path: root/app/helpers/markup_helper.rb

# frozen_string_literal: true

require 'nokogiri'

module MarkupHelper
  include ActionView::Helpers::TextHelper
  include ActionView::Context

  def plain?(filename)
    Gitlab::MarkupHelper.plain?(filename)
  end

  def markup?(filename)
    Gitlab::MarkupHelper.markup?(filename)
  end

  def gitlab_markdown?(filename)
    Gitlab::MarkupHelper.gitlab_markdown?(filename)
  end

  def asciidoc?(filename)
    Gitlab::MarkupHelper.asciidoc?(filename)
  end

  # Use this in places where you would normally use link_to(gfm(...), ...).
  def link_to_markdown(body, url, html_options = {})
    return '' if body.blank?

    link_to_html(markdown(body, pipeline: :single_line), url, html_options)
  end

  def link_to_markdown_field(object, field, url, html_options = {})
    rendered_field = markdown_field(object, field)

    link_to_html(rendered_field, url, html_options)
  end

  # It solves a problem occurring with nested links (i.e.
  # "<a>outer text <a>gfm ref</a> more outer text</a>"). This will not be
  # interpreted as intended. Browsers will parse something like
  # "<a>outer text </a><a>gfm ref</a> more outer text" (notice the last part is
  # not linked any more). link_to_html corrects that. It wraps all parts to
  # explicitly produce the correct linking behavior (i.e.
  # "<a>outer text </a><a>gfm ref</a><a> more outer text</a>").
  def link_to_html(redacted, url, html_options = {})
    fragment = Nokogiri::HTML::DocumentFragment.parse(redacted)

    if fragment.children.size == 1 && fragment.children[0].name == 'a'
      # Fragment has only one node, and it's a link generated by `gfm`.
      # Replace it with our requested link.
      text = fragment.children[0].text
      fragment.children[0].replace(link_to(text, url, html_options))
    else
      # Traverse the fragment's first generation of children looking for
      # either pure text or emojis, wrapping anything found in the
      # requested link
      fragment.children.each do |node|
        if node.text?
          node.replace(link_to(node.text, url, html_options))
        elsif node.name == 'gl-emoji'
          node.replace(link_to(node.to_html.html_safe, url, html_options))
        end
      end
    end

    # Add any custom CSS classes to the GFM-generated reference links
    if html_options[:class]
      fragment.css('a.gfm').add_class(html_options[:class])
    end

    fragment.to_html.html_safe
  end

  # Return the first line of +text+, up to +max_chars+, after parsing the line
  # as Markdown.  HTML tags in the parsed output are not counted toward the
  # +max_chars+ limit.  If the length limit falls within a tag's contents, then
  # the tag contents are truncated without removing the closing tag.
  def first_line_in_markdown(object, attribute, max_chars = nil, options = {})
    md = markdown_field(object, attribute, options.merge(post_process: false))
    return unless md.present?

    tags = %w(a gl-emoji b strong i em pre code p span)
    tags << 'img' if options[:allow_images]

    text = truncate_visible(md, max_chars || md.length)
    text = prepare_for_rendering(text, markdown_field_render_context(object, attribute, options))
    text = sanitize(
      text,
      tags: tags,
      attributes: Rails::Html::WhiteListSanitizer.allowed_attributes +
          %w(style data-src data-name data-unicode-version data-iid data-project-path data-mr-title data-html)
    )

    # since <img> tags are stripped, this can leave empty <a> tags hanging around
    # (as our markdown wraps images in links)
    options[:allow_images] ? text : strip_empty_link_tags(text).html_safe
  end

  def markdown(text, context = {})
    return '' unless text.present?

    context[:project] ||= @project
    context[:group] ||= @group

    html = markdown_unsafe(text, context)
    prepare_for_rendering(html, context)
  end

  def markdown_field(object, field, context = {})
    object = object.for_display if object.respond_to?(:for_display)
    return '' unless object.present?

    redacted_field_html = object.try(:"redacted_#{field}_html")
    return redacted_field_html if redacted_field_html

    render_markdown_field(object, field, context)
  end

  def markup(file_name, text, context = {})
    context[:project] ||= @project
    html = context.delete(:rendered) || markup_unsafe(file_name, text, context)
    prepare_for_rendering(html, context)
  end

  def render_wiki_content(wiki_page, context = {})
    text = wiki_page.content
    return '' unless text.present?

    context.merge!(
      pipeline: :wiki,
      project: @project,
      wiki: @wiki,
      repository: @wiki.repository,
      page_slug: wiki_page.slug,
      issuable_state_filter_enabled: true
    )

    html = markup_unsafe(wiki_page.path, text, context)

    prepare_for_rendering(html, context)
  end

  def markup_unsafe(file_name, text, context = {})
    return '' unless text.present?

    if gitlab_markdown?(file_name)
      markdown_unsafe(text, context)
    elsif asciidoc?(file_name)
      asciidoc_unsafe(text, context)
    elsif plain?(file_name)
      plain_unsafe(text)
    else
      other_markup_unsafe(file_name, text, context)
    end
  rescue StandardError => e
    Gitlab::ErrorTracking.track_exception(e, project_id: @project&.id, file_name: file_name)

    simple_format(text)
  end

  # Returns the text necessary to reference `entity` across projects
  #
  # project - Project to reference
  # entity  - Object that responds to `to_reference`
  #
  # Examples:
  #
  #   cross_project_reference(project, project.issues.first)
  #   # => 'namespace1/project1#123'
  #
  #   cross_project_reference(project, project.merge_requests.first)
  #   # => 'namespace1/project1!345'
  #
  # Returns a String
  def cross_project_reference(project, entity)
    if entity.respond_to?(:to_reference)
      entity.to_reference(project, full: true)
    else
      ''
    end
  end

  private

  # Return +text+, truncated to +max_chars+ characters, excluding any HTML
  # tags.
  def truncate_visible(text, max_chars)
    doc = Nokogiri::HTML.fragment(text)
    content_length = 0
    truncated = false

    doc.traverse do |node|
      if node.text? || node.content.empty?
        if truncated
          node.remove
          next
        end

        # Handle line breaks within a node
        if node.content.strip.lines.length > 1
          node.content = "#{node.content.lines.first.chomp}..."
          truncated = true
        end

        num_remaining = max_chars - content_length
        if node.content.length > num_remaining
          node.content = node.content.truncate(num_remaining)
          truncated = true
        end

        content_length += node.content.length
      end

      truncated = truncate_if_block(node, truncated)
    end

    doc.to_html
  end

  # Used by #truncate_visible.  If +node+ is the first block element, and the
  # text hasn't already been truncated, then append "..." to the node contents
  # and return true.  Otherwise return false.
  def truncate_if_block(node, truncated)
    return true if truncated

    if node.element? && (node.description&.block? || node.matches?('pre > code > .line'))
      node.inner_html = "#{node.inner_html}..." if node.next_sibling
      true
    else
      truncated
    end
  end

  def strip_empty_link_tags(text)
    scrubber = Loofah::Scrubber.new do |node|
      node.remove if node.name == 'a' && node.children.empty?
    end

    sanitize text, scrubber: scrubber
  end

  def markdown_toolbar_button(options = {})
    data = options[:data].merge({ container: 'body' })
    content_tag :button,
      type: 'button',
      class: 'toolbar-btn js-md has-tooltip',
      tabindex: -1,
      data: data,
      title: options[:title],
      aria: { label: options[:title] } do
      sprite_icon(options[:icon])
    end
  end

  def markdown_unsafe(text, context = {})
    Banzai.render(text, context)
  end

  def asciidoc_unsafe(text, context = {})
    context.merge!(
      commit:         @commit,
      ref:            @ref,
      requested_path: @path
    )
    Gitlab::Asciidoc.render(text, context)
  end

  def plain_unsafe(text)
    content_tag :pre, class: 'plain-readme' do
      text
    end
  end

  def other_markup_unsafe(file_name, text, context = {})
    Gitlab::OtherMarkup.render(file_name, text, context)
  end

  def render_markdown_field(object, field, context = {})
    post_process = context.delete(:post_process)
    post_process = true if post_process.nil?

    html = Banzai.render_field(object, field, context)

    return html unless post_process

    prepare_for_rendering(html, markdown_field_render_context(object, field, context))
  end

  def markdown_field_render_context(object, field, base_context = {})
    return base_context unless object.respond_to?(:banzai_render_context)

    base_context.reverse_merge(object.banzai_render_context(field))
  end

  def prepare_for_rendering(html, context = {})
    return '' unless html.present?

    context.reverse_merge!(
      current_user: (current_user if defined?(current_user)),

      # RepositoryLinkFilter and UploadLinkFilter
      commit:         @commit,
      wiki:           @wiki,
      ref:            @ref,
      requested_path: @path
    )

    html = Banzai.post_process(html, context)

    Hamlit::RailsHelpers.preserve(html)
  end

  extend self
end