X-Git-Url: http://git.squeep.com/?a=blobdiff_plain;f=lib%2Fpleroma%2Fweb%2Frich_media%2Fhelpers.ex;h=2fb482b51806a7507ccff42de9738270ccfc97ec;hb=119e11f455e59bcc5ae890c24bb0fd711f660ed1;hp=44e876777c92ce93b1cbdd4c1ae3a9c6fb466130;hpb=fb9730010e794a7f80c94b4320fd0591f2e7c93c;p=akkoma

diff --git a/lib/pleroma/web/rich_media/helpers.ex b/lib/pleroma/web/rich_media/helpers.ex
index 44e876777..2fb482b51 100644
--- a/lib/pleroma/web/rich_media/helpers.ex
+++ b/lib/pleroma/web/rich_media/helpers.ex
@@ -1,19 +1,101 @@
 # Pleroma: A lightweight social networking server
-# Copyright _ 2017-2019 Pleroma Authors <https://pleroma.social/>
+# Copyright _ 2017-2020 Pleroma Authors <https://pleroma.social/>
 # SPDX-License-Identifier: AGPL-3.0-only
 
 defmodule Pleroma.Web.RichMedia.Helpers do
-  alias Pleroma.{Activity, Object, HTML}
+  alias Pleroma.Activity
+  alias Pleroma.Config
+  alias Pleroma.HTML
+  alias Pleroma.Object
   alias Pleroma.Web.RichMedia.Parser
 
-  def fetch_data_for_activity(%Activity{} = activity) do
-    with true <- Pleroma.Config.get([:rich_media, :enabled], true),
-         %Object{} = object <- Object.normalize(activity.data["object"]),
-         {:ok, page_url} <- HTML.extract_first_external_url(object, object.data["content"]),
+  @rich_media_options [
+    pool: :media,
+    max_body: 2_000_000
+  ]
+
+  @spec validate_page_url(URI.t() | binary()) :: :ok | :error
+  defp validate_page_url(page_url) when is_binary(page_url) do
+    validate_tld = Pleroma.Config.get([Pleroma.Formatter, :validate_tld])
+
+    page_url
+    |> Linkify.Parser.url?(validate_tld: validate_tld)
+    |> parse_uri(page_url)
+  end
+
+  defp validate_page_url(%URI{host: host, scheme: "https", authority: authority})
+       when is_binary(authority) do
+    cond do
+      host in Config.get([:rich_media, :ignore_hosts], []) ->
+        :error
+
+      get_tld(host) in Config.get([:rich_media, :ignore_tld], []) ->
+        :error
+
+      true ->
+        :ok
+    end
+  end
+
+  defp validate_page_url(_), do: :error
+
+  defp parse_uri(true, url) do
+    url
+    |> URI.parse()
+    |> validate_page_url
+  end
+
+  defp parse_uri(_, _), do: :error
+
+  defp get_tld(host) do
+    host
+    |> String.split(".")
+    |> Enum.reverse()
+    |> hd
+  end
+
+  def fetch_data_for_object(object) do
+    with true <- Config.get([:rich_media, :enabled]),
+         false <- object.data["sensitive"] || false,
+         {:ok, page_url} <-
+           HTML.extract_first_external_url(object, object.data["content"]),
+         :ok <- validate_page_url(page_url),
          {:ok, rich_media} <- Parser.parse(page_url) do
       %{page_url: page_url, rich_media: rich_media}
     else
       _ -> %{}
     end
   end
+
+  def fetch_data_for_activity(%Activity{data: %{"type" => "Create"}} = activity) do
+    with true <- Config.get([:rich_media, :enabled]),
+         %Object{} = object <- Object.normalize(activity) do
+      fetch_data_for_object(object)
+    else
+      _ -> %{}
+    end
+  end
+
+  def fetch_data_for_activity(_), do: %{}
+
+  def perform(:fetch, %Activity{} = activity) do
+    fetch_data_for_activity(activity)
+    :ok
+  end
+
+  def rich_media_get(url) do
+    headers = [{"user-agent", Pleroma.Application.user_agent() <> "; Bot"}]
+
+    options =
+      if Application.get_env(:tesla, :adapter) == Tesla.Adapter.Hackney do
+        Keyword.merge(@rich_media_options,
+          recv_timeout: 2_000,
+          with_body: true
+        )
+      else
+        @rich_media_options
+      end
+
+    Pleroma.HTTP.get(url, headers, adapter: options)
+  end
 end