From 9172540f8d3965a40a65f4431873d903e47f4665 Mon Sep 17 00:00:00 2001 From: rinpatch Date: Mon, 10 Feb 2020 21:52:01 +0300 Subject: [PATCH 1/3] deps: update floki to 0.25.0 needed to make it use fast_html --- mix.exs | 2 +- mix.lock | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/mix.exs b/mix.exs index 3e3eac521..4083da01e 100644 --- a/mix.exs +++ b/mix.exs @@ -139,7 +139,7 @@ defp deps do {:phoenix_swoosh, "~> 0.2"}, {:gen_smtp, "~> 0.13"}, {:websocket_client, git: "https://github.com/jeremyong/websocket_client.git", only: :test}, - {:floki, "~> 0.23.0"}, + {:floki, "~> 0.25"}, {:ex_syslogger, github: "slashmili/ex_syslogger", tag: "1.4.0"}, {:timex, "~> 3.5"}, {:ueberauth, "~> 0.4"}, diff --git a/mix.lock b/mix.lock index 69eec5431..2855ac592 100644 --- a/mix.lock +++ b/mix.lock @@ -40,13 +40,13 @@ "fast_html": {:hex, :fast_html, "0.99.4", "d80812664f0429607e1d880fba0ef04da87a2e4fa596701bcaae17953535695c", [:make, :mix], [], "hexpm"}, "fast_sanitize": {:hex, :fast_sanitize, "0.1.4", "6c2e7203ca2f8275527a3021ba6e9d5d4ee213a47dc214a97c128737c9e56df1", [:mix], [{:fast_html, "~> 0.99", [hex: :fast_html, repo: "hexpm", optional: false]}, {:plug, "~> 1.8", [hex: :plug, repo: "hexpm", optional: false]}], "hexpm"}, "flake_id": {:hex, :flake_id, "0.1.0", "7716b086d2e405d09b647121a166498a0d93d1a623bead243e1f74216079ccb3", [:mix], [{:base62, "~> 1.2", [hex: :base62, repo: "hexpm", optional: false]}, {:ecto, ">= 2.0.0", [hex: :ecto, repo: "hexpm", optional: true]}], "hexpm"}, - "floki": {:hex, :floki, "0.23.1", "e100306ce7d8841d70a559748e5091542e2cfc67ffb3ade92b89a8435034dab1", [:mix], [{:html_entities, "~> 0.5.0", [hex: :html_entities, repo: "hexpm", optional: false]}], "hexpm"}, + "floki": {:hex, :floki, "0.25.0", "b1c9ddf5f32a3a90b43b76f3386ca054325dc2478af020e87b5111c19f2284ac", [:mix], [{:html_entities, "~> 0.5.0", [hex: :html_entities, repo: "hexpm", optional: false]}], "hexpm"}, "gen_smtp": {:hex, :gen_smtp, "0.15.0", "9f51960c17769b26833b50df0b96123605a8024738b62db747fece14eb2fbfcc", [:rebar3], [], "hexpm"}, "gen_stage": {:hex, :gen_stage, "0.14.3", "d0c66f1c87faa301c1a85a809a3ee9097a4264b2edf7644bf5c123237ef732bf", [:mix], [], "hexpm"}, "gen_state_machine": {:hex, :gen_state_machine, "2.0.5", "9ac15ec6e66acac994cc442dcc2c6f9796cf380ec4b08267223014be1c728a95", [:mix], [], "hexpm"}, "gettext": {:hex, :gettext, "0.17.1", "8baab33482df4907b3eae22f719da492cee3981a26e649b9c2be1c0192616962", [:mix], [], "hexpm"}, "hackney": {:hex, :hackney, "1.15.2", "07e33c794f8f8964ee86cebec1a8ed88db5070e52e904b8f12209773c1036085", [:rebar3], [{:certifi, "2.5.1", [hex: :certifi, repo: "hexpm", optional: false]}, {:idna, "6.0.0", [hex: :idna, repo: "hexpm", optional: false]}, {:metrics, "1.0.1", [hex: :metrics, repo: "hexpm", optional: false]}, {:mimerl, "~>1.1", [hex: :mimerl, repo: "hexpm", optional: false]}, {:ssl_verify_fun, "1.1.5", [hex: :ssl_verify_fun, repo: "hexpm", optional: false]}], "hexpm"}, - "html_entities": {:hex, :html_entities, "0.5.0", "40f5c5b9cbe23073b48a4e69c67b6c11974f623a76165e2b92d098c0e88ccb1d", [:mix], [], "hexpm"}, + "html_entities": {:hex, :html_entities, "0.5.1", "1c9715058b42c35a2ab65edc5b36d0ea66dd083767bef6e3edb57870ef556549", [:mix], [], "hexpm"}, "html_sanitize_ex": {:hex, :html_sanitize_ex, "1.3.0", "f005ad692b717691203f940c686208aa3d8ffd9dd4bb3699240096a51fa9564e", [:mix], [{:mochiweb, "~> 2.15", [hex: :mochiweb, repo: "hexpm", optional: false]}], "hexpm"}, "http_signatures": {:git, "https://git.pleroma.social/pleroma/http_signatures.git", "293d77bb6f4a67ac8bde1428735c3b42f22cbb30", [ref: "293d77bb6f4a67ac8bde1428735c3b42f22cbb30"]}, "httpoison": {:hex, :httpoison, "1.6.1", "2ce5bf6e535cd0ab02e905ba8c276580bab80052c5c549f53ddea52d72e81f33", [:mix], [{:hackney, "~> 1.15 and >= 1.15.2", [hex: :hackney, repo: "hexpm", optional: false]}], "hexpm"}, From 58299fcfb4c9f4ed63c2f791b6ebf9920ebb7659 Mon Sep 17 00:00:00 2001 From: rinpatch Date: Tue, 11 Feb 2020 00:16:13 +0300 Subject: [PATCH 2/3] deps: update fast_sanitize to 0.1.7 needed to get new fast_html version since 1.0.1 had an attribute parsing bug --- mix.lock | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/mix.lock b/mix.lock index 2855ac592..5c26bdd35 100644 --- a/mix.lock +++ b/mix.lock @@ -37,8 +37,8 @@ "ex_machina": {:hex, :ex_machina, "2.3.0", "92a5ad0a8b10ea6314b876a99c8c9e3f25f4dde71a2a835845b136b9adaf199a", [:mix], [{:ecto, "~> 2.2 or ~> 3.0", [hex: :ecto, repo: "hexpm", optional: true]}, {:ecto_sql, "~> 3.0", [hex: :ecto_sql, repo: "hexpm", optional: true]}], "hexpm"}, "ex_syslogger": {:git, "https://github.com/slashmili/ex_syslogger.git", "f3963399047af17e038897c69e20d552e6899e1d", [tag: "1.4.0"]}, "excoveralls": {:hex, :excoveralls, "0.12.1", "a553c59f6850d0aff3770e4729515762ba7c8e41eedde03208182a8dc9d0ce07", [:mix], [{:hackney, "~> 1.0", [hex: :hackney, repo: "hexpm", optional: false]}, {:jason, "~> 1.0", [hex: :jason, repo: "hexpm", optional: false]}], "hexpm"}, - "fast_html": {:hex, :fast_html, "0.99.4", "d80812664f0429607e1d880fba0ef04da87a2e4fa596701bcaae17953535695c", [:make, :mix], [], "hexpm"}, - "fast_sanitize": {:hex, :fast_sanitize, "0.1.4", "6c2e7203ca2f8275527a3021ba6e9d5d4ee213a47dc214a97c128737c9e56df1", [:mix], [{:fast_html, "~> 0.99", [hex: :fast_html, repo: "hexpm", optional: false]}, {:plug, "~> 1.8", [hex: :plug, repo: "hexpm", optional: false]}], "hexpm"}, + "fast_html": {:hex, :fast_html, "1.0.2", "b2a32022741699421e90762ce904cacb4faf12c10129acc3674262dd7fa5d2b6", [:make, :mix], [], "hexpm"}, + "fast_sanitize": {:hex, :fast_sanitize, "0.1.7", "2a7cd8734c88a2de6de55022104f8a3b87f1fdbe8bbf131d9049764b53d50d0d", [:mix], [{:fast_html, "~> 1.0", [hex: :fast_html, repo: "hexpm", optional: false]}, {:plug, "~> 1.8", [hex: :plug, repo: "hexpm", optional: false]}], "hexpm"}, "flake_id": {:hex, :flake_id, "0.1.0", "7716b086d2e405d09b647121a166498a0d93d1a623bead243e1f74216079ccb3", [:mix], [{:base62, "~> 1.2", [hex: :base62, repo: "hexpm", optional: false]}, {:ecto, ">= 2.0.0", [hex: :ecto, repo: "hexpm", optional: true]}], "hexpm"}, "floki": {:hex, :floki, "0.25.0", "b1c9ddf5f32a3a90b43b76f3386ca054325dc2478af020e87b5111c19f2284ac", [:mix], [{:html_entities, "~> 0.5.0", [hex: :html_entities, repo: "hexpm", optional: false]}], "hexpm"}, "gen_smtp": {:hex, :gen_smtp, "0.15.0", "9f51960c17769b26833b50df0b96123605a8024738b62db747fece14eb2fbfcc", [:rebar3], [], "hexpm"}, From ea1631d7e67e22eb49d608e066ef4a3555bf25f7 Mon Sep 17 00:00:00 2001 From: rinpatch Date: Tue, 11 Feb 2020 00:29:25 +0300 Subject: [PATCH 3/3] Make Floki use fast_html --- config/config.exs | 2 ++ lib/pleroma/html.ex | 1 + .../activity_pub/mrf/anti_link_spam_policy.ex | 1 + lib/pleroma/web/metadata/rel_me.ex | 6 +++-- lib/pleroma/web/rel_me.ex | 5 +++-- lib/pleroma/web/rich_media/parser.ex | 6 ++--- .../rich_media/parsers/twitter_card_test.exs | 22 ++++++++++++++----- 7 files changed, 30 insertions(+), 13 deletions(-) diff --git a/config/config.exs b/config/config.exs index 41c1ff637..364aaf776 100644 --- a/config/config.exs +++ b/config/config.exs @@ -612,6 +612,8 @@ config :pleroma, configurable_from_database: false +config :floki, :html_parser, Floki.HTMLParser.FastHtml + # Import environment specific config. This must remain at the bottom # of this file so it overrides the configuration defined above. import_config "#{Mix.env()}.exs" diff --git a/lib/pleroma/html.ex b/lib/pleroma/html.ex index 11513106e..05946aa96 100644 --- a/lib/pleroma/html.ex +++ b/lib/pleroma/html.ex @@ -108,6 +108,7 @@ def extract_first_external_url(object, content) do Cachex.fetch!(:scrubber_cache, key, fn _key -> result = content + |> Floki.parse_fragment!() |> Floki.filter_out("a.mention,a.hashtag,a[rel~=\"tag\"]") |> Floki.attribute("a", "href") |> Enum.at(0) diff --git a/lib/pleroma/web/activity_pub/mrf/anti_link_spam_policy.ex b/lib/pleroma/web/activity_pub/mrf/anti_link_spam_policy.ex index 8abe18e29..802d10edc 100644 --- a/lib/pleroma/web/activity_pub/mrf/anti_link_spam_policy.ex +++ b/lib/pleroma/web/activity_pub/mrf/anti_link_spam_policy.ex @@ -17,6 +17,7 @@ defp old_user?(%User{} = u) do # does the post contain links? defp contains_links?(%{"content" => content} = _object) do content + |> Floki.parse_fragment!() |> Floki.filter_out("a.mention,a.hashtag,a[rel~=\"tag\"],a.zrl") |> Floki.attribute("a", "href") |> length() > 0 diff --git a/lib/pleroma/web/metadata/rel_me.ex b/lib/pleroma/web/metadata/rel_me.ex index f87fc1973..86dcc1a3b 100644 --- a/lib/pleroma/web/metadata/rel_me.ex +++ b/lib/pleroma/web/metadata/rel_me.ex @@ -8,8 +8,10 @@ defmodule Pleroma.Web.Metadata.Providers.RelMe do @impl Provider def build_tags(%{user: user}) do - (Floki.attribute(user.bio, "link[rel~=me]", "href") ++ - Floki.attribute(user.bio, "a[rel~=me]", "href")) + bio_tree = Floki.parse_fragment!(user.bio) + + (Floki.attribute(bio_tree, "link[rel~=me]", "href") ++ + Floki.attribute(bio_tree, "a[rel~=me]", "href")) |> Enum.map(fn link -> {:link, [rel: "me", href: link], []} end) diff --git a/lib/pleroma/web/rel_me.ex b/lib/pleroma/web/rel_me.ex index 16b1a53d2..540fa65df 100644 --- a/lib/pleroma/web/rel_me.ex +++ b/lib/pleroma/web/rel_me.ex @@ -27,9 +27,10 @@ def parse(_), do: {:error, "No URL provided"} defp parse_url(url) do with {:ok, %Tesla.Env{body: html, status: status}} when status in 200..299 <- Pleroma.HTTP.get(url, [], adapter: @hackney_options), + {:ok, html_tree} <- Floki.parse_document(html), data <- - Floki.attribute(html, "link[rel~=me]", "href") ++ - Floki.attribute(html, "a[rel~=me]", "href") do + Floki.attribute(html_tree, "link[rel~=me]", "href") ++ + Floki.attribute(html_tree, "a[rel~=me]", "href") do {:ok, data} end rescue diff --git a/lib/pleroma/web/rich_media/parser.ex b/lib/pleroma/web/rich_media/parser.ex index c06b0a0f2..9702e90f1 100644 --- a/lib/pleroma/web/rich_media/parser.ex +++ b/lib/pleroma/web/rich_media/parser.ex @@ -81,18 +81,18 @@ defp parse_url(url) do {:ok, %Tesla.Env{body: html}} = Pleroma.HTTP.get(url, [], adapter: @hackney_options) html - |> parse_html + |> parse_html() |> maybe_parse() |> Map.put(:url, url) |> clean_parsed_data() |> check_parsed_data() rescue e -> - {:error, "Parsing error: #{inspect(e)}"} + {:error, "Parsing error: #{inspect(e)} #{inspect(__STACKTRACE__)}"} end end - defp parse_html(html), do: Floki.parse(html) + defp parse_html(html), do: Floki.parse_document!(html) defp maybe_parse(html) do Enum.reduce_while(parsers(), %{}, fn parser, acc -> diff --git a/test/web/rich_media/parsers/twitter_card_test.exs b/test/web/rich_media/parsers/twitter_card_test.exs index 751ca614c..f2ebbde7e 100644 --- a/test/web/rich_media/parsers/twitter_card_test.exs +++ b/test/web/rich_media/parsers/twitter_card_test.exs @@ -7,11 +7,14 @@ defmodule Pleroma.Web.RichMedia.Parsers.TwitterCardTest do alias Pleroma.Web.RichMedia.Parsers.TwitterCard test "returns error when html not contains twitter card" do - assert TwitterCard.parse("", %{}) == {:error, "No twitter card metadata found"} + assert TwitterCard.parse([{"html", [], [{"head", [], []}, {"body", [], []}]}], %{}) == + {:error, "No twitter card metadata found"} end test "parses twitter card with only name attributes" do - html = File.read!("test/fixtures/nypd-facial-recognition-children-teenagers3.html") + html = + File.read!("test/fixtures/nypd-facial-recognition-children-teenagers3.html") + |> Floki.parse_document!() assert TwitterCard.parse(html, %{}) == {:ok, @@ -26,7 +29,9 @@ test "parses twitter card with only name attributes" do end test "parses twitter card with only property attributes" do - html = File.read!("test/fixtures/nypd-facial-recognition-children-teenagers2.html") + html = + File.read!("test/fixtures/nypd-facial-recognition-children-teenagers2.html") + |> Floki.parse_document!() assert TwitterCard.parse(html, %{}) == {:ok, @@ -45,7 +50,9 @@ test "parses twitter card with only property attributes" do end test "parses twitter card with name & property attributes" do - html = File.read!("test/fixtures/nypd-facial-recognition-children-teenagers.html") + html = + File.read!("test/fixtures/nypd-facial-recognition-children-teenagers.html") + |> Floki.parse_document!() assert TwitterCard.parse(html, %{}) == {:ok, @@ -73,7 +80,8 @@ test "respect only first title tag on the page" do "YTQ5MF9EQVIgZXhodW1hdGlvbiBvZiBNYXJnYXJldCBDb3JiaW4gZ3JhdmUgMTkyNi5qcGciXSxbInAiLCJjb252ZXJ0IiwiIl0sWyJwIiwiY29udmVydCIsIi1xdWFsaXR5IDgxIC1hdXRvLW9" <> "yaWVudCJdLFsicCIsInRodW1iIiwiNjAweD4iXV0/DAR%20exhumation%20of%20Margaret%20Corbin%20grave%201926.jpg" - html = File.read!("test/fixtures/margaret-corbin-grave-west-point.html") + html = + File.read!("test/fixtures/margaret-corbin-grave-west-point.html") |> Floki.parse_document!() assert TwitterCard.parse(html, %{}) == {:ok, @@ -87,7 +95,9 @@ test "respect only first title tag on the page" do end test "takes first founded title in html head if there is html markup error" do - html = File.read!("test/fixtures/nypd-facial-recognition-children-teenagers4.html") + html = + File.read!("test/fixtures/nypd-facial-recognition-children-teenagers4.html") + |> Floki.parse_document!() assert TwitterCard.parse(html, %{}) == {:ok,