sub! '/std/std/', '/std/'      url    end     options[:attribution] = <<-HTML      &copy; 2010 The Rust Project Developers<br>      Licensed under the Apache License, Version 2.0 or the MIT license, at your option.    HTML     def get_latest_version(opts)      doc = fetch_doc('https://www.rust-lang.org/', opts)      label = doc.at_css('.button-download + p > a').content      label.sub(/Version /, '')    end     private     REDIRECT_RGX = /http-equiv="refresh"/i    NOT_FOUND_RGX = /<title>Not Found<\/title>/     def process_response?(response)      !(response.body =~ REDIRECT_RGX || response.body =~ NOT_FOUND_RGX || response.body.blank?)    end     def parse(response) # Hook here because Nokogori removes whitespace from headings      response.body.gsub! %r{<h[1-6] class="code-header">}, '<pre class="code-header">'      super    end  endend