tags with newlines, to help with line-wise postprocessing
- # and ensure markdown paragraphs
- doc.xpath('//p').each do |p|
- p.before(new_linebreak_node doc)
- p.after(new_linebreak_node doc, 2)
- end
- end
-
- def new_linebreak_node(doc, count=1)
- Nokogiri::XML::Text.new("\n" * count, doc)
- end
-
- def trim_process_node(node)
- if should_hoist?(node)
- hoisted = trim_hoist_element node
- hoisted.each { |child| trim_process_node child }
- elsif should_delete?(node)
- node.remove
- else
- if children = node.children
- children.each { |child| trim_process_node child }
- end
- end
-
- node
- end
-
- def trim_hoist_element(element)
- hoisted = []
- element.children.each do |child|
- element.before(child)
- hoisted << child
- end
- element.remove
- hoisted
- end
-
- def should_hoist?(node)
- return false unless node.element?
- HTML_HOIST_ELEMENTS.include? node.name
- end
-
- def should_delete?(node)
- return true if HTML_DELETE_ELEMENT_TYPES.include? node.type
- return true if node.element? && node.name == 'head'
- return true if node.text? && node.text.strip.blank?
-
- false
- end
- end
-end
diff --git a/lib/email/receiver.rb b/lib/email/receiver.rb
index b629fdb5587..f87ca1bab64 100644
--- a/lib/email/receiver.rb
+++ b/lib/email/receiver.rb
@@ -1,7 +1,7 @@
require "digest"
require_dependency "new_post_manager"
require_dependency "post_action_creator"
-require_dependency "email/html_cleaner"
+require_dependency "html_to_markdown"
module Email
@@ -188,18 +188,18 @@ module Email
text = fix_charset(@mail)
end
- if html.present? && (SiteSetting.incoming_email_prefer_html || text.blank?)
- html = Email::HtmlCleaner.new(html).output_html
- html = trim_discourse_markers(html)
- html, elided = EmailReplyTrimmer.trim(html, true)
- return [html, elided]
- end
-
if text.present?
text = trim_discourse_markers(text)
text, elided = EmailReplyTrimmer.trim(text, true)
return [text, elided]
end
+
+ if html.present?
+ markdown = HtmlToMarkdown.new(html).to_markdown
+ markdown = trim_discourse_markers(markdown)
+ markdown, elided = EmailReplyTrimmer.trim(markdown, true)
+ return [markdown, elided]
+ end
end
def fix_charset(mail_part)
diff --git a/lib/html_to_markdown.rb b/lib/html_to_markdown.rb
index 7cfa0d6df01..6a5112c8270 100644
--- a/lib/html_to_markdown.rb
+++ b/lib/html_to_markdown.rb
@@ -115,7 +115,7 @@ class HtmlToMarkdown
RUBY
end
- WHITELISTED ||= %w{del ins kbd s small strike sub sup table tbody td tfoot th thead tr}
+ WHITELISTED ||= %w{del ins kbd s small strike sub sup}
WHITELISTED.each do |tag|
class_eval <<-RUBY
def visit_#{tag}(node)
diff --git a/spec/components/email/receiver_spec.rb b/spec/components/email/receiver_spec.rb
index 5f0ea2fcf63..e00fcc353a8 100644
--- a/spec/components/email/receiver_spec.rb
+++ b/spec/components/email/receiver_spec.rb
@@ -157,7 +157,7 @@ describe Email::Receiver do
expect(topic.posts.last.cooked).not_to match(/
HTML reply ;)")
+ expect(topic.posts.last.raw).to eq("This is a **HTML** reply ;)")
expect { process(:hebrew_reply) }.to change { topic.posts.count }
expect(topic.posts.last.raw).to eq("שלום! מה שלומך היום?")
@@ -174,18 +174,6 @@ describe Email::Receiver do
expect(topic.posts.last.raw).to eq("This is the *text* part.")
end
- it "prefers html over text when site setting is enabled" do
- SiteSetting.incoming_email_prefer_html = true
- expect { process(:text_and_html_reply) }.to change { topic.posts.count }
- expect(topic.posts.last.raw).to eq('This is the html part.')
- end
-
- it "uses text when prefer_html site setting is enabled but no html is available" do
- SiteSetting.incoming_email_prefer_html = true
- expect { process(:text_reply) }.to change { topic.posts.count }
- expect(topic.posts.last.raw).to eq("This is a text reply :)")
- end
-
it "removes the 'on