Question

我需要解析一个HTML文档来计算Ruby中两个标签（包括属性）和文本的字符数。出于性能原因，我不想使用DOM解析器。我看过Nokogiri的SAX和Reader解析器以及SaxMachine，但似乎都没有为我提供跟踪解析器在输入HTML中的位置的方法。

有没有人知道在Ruby中访问此信息的方法？提前致谢

Answer 1

输入字符串

html = <<-HTML
<html>

<head>
  <title>Title</title>
</head>

<body>
  Hello world!
</body>

</html>
HTML

愚蠢的解决方案

原始解决方案，它计算每个字母字符（即。</html>计数4个字符。）

tag_count = 0
text_count = 0

in_tag = false

html.each_char do |char|
  case char
  when '<'
    in_tag = true
  when '>'
    in_tag = false
  when /\w/
    in_tag ? tag_count += 1 : text_count += 1
  end
end

puts "Text char count: #{text_count}"
puts "Tag char count: #{tag_count}"

Nokogiri SAX解决方案

这个可以很容易地翻译成另一种语言（例如Java）。

require 'nokogiri'

class HtmlCounter < Nokogiri::XML::SAX::Document

  attr_accessor :tag_count, :text_count, :comment_count

  def initialize(filtered_tags = [])
    @filtered_tags = filtered_tags
  end

  def start_document
    @tag_count = Hash.new(0)
    @text_count = Hash.new(0)
    @comment_count = 0
    @current_tags = []
  end

  def start_element(name, attrs)
    # Keep track of the nesting
    @current_tags.push(name)

    if should_count? 
      # Count the end element as well
      count_tag(name.length * 2)
      count_tag(attrs.flatten.map(&:length).inject(0) {|sum, length| sum + length})
    end
  end

  def end_element(name)
    @current_tags.pop
  end

  def comment(string)
    count_comment(string.length) if should_count?
  end

  def characters(string)
    count_text(string.strip.length) if should_count?
  end

  def should_count?
    # Are we in a filtered tag ?
    (@current_tags & @filtered_tags).empty?
  end

  def count_text(count)
    @text_count[@current_tags.last] += count
  end

  def count_tag(count)
    @tag_count[@current_tags.last] += count
  end

  def count_comment(count)
    @comment_count[@current_tags.last] += count
  end
end

# Don't count things in title tags
counter = HtmlCounter.new(["title"])
parser = Nokogiri::HTML::SAX::Parser.new(counter)
parser.parse(html)

puts "Text char count: #{counter.text_count}"
puts "Tag char count: #{counter.tag_count}"

输出：

Text char count: {"body"=>12}
Tag char count: {"html"=>8, "head"=>8, "body"=>8}

希望这有帮助。

在Ruby中计算HTML文档中的标记和文本字符

1 个答案: