semtex

defmodule Semtex do
  @config %{
    "allowed_attributes" => %{
      "all_tags" => ["lang", "title"],
      "by_tag" => %{
        "a" => ["href", "hreflang"],
        "bdo" => ["dir"],
        "blockquote" => ["cite"],
        "col" => ["align", "char", "charoff", "span"],
        "colgroup" => ["align", "char", "charoff", "span"],
        "del" => ["cite", "datetime"],
        "hr" => ["align", "size", "width"],
        "img" => ["align", "alt", "height", "src", "width"],
        "ins" => ["cite", "datetime"],
        "ol" => ["start"],
        "q" => ["cite"],
        "table" => ["align", "char", "charoff", "summary"],
        "tbody" => ["align", "char", "charoff"],
        "td" => ["align", "char", "charoff", "colspan", "headers", "rowspan"],
        "tfoot" => ["align", "char", "charoff"],
        "th" => ["align", "char", "charoff", "colspan", "headers", "rowspan", "scope"],
        "thead" => ["align", "char", "charoff"],
        "tr" => ["align", "char", "charoff"]
      }
    },
    "allowed_tags" => [
      "a",
      "abbr",
      "acronym",
      "area",
      "article",
      "aside",
      "b",
      "bdi",
      "bdo",
      "blockquote",
      "br",
      "caption",
      "center",
      "cite",
      "code",
      "col",
      "colgroup",
      "data",
      "dd",
      "del",
      "details",
      "dfn",
      "div",
      "dl",
      "dt",
      "em",
      "figcaption",
      "figure",
      "footer",
      "h1",
      "h2",
      "h3",
      "h4",
      "h5",
      "h6",
      "header",
      "hgroup",
      "hr",
      "i",
      "img",
      "ins",
      "kbd",
      "kbd",
      "li",
      "map",
      "mark",
      "nav",
      "ol",
      "p",
      "pre",
      "q",
      "rp",
      "rt",
      "rtc",
      "ruby",
      "s",
      "samp",
      "small",
      "span",
      "strike",
      "strong",
      "sub",
      "summary",
      "sup",
      "table",
      "tbody",
      "td",
      "th",
      "thead",
      "time",
      "tr",
      "tt",
      "u",
      "ul",
      "var",
      "wbr"
    ],
    "url_encode_attributes" => ["href", "src"],
    "allowed_url_scheme_attributes" => ["href", "src"],
    "allowed_url_schemes" => [
      "bitcoin",
      "ftp",
      "ftps",
      "geo",
      "http",
      "https",
      "im",
      "irc",
      "ircs",
      "magnet",
      "mailto",
      "mms",
      "mx",
      "news",
      "nntp",
      "openpgp4fpr",
      "sip",
      "sms",
      "smsto",
      "ssh",
      "tel",
      "url",
      "webcal",
      "wtai",
      "xmpp"
    ],
    "anchor_rel_values" => ["noopener", "noreferrer", "nofollow"],
    "strip_comments" => true,
    "escape_serialization" => true
  }

  def config() do
    @config
  end

  def sanitize_str_fragment!(raw_html, config \\ @config) when is_binary(raw_html) do
    raw_html
    |> String.trim()
    |> Semtex.Parser.parse_fragment!(config)
    |> Semtex.Sanitizer.sanitize!(config)
    |> Semtex.Serializer.serialize!(config)
  end
end