mirror of https://github.com/docusealco/docuseal
You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
133 lines
3.1 KiB
133 lines
3.1 KiB
# frozen_string_literal: true
|
|
|
|
module MarkdownToHtml
|
|
TAGS = {
|
|
'' => %w[<em> </em>],
|
|
'*' => %w[<strong> </strong>],
|
|
'~' => %w[<s> </s>]
|
|
}.freeze
|
|
|
|
INLINE_TOKENIZER = /(\[)|(\]\(([^)]+?)\))|(?:`([^`].*?)`)|(\*\*\*|\*\*|\*|~~)/
|
|
|
|
ALLOWED_TAGS = %w[p br strong b em i u a].freeze
|
|
ALLOWED_ATTRIBUTES = %w[href].freeze
|
|
|
|
module_function
|
|
|
|
def call(markdown)
|
|
return '' if markdown.blank?
|
|
|
|
text = auto_link_urls(markdown)
|
|
html = render_markdown(text)
|
|
|
|
ActionController::Base.helpers.sanitize(html, tags: ALLOWED_TAGS, attributes: ALLOWED_ATTRIBUTES)
|
|
end
|
|
|
|
BRACKETS = { ')' => '(', ']' => '[', '}' => '{' }.freeze
|
|
|
|
def auto_link_urls(text)
|
|
link_parts = text.split(%r{((?:https?://|www\.)[^\s<>\u00A0"]+)})
|
|
|
|
link_parts.map.with_index do |part, index|
|
|
if part.match?(%r{\A(?:https?://|www\.)}) && !(index > 0 && link_parts[index - 1]&.match?(/\]\(\s*\z/))
|
|
url_part = part.dup
|
|
punctuation = []
|
|
|
|
while url_part.sub!(%r{[^\p{Word}/\-=;]\z}, '')
|
|
punctuation.push(::Regexp.last_match(0))
|
|
|
|
opening = BRACKETS[punctuation.last]
|
|
|
|
next unless opening && url_part.count(opening) > url_part.count(punctuation.last)
|
|
|
|
url_part << punctuation.pop
|
|
|
|
break
|
|
end
|
|
|
|
trail = punctuation.reverse.join
|
|
url = url_part.start_with?('www.') ? "https://#{url_part}" : url_part
|
|
|
|
"[#{url_part}](#{url})#{trail}"
|
|
else
|
|
part
|
|
end
|
|
end.join
|
|
end
|
|
|
|
def render_markdown(text)
|
|
text = text.gsub(/\+\+([^+]+)\+\+/, '<u>\1</u>')
|
|
|
|
paragraphs = text.split(/\n{2,}/)
|
|
|
|
html = paragraphs.filter_map do |para|
|
|
content = para.strip
|
|
|
|
next if content.empty?
|
|
|
|
next '<p><br></p>' if [' ', '&nbsp;'].include?(content)
|
|
|
|
content = content.gsub(/ *\n/, '<br>')
|
|
|
|
"<p>#{parse_inline(content)}</p>"
|
|
end.join
|
|
|
|
html.presence || '<p></p>'
|
|
end
|
|
|
|
# rubocop:disable Metrics
|
|
def parse_inline(text)
|
|
context = []
|
|
out = ''
|
|
last = 0
|
|
|
|
tag = lambda do |t|
|
|
desc = TAGS[t[1] || '']
|
|
|
|
return t unless desc
|
|
|
|
is_end = context.last == t
|
|
is_end ? context.pop : context.push(t)
|
|
desc[is_end ? 1 : 0]
|
|
end
|
|
|
|
flush = lambda do
|
|
str = ''
|
|
str += tag.call(context.last) while context.any?
|
|
str
|
|
end
|
|
|
|
while last <= text.length && (m = INLINE_TOKENIZER.match(text, last))
|
|
prev = text[last...m.begin(0)]
|
|
last = m.end(0)
|
|
chunk = m[0]
|
|
|
|
if m[4]
|
|
chunk = "<code>#{ERB::Util.html_escape(m[4])}</code>"
|
|
elsif m[2]
|
|
out = out.sub(/\A(.*)<a>/m, "\\1<a href=\"#{ERB::Util.html_escape(m[3])}\">")
|
|
out = out.gsub('<a>', '[')
|
|
chunk = "#{flush.call}</a>"
|
|
elsif m[1]
|
|
chunk = '<a>'
|
|
elsif m[5]
|
|
chunk =
|
|
if m[5] == '***'
|
|
if context.include?('*') && context.include?('**')
|
|
tag.call('*') + tag.call('**')
|
|
else
|
|
tag.call('**') + tag.call('*')
|
|
end
|
|
else
|
|
tag.call(m[5])
|
|
end
|
|
end
|
|
|
|
out += prev.to_s + chunk
|
|
end
|
|
|
|
(out + text[last..].to_s + flush.call).gsub('<a>', '[')
|
|
end
|
|
# rubocop:enable Metrics
|
|
end
|