Fix Rich Media Previews for updated activities

The Rich Media Previews were not regenerated when a post was updated due to a cache invalidation issue. They are now cached by the activity id so they can be evicted with the other activity cache objects in the :scrubber_cache.
This commit is contained in:
Mark Felder 2024-02-04 19:24:52 -05:00
commit 04fc4eddaa
8 changed files with 96 additions and 23 deletions

View file

@ -6,8 +6,6 @@ defmodule Pleroma.HTML do
# Scrubbers are compiled on boot so they can be configured in OTP releases
# @on_load :compile_scrubbers
@cachex Pleroma.Config.get([:cachex, :provider], Cachex)
def compile_scrubbers do
dir = Path.join(:code.priv_dir(:pleroma), "scrubbers")
@ -67,27 +65,20 @@ defmodule Pleroma.HTML do
end
end
def extract_first_external_url_from_object(%{data: %{"content" => content}} = object)
@spec extract_first_external_url_from_object(Pleroma.Object.t()) ::
{:ok, String.t()} | {:error, :no_content}
def extract_first_external_url_from_object(%{data: %{"content" => content}})
when is_binary(content) do
unless object.data["fake"] do
key = "URL|#{object.id}"
url =
content
|> Floki.parse_fragment!()
|> Floki.find("a:not(.mention,.hashtag,.attachment,[rel~=\"tag\"])")
|> Enum.take(1)
|> Floki.attribute("href")
|> Enum.at(0)
@cachex.fetch!(:scrubber_cache, key, fn _key ->
{:commit, {:ok, extract_first_external_url(content)}}
end)
else
{:ok, extract_first_external_url(content)}
end
{:ok, url}
end
def extract_first_external_url_from_object(_), do: {:error, :no_content}
def extract_first_external_url(content) do
content
|> Floki.parse_fragment!()
|> Floki.find("a:not(.mention,.hashtag,.attachment,[rel~=\"tag\"])")
|> Enum.take(1)
|> Floki.attribute("href")
|> Enum.at(0)
end
end