AutoLinker --> Linkify, update to latest version

https://git.pleroma.social/pleroma/elixir-libraries/linkify

CHANGELOG.md

@@ -12,6 +12,7 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/).
 - MFR policy to set global expiration for all local Create activities
 - OGP rich media parser merged with TwitterCard
 - Configuration: `:instance, rewrite_policy` moved to `:mrf, policies`, `:instance, :mrf_transparency` moved to `:mrf, :transparency`, `:instance, :mrf_transparency_exclusions` moved to `:mrf, :transparency_exclusions`. Old config namespace is deprecated.
+- **Breaking:** Configuration: `:auto_linker, :opts` moved to `:pleroma, Pleroma.Formatter`. Old config namespace is deprecated.
 
 <details>
   <summary>API Changes</summary>

config/config.exs

@@ -520,16 +520,14 @@ config :pleroma, :workers,
     federator_outgoing: 5
   ]
 
-config :auto_linker,
-  opts: [
-    extra: true,
-    # TODO: Set to :no_scheme when it works properly
-    validate_tld: true,
+config :pleroma, Pleroma.Formatter,
   class: false,
-    strip_prefix: false,
+  rel: "ugc",
   new_window: false,
-    rel: "ugc"
-  ]
+  truncate: false,
+  strip_prefix: false,
+  extra: true,
+  validate_tld: :no_scheme
 
 config :pleroma, :ldap,
   enabled: System.get_env("LDAP_ENABLED") == "true",

config/description.exs

@@ -2157,44 +2157,52 @@ config :pleroma, :config_description, [
     ]
   },
   %{
-    group: :auto_linker,
-    key: :opts,
+    group: :pleroma,
+    key: Pleroma.Formatter,
     type: :group,
-    description: "Configuration for the auto_linker library",
+    description:
+      "Configuration for Pleroma's link formatter which parses mentions, hashtags, and URLs.",
     children: [
       %{
         key: :class,
         type: [:string, false],
-        description: "Specify the class to be added to the generated link. Disable to clear",
+        description: "Specify the class to be added to the generated link. Disable to clear.",
         suggestions: ["auto-linker", false]
       },
       %{
         key: :rel,
         type: [:string, false],
-        description: "Override the rel attribute. Disable to clear",
+        description: "Override the rel attribute. Disable to clear.",
         suggestions: ["ugc", "noopener noreferrer", false]
       },
       %{
         key: :new_window,
         type: :boolean,
-        description: "Link urls will open in new window/tab"
+        description: "Link URLs will open in new window/tab."
       },
       %{
         key: :truncate,
         type: [:integer, false],
         description:
-          "Set to a number to truncate urls longer then the number. Truncated urls will end in `..`",
+          "Set to a number to truncate URLs longer then the number. Truncated URLs will end in `...`",
         suggestions: [15, false]
       },
       %{
         key: :strip_prefix,
         type: :boolean,
-        description: "Strip the scheme prefix"
+        description: "Strip the scheme prefix."
       },
       %{
         key: :extra,
         type: :boolean,
-        description: "Link urls with rarely used schemes (magnet, ipfs, irc, etc.)"
+        description: "Link URLs with rarely used schemes (magnet, ipfs, irc, etc.)"
+      },
+      %{
+        key: :validate_tld,
+        type: [:atom, :boolean],
+        description:
+          "Set to false to disable TLD validation for URLs/emails. Can be set to :no_scheme to validate TLDs only for URLs without a scheme (e.g `example.com` will be validated, but `http://example.loki` won't)",
+        suggestions: [:no_scheme, true]
       }
     ]
   },

docs/configuration/cheatsheet.md

@@ -908,30 +908,29 @@ Configure OAuth 2 provider capabilities:
 ### :uri_schemes
 * `valid_schemes`: List of the scheme part that is considered valid to be an URL.
 
-### :auto_linker
+### Pleroma.Formatter
 
-Configuration for the `auto_linker` library:
+Configuration for Pleroma's link formatter which parses mentions, hashtags, and URLs.
 
-* `class: "auto-linker"` - specify the class to be added to the generated link. false to clear.
-* `rel: "noopener noreferrer"` - override the rel attribute. false to clear.
-* `new_window: true` - set to false to remove `target='_blank'` attribute.
-* `scheme: false` - Set to true to link urls with schema `http://google.com`.
-* `truncate: false` - Set to a number to truncate urls longer then the number. Truncated urls will end in `..`.
-* `strip_prefix: true` - Strip the scheme prefix.
-* `extra: false` - link urls with rarely used schemes (magnet, ipfs, irc, etc.).
+* `class` - specify the class to be added to the generated link (default: `false`)
+* `rel` - specify the rel attribute (default: `ugc`)
+* `new_window` - adds `target="_blank"` attribute (default: `false`)
+* `truncate` - Set to a number to truncate URLs longer then the number. Truncated URLs will end in `...` (default: `false`)
+* `strip_prefix` - Strip the scheme prefix (default: `false`)
+* `extra` - link URLs with rarely used schemes (magnet, ipfs, irc, etc.) (default: `true`)
+* `validate_tld` - Set to false to disable TLD validation for URLs/emails. Can be set to :no_scheme to validate TLDs only for urls without a scheme (e.g `example.com` will be validated, but `http://example.loki` won't) (default: `:no_scheme`)
 
 Example:
 
 ```elixir
-config :auto_linker,
-  opts: [
-    scheme: true,
-    extra: true,
+config :pleroma, Pleroma.Formatter,
   class: false,
-    strip_prefix: false,
+  rel: "ugc",
   new_window: false,
-    rel: "ugc"
-  ]
+  truncate: false,
+  strip_prefix: false,
+  extra: true,
+  validate_tld: :no_scheme
 ```
 
 ## Custom Runtime Modules (`:modules`)

lib/pleroma/config/config_db.ex

@@ -156,7 +156,7 @@ defmodule Pleroma.ConfigDB do
       {:quack, :meta},
       {:mime, :types},
       {:cors_plug, [:max_age, :methods, :expose, :headers]},
-      {:auto_linker, :opts},
+      {:linkify, :opts},
       {:swarm, :node_blacklist},
       {:logger, :backends}
     ]

lib/pleroma/formatter.ex

@@ -10,11 +10,15 @@ defmodule Pleroma.Formatter do
   @link_regex ~r"((?:http(s)?:\/\/)?[\w.-]+(?:\.[\w\.-]+)+[\w\-\._~%:/?#[\]@!\$&'\(\)\*\+,;=.]+)|[0-9a-z+\-\.]+:[0-9a-z$-_.+!*'(),]+"ui
   @markdown_characters_regex ~r/(`|\*|_|{|}|[|]|\(|\)|#|\+|-|\.|!)/
 
-  @auto_linker_config hashtag: true,
+  defp linkify_opts do
+    Pleroma.Config.get(Pleroma.Formatter) ++
+      [
+        hashtag: true,
         hashtag_handler: &Pleroma.Formatter.hashtag_handler/4,
         mention: true,
-                      mention_handler: &Pleroma.Formatter.mention_handler/4,
-                      scheme: true
+        mention_handler: &Pleroma.Formatter.mention_handler/4
+      ]
+  end
 
   def escape_mention_handler("@" <> nickname = mention, buffer, _, _) do
     case User.get_cached_by_nickname(nickname) do
@@ -80,19 +84,19 @@ defmodule Pleroma.Formatter do
   @spec linkify(String.t(), keyword()) ::
           {String.t(), [{String.t(), User.t()}], [{String.t(), String.t()}]}
   def linkify(text, options \\ []) do
-    options = options ++ @auto_linker_config
+    options = linkify_opts() ++ options
 
     if options[:safe_mention] && Regex.named_captures(@safe_mention_regex, text) do
       %{"mentions" => mentions, "rest" => rest} = Regex.named_captures(@safe_mention_regex, text)
       acc = %{mentions: MapSet.new(), tags: MapSet.new()}
 
-      {text_mentions, %{mentions: mentions}} = AutoLinker.link_map(mentions, acc, options)
-      {text_rest, %{tags: tags}} = AutoLinker.link_map(rest, acc, options)
+      {text_mentions, %{mentions: mentions}} = Linkify.link_map(mentions, acc, options)
+      {text_rest, %{tags: tags}} = Linkify.link_map(rest, acc, options)
 
       {text_mentions <> text_rest, MapSet.to_list(mentions), MapSet.to_list(tags)}
     else
       acc = %{mentions: MapSet.new(), tags: MapSet.new()}
-      {text, %{mentions: mentions, tags: tags}} = AutoLinker.link_map(text, acc, options)
+      {text, %{mentions: mentions, tags: tags}} = Linkify.link_map(text, acc, options)
 
       {text, MapSet.to_list(mentions), MapSet.to_list(tags)}
     end
@@ -111,9 +115,9 @@ defmodule Pleroma.Formatter do
 
     if options[:safe_mention] && Regex.named_captures(@safe_mention_regex, text) do
       %{"mentions" => mentions, "rest" => rest} = Regex.named_captures(@safe_mention_regex, text)
-      AutoLinker.link(mentions, options) <> AutoLinker.link(rest, options)
+      Linkify.link(mentions, options) <> Linkify.link(rest, options)
     else
-      AutoLinker.link(text, options)
+      Linkify.link(text, options)
     end
   end
 

lib/pleroma/web/rich_media/helpers.ex

@@ -11,10 +11,10 @@ defmodule Pleroma.Web.RichMedia.Helpers do
 
   @spec validate_page_url(URI.t() | binary()) :: :ok | :error
   defp validate_page_url(page_url) when is_binary(page_url) do
-    validate_tld = Application.get_env(:auto_linker, :opts)[:validate_tld]
+    validate_tld = Pleroma.Config.get([Pleroma.Formatter, :validate_tld])
 
     page_url
-    |> AutoLinker.Parser.url?(scheme: true, validate_tld: validate_tld)
+    |> Linkify.Parser.url?(validate_tld: validate_tld)
     |> parse_uri(page_url)
   end
 

mix.exs

@@ -167,9 +167,7 @@ defmodule Pleroma.Mixfile do
       {:floki, "~> 0.25"},
       {:timex, "~> 3.5"},
       {:ueberauth, "~> 0.4"},
-      {:auto_linker,
-       git: "https://git.pleroma.social/pleroma/auto_linker.git",
-       ref: "95e8188490e97505c56636c1379ffdf036c1fdde"},
+      {:linkify, "~> 0.1.0"},
       {:http_signatures,
        git: "https://git.pleroma.social/pleroma/http_signatures.git",
        ref: "293d77bb6f4a67ac8bde1428735c3b42f22cbb30"},

mix.lock

@@ -1,6 +1,5 @@
 %{
   "accept": {:hex, :accept, "0.3.5", "b33b127abca7cc948bbe6caa4c263369abf1347cfa9d8e699c6d214660f10cd1", [:rebar3], [], "hexpm", "11b18c220bcc2eab63b5470c038ef10eb6783bcb1fcdb11aa4137defa5ac1bb8"},
-  "auto_linker": {:git, "https://git.pleroma.social/pleroma/auto_linker.git", "95e8188490e97505c56636c1379ffdf036c1fdde", [ref: "95e8188490e97505c56636c1379ffdf036c1fdde"]},
   "base62": {:hex, :base62, "1.2.1", "4866763e08555a7b3917064e9eef9194c41667276c51b59de2bc42c6ea65f806", [:mix], [{:custom_base, "~> 0.2.1", [hex: :custom_base, repo: "hexpm", optional: false]}], "hexpm", "3b29948de2013d3f93aa898c884a9dff847e7aec75d9d6d8c1dc4c61c2716c42"},
   "base64url": {:hex, :base64url, "0.0.1", "36a90125f5948e3afd7be97662a1504b934dd5dac78451ca6e9abf85a10286be", [:rebar], [], "hexpm"},
   "bbcode": {:git, "https://git.pleroma.social/pleroma/elixir-libraries/bbcode.git", "f2d267675e9a7e1ad1ea9beb4cc23382762b66c2", [ref: "v0.2.0"]},
@@ -62,6 +61,7 @@
   "jose": {:hex, :jose, "1.10.1", "16d8e460dae7203c6d1efa3f277e25b5af8b659febfc2f2eb4bacf87f128b80a", [:mix, :rebar3], [], "hexpm", "3c7ddc8a9394b92891db7c2771da94bf819834a1a4c92e30857b7d582e2f8257"},
   "jumper": {:hex, :jumper, "1.0.1", "3c00542ef1a83532b72269fab9f0f0c82bf23a35e27d278bfd9ed0865cecabff", [:mix], [], "hexpm", "318c59078ac220e966d27af3646026db9b5a5e6703cb2aa3e26bcfaba65b7433"},
   "libring": {:hex, :libring, "1.4.0", "41246ba2f3fbc76b3971f6bce83119dfec1eee17e977a48d8a9cfaaf58c2a8d6", [:mix], [], "hexpm"},
+  "linkify": {:hex, :linkify, "0.1.0", "a2d35de64271c7fbbc7d8773adb9f595592b7fbaa581271c7733f39d3058bfa4", [:mix], [], "hexpm", "d3140ef8dbdcc53ef93a6a5374c11fffe0189f00d132161e9d020a417780bee7"},
   "makeup": {:hex, :makeup, "1.0.0", "671df94cf5a594b739ce03b0d0316aa64312cee2574b6a44becb83cd90fb05dc", [:mix], [{:nimble_parsec, "~> 0.5.0", [hex: :nimble_parsec, repo: "hexpm", optional: false]}], "hexpm", "a10c6eb62cca416019663129699769f0c2ccf39428b3bb3c0cb38c718a0c186d"},
   "makeup_elixir": {:hex, :makeup_elixir, "0.14.0", "cf8b7c66ad1cff4c14679698d532f0b5d45a3968ffbcbfd590339cb57742f1ae", [:mix], [{:makeup, "~> 1.0", [hex: :makeup, repo: "hexpm", optional: false]}], "hexpm", "d4b316c7222a85bbaa2fd7c6e90e37e953257ad196dc229505137c5e505e9eff"},
   "meck": {:hex, :meck, "0.8.13", "ffedb39f99b0b99703b8601c6f17c7f76313ee12de6b646e671e3188401f7866", [:rebar3], [], "hexpm", "d34f013c156db51ad57cc556891b9720e6a1c1df5fe2e15af999c84d6cebeb1a"},