require 'v8' require 'nokogiri' require_dependency 'excerpt_parser' require_dependency 'post' module PrettyText class Helpers def t(key, opts) key = "js." + key unless opts return I18n.t(key) else str = I18n.t(key, Hash[opts.entries].symbolize_keys).dup opts.each {|k,v| str.gsub!("{{#{k.to_s}}}", v.to_s) } return str end end # function here are available to v8 def avatar_template(username) return "" unless username user = User.find_by(username_lower: username.downcase) user.avatar_template if user.present? end def is_username_valid(username) return false unless username username = username.downcase return User.exec_sql('SELECT 1 FROM users WHERE username_lower = ?', username).values.length == 1 end end @mutex = Mutex.new @ctx_init = Mutex.new def self.mention_matcher Regexp.new("(\@[a-zA-Z0-9_]{#{User.username_length.begin},#{User.username_length.end}})") end def self.app_root Rails.root end def self.create_new_context # timeout any eval that takes longer that 5 seconds ctx = V8::Context.new(timeout: 5000) ctx["helpers"] = Helpers.new ctx_load(ctx, "vendor/assets/javascripts/md5.js", "vendor/assets/javascripts/lodash.js", "vendor/assets/javascripts/Markdown.Converter.js", "lib/headless-ember.js", "vendor/assets/javascripts/rsvp.js", Rails.configuration.ember.handlebars_location ) ctx.eval("var Discourse = {}; Discourse.SiteSettings = {};") ctx.eval("var window = {}; window.devicePixelRatio = 2;") # hack to make code think stuff is retina ctx.eval("var I18n = {}; I18n.t = function(a,b){ return helpers.t(a,b); }"); decorate_context(ctx) ctx_load(ctx, "public/javascripts/highlight.pack.js", "vendor/assets/javascripts/better_markdown.js", "app/assets/javascripts/defer/html-sanitizer-bundle.js", "app/assets/javascripts/discourse/dialects/dialect.js", "app/assets/javascripts/discourse/lib/utilities.js", "app/assets/javascripts/discourse/lib/html.js", "app/assets/javascripts/discourse/lib/markdown.js" ) Dir["#{Rails.root}/app/assets/javascripts/discourse/dialects/**.js"].sort.each do |dialect| unless dialect =~ /\/dialect\.js$/ ctx.load(dialect) end end # Load server side javascripts if DiscoursePluginRegistry.server_side_javascripts.present? DiscoursePluginRegistry.server_side_javascripts.each do |ssjs| if(ssjs =~ /\.erb/) erb = ERB.new(File.read(ssjs)) erb.filename = ssjs ctx.eval(erb.result) else ctx.load(ssjs) end end end ctx['quoteTemplate'] = File.open(app_root + 'app/assets/javascripts/discourse/templates/quote.hbs') {|f| f.read} ctx['quoteEmailTemplate'] = File.open(app_root + 'lib/assets/quote_email.hbs') {|f| f.read} ctx.eval("HANDLEBARS_TEMPLATES = { 'quote': Handlebars.compile(quoteTemplate), 'quote_email': Handlebars.compile(quoteEmailTemplate), };") ctx end def self.v8 return @ctx if @ctx # ensure we only init one of these @ctx_init.synchronize do return @ctx if @ctx @ctx = create_new_context end @ctx end def self.decorate_context(context) context.eval("Discourse.SiteSettings = #{SiteSetting.client_settings_json};") context.eval("Discourse.CDN = '#{Rails.configuration.action_controller.asset_host}';") context.eval("Discourse.BaseUrl = 'http://#{RailsMultisite::ConnectionManagement.current_hostname}';") context.eval("Discourse.getURL = function(url) {return '#{Discourse::base_uri}' + url};") end def self.markdown(text, opts=nil) # we use the exact same markdown converter as the client # TODO: use the same extensions on both client and server (in particular the template for mentions) baked = nil protect do context = v8 # we need to do this to work in a multi site environment, many sites, many settings decorate_context(context) context_opts = opts || {} context_opts[:sanitize] ||= true context['opts'] = context_opts context['raw'] = text if Post.white_listed_image_classes.present? Post.white_listed_image_classes.each do |klass| context.eval("Discourse.Markdown.whiteListClass('#{klass}')") end end context.eval('opts["mentionLookup"] = function(u){return helpers.is_username_valid(u);}') context.eval('opts["lookupAvatar"] = function(p){return Discourse.Utilities.avatarImg({size: "tiny", avatarTemplate: helpers.avatar_template(p)});}') baked = context.eval('Discourse.Markdown.markdownConverter(opts).makeHtml(raw)') end baked end # leaving this here, cause it invokes v8, don't want to implement twice def self.avatar_img(avatar_template, size) protect do v8['avatarTemplate'] = avatar_template v8['size'] = size decorate_context(v8) v8.eval("Discourse.Utilities.avatarImg({ avatarTemplate: avatarTemplate, size: size });") end end def self.cook(text, opts={}) cloned = opts.dup # we have a minor inconsistency cloned[:topicId] = opts[:topic_id] sanitized = markdown(text.dup, cloned) sanitized = add_rel_nofollow_to_user_content(sanitized) if !cloned[:omit_nofollow] && SiteSetting.add_rel_nofollow_to_user_content sanitized end def self.add_rel_nofollow_to_user_content(html) whitelist = [] domains = SiteSetting.exclude_rel_nofollow_domains whitelist = domains.split('|') if domains.present? site_uri = nil doc = Nokogiri::HTML.fragment(html) doc.css("a").each do |l| href = l["href"].to_s begin uri = URI(href) site_uri ||= URI(Discourse.base_url) if !uri.host.present? || uri.host.ends_with?(site_uri.host) || whitelist.any?{|u| uri.host.ends_with?(u)} # we are good no need for nofollow else l["rel"] = "nofollow" end rescue URI::InvalidURIError, URI::InvalidComponentError # add a nofollow anyway l["rel"] = "nofollow" end end doc.to_html end class DetectedLink attr_accessor :is_quote, :url def initialize(url, is_quote=false) @url = url @is_quote = is_quote end end def self.extract_links(html) links = [] doc = Nokogiri::HTML.fragment(html) # remove href inside quotes doc.css("aside.quote a").each { |l| l["href"] = "" } # extract all links from the post doc.css("a").each { |l| unless l["href"].blank? links << DetectedLink.new(l["href"]) end } # extract links to quotes doc.css("aside.quote[data-topic]").each do |a| topic_id = a['data-topic'] url = "/t/topic/#{topic_id}" if post_number = a['data-post'] url << "/#{post_number}" end links << DetectedLink.new(url, true) end links end def self.excerpt(html, max_length, options={}) # TODO: properly fix this HACK in ExcerptParser without introducing XSS doc = Nokogiri::HTML.fragment(html) strip_image_wrapping(doc) html = doc.to_html ExcerptParser.get_excerpt(html, max_length, options) end def self.strip_links(string) return string if string.blank? # If the user is not basic, strip links from their bio fragment = Nokogiri::HTML.fragment(string) fragment.css('a').each {|a| a.replace(a.inner_html) } fragment.to_html end # Given a Nokogiri doc, convert all links to absolute def self.make_all_links_absolute(doc) site_uri = nil doc.css("a").each do |link| href = link["href"].to_s begin uri = URI(href) site_uri ||= URI(Discourse.base_url) link["href"] = "#{site_uri}#{link['href']}" unless uri.host.present? rescue URI::InvalidURIError, URI::InvalidComponentError # leave it end end end def self.strip_image_wrapping(doc) doc.css(".lightbox-wrapper .meta").remove end def self.format_for_email(html) doc = Nokogiri::HTML.fragment(html) make_all_links_absolute(doc) strip_image_wrapping(doc) doc.to_html end protected class JavaScriptError < StandardError attr_accessor :message, :backtrace def initialize(message, backtrace) @message = message @backtrace = backtrace end end def self.protect rval = nil @mutex.synchronize do begin rval = yield # This may seem a bit odd, but we don't want to leak out # objects that require locks on the v8 vm, to get a backtrace # you need a lock, if this happens in the wrong spot you can # deadlock a process rescue V8::Error => e raise JavaScriptError.new(e.message, e.backtrace) end end rval end def self.ctx_load(ctx, *files) files.each do |file| ctx.load(app_root + file) end end end