discourse/lib/excerpt_parser.rb

143 lines
4.1 KiB
Ruby
Raw Normal View History

2013-05-28 07:48:47 +08:00
class ExcerptParser < Nokogiri::XML::SAX::Document
attr_reader :excerpt
SPAN_REGEX = /<\s*span[^>]*class\s*=\s*['|"]excerpt['|"][^>]*>/
def initialize(length, options=nil)
2013-05-28 07:48:47 +08:00
@length = length
@excerpt = ""
@current_length = 0
options || {}
2013-05-28 07:48:47 +08:00
@strip_links = options[:strip_links] == true
@text_entities = options[:text_entities] == true
@markdown_images = options[:markdown_images] == true
@keep_newlines = options[:keep_newlines] == true
@keep_emoji_images = options[:keep_emoji_images] == true
@keep_onebox_source = options[:keep_onebox_source] == true
@remap_emoji = options[:remap_emoji] == true
@start_excerpt = false
2013-05-28 07:48:47 +08:00
end
def self.get_excerpt(html, length, options)
html ||= ''
length = html.length if html.include?('excerpt') && SPAN_REGEX === html
me = self.new(length, options)
2013-05-28 07:48:47 +08:00
parser = Nokogiri::HTML::SAX::Parser.new(me)
catch(:done) do
parser.parse(html)
2013-05-28 07:48:47 +08:00
end
excerpt = me.excerpt.strip
excerpt = excerpt.gsub(/\s*\n+\s*/, "\n\n") if options[:keep_onebox_source]
excerpt = CGI.unescapeHTML(excerpt) if options[:text_entities] == true
excerpt
2013-05-28 07:48:47 +08:00
end
2014-07-25 10:15:43 +08:00
def escape_attribute(v)
return "" unless v
v = v.dup
v.gsub!("&", "&amp;")
v.gsub!("\"", "&#34;")
v.gsub!("<", "&lt;")
v.gsub!(">", "&gt;")
v
2014-07-25 10:15:43 +08:00
end
def include_tag(name, attributes)
2014-07-25 10:15:43 +08:00
characters("<#{name} #{attributes.map{|k,v| "#{k}=\"#{escape_attribute(v)}\""}.join(' ')}>", false, false, false)
end
2013-05-28 07:48:47 +08:00
def start_element(name, attributes=[])
case name
when "img"
attributes = Hash[*attributes.flatten]
if attributes["class"] == 'emoji'
if @remap_emoji
title = (attributes["alt"] || "").gsub(":", "")
title = Emoji.lookup_unicode(title) || attributes["alt"]
return characters(title)
elsif @keep_emoji_images
return include_tag(name, attributes)
else
return characters(attributes["alt"])
end
end
# If include_images is set, include the image in markdown
characters("!") if @markdown_images
2013-05-28 07:48:47 +08:00
if attributes["alt"]
characters("[#{attributes["alt"]}]")
elsif attributes["title"]
characters("[#{attributes["title"]}]")
else
2015-01-23 17:57:01 +08:00
characters("[#{I18n.t 'excerpt_image'}]")
2013-05-28 07:48:47 +08:00
end
characters("(#{attributes['src']})") if @markdown_images
2013-05-28 07:48:47 +08:00
when "a"
unless @strip_links
include_tag(name, attributes)
2013-05-28 07:48:47 +08:00
@in_a = true
end
2013-05-28 07:48:47 +08:00
when "aside"
@in_quote = true unless @keep_onebox_source
when 'article'
if @keep_onebox_source && attributes.include?(['class', 'onebox-body'])
@in_quote = true
end
when "div", "span"
if attributes.include?(["class", "excerpt"])
@excerpt = ""
@current_length = 0
@start_excerpt = true
end
# Preserve spoilers
if attributes.include?(["class", "spoiler"])
include_tag("span", attributes)
@in_spoiler = true
end
2013-05-28 07:48:47 +08:00
end
end
def end_element(name)
case name
when "a"
unless @strip_links
characters("</a>",false, false, false)
@in_a = false
end
when "p", "br"
if @keep_newlines
characters("<br>", false, false, false)
else
characters(" ")
end
2013-05-28 07:48:47 +08:00
when "aside"
@in_quote = false
when "div", "span"
throw :done if @start_excerpt
characters("</span>", false, false, false) if @in_spoiler
@in_spoiler = false
2013-05-28 07:48:47 +08:00
end
end
def characters(string, truncate = true, count_it = true, encode = true)
return if @in_quote
encode = encode ? lambda{|s| ERB::Util.html_escape(s)} : lambda {|s| s}
if count_it && @current_length + string.length > @length
length = [0, @length - @current_length - 1].max
@excerpt << encode.call(string[0..length]) if truncate
@excerpt << (@text_entities ? "..." : "&hellip;")
2013-05-28 07:48:47 +08:00
@excerpt << "</a>" if @in_a
throw :done
end
@excerpt << encode.call(string)
@current_length += string.length if count_it
end
end