summaryrefslogtreecommitdiff
path: root/lib/pleroma/web/rich_media/parser.ex
blob: ca592833f3d7574973eb5776126741de3bb42daf (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
# Pleroma: A lightweight social networking server
# Copyright © 2017-2020 Pleroma Authors <https://pleroma.social/>
# SPDX-License-Identifier: AGPL-3.0-only

defmodule Pleroma.Web.RichMedia.Parser do
  defp parsers do
    Pleroma.Config.get([:rich_media, :parsers])
  end

  def parse(nil), do: {:error, "No URL provided"}

  if Pleroma.Config.get(:env) == :test do
    def parse(url), do: parse_url(url)
  else
    def parse(url) do
      try do
        Cachex.fetch!(:rich_media_cache, url, fn _ ->
          {:commit, parse_url(url)}
        end)
        |> set_ttl_based_on_image(url)
      rescue
        e ->
          {:error, "Cachex error: #{inspect(e)}"}
      end
    end
  end

  @doc """
  Set the rich media cache based on the expiration time of image.

  Adopt behaviour `Pleroma.Web.RichMedia.Parser.TTL`

  ## Example

      defmodule MyModule do
        @behaviour Pleroma.Web.RichMedia.Parser.TTL
        def ttl(data, url) do
          image_url = Map.get(data, :image)
          # do some parsing in the url and get the ttl of the image
          # and return ttl is unix time
          parse_ttl_from_url(image_url)
        end
      end

  Define the module in the config

      config :pleroma, :rich_media,
        ttl_setters: [MyModule]
  """
  def set_ttl_based_on_image({:ok, data}, url) do
    with {:ok, nil} <- Cachex.ttl(:rich_media_cache, url),
         ttl when is_number(ttl) <- get_ttl_from_image(data, url) do
      Cachex.expire_at(:rich_media_cache, url, ttl * 1000)
      {:ok, data}
    else
      _ ->
        {:ok, data}
    end
  end

  defp get_ttl_from_image(data, url) do
    Pleroma.Config.get([:rich_media, :ttl_setters])
    |> Enum.reduce({:ok, nil}, fn
      module, {:ok, _ttl} ->
        module.ttl(data, url)

      _, error ->
        error
    end)
  end

  defp parse_url(url) do
    try do
      {:ok, %Tesla.Env{body: html}} = Pleroma.Web.RichMedia.Helpers.rich_media_get(url)

      html
      |> parse_html()
      |> maybe_parse()
      |> Map.put("url", url)
      |> clean_parsed_data()
      |> check_parsed_data()
    rescue
      e ->
        {:error, "Parsing error: #{inspect(e)} #{inspect(__STACKTRACE__)}"}
    end
  end

  defp parse_html(html), do: Floki.parse_document!(html)

  defp maybe_parse(html) do
    Enum.reduce_while(parsers(), %{}, fn parser, acc ->
      case parser.parse(html, acc) do
        data when data != %{} -> {:halt, data}
        _ -> {:cont, acc}
      end
    end)
  end

  defp check_parsed_data(%{"title" => title} = data)
       when is_binary(title) and title != "" do
    {:ok, data}
  end

  defp check_parsed_data(data) do
    {:error, "Found metadata was invalid or incomplete: #{inspect(data)}"}
  end

  defp clean_parsed_data(data) do
    data
    |> Enum.reject(fn {key, val} ->
      not match?({:ok, _}, Jason.encode(%{key => val}))
    end)
    |> Map.new()
  end
end