parse posted_at from fa static file url

app/helpers/fa_uri_helper.rb

@@ -0,0 +1,57 @@
+# typed: strict
+
+module FaUriHelper
+  extend T::Sig
+
+  FA_CDN_HOSTS = %w[d.facdn.net d.furaffinity.net].freeze
+
+  class FaMediaUrlInfo < T::ImmutableStruct
+    extend T::Sig
+    include T::Struct::ActsAsComparable
+
+    const :url_name, String
+    const :original_file_posted, Integer
+    const :latest_file_posted, Integer
+    const :filename, String
+
+    sig { returns(Time) }
+    def original_file_posted_at
+      Time.at(original_file_posted)
+    end
+
+    sig { returns(Time) }
+    def latest_file_posted_at
+      Time.at(latest_file_posted)
+    end
+  end
+
+  sig { params(url_str: String).returns(T.nilable(FaMediaUrlInfo)) }
+  def self.parse_fa_media_url(url_str)
+    uri = Addressable::URI.parse(url_str)
+    return nil unless is_fa_cdn_host?(uri.host)
+    # paths are in the form of `art/<user.url_name>/<latest_file_ts>/<og_file_ts>.<rest_of_filename>`
+    # latest_file_ts is the timestamp of the most up to date file that has been uploaded for the post
+    # og_file_ts is the timestamp of when the post was originally made
+    path = uri.path
+    match =
+      path.match(
+        %r{/art/(?<url_name>[^/]+)/(?<latest_ts>\d+)/(?<original_ts>\d+)\.(?<filename>.*)},
+      )
+    return nil unless match
+    url_name = match[:url_name]
+    latest_ts = match[:latest_ts].to_i
+    original_ts = match[:original_ts].to_i
+    filename = match[:filename]
+    FaMediaUrlInfo.new(
+      url_name:,
+      original_file_posted: original_ts,
+      latest_file_posted: latest_ts,
+      filename:,
+    )
+  end
+
+  sig { params(host: String).returns(T::Boolean) }
+  def self.is_fa_cdn_host?(host)
+    FA_CDN_HOSTS.include?(host)
+  end
+end

app/jobs/domain/fa/job/scan_post_job.rb

@@ -157,7 +157,6 @@ class Domain::Fa::Job::ScanPostJob < Domain::Fa::Job::Base
     post.scanned_at = Time.now
   end
 
-  FA_CDN_HOSTS = %w[d.facdn.net d.furaffinity.net].freeze
   sig { params(url_str: String, new_url_str: String).returns(T::Boolean) }
   def self.uri_same_with_normalized_facdn_host?(url_str, new_url_str)
     uri = Addressable::URI.parse(url_str)
@@ -165,7 +164,7 @@ class Domain::Fa::Job::ScanPostJob < Domain::Fa::Job::Base
     uri.scheme = nil
     new_uri.scheme = nil
 
-    if [uri, new_uri].all? { |uri| FA_CDN_HOSTS.include?(uri.host) }
+    if [uri, new_uri].all? { |uri| FaUriHelper.is_fa_cdn_host?(uri.host) }
       # both URIs have an facdn host, so compare them but ignore the host
       uri.host = nil
       new_uri.host = nil

app/models/domain/post/fa_post.rb

@@ -177,7 +177,12 @@ class Domain::Post::FaPost < Domain::Post
     pa = super
     return pa unless pa.nil?
 
-    log_entry = guess_last_submission_log_entry
+    if file_url_str = file&.url_str
+      parsed = FaUriHelper.parse_fa_media_url(file_url_str)
+      return parsed.original_file_posted_at.in_time_zone("UTC") if parsed
+    end
+
+    log_entry = last_submission_log_entry || guess_last_submission_log_entry
     if log_entry&.response_bytes
       parser =
         Domain::Fa::Parser::Page.from_log_entry(

app/views/domain/posts/default/_section_post_title.erb

@@ -28,7 +28,7 @@
     </div>
   </div>
   <div class="mt-2 flex flex-wrap gap-x-4  text-slate-600 leading-relaxed justify-between">
-    <span title="<%= post.posted_at&.strftime("%Y-%m-%d") %>">
+    <span title="<%= post.posted_at&.strftime("%Y-%m-%d") || 'Unknown' %>">
       <i class="fa-regular fa-calendar mr-1"></i>
       <% if posted_at = post.posted_at %>
         Posted <%= time_ago_in_words(posted_at) %> ago

spec/helpers/fa_uri_helper_spec.rb

@@ -0,0 +1,38 @@
+# typed: false
+require "rails_helper"
+
+RSpec.describe FaUriHelper do
+  describe "#parse_fa_media_url" do
+    it "parses a media URL with same timestamps" do
+      url =
+        "https://d.furaffinity.net/art/zzreg/1740700581/1740700581.zzreg_stippling-crop.jpg"
+      parsed = described_class.parse_fa_media_url(url)
+      expect(parsed).to eq(
+        FaUriHelper::FaMediaUrlInfo.new(
+          url_name: "zzreg",
+          original_file_posted: 1_740_700_581,
+          latest_file_posted: 1_740_700_581,
+          filename: "zzreg_stippling-crop.jpg",
+        ),
+      )
+      expect(parsed.original_file_posted_at).to eq(Time.at(1_740_700_581))
+      expect(parsed.latest_file_posted_at).to eq(Time.at(1_740_700_581))
+    end
+
+    it "parses a media URL with different timestamps" do
+      url =
+        "https://d.furaffinity.net/art/zzreg/1753374875/1740700581.zzreg_stippling-crop.jpg"
+      parsed = described_class.parse_fa_media_url(url)
+      expect(parsed).to eq(
+        FaUriHelper::FaMediaUrlInfo.new(
+          url_name: "zzreg",
+          original_file_posted: 1_740_700_581,
+          latest_file_posted: 1_753_374_875,
+          filename: "zzreg_stippling-crop.jpg",
+        ),
+      )
+      expect(parsed.original_file_posted_at).to eq(Time.at(1_740_700_581))
+      expect(parsed.latest_file_posted_at).to eq(Time.at(1_753_374_875))
+    end
+  end
+end

spec/models/domain/post/fa_post_spec.rb

@@ -48,6 +48,53 @@ RSpec.describe Domain::Post::FaPost do
     end
   end
 
+  describe "posted_at" do
+    let(:post_url_str) do
+      "https://d.furaffinity.net/art/zzreg/1740700581/1740700581.zzreg_stippling-crop.jpg"
+    end
+    let(:post) { create(:domain_post_fa_post, posted_at: 1.day.ago) }
+    let(:post_file) do
+      create(:domain_post_file, post: post, url_str: post_url_str)
+    end
+
+    it "can be extracted from file uri if no explicit posted_at is set" do
+      post_file # ensure the file is created
+      post.posted_at = nil
+      post.save!
+      post.reload
+      expect(post.posted_at).to eq(Time.at(1_740_700_581))
+    end
+
+    it "uses the existing posted_at if it is set" do
+      post_file # ensure the file is created
+      expect(post.posted_at).to be_within(10.seconds).of(1.day.ago)
+    end
+
+    it "guesses it from the submission log entry if all else fails" do
+      # do not create the post file
+      post.posted_at = nil
+      post.last_submission_log_entry =
+        create(
+          :http_log_entry,
+          response:
+            create(
+              :blob_file,
+              contents:
+                File.read(
+                  Rails.root.join(
+                    "test/fixtures/files/domain/fa/submission/submission_page_59723907.html",
+                  ),
+                ),
+            ),
+        )
+      post.save!
+      post.reload
+      expect(post.posted_at).to be_within(1.minute).of(
+        Time.parse("Feb 1, 2025 07:15 AM PST"),
+      )
+    end
+  end
+
   describe "attributes" do
     let(:post) { build(:domain_post_fa_post) }
     let(:time) { Time.now }