Merge branch 'feat/floki-fast-html-2' into 'develop'

Update to fast_html 2 and use it for Floki parsing

See merge request pleroma/pleroma!2861

.gitlab-ci.yml

@@ -22,6 +22,7 @@ stages:
   - docker
 
 before_script:
+  - apt-get update && apt-get install -y cmake
   - mix local.hex --force
   - mix local.rebar --force
 

config/config.exs

@@ -743,6 +743,8 @@
 
 config :pleroma, :instances_favicons, enabled: false
 
+config :floki, :html_parser, Floki.HTMLParser.FastHtml
+
 config :pleroma, Pleroma.Web.Auth.Authenticator, Pleroma.Web.Auth.PleromaAuthenticator
 
 # Import environment specific config. This must remain at the bottom

mix.exs

@@ -127,7 +127,7 @@ defp deps do
       {:pbkdf2_elixir, "~> 1.2"},
       {:bcrypt_elixir, "~> 2.2"},
       {:trailing_format_plug, "~> 0.0.7"},
-      {:fast_sanitize, "~> 0.1"},
+      {:fast_sanitize, "~> 0.2.0"},
       {:html_entities, "~> 0.5", override: true},
       {:phoenix_html, "~> 2.14"},
       {:calendar, "~> 1.0"},

mix.lock

@@ -42,8 +42,8 @@
   "ex_machina": {:hex, :ex_machina, "2.4.0", "09a34c5d371bfb5f78399029194a8ff67aff340ebe8ba19040181af35315eabb", [: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", "a20bc9ddc721b33ea913b93666c5d0bdca5cbad7a67540784ae277228832d72c"},
   "ex_syslogger": {:hex, :ex_syslogger, "1.5.2", "72b6aa2d47a236e999171f2e1ec18698740f40af0bd02c8c650bf5f1fd1bac79", [:mix], [{:poison, ">= 1.5.0", [hex: :poison, repo: "hexpm", optional: true]}, {:syslog, "~> 1.1.0", [hex: :syslog, repo: "hexpm", optional: false]}], "hexpm", "ab9fab4136dbc62651ec6f16fa4842f10cf02ab4433fa3d0976c01be99398399"},
   "excoveralls": {:hex, :excoveralls, "0.13.1", "b9f1697f7c9e0cfe15d1a1d737fb169c398803ffcbc57e672aa007e9fd42864c", [:mix], [{:hackney, "~> 1.16", [hex: :hackney, repo: "hexpm", optional: false]}, {:jason, "~> 1.0", [hex: :jason, repo: "hexpm", optional: false]}], "hexpm", "b4bb550e045def1b4d531a37fb766cbbe1307f7628bf8f0414168b3f52021cce"},
-  "fast_html": {:hex, :fast_html, "1.0.3", "2cc0d4b68496266a1530e0c852cafeaede0bd10cfdee26fda50dc696c203162f", [:make, :mix], [], "hexpm", "ab3d782b639d3c4655fbaec0f9d032c91f8cab8dd791ac7469c2381bc7c32f85"},
-  "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", "f39fe8ea08fbac17487c30bf09b7d9f3e12472e51fb07a88ffeb8fd17da8ab67"},
+  "fast_html": {:hex, :fast_html, "2.0.1", "e126c74d287768ae78c48938da6711164517300d108a78f8a38993df8d588335", [:make, :mix], [{:elixir_make, "~> 0.4", [hex: :elixir_make, repo: "hexpm", optional: false]}, {:nimble_pool, "~> 0.1.0", [hex: :nimble_pool, repo: "hexpm", optional: false]}], "hexpm", "bdd6f8525c95ad391a4f10d9a1b3da4cea94078ec8638487aa8c24015ad9393a"},
+  "fast_sanitize": {:hex, :fast_sanitize, "0.2.0", "004b40d5bbecda182b6fdba762a51fffd3501e689e8eafe196e1a97eb0caf733", [:mix], [{:fast_html, "~> 2.0", [hex: :fast_html, repo: "hexpm", optional: false]}, {:plug, "~> 1.8", [hex: :plug, repo: "hexpm", optional: false]}], "hexpm", "11fcb37f26d272a3a2aff861872bf100be4eeacea69505908b8cdbcea5b0813a"},
   "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", "31fc8090fde1acd267c07c36ea7365b8604055f897d3a53dd967658c691bd827"},
   "floki": {:hex, :floki, "0.27.0", "6b29a14283f1e2e8fad824bc930eaa9477c462022075df6bea8f0ad811c13599", [:mix], [{:html_entities, "~> 0.5.0", [hex: :html_entities, repo: "hexpm", optional: false]}], "hexpm", "583b8c13697c37179f1f82443bcc7ad2f76fbc0bf4c186606eebd658f7f2631b"},
   "gen_smtp": {:hex, :gen_smtp, "0.15.0", "9f51960c17769b26833b50df0b96123605a8024738b62db747fece14eb2fbfcc", [:rebar3], [], "hexpm", "29bd14a88030980849c7ed2447b8db6d6c9278a28b11a44cafe41b791205440f"},
@@ -76,6 +76,7 @@
   "mox": {:hex, :mox, "0.5.2", "55a0a5ba9ccc671518d068c8dddd20eeb436909ea79d1799e2209df7eaa98b6c", [:mix], [], "hexpm", "df4310628cd628ee181df93f50ddfd07be3e5ecc30232d3b6aadf30bdfe6092b"},
   "myhtmlex": {:git, "https://git.pleroma.social/pleroma/myhtmlex.git", "ad0097e2f61d4953bfef20fb6abddf23b87111e6", [ref: "ad0097e2f61d4953bfef20fb6abddf23b87111e6", submodules: true]},
   "nimble_parsec": {:hex, :nimble_parsec, "0.6.0", "32111b3bf39137144abd7ba1cce0914533b2d16ef35e8abc5ec8be6122944263", [:mix], [], "hexpm", "27eac315a94909d4dc68bc07a4a83e06c8379237c5ea528a9acff4ca1c873c52"},
+  "nimble_pool": {:hex, :nimble_pool, "0.1.0", "ffa9d5be27eee2b00b0c634eb649aa27f97b39186fec3c493716c2a33e784ec6", [:mix], [], "hexpm", "343a1eaa620ddcf3430a83f39f2af499fe2370390d4f785cd475b4df5acaf3f9"},
   "nodex": {:git, "https://git.pleroma.social/pleroma/nodex", "cb6730f943cfc6aad674c92161be23a8411f15d1", [ref: "cb6730f943cfc6aad674c92161be23a8411f15d1"]},
   "oban": {:hex, :oban, "2.0.0", "e6ce70d94dd46815ec0882a1ffb7356df9a9d5b8a40a64ce5c2536617a447379", [:mix], [{:ecto_sql, ">= 3.4.3", [hex: :ecto_sql, repo: "hexpm", optional: false]}, {:jason, "~> 1.1", [hex: :jason, repo: "hexpm", optional: false]}, {:postgrex, "~> 0.14", [hex: :postgrex, repo: "hexpm", optional: false]}, {:telemetry, "~> 0.4", [hex: :telemetry, repo: "hexpm", optional: false]}], "hexpm", "cf574813bd048b98a698aa587c21367d2e06842d4e1b1993dcd6a696e9e633bd"},
   "open_api_spex": {:git, "https://git.pleroma.social/pleroma/elixir-libraries/open_api_spex.git", "f296ac0924ba3cf79c7a588c4c252889df4c2edd", [ref: "f296ac0924ba3cf79c7a588c4c252889df4c2edd"]},

test/web/metadata/rel_me_test.exs

@@ -9,13 +9,12 @@ defmodule Pleroma.Web.Metadata.Providers.RelMeTest do
 
   test "it renders all links with rel='me' from user bio" do
     bio =
-      ~s(<a href="https://some-link.com">https://some-link.com</a> <a rel="me" href="https://another-link.com">https://another-link.com</a>
-    <link href="http://some.com"> <link rel="me" href="http://some3.com>")
+      ~s(<a href="https://some-link.com">https://some-link.com</a> <a rel="me" href="https://another-link.com">https://another-link.com</a> <link href="http://some.com"> <link rel="me" href="http://some3.com">)
 
     user = insert(:user, %{bio: bio})
 
     assert RelMe.build_tags(%{user: user}) == [
-             {:link, [rel: "me", href: "http://some3.com>"], []},
+             {:link, [rel: "me", href: "http://some3.com"], []},
              {:link, [rel: "me", href: "https://another-link.com"], []}
            ]
   end