Page MenuHomePhorge
Files F85710329

No OneTemporary
Actions

Size
15 KB
Referenced Files
None
Subscribers
None

View Options

diff --git a/lib/pleroma/web/rich_media/parser/card.ex b/lib/pleroma/web/rich_media/parser/card.ex
index 2e06aa562..1eb3ddbdd 100644
--- a/lib/pleroma/web/rich_media/parser/card.ex
+++ b/lib/pleroma/web/rich_media/parser/card.ex
@@ -1,112 +1,112 @@
# Pleroma: A lightweight social networking server
# Copyright © 2017-2021 Pleroma Authors <https://pleroma.social/>
# SPDX-License-Identifier: AGPL-3.0-only
defmodule Pleroma.Web.RichMedia.Parser.Card do
alias Pleroma.Web.RichMedia.Parser.Card
alias Pleroma.Web.RichMedia.Parser.Embed
@types ["link", "photo", "video", "rich"]
# https://docs.joinmastodon.org/entities/card/
defstruct url: nil,
title: nil,
description: "",
type: "link",
author_name: "",
author_url: "",
provider_name: "",
provider_url: "",
html: "",
width: 0,
height: 0,
image: nil,
embed_url: "",
blurhash: nil
def parse(%{url: url, oembed: %{"type" => type, "title" => title} = oembed} = embed)
when type in @types do
%Card{
url: url,
title: title,
description: get_description(embed),
type: oembed["type"],
author_name: oembed["author_name"],
author_url: oembed["author_url"],
provider_name: oembed["provider_name"],
provider_url: oembed["provider_url"],
html: oembed["html"],
width: oembed["width"],
height: oembed["height"],
image: oembed["thumbnail_url"] |> proxy(),
embed_url: oembed["url"] |> proxy()
}
|> validate()
end
def parse(%{url: url} = embed) do
%Card{
url: url,
title: get_title(embed),
description: get_description(embed),
type: "link",
image: get_image(embed) |> proxy()
}
|> validate()
end
def parse(card), do: {:error, {:invalid_metadata, card}}
defp get_title(embed) do
case embed do
%{meta: %{"twitter:title" => title}} when is_binary(title) and title != "" -> title
%{meta: %{"og:title" => title}} when is_binary(title) and title != "" -> title
%{title: title} when is_binary(title) and title != "" -> title
_ -> nil
end
end
defp get_description(%{meta: meta}) do
case meta do
%{"twitter:description" => desc} when is_binary(desc) and desc != "" -> desc
%{"og:description" => desc} when is_binary(desc) and desc != "" -> desc
%{"description" => desc} when is_binary(desc) and desc != "" -> desc
_ -> ""
end
end
defp get_image(%{meta: meta}) do
case meta do
%{"twitter:image" => image} when is_binary(image) and image != "" -> image
%{"og:image" => image} when is_binary(image) and image != "" -> image
_ -> ""
end
end
def to_map(%Card{} = card) do
card
|> Map.from_struct()
|> stringify_keys()
end
def to_map(%{} = card), do: stringify_keys(card)
defp stringify_keys(%{} = map), do: Map.new(map, fn {k, v} -> {Atom.to_string(k), v} end)
defp proxy(url) when is_binary(url), do: Pleroma.Web.MediaProxy.url(url)
defp proxy(_), do: nil
def validate(%Card{type: type, title: title} = card)
when type in @types and is_binary(title) and title != "" do
{:ok, card}
end
def validate(%Embed{} = embed) do
case Card.parse(embed) do
- %Card{} = card -> validate(card)
+ {:ok, %Card{} = card} -> validate(card)
card -> {:error, {:invalid_metadata, card}}
end
end
def validate(card), do: {:error, {:invalid_metadata, card}}
end
diff --git a/lib/pleroma/web/rich_media/parser/meta_tags.ex b/lib/pleroma/web/rich_media/parser/meta_tags.ex
index e5c6b448d..ea544b110 100644
--- a/lib/pleroma/web/rich_media/parser/meta_tags.ex
+++ b/lib/pleroma/web/rich_media/parser/meta_tags.ex
@@ -1,30 +1,39 @@
# Pleroma: A lightweight social networking server
# Copyright © 2017-2021 Pleroma Authors <https://pleroma.social/>
# SPDX-License-Identifier: AGPL-3.0-only
defmodule Pleroma.Web.RichMedia.Parser.MetaTags do
@doc """
Parses a `Floki.html_tree/0` and returns a map of raw `<meta>` tag values.
"""
@spec parse(html_tree :: Floki.html_tree()) :: map()
def parse(html_tree) do
html_tree
|> Floki.find("meta")
|> Enum.reduce(%{}, fn html_node, acc ->
case parse_node(html_node) do
{:ok, {name, content}} -> Map.put(acc, name, content)
_ -> acc
end
end)
+ |> clean_data()
end
defp parse_node({_tag, attrs, _children}) when is_list(attrs) do
case Map.new(attrs) do
%{"name" => name, "content" => content} -> {:ok, {name, content}}
%{"property" => name, "content" => content} -> {:ok, {name, content}}
_ -> {:error, :invalid_meta_tag}
end
end
defp parse_node(_), do: {:error, :invalid_meta_tag}
+
+ defp clean_data(data) do
+ data
+ |> Enum.reject(fn {key, val} ->
+ not match?({:ok, _}, Jason.encode(%{key => val}))
+ end)
+ |> Map.new()
+ end
end
diff --git a/lib/pleroma/web/rich_media/parsers/twitter_card.ex b/lib/pleroma/web/rich_media/parsers/twitter_card.ex
index a892d16ea..4785ac6df 100644
--- a/lib/pleroma/web/rich_media/parsers/twitter_card.ex
+++ b/lib/pleroma/web/rich_media/parsers/twitter_card.ex
@@ -1,23 +1,24 @@
# Pleroma: A lightweight social networking server
# Copyright © 2017-2021 Pleroma Authors <https://pleroma.social/>
# SPDX-License-Identifier: AGPL-3.0-only
defmodule Pleroma.Web.RichMedia.Parsers.TwitterCard do
alias Pleroma.Web.RichMedia.Parser.MetaTags
@spec parse(Floki.html_tree(), map()) :: map()
def parse(html, data) do
data
|> Map.put(:title, get_page_title(html))
|> Map.put(:meta, MetaTags.parse(html))
end
def get_page_title(html) do
with [node | _] <- Floki.find(html, "html head title"),
- title when is_binary(title) and title != "" <- Floki.text(node) do
+ title when is_binary(title) and title != "" <- Floki.text(node),
+ true <- String.valid?(title) do
title
else
_ -> nil
end
end
end
diff --git a/test/pleroma/web/rich_media/parser_test.exs b/test/pleroma/web/rich_media/parser_test.exs
index 2f363b012..f7d010219 100644
--- a/test/pleroma/web/rich_media/parser_test.exs
+++ b/test/pleroma/web/rich_media/parser_test.exs
@@ -1,176 +1,184 @@
# Pleroma: A lightweight social networking server
# Copyright © 2017-2021 Pleroma Authors <https://pleroma.social/>
# SPDX-License-Identifier: AGPL-3.0-only
defmodule Pleroma.Web.RichMedia.ParserTest do
use ExUnit.Case, async: true
alias Pleroma.Web.RichMedia.Parser
+ alias Pleroma.Web.RichMedia.Parser.Embed
setup do
Tesla.Mock.mock(fn
%{
method: :get,
url: "http://example.com/ogp"
} ->
%Tesla.Env{status: 200, body: File.read!("test/fixtures/rich_media/ogp.html")}
%{
method: :get,
url: "http://example.com/non-ogp"
} ->
%Tesla.Env{status: 200, body: File.read!("test/fixtures/rich_media/non_ogp_embed.html")}
%{
method: :get,
url: "http://example.com/ogp-missing-title"
} ->
%Tesla.Env{
status: 200,
body: File.read!("test/fixtures/rich_media/ogp-missing-title.html")
}
%{
method: :get,
url: "http://example.com/twitter-card"
} ->
%Tesla.Env{status: 200, body: File.read!("test/fixtures/rich_media/twitter_card.html")}
%{
method: :get,
url: "http://example.com/oembed"
} ->
%Tesla.Env{status: 200, body: File.read!("test/fixtures/rich_media/oembed.html")}
%{
method: :get,
url: "http://example.com/oembed.json"
} ->
%Tesla.Env{status: 200, body: File.read!("test/fixtures/rich_media/oembed.json")}
%{method: :get, url: "http://example.com/empty"} ->
%Tesla.Env{status: 200, body: "hello"}
%{method: :get, url: "http://example.com/malformed"} ->
%Tesla.Env{status: 200, body: File.read!("test/fixtures/rich_media/malformed-data.html")}
%{method: :get, url: "http://example.com/error"} ->
{:error, :overload}
%{
method: :head,
url: "http://example.com/huge-page"
} ->
%Tesla.Env{
status: 200,
headers: [{"content-length", "2000001"}, {"content-type", "text/html"}]
}
%{
method: :head,
url: "http://example.com/pdf-file"
} ->
%Tesla.Env{
status: 200,
headers: [{"content-length", "1000000"}, {"content-type", "application/pdf"}]
}
%{method: :head} ->
%Tesla.Env{status: 404, body: "", headers: []}
end)
:ok
end
test "returns error when no metadata present" do
assert {:error, _} = Parser.parse("http://example.com/empty")
end
- test "doesn't just add a title" do
- assert {:error, {:invalid_metadata, _}} = Parser.parse("http://example.com/non-ogp")
- end
-
test "parses ogp" do
- assert Parser.parse("http://example.com/ogp") ==
- {:ok,
- %{
- "image" => "http://ia.media-imdb.com/images/rock.jpg",
- "title" => "The Rock",
- "description" =>
- "Directed by Michael Bay. With Sean Connery, Nicolas Cage, Ed Harris, John Spencer.",
- "type" => "video.movie",
- "url" => "http://example.com/ogp"
- }}
+ url = "http://example.com/ogp"
+
+ expected = %Embed{
+ meta: %{
+ "og:image" => "http://ia.media-imdb.com/images/rock.jpg",
+ "og:title" => "The Rock",
+ "og:description" =>
+ "Directed by Michael Bay. With Sean Connery, Nicolas Cage, Ed Harris, John Spencer.",
+ "og:type" => "video.movie",
+ "og:url" => "http://www.imdb.com/title/tt0117500/"
+ },
+ oembed: nil,
+ title: "The Rock (1996)",
+ url: "http://example.com/ogp"
+ }
+
+ assert Parser.parse(url) == {:ok, expected}
end
- test "falls back to <title> when ogp:title is missing" do
- assert Parser.parse("http://example.com/ogp-missing-title") ==
- {:ok,
- %{
- "image" => "http://ia.media-imdb.com/images/rock.jpg",
- "title" => "The Rock (1996)",
- "description" =>
- "Directed by Michael Bay. With Sean Connery, Nicolas Cage, Ed Harris, John Spencer.",
- "type" => "video.movie",
- "url" => "http://example.com/ogp-missing-title"
- }}
+ test "gets <title> tag" do
+ url = "http://example.com/ogp-missing-title"
+ expected = "The Rock (1996)"
+ assert {:ok, %Embed{title: ^expected}} = Parser.parse(url)
end
test "parses twitter card" do
- assert Parser.parse("http://example.com/twitter-card") ==
- {:ok,
- %{
- "card" => "summary",
- "site" => "@flickr",
- "image" => "https://farm6.staticflickr.com/5510/14338202952_93595258ff_z.jpg",
- "title" => "Small Island Developing States Photo Submission",
- "description" => "View the album on Flickr.",
- "url" => "http://example.com/twitter-card"
- }}
+ url = "http://example.com/twitter-card"
+
+ expected = %Embed{
+ meta: %{
+ "twitter:card" => "summary",
+ "twitter:description" => "View the album on Flickr.",
+ "twitter:image" => "https://farm6.staticflickr.com/5510/14338202952_93595258ff_z.jpg",
+ "twitter:site" => "@flickr",
+ "twitter:title" => "Small Island Developing States Photo Submission"
+ },
+ oembed: nil,
+ title: nil,
+ url: "http://example.com/twitter-card"
+ }
+
+ assert Parser.parse(url) == {:ok, expected}
end
test "parses OEmbed" do
- assert Parser.parse("http://example.com/oembed") ==
- {:ok,
- %{
- "author_name" => "‮‭‬bees‬",
- "author_url" => "https://www.flickr.com/photos/bees/",
- "cache_age" => 3600,
- "flickr_type" => "photo",
- "height" => "768",
- "html" =>
- "<a data-flickr-embed=\"true\" href=\"https://www.flickr.com/photos/bees/2362225867/\" title=\"Bacon Lollys by ‮‭‬bees‬, on Flickr\"><img src=\"https://farm4.staticflickr.com/3040/2362225867_4a87ab8baf_b.jpg\" width=\"1024\" height=\"768\" alt=\"Bacon Lollys\"></a><script async src=\"https://embedr.flickr.com/assets/client-code.js\" charset=\"utf-8\"></script>",
- "license" => "All Rights Reserved",
- "license_id" => 0,
- "provider_name" => "Flickr",
- "provider_url" => "https://www.flickr.com/",
- "thumbnail_height" => 150,
- "thumbnail_url" =>
- "https://farm4.staticflickr.com/3040/2362225867_4a87ab8baf_q.jpg",
- "thumbnail_width" => 150,
- "title" => "Bacon Lollys",
- "type" => "photo",
- "url" => "http://example.com/oembed",
- "version" => "1.0",
- "web_page" => "https://www.flickr.com/photos/bees/2362225867/",
- "web_page_short_url" => "https://flic.kr/p/4AK2sc",
- "width" => "1024"
- }}
+ url = "http://example.com/oembed"
+
+ expected = %Embed{
+ meta: %{},
+ oembed: %{
+ "author_name" => "‮‭‬bees‬",
+ "author_url" => "https://www.flickr.com/photos/bees/",
+ "cache_age" => 3600,
+ "flickr_type" => "photo",
+ "height" => "768",
+ "html" =>
+ "<a data-flickr-embed=\"true\" href=\"https://www.flickr.com/photos/bees/2362225867/\" title=\"Bacon Lollys by ‮‭‬bees‬, on Flickr\"><img src=\"https://farm4.staticflickr.com/3040/2362225867_4a87ab8baf_b.jpg\" width=\"1024\" height=\"768\" alt=\"Bacon Lollys\"></a><script async src=\"https://embedr.flickr.com/assets/client-code.js\" charset=\"utf-8\"></script>",
+ "license" => "All Rights Reserved",
+ "license_id" => 0,
+ "provider_name" => "Flickr",
+ "provider_url" => "https://www.flickr.com/",
+ "thumbnail_height" => 150,
+ "thumbnail_url" => "https://farm4.staticflickr.com/3040/2362225867_4a87ab8baf_q.jpg",
+ "thumbnail_width" => 150,
+ "title" => "Bacon Lollys",
+ "type" => "photo",
+ "url" => "https://farm4.staticflickr.com/3040/2362225867_4a87ab8baf_b.jpg",
+ "version" => "1.0",
+ "web_page" => "https://www.flickr.com/photos/bees/2362225867/",
+ "web_page_short_url" => "https://flic.kr/p/4AK2sc",
+ "width" => "1024"
+ },
+ url: "http://example.com/oembed"
+ }
+
+ assert Parser.parse(url) == {:ok, expected}
end
test "rejects invalid OGP data" do
assert {:error, _} = Parser.parse("http://example.com/malformed")
end
test "returns error if getting page was not successful" do
assert {:error, :overload} = Parser.parse("http://example.com/error")
end
test "does a HEAD request to check if the body is too large" do
assert {:error, :body_too_large} = Parser.parse("http://example.com/huge-page")
end
test "does a HEAD request to check if the body is html" do
assert {:error, {:content_type, _}} = Parser.parse("http://example.com/pdf-file")
end
end

File Metadata

Mime Type
text/x-diff
Expires
Fri, Sep 18, 11:26 PM (6 h, 58 s)
Storage Engine
blob
Storage Format
Raw Data
Storage Handle
1768450
Default Alt Text
(15 KB)

Event Timeline