parse_source option to extract title

config/config.exs

@@ -283,9 +283,10 @@
 config :pleroma, :feed,
   post_title: %{
     # Set max_length to 0 to suppress titles (Dave Winer suggestion)
-    # Previously 100
+    max_length: 100,
-    max_length: 0,
+    omission: "...",
-    omission: "..."
+    # New method to extract title
+    parse_source: true
   }
 
 config :pleroma, :markup,

config/description.exs

@@ -2938,7 +2938,13 @@
             type: :string,
             description: "Replacement which will be used after truncating string",
             suggestions: ["..."]
-          }
+          },
+          %{
+            key: :parse_source,
+            type: :boolean,
+            description: "Use content type-specific parsers to extract title (ignores max_length)",
+            suggestions: [true]
+          },
         ]
       }
     ]

lib/pleroma/web/feed/feed_view.ex

@@ -70,14 +70,23 @@ def logo(user) do
 
   def last_activity(activities), do: List.last(activities)
 
-  def maybe_activity_title(activity_data, opts \\ %{}) do
+  def maybe_activity_title(activity, opts \\ %{}) do
-    case activity_title(activity_data, opts) do
+    case activity_title(activity, opts) do
       "" -> ""
       title -> "<title>#{title}</title>"
     end
   end
 
-  def activity_title(%{"content" => content}, opts \\ %{}) do
+  def activity_title(activity, opts \\ %{})
+
+  def activity_title(
+        %{"source" => %{"mediaType" => content_type, "content" => content}},
+        %{parse_source: true} = opts
+      ) do
+    split_content(content, content_type, opts) |> elem(0)
+  end
+
+  def activity_title(%{"content" => content}, opts) do
     content
     |> Pleroma.Web.Metadata.Utils.scrub_html()
     |> Pleroma.Emoji.Formatter.demojify()
@@ -85,13 +94,36 @@ def activity_title(%{"content" => content}, opts \\ %{}) do
     |> escape()
   end
 
-  def activity_content(%{"content" => content}) do
+  def activity_title(_, _), do: ""
-    content
+
-    |> String.replace(~r/[\n\r]/, "")
+  def activity_content(activity, opts \\ %{})
+
+  def activity_content(
+        %{"source" => %{"mediaType" => content_type, "content" => content}},
+        %{parse_source: true} = opts
+      ) do
+    start = split_content(content, content_type, opts) |> elem(1)
+    length = String.length(content)
+
+    {text, _mentions, _tags} =
+      String.slice(content, start, length)
+      |> Pleroma.Web.CommonAPI.Utils.format_input(content_type)
+
+    text
+    |> String.replace(~r/(\r?\n)+/, " ")
+    |> String.trim()
     |> escape()
   end
 
-  def activity_content(_), do: ""
+  def activity_content(%{"content" => content}, _opts) do
+    # Replace 1 or more newlines with 1 space
+    content
+    |> String.replace(~r/(\r?\n)+/, " ")
+    |> String.trim()
+    |> escape()
+  end
+
+  def activity_content(_, _), do: ""
 
   def activity_context(activity), do: escape(activity.data["context"])
 
@@ -111,15 +143,15 @@ def source_content(%{"source" => %{"mediaType" => _, "content" => content}}) do
     xml_escape(content)
   end
 
-  def parse_title(activity_data, opts \\ %{})
+  def parse_title(activity, opts \\ %{})
 
   def parse_title(%{"source" => %{"mediaType" => content_type, "content" => content}}, opts) do
     split_content(content, content_type, opts) |> elem(0)
   end
 
-  def parse_title(_activity_data, _opts), do: 0
+  def parse_title(_activity, _opts), do: 0
 
-  def parse_description_offset(activity_data, opts \\ %{})
+  def parse_description_offset(activity, opts \\ %{})
 
   def parse_description_offset(
         %{"source" => %{"mediaType" => content_type, "content" => content}},
@@ -128,7 +160,7 @@ def parse_description_offset(
     split_content(content, content_type, opts) |> elem(1)
   end
 
-  def parse_description_offset(_activity_data, _opts), do: 0
+  def parse_description_offset(_activity, _opts), do: 0
 
   @spec split_content(binary(), binary(), any()) :: {binary(), non_neg_integer()}
   def split_content(content, "text/html", _opts) do

lib/pleroma/web/templates/feed/feed/_activity.atom.eex

@@ -5,7 +5,7 @@
 
   <%= maybe_activity_title(@data, Keyword.get(@feed_config, :post_title, %{})) %>
 
-  <content type="html"><%= activity_content(@data) %></content>
+  <content type="html"><%= activity_content(@data, Keyword.get(@feed_config, :post_title, %{})) %></content>
   <published><%= @activity.data["published"] %></published>
   <updated><%= @activity.data["published"] %></updated>
   <ostatus:conversation ref="<%= activity_context(@activity) %>">

lib/pleroma/web/templates/feed/feed/_activity.rss.eex

@@ -5,7 +5,7 @@
 
   <%= maybe_activity_title(@data, Keyword.get(@feed_config, :post_title, %{})) %>
 
-  <description><%= activity_content(@data) %></description>
+  <description><%= activity_content(@data, Keyword.get(@feed_config, :post_title, %{})) %></description>
   <pubDate><%= @activity.data["published"] %></pubDate>
   <updated><%= @activity.data["published"] %></updated>
   <ostatus:conversation ref="<%= activity_context(@activity) %>">

lib/pleroma/web/templates/feed/feed/_tag_activity.atom.eex

@@ -8,7 +8,7 @@
 
     <%= maybe_activity_title(@data, Keyword.get(@feed_config, :post_title, %{})) %>
 
-    <content type="html"><%= activity_content(@data) %></content>
+    <content type="html"><%= activity_content(@data, Keyword.get(@feed_config, :post_title, %{})) %></content>
 
   <%= if @activity.local do %>
     <link type="application/atom+xml" href='<%= @data["id"] %>' rel="self"/>

lib/pleroma/web/templates/feed/feed/_tag_activity.xml.eex

@@ -5,7 +5,7 @@
 
   <%= maybe_activity_title(@data, Keyword.get(@feed_config, :post_title, %{})) %>
 
-  <description><%= activity_content(@data) %></description>
+  <description><%= activity_content(@data, Keyword.get(@feed_config, :post_title, %{})) %></description>
   <%= for attachment <- @data["attachment"] || [] do %>
     <enclosure url="<%= attachment_href(attachment) %>" type="<%= attachment_type(attachment) %>"/>
   <% end %>