2019-05-02 18:17:27 -04:00
|
|
|
# frozen_string_literal: true
|
|
|
|
|
2016-05-19 08:25:08 -04:00
|
|
|
require 'mini_racer'
|
2013-02-05 14:16:51 -05:00
|
|
|
require 'nokogiri'
|
2016-06-14 14:31:51 -04:00
|
|
|
require 'erb'
|
2013-02-05 14:16:51 -05:00
|
|
|
|
|
|
|
module PrettyText
|
2016-06-14 14:31:51 -04:00
|
|
|
@mutex = Mutex.new
|
|
|
|
@ctx_init = Mutex.new
|
2013-02-05 14:16:51 -05:00
|
|
|
|
2016-06-14 14:31:51 -04:00
|
|
|
def self.app_root
|
|
|
|
Rails.root
|
|
|
|
end
|
2013-07-16 03:48:48 -04:00
|
|
|
|
2016-06-14 14:31:51 -04:00
|
|
|
def self.find_file(root, filename)
|
|
|
|
return filename if File.file?("#{root}#{filename}")
|
2015-03-12 15:51:28 -04:00
|
|
|
|
2016-06-14 14:31:51 -04:00
|
|
|
es6_name = "#{filename}.js.es6"
|
|
|
|
return es6_name if File.file?("#{root}#{es6_name}")
|
2013-02-05 14:16:51 -05:00
|
|
|
|
2016-06-14 14:31:51 -04:00
|
|
|
js_name = "#{filename}.js"
|
|
|
|
return js_name if File.file?("#{root}#{js_name}")
|
2015-09-24 23:35:14 -04:00
|
|
|
|
2016-06-14 14:31:51 -04:00
|
|
|
erb_name = "#{filename}.js.es6.erb"
|
|
|
|
return erb_name if File.file?("#{root}#{erb_name}")
|
2020-03-20 09:55:42 -04:00
|
|
|
|
2016-06-14 14:31:51 -04:00
|
|
|
erb_name = "#{filename}.js.erb"
|
|
|
|
return erb_name if File.file?("#{root}#{erb_name}")
|
|
|
|
end
|
2015-12-28 01:28:16 -05:00
|
|
|
|
2016-06-14 14:31:51 -04:00
|
|
|
def self.apply_es6_file(ctx, root_path, part_name)
|
|
|
|
filename = find_file(root_path, part_name)
|
|
|
|
if filename
|
|
|
|
source = File.read("#{root_path}#{filename}")
|
2020-03-11 09:43:55 -04:00
|
|
|
source = ERB.new(source).result(binding) if filename =~ /\.erb$/
|
2016-04-25 15:55:15 -04:00
|
|
|
|
2020-03-11 09:43:55 -04:00
|
|
|
transpiler = DiscourseJsProcessor::Transpiler.new
|
|
|
|
transpiled = transpiler.perform(source, "#{Rails.root}/app/assets/javascripts/", part_name)
|
2016-06-14 14:31:51 -04:00
|
|
|
ctx.eval(transpiled)
|
|
|
|
else
|
|
|
|
# Look for vendored stuff
|
|
|
|
vendor_root = "#{Rails.root}/vendor/assets/javascripts/"
|
|
|
|
filename = find_file(vendor_root, part_name)
|
|
|
|
if filename
|
|
|
|
ctx.eval(File.read("#{vendor_root}#{filename}"))
|
2016-04-25 15:55:15 -04:00
|
|
|
end
|
|
|
|
end
|
2013-02-05 14:16:51 -05:00
|
|
|
end
|
|
|
|
|
2017-06-08 18:02:30 -04:00
|
|
|
def self.ctx_load_manifest(ctx, name)
|
|
|
|
manifest = File.read("#{Rails.root}/app/assets/javascripts/#{name}")
|
2016-06-14 14:31:51 -04:00
|
|
|
root_path = "#{Rails.root}/app/assets/javascripts/"
|
2017-06-08 18:02:30 -04:00
|
|
|
|
2016-06-14 14:31:51 -04:00
|
|
|
manifest.each_line do |l|
|
2016-08-25 07:45:29 -04:00
|
|
|
l = l.chomp
|
2016-06-14 14:31:51 -04:00
|
|
|
if l =~ /\/\/= require (\.\/)?(.*)$/
|
|
|
|
apply_es6_file(ctx, root_path, Regexp.last_match[2])
|
|
|
|
elsif l =~ /\/\/= require_tree (\.\/)?(.*)$/
|
|
|
|
path = Regexp.last_match[2]
|
2016-08-10 13:20:39 -04:00
|
|
|
Dir["#{root_path}/#{path}/**"].sort.each do |f|
|
2016-06-14 14:31:51 -04:00
|
|
|
apply_es6_file(ctx, root_path, f.sub(root_path, '')[1..-1].sub(/\.js(.es6)?$/, ''))
|
|
|
|
end
|
|
|
|
end
|
2016-05-19 08:25:08 -04:00
|
|
|
end
|
2017-06-08 18:02:30 -04:00
|
|
|
end
|
|
|
|
|
|
|
|
def self.create_es6_context
|
2020-05-15 00:01:54 -04:00
|
|
|
ctx = MiniRacer::Context.new(timeout: 25000, ensure_gc_after_idle: 2000)
|
2017-06-08 18:02:30 -04:00
|
|
|
|
|
|
|
ctx.eval("window = {}; window.devicePixelRatio = 2;") # hack to make code think stuff is retina
|
|
|
|
|
|
|
|
if Rails.env.development? || Rails.env.test?
|
|
|
|
ctx.attach("console.log", proc { |l| p l })
|
2017-06-23 15:24:11 -04:00
|
|
|
ctx.eval('window.console = console;')
|
2017-06-08 18:02:30 -04:00
|
|
|
end
|
2019-10-30 09:48:24 -04:00
|
|
|
ctx.eval("__PRETTY_TEXT = true")
|
2017-06-08 18:02:30 -04:00
|
|
|
|
2020-05-13 16:23:41 -04:00
|
|
|
PrettyText::Helpers.instance_methods.each do |method|
|
|
|
|
ctx.attach("__helpers.#{method}", PrettyText::Helpers.method(method))
|
|
|
|
end
|
|
|
|
|
2017-06-08 18:02:30 -04:00
|
|
|
ctx_load(ctx, "#{Rails.root}/app/assets/javascripts/discourse-loader.js")
|
2020-04-30 16:41:02 -04:00
|
|
|
ctx_load(ctx, "#{Rails.root}/app/assets/javascripts/handlebars-shim.js")
|
2017-06-08 18:02:30 -04:00
|
|
|
ctx_load(ctx, "vendor/assets/javascripts/lodash.js")
|
2020-09-15 10:59:41 -04:00
|
|
|
ctx_load(ctx, "vendor/assets/javascripts/xss.min.js")
|
|
|
|
ctx.load("#{Rails.root}/lib/pretty_text/vendor-shims.js")
|
2017-06-08 18:02:30 -04:00
|
|
|
ctx_load_manifest(ctx, "pretty-text-bundle.js")
|
2017-07-14 08:27:28 -04:00
|
|
|
ctx_load_manifest(ctx, "markdown-it-bundle.js")
|
2017-06-08 18:02:30 -04:00
|
|
|
root_path = "#{Rails.root}/app/assets/javascripts/"
|
2013-02-05 14:16:51 -05:00
|
|
|
|
2020-06-03 12:45:26 -04:00
|
|
|
apply_es6_file(ctx, root_path, "discourse-common/addon/lib/get-url")
|
2020-09-02 11:52:54 -04:00
|
|
|
apply_es6_file(ctx, root_path, "discourse-common/addon/lib/object")
|
2020-10-27 22:22:06 -04:00
|
|
|
apply_es6_file(ctx, root_path, "discourse-common/addon/lib/deprecated")
|
2018-05-05 05:21:07 -04:00
|
|
|
apply_es6_file(ctx, root_path, "discourse/app/lib/to-markdown")
|
2016-06-14 14:31:51 -04:00
|
|
|
apply_es6_file(ctx, root_path, "discourse/app/lib/utilities")
|
|
|
|
|
|
|
|
ctx.load("#{Rails.root}/lib/pretty_text/shims.js")
|
|
|
|
ctx.eval("__setUnicode(#{Emoji.unicode_replacements_json})")
|
2013-08-08 18:14:12 -04:00
|
|
|
|
2016-06-14 14:31:51 -04:00
|
|
|
to_load = []
|
|
|
|
DiscoursePluginRegistry.each_globbed_asset do |a|
|
|
|
|
to_load << a if File.file?(a) && a =~ /discourse-markdown/
|
|
|
|
end
|
|
|
|
to_load.uniq.each do |f|
|
|
|
|
if f =~ /^.+assets\/javascripts\//
|
|
|
|
root = Regexp.last_match[0]
|
|
|
|
apply_es6_file(ctx, root, f.sub(root, '').sub(/\.js(\.es6)?$/, ''))
|
2013-02-05 14:16:51 -05:00
|
|
|
end
|
|
|
|
end
|
|
|
|
|
2018-04-10 02:37:16 -04:00
|
|
|
DiscoursePluginRegistry.vendored_core_pretty_text.each do |vpt|
|
|
|
|
ctx.eval(File.read(vpt))
|
|
|
|
end
|
|
|
|
|
2017-04-18 17:49:56 -04:00
|
|
|
DiscoursePluginRegistry.vendored_pretty_text.each do |vpt|
|
|
|
|
ctx.eval(File.read(vpt))
|
|
|
|
end
|
|
|
|
|
2013-08-15 18:12:10 -04:00
|
|
|
ctx
|
|
|
|
end
|
|
|
|
|
|
|
|
def self.v8
|
|
|
|
return @ctx if @ctx
|
|
|
|
|
|
|
|
# ensure we only init one of these
|
|
|
|
@ctx_init.synchronize do
|
|
|
|
return @ctx if @ctx
|
2016-06-14 14:31:51 -04:00
|
|
|
@ctx = create_es6_context
|
2013-08-15 18:12:10 -04:00
|
|
|
end
|
2014-04-14 16:55:57 -04:00
|
|
|
|
2013-02-05 14:16:51 -05:00
|
|
|
@ctx
|
|
|
|
end
|
|
|
|
|
2020-05-27 14:11:52 -04:00
|
|
|
def self.reset_translations
|
|
|
|
v8.eval("__resetTranslationTree()")
|
|
|
|
end
|
|
|
|
|
2014-11-14 01:51:04 -05:00
|
|
|
def self.reset_context
|
|
|
|
@ctx_init.synchronize do
|
2017-07-20 00:17:45 -04:00
|
|
|
@ctx&.dispose
|
2014-11-14 01:51:04 -05:00
|
|
|
@ctx = nil
|
|
|
|
end
|
|
|
|
end
|
|
|
|
|
2016-07-07 03:52:56 -04:00
|
|
|
def self.markdown(text, opts = {})
|
2013-02-05 14:16:51 -05:00
|
|
|
# we use the exact same markdown converter as the client
|
2013-02-25 11:42:20 -05:00
|
|
|
# TODO: use the same extensions on both client and server (in particular the template for mentions)
|
2013-02-05 14:16:51 -05:00
|
|
|
baked = nil
|
2016-05-19 08:25:08 -04:00
|
|
|
text = text || ""
|
2013-02-05 14:16:51 -05:00
|
|
|
|
2014-02-03 19:12:53 -05:00
|
|
|
protect do
|
2013-08-15 23:03:47 -04:00
|
|
|
context = v8
|
2013-10-11 16:24:27 -04:00
|
|
|
|
2016-06-14 14:31:51 -04:00
|
|
|
custom_emoji = {}
|
2016-11-17 13:35:39 -05:00
|
|
|
Emoji.custom.map { |e| custom_emoji[e.name] = e.url }
|
2016-06-14 14:31:51 -04:00
|
|
|
|
2019-05-02 18:17:27 -04:00
|
|
|
buffer = +<<~JS
|
2017-06-08 18:02:30 -04:00
|
|
|
__optInput = {};
|
|
|
|
__optInput.siteSettings = #{SiteSetting.client_settings_json};
|
2019-06-03 03:41:26 -04:00
|
|
|
#{"__optInput.disableEmojis = true" if opts[:disable_emojis]}
|
2018-05-23 10:47:09 -04:00
|
|
|
__paths = #{paths_json};
|
2017-06-08 18:02:30 -04:00
|
|
|
__optInput.getURL = __getURL;
|
2019-07-09 07:42:02 -04:00
|
|
|
#{"__optInput.features = #{opts[:features].to_json};" if opts[:features]}
|
2017-06-08 18:02:30 -04:00
|
|
|
__optInput.getCurrentUser = __getCurrentUser;
|
|
|
|
__optInput.lookupAvatar = __lookupAvatar;
|
2017-11-03 09:51:40 -04:00
|
|
|
__optInput.lookupPrimaryUserGroup = __lookupPrimaryUserGroup;
|
2017-11-20 16:28:03 -05:00
|
|
|
__optInput.formatUsername = __formatUsername;
|
2017-06-08 18:02:30 -04:00
|
|
|
__optInput.getTopicInfo = __getTopicInfo;
|
|
|
|
__optInput.categoryHashtagLookup = __categoryLookup;
|
|
|
|
__optInput.customEmoji = #{custom_emoji.to_json};
|
2020-05-27 14:11:52 -04:00
|
|
|
__optInput.customEmojiTranslation = #{Plugin::CustomEmoji.translations.to_json};
|
2017-06-28 13:47:22 -04:00
|
|
|
__optInput.emojiUnicodeReplacer = __emojiUnicodeReplacer;
|
2019-05-28 21:00:25 -04:00
|
|
|
__optInput.lookupUploadUrls = __lookupUploadUrls;
|
2019-07-31 13:33:49 -04:00
|
|
|
__optInput.censoredRegexp = #{WordWatcher.word_matcher_regexp(:censor)&.source.to_json};
|
2017-06-08 18:02:30 -04:00
|
|
|
JS
|
|
|
|
|
|
|
|
if opts[:topicId]
|
|
|
|
buffer << "__optInput.topicId = #{opts[:topicId].to_i};\n"
|
|
|
|
end
|
|
|
|
|
|
|
|
if opts[:user_id]
|
|
|
|
buffer << "__optInput.userId = #{opts[:user_id].to_i};\n"
|
|
|
|
end
|
|
|
|
|
|
|
|
buffer << "__textOptions = __buildOptions(__optInput);\n"
|
2017-07-14 08:27:28 -04:00
|
|
|
buffer << ("__pt = new __PrettyText(__textOptions);")
|
|
|
|
|
2016-08-11 14:59:20 -04:00
|
|
|
# Be careful disabling sanitization. We allow for custom emails
|
|
|
|
if opts[:sanitize] == false
|
2017-07-14 08:27:28 -04:00
|
|
|
buffer << ('__pt.disableSanitizer();')
|
2016-08-11 14:59:20 -04:00
|
|
|
end
|
|
|
|
|
2017-06-08 18:02:30 -04:00
|
|
|
opts = context.eval(buffer)
|
2016-06-14 14:31:51 -04:00
|
|
|
|
2016-01-29 09:59:15 -05:00
|
|
|
DiscourseEvent.trigger(:markdown_context, context)
|
2016-06-14 14:31:51 -04:00
|
|
|
baked = context.eval("__pt.cook(#{text.inspect})")
|
2013-02-05 14:16:51 -05:00
|
|
|
end
|
|
|
|
|
|
|
|
baked
|
|
|
|
end
|
|
|
|
|
2018-05-23 10:47:09 -04:00
|
|
|
def self.paths_json
|
|
|
|
paths = {
|
2020-10-09 07:51:24 -04:00
|
|
|
baseUri: Discourse.base_path,
|
2018-05-23 10:47:09 -04:00
|
|
|
CDN: Rails.configuration.action_controller.asset_host,
|
|
|
|
}
|
|
|
|
|
|
|
|
if SiteSetting.Upload.enable_s3_uploads
|
|
|
|
if SiteSetting.Upload.s3_cdn_url.present?
|
|
|
|
paths[:S3CDN] = SiteSetting.Upload.s3_cdn_url
|
|
|
|
end
|
|
|
|
paths[:S3BaseUrl] = Discourse.store.absolute_base_url
|
|
|
|
end
|
|
|
|
|
|
|
|
paths.to_json
|
|
|
|
end
|
|
|
|
|
2013-02-05 14:16:51 -05:00
|
|
|
# leaving this here, cause it invokes v8, don't want to implement twice
|
2013-08-13 16:08:29 -04:00
|
|
|
def self.avatar_img(avatar_template, size)
|
2014-02-03 19:12:53 -05:00
|
|
|
protect do
|
2018-05-23 10:47:09 -04:00
|
|
|
v8.eval(<<~JS)
|
|
|
|
__paths = #{paths_json};
|
|
|
|
__utils.avatarImg({size: #{size.inspect}, avatarTemplate: #{avatar_template.inspect}}, __getURL);
|
|
|
|
JS
|
2013-02-05 14:16:51 -05:00
|
|
|
end
|
|
|
|
end
|
|
|
|
|
2015-10-15 03:59:29 -04:00
|
|
|
def self.unescape_emoji(title)
|
2019-03-21 04:11:33 -04:00
|
|
|
return title unless SiteSetting.enable_emoji? && title
|
2016-06-14 14:31:51 -04:00
|
|
|
|
|
|
|
set = SiteSetting.emoji_set.inspect
|
2019-03-06 06:49:17 -05:00
|
|
|
custom = Emoji.custom.map { |e| [e.name, e.url] }.to_h.to_json
|
2019-12-03 11:32:33 -05:00
|
|
|
|
2015-10-15 03:59:29 -04:00
|
|
|
protect do
|
2018-05-23 10:47:09 -04:00
|
|
|
v8.eval(<<~JS)
|
|
|
|
__paths = #{paths_json};
|
2019-05-21 10:56:51 -04:00
|
|
|
__performEmojiUnescape(#{title.inspect}, {
|
|
|
|
getURL: __getURL,
|
|
|
|
emojiSet: #{set},
|
|
|
|
customEmoji: #{custom},
|
2019-12-03 11:32:33 -05:00
|
|
|
enableEmojiShortcuts: #{SiteSetting.enable_emoji_shortcuts},
|
|
|
|
inlineEmoji: #{SiteSetting.enable_inline_emoji_translation}
|
2019-05-21 10:56:51 -04:00
|
|
|
});
|
2018-05-23 10:47:09 -04:00
|
|
|
JS
|
2015-10-15 03:59:29 -04:00
|
|
|
end
|
|
|
|
end
|
|
|
|
|
2019-03-21 04:11:33 -04:00
|
|
|
def self.escape_emoji(title)
|
|
|
|
return unless title
|
|
|
|
|
2019-05-21 10:56:51 -04:00
|
|
|
replace_emoji_shortcuts = SiteSetting.enable_emoji && SiteSetting.enable_emoji_shortcuts
|
|
|
|
|
2019-03-21 04:11:33 -04:00
|
|
|
protect do
|
|
|
|
v8.eval(<<~JS)
|
2019-12-03 11:32:33 -05:00
|
|
|
__performEmojiEscape(#{title.inspect}, {
|
|
|
|
emojiShortcuts: #{replace_emoji_shortcuts},
|
|
|
|
inlineEmoji: #{SiteSetting.enable_inline_emoji_translation}
|
|
|
|
});
|
2019-03-21 04:11:33 -04:00
|
|
|
JS
|
|
|
|
end
|
|
|
|
end
|
|
|
|
|
2013-02-05 14:16:51 -05:00
|
|
|
def self.cook(text, opts = {})
|
2015-04-23 13:33:29 -04:00
|
|
|
options = opts.dup
|
|
|
|
|
2013-02-05 14:16:51 -05:00
|
|
|
# we have a minor inconsistency
|
2015-04-23 13:33:29 -04:00
|
|
|
options[:topicId] = opts[:topic_id]
|
|
|
|
|
2015-12-30 14:35:25 -05:00
|
|
|
working_text = text.dup
|
2016-11-08 16:36:34 -05:00
|
|
|
|
2018-02-26 15:48:59 -05:00
|
|
|
sanitized = markdown(working_text, options)
|
2015-05-25 21:13:12 -04:00
|
|
|
|
2020-05-04 23:46:57 -04:00
|
|
|
doc = Nokogiri::HTML5.fragment(sanitized)
|
2015-05-25 21:13:12 -04:00
|
|
|
|
2020-09-10 11:59:51 -04:00
|
|
|
add_nofollow = !options[:omit_nofollow] && SiteSetting.add_rel_nofollow_to_user_content
|
|
|
|
add_rel_attributes_to_user_content(doc, add_nofollow)
|
2015-05-25 21:13:12 -04:00
|
|
|
|
2018-11-22 03:01:03 -05:00
|
|
|
if SiteSetting.enable_mentions
|
|
|
|
add_mentions(doc, user_id: opts[:user_id])
|
|
|
|
end
|
2018-11-22 01:28:48 -05:00
|
|
|
|
2020-05-04 23:46:57 -04:00
|
|
|
scrubber = Loofah::Scrubber.new do |node|
|
|
|
|
node.remove if node.name == 'script'
|
|
|
|
end
|
|
|
|
loofah_fragment = Loofah.fragment(doc.to_html)
|
|
|
|
loofah_fragment.scrub!(scrubber).to_html
|
2015-05-25 21:13:12 -04:00
|
|
|
end
|
|
|
|
|
2020-09-10 11:59:51 -04:00
|
|
|
def self.add_rel_attributes_to_user_content(doc, add_nofollow)
|
2020-07-26 20:23:54 -04:00
|
|
|
allowlist = []
|
2013-02-11 03:01:33 -05:00
|
|
|
|
2013-11-19 22:38:21 -05:00
|
|
|
domains = SiteSetting.exclude_rel_nofollow_domains
|
2020-07-26 20:23:54 -04:00
|
|
|
allowlist = domains.split('|') if domains.present?
|
2013-02-11 03:01:33 -05:00
|
|
|
|
2013-02-10 19:43:07 -05:00
|
|
|
site_uri = nil
|
|
|
|
doc.css("a").each do |l|
|
|
|
|
href = l["href"].to_s
|
2020-09-10 11:59:51 -04:00
|
|
|
l["rel"] = "noopener" if l["target"] == "_blank"
|
|
|
|
|
2013-02-25 11:42:20 -05:00
|
|
|
begin
|
2020-05-07 21:14:59 -04:00
|
|
|
uri = URI(UrlHelper.encode_component(href))
|
2013-02-10 19:43:07 -05:00
|
|
|
site_uri ||= URI(Discourse.base_url)
|
2013-02-25 11:42:20 -05:00
|
|
|
|
2020-09-10 11:59:51 -04:00
|
|
|
same_domain = !uri.host.present? ||
|
|
|
|
uri.host == site_uri.host ||
|
|
|
|
uri.host.ends_with?(".#{site_uri.host}") ||
|
|
|
|
allowlist.any? { |u| uri.host == u || uri.host.ends_with?(".#{u}") }
|
|
|
|
|
|
|
|
l["rel"] = "noopener nofollow ugc" if add_nofollow && !same_domain
|
2018-08-14 06:23:32 -04:00
|
|
|
rescue URI::Error
|
2013-02-25 11:42:20 -05:00
|
|
|
# add a nofollow anyway
|
2020-09-10 11:59:51 -04:00
|
|
|
l["rel"] = "noopener nofollow ugc"
|
2013-02-10 19:43:07 -05:00
|
|
|
end
|
|
|
|
end
|
2013-02-05 14:16:51 -05:00
|
|
|
end
|
|
|
|
|
2017-02-06 08:45:04 -05:00
|
|
|
class DetectedLink < Struct.new(:url, :is_quote); end
|
2014-07-11 00:17:01 -04:00
|
|
|
|
2013-02-05 14:16:51 -05:00
|
|
|
def self.extract_links(html)
|
|
|
|
links = []
|
2020-05-04 23:46:57 -04:00
|
|
|
doc = Nokogiri::HTML5.fragment(html)
|
2017-02-06 08:45:04 -05:00
|
|
|
|
2016-03-16 17:35:08 -04:00
|
|
|
# remove href inside quotes & elided part
|
2017-02-06 08:45:04 -05:00
|
|
|
doc.css("aside.quote a, .elided a").each { |a| a["href"] = "" }
|
2014-07-11 00:17:01 -04:00
|
|
|
|
2017-02-06 08:45:04 -05:00
|
|
|
# extract all links
|
|
|
|
doc.css("a").each do |a|
|
2020-04-30 02:48:34 -04:00
|
|
|
if a["href"].present? && a["href"][0] != "#"
|
2017-02-06 08:45:04 -05:00
|
|
|
links << DetectedLink.new(a["href"], false)
|
2014-07-11 00:17:01 -04:00
|
|
|
end
|
2017-02-06 08:45:04 -05:00
|
|
|
end
|
2013-02-25 11:42:20 -05:00
|
|
|
|
2017-02-06 08:45:04 -05:00
|
|
|
# extract quotes
|
|
|
|
doc.css("aside.quote[data-topic]").each do |aside|
|
|
|
|
if aside["data-topic"].present?
|
2019-05-02 18:17:27 -04:00
|
|
|
url = +"/t/topic/#{aside["data-topic"]}"
|
2017-02-06 08:45:04 -05:00
|
|
|
url << "/#{aside["data-post"]}" if aside["data-post"].present?
|
|
|
|
links << DetectedLink.new(url, true)
|
2013-02-13 15:22:04 -05:00
|
|
|
end
|
|
|
|
end
|
|
|
|
|
2017-02-06 08:45:04 -05:00
|
|
|
# extract Youtube links
|
|
|
|
doc.css("div[data-youtube-id]").each do |div|
|
|
|
|
if div["data-youtube-id"].present?
|
|
|
|
links << DetectedLink.new("https://www.youtube.com/watch?v=#{div['data-youtube-id']}", false)
|
|
|
|
end
|
2016-09-22 16:50:05 -04:00
|
|
|
end
|
|
|
|
|
2013-02-05 14:16:51 -05:00
|
|
|
links
|
|
|
|
end
|
|
|
|
|
2013-05-27 19:48:47 -04:00
|
|
|
def self.excerpt(html, max_length, options = {})
|
2014-11-05 14:37:00 -05:00
|
|
|
# TODO: properly fix this HACK in ExcerptParser without introducing XSS
|
2020-05-04 23:46:57 -04:00
|
|
|
doc = Nokogiri::HTML5.fragment(html)
|
2019-05-29 11:05:52 -04:00
|
|
|
DiscourseEvent.trigger(:reduce_excerpt, doc, options)
|
2014-11-05 14:37:00 -05:00
|
|
|
strip_image_wrapping(doc)
|
2020-02-06 15:08:13 -05:00
|
|
|
strip_oneboxed_media(doc)
|
2014-11-05 14:37:00 -05:00
|
|
|
html = doc.to_html
|
2013-05-27 19:48:47 -04:00
|
|
|
ExcerptParser.get_excerpt(html, max_length, options)
|
|
|
|
end
|
2013-02-05 14:16:51 -05:00
|
|
|
|
2013-06-05 15:28:10 -04:00
|
|
|
def self.strip_links(string)
|
|
|
|
return string if string.blank?
|
|
|
|
|
|
|
|
# If the user is not basic, strip links from their bio
|
2020-05-04 23:46:57 -04:00
|
|
|
fragment = Nokogiri::HTML5.fragment(string)
|
2014-09-17 12:08:00 -04:00
|
|
|
fragment.css('a').each { |a| a.replace(a.inner_html) }
|
2013-06-05 15:28:10 -04:00
|
|
|
fragment.to_html
|
|
|
|
end
|
|
|
|
|
2018-05-09 13:24:44 -04:00
|
|
|
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)
|
2018-06-08 13:11:52 -04:00
|
|
|
unless uri.host.present? || href.start_with?('mailto')
|
|
|
|
link["href"] = "#{site_uri}#{link['href']}"
|
2018-06-08 13:56:20 -04:00
|
|
|
end
|
2018-08-14 06:23:32 -04:00
|
|
|
rescue URI::Error
|
2018-05-09 13:24:44 -04:00
|
|
|
# leave it
|
|
|
|
end
|
|
|
|
end
|
|
|
|
end
|
2016-06-21 11:12:30 -04:00
|
|
|
|
2014-04-17 12:32:51 -04:00
|
|
|
def self.strip_image_wrapping(doc)
|
|
|
|
doc.css(".lightbox-wrapper .meta").remove
|
|
|
|
end
|
|
|
|
|
2020-02-06 15:08:13 -05:00
|
|
|
def self.strip_oneboxed_media(doc)
|
|
|
|
doc.css("audio").remove
|
2020-02-17 13:52:23 -05:00
|
|
|
doc.css(".video-onebox,video").remove
|
2020-02-06 15:08:13 -05:00
|
|
|
end
|
|
|
|
|
2018-05-09 13:24:44 -04:00
|
|
|
def self.convert_vimeo_iframes(doc)
|
|
|
|
doc.css("iframe[src*='player.vimeo.com']").each do |iframe|
|
2019-04-26 07:39:18 -04:00
|
|
|
if iframe["data-original-href"].present?
|
2019-05-09 11:37:55 -04:00
|
|
|
vimeo_url = UrlHelper.escape_uri(iframe["data-original-href"])
|
2019-04-26 07:39:18 -04:00
|
|
|
else
|
|
|
|
vimeo_id = iframe['src'].split('/').last
|
|
|
|
vimeo_url = "https://vimeo.com/#{vimeo_id}"
|
|
|
|
end
|
|
|
|
iframe.replace "<p><a href='#{vimeo_url}'>#{vimeo_url}</a></p>"
|
2018-05-09 13:24:44 -04:00
|
|
|
end
|
|
|
|
end
|
|
|
|
|
2019-11-17 20:25:42 -05:00
|
|
|
def self.strip_secure_media(doc)
|
2020-11-01 18:52:21 -05:00
|
|
|
# images inside a lightbox or other link
|
|
|
|
doc.css('a[href]').each do |a|
|
|
|
|
next if !Upload.secure_media_url?(a['href'])
|
|
|
|
|
|
|
|
non_image_media = %w(video audio).include?(a&.parent&.name)
|
|
|
|
target = non_image_media ? a.parent : a
|
|
|
|
next if target.to_s.include?('stripped-secure-view-media')
|
|
|
|
|
2020-11-04 15:45:50 -05:00
|
|
|
next if a.css('img[src]').empty? && !non_image_media
|
|
|
|
|
2020-11-01 18:52:21 -05:00
|
|
|
if a.classes.include?('lightbox')
|
|
|
|
img = a.css('img[src]').first
|
2020-11-04 15:45:50 -05:00
|
|
|
srcset = img&.attributes['srcset']&.value
|
|
|
|
if srcset
|
|
|
|
# if available, use the first image from the srcset here
|
|
|
|
# so we get the optimized image instead of the possibly huge original
|
|
|
|
url = srcset.split(',').first
|
|
|
|
else
|
|
|
|
url = img['src']
|
|
|
|
end
|
2020-11-01 18:52:21 -05:00
|
|
|
a.add_next_sibling secure_media_placeholder(doc, url, width: img['width'], height: img['height'])
|
|
|
|
a.remove
|
|
|
|
else
|
|
|
|
width = non_image_media ? nil : a.at_css('img').attr('width')
|
|
|
|
height = non_image_media ? nil : a.at_css('img').attr('height')
|
2020-10-21 22:25:09 -04:00
|
|
|
target.add_next_sibling secure_media_placeholder(doc, a['href'], width: width, height: height)
|
2020-09-09 19:50:16 -04:00
|
|
|
target.remove
|
2019-11-17 20:25:42 -05:00
|
|
|
end
|
|
|
|
end
|
2020-11-01 18:52:21 -05:00
|
|
|
|
|
|
|
# images by themselves or inside a onebox
|
2020-09-09 19:50:16 -04:00
|
|
|
doc.css('img[src]').each do |img|
|
2020-11-03 11:53:15 -05:00
|
|
|
url = if img.parent.classes.include?("aspect-image") && img.attributes["srcset"].present?
|
2020-11-01 18:52:21 -05:00
|
|
|
|
|
|
|
# we are using the first image from the srcset here so we get the
|
|
|
|
# optimized image instead of the original, because an optimized
|
|
|
|
# image may be used for the onebox thumbnail
|
|
|
|
srcset = img.attributes["srcset"].value
|
|
|
|
srcset.split(",").first
|
|
|
|
else
|
|
|
|
img['src']
|
|
|
|
end
|
|
|
|
|
2020-11-09 21:55:18 -05:00
|
|
|
width = img['width']
|
|
|
|
height = img['height']
|
2020-11-15 18:58:40 -05:00
|
|
|
onebox_type = nil
|
|
|
|
|
|
|
|
if img.ancestors.css(".onebox-body").any?
|
|
|
|
if img.classes.include?("onebox-avatar-inline")
|
|
|
|
onebox_type = "avatar-inline"
|
|
|
|
else
|
|
|
|
onebox_type = "thumbnail"
|
|
|
|
end
|
|
|
|
end
|
2020-11-09 21:55:18 -05:00
|
|
|
|
|
|
|
# we always want this to be tiny and without any special styles
|
|
|
|
if img.classes.include?('site-icon')
|
2020-11-15 18:58:40 -05:00
|
|
|
onebox_type = nil
|
2020-11-09 21:55:18 -05:00
|
|
|
width = 16
|
|
|
|
height = 16
|
|
|
|
end
|
2020-11-01 18:52:21 -05:00
|
|
|
|
|
|
|
if Upload.secure_media_url?(url)
|
2020-11-15 18:58:40 -05:00
|
|
|
img.add_next_sibling secure_media_placeholder(doc, url, onebox_type: onebox_type, width: width, height: height)
|
2020-09-09 19:50:16 -04:00
|
|
|
img.remove
|
|
|
|
end
|
|
|
|
end
|
|
|
|
end
|
|
|
|
|
2020-11-15 18:58:40 -05:00
|
|
|
def self.secure_media_placeholder(doc, url, onebox_type: false, width: nil, height: nil)
|
2020-10-21 22:25:09 -04:00
|
|
|
data_width = width ? "data-width=#{width}" : ''
|
|
|
|
data_height = height ? "data-height=#{height}" : ''
|
2020-11-15 18:58:40 -05:00
|
|
|
data_onebox_type = onebox_type ? "data-onebox-type='#{onebox_type}'" : ''
|
2020-09-09 19:50:16 -04:00
|
|
|
<<~HTML
|
2020-11-15 18:58:40 -05:00
|
|
|
<div class="secure-media-notice" data-stripped-secure-media="#{url}" #{data_onebox_type} #{data_width} #{data_height}>
|
2020-09-09 19:50:16 -04:00
|
|
|
#{I18n.t('emails.secure_media_placeholder')} <a class='stripped-secure-view-media' href="#{url}">#{I18n.t("emails.view_redacted_media")}</a>.
|
|
|
|
</div>
|
|
|
|
HTML
|
2019-11-17 20:25:42 -05:00
|
|
|
end
|
|
|
|
|
2016-06-21 11:12:30 -04:00
|
|
|
def self.format_for_email(html, post = nil)
|
2020-05-04 23:46:57 -04:00
|
|
|
doc = Nokogiri::HTML5.fragment(html)
|
2016-06-21 11:12:30 -04:00
|
|
|
DiscourseEvent.trigger(:reduce_cooked, doc, post)
|
2019-11-17 20:25:42 -05:00
|
|
|
strip_secure_media(doc) if post&.with_secure_media?
|
2016-06-21 11:12:30 -04:00
|
|
|
strip_image_wrapping(doc)
|
2018-05-09 13:24:44 -04:00
|
|
|
convert_vimeo_iframes(doc)
|
2016-06-21 11:12:30 -04:00
|
|
|
make_all_links_absolute(doc)
|
|
|
|
doc.to_html
|
2013-11-28 15:57:21 -05:00
|
|
|
end
|
|
|
|
|
2013-05-27 19:48:47 -04:00
|
|
|
protected
|
2013-02-05 14:16:51 -05:00
|
|
|
|
2014-02-03 19:12:53 -05:00
|
|
|
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
|
2016-05-19 08:25:08 -04:00
|
|
|
rval = yield
|
2014-02-03 19:12:53 -05:00
|
|
|
end
|
|
|
|
rval
|
|
|
|
end
|
|
|
|
|
2013-08-15 18:12:10 -04:00
|
|
|
def self.ctx_load(ctx, *files)
|
2013-05-27 19:48:47 -04:00
|
|
|
files.each do |file|
|
2013-08-15 18:12:10 -04:00
|
|
|
ctx.load(app_root + file)
|
2013-02-05 14:16:51 -05:00
|
|
|
end
|
|
|
|
end
|
|
|
|
|
2018-11-22 01:28:48 -05:00
|
|
|
private
|
|
|
|
|
|
|
|
USER_TYPE ||= 'user'
|
|
|
|
GROUP_TYPE ||= 'group'
|
2020-02-18 15:45:02 -05:00
|
|
|
GROUP_MENTIONABLE_TYPE ||= 'group-mentionable'
|
2018-11-22 01:28:48 -05:00
|
|
|
|
2018-11-22 03:01:03 -05:00
|
|
|
def self.add_mentions(doc, user_id: nil)
|
2018-11-22 01:28:48 -05:00
|
|
|
elements = doc.css("span.mention")
|
2018-11-22 19:31:52 -05:00
|
|
|
names = elements.map { |element| element.text[1..-1] }
|
2018-11-22 01:28:48 -05:00
|
|
|
|
2018-11-22 03:01:03 -05:00
|
|
|
mentions = lookup_mentions(names, user_id: user_id)
|
2018-11-22 01:28:48 -05:00
|
|
|
|
2020-02-10 12:31:42 -05:00
|
|
|
elements.each do |element|
|
2018-11-22 01:28:48 -05:00
|
|
|
name = element.text[1..-1]
|
|
|
|
name.downcase!
|
|
|
|
|
|
|
|
if type = mentions[name]
|
|
|
|
element.name = 'a'
|
|
|
|
|
|
|
|
element.children = PrettyText::Helpers.format_username(
|
|
|
|
element.children.text
|
|
|
|
)
|
|
|
|
|
|
|
|
case type
|
|
|
|
when USER_TYPE
|
2020-10-09 07:51:24 -04:00
|
|
|
element['href'] = "#{Discourse.base_path}/u/#{UrlHelper.encode_component(name)}"
|
2020-02-18 15:45:02 -05:00
|
|
|
when GROUP_MENTIONABLE_TYPE
|
|
|
|
element['class'] = 'mention-group notify'
|
2020-10-09 07:51:24 -04:00
|
|
|
element['href'] = "#{Discourse.base_path}/groups/#{UrlHelper.encode_component(name)}"
|
2018-11-22 01:28:48 -05:00
|
|
|
when GROUP_TYPE
|
|
|
|
element['class'] = 'mention-group'
|
2020-10-09 07:51:24 -04:00
|
|
|
element['href'] = "#{Discourse.base_path}/groups/#{UrlHelper.encode_component(name)}"
|
2018-11-22 01:28:48 -05:00
|
|
|
end
|
|
|
|
end
|
|
|
|
end
|
|
|
|
end
|
|
|
|
|
2018-11-22 03:01:03 -05:00
|
|
|
def self.lookup_mentions(names, user_id: nil)
|
2018-11-22 03:42:56 -05:00
|
|
|
return {} if names.blank?
|
|
|
|
|
2018-11-22 01:28:48 -05:00
|
|
|
sql = <<~SQL
|
|
|
|
(
|
|
|
|
SELECT
|
|
|
|
:user_type AS type,
|
|
|
|
username_lower AS name
|
|
|
|
FROM users
|
2018-11-22 02:00:46 -05:00
|
|
|
WHERE username_lower IN (:names) AND staged = false
|
2018-11-22 01:28:48 -05:00
|
|
|
)
|
|
|
|
UNION
|
|
|
|
(
|
|
|
|
SELECT
|
|
|
|
:group_type AS type,
|
2018-11-26 10:34:56 -05:00
|
|
|
lower(name) AS name
|
2018-11-22 01:28:48 -05:00
|
|
|
FROM groups
|
|
|
|
)
|
2020-02-18 15:45:02 -05:00
|
|
|
UNION
|
|
|
|
(
|
|
|
|
SELECT
|
|
|
|
:group_mentionable_type AS type,
|
|
|
|
lower(name) AS name
|
|
|
|
FROM groups
|
|
|
|
WHERE lower(name) IN (:names) AND (#{Group.mentionable_sql_clause(include_public: false)})
|
|
|
|
)
|
|
|
|
ORDER BY type
|
2018-11-22 01:28:48 -05:00
|
|
|
SQL
|
|
|
|
|
2018-11-22 03:01:03 -05:00
|
|
|
user = User.find_by(id: user_id)
|
2018-11-22 03:32:56 -05:00
|
|
|
names.each(&:downcase!)
|
2018-11-22 03:01:03 -05:00
|
|
|
|
2018-11-22 01:28:48 -05:00
|
|
|
results = DB.query(sql,
|
|
|
|
names: names,
|
|
|
|
user_type: USER_TYPE,
|
2018-11-22 03:01:03 -05:00
|
|
|
group_type: GROUP_TYPE,
|
2020-02-18 15:45:02 -05:00
|
|
|
group_mentionable_type: GROUP_MENTIONABLE_TYPE,
|
2018-11-22 03:01:03 -05:00
|
|
|
levels: Group.alias_levels(user),
|
|
|
|
user_id: user_id
|
2018-11-22 01:28:48 -05:00
|
|
|
)
|
|
|
|
|
|
|
|
mentions = {}
|
|
|
|
results.each { |result| mentions[result.name] = result.type }
|
|
|
|
mentions
|
|
|
|
end
|
|
|
|
|
2013-02-05 14:16:51 -05:00
|
|
|
end
|