refactor(ingest): per-post handling into run stdout via a downloader outcome (#842)

Two corrections from operator review:
1. Reuse the existing 'Raw stdout' panel instead of a bespoke structured UI
   section — the native ingester now writes a per-post line into the run stdout
   (parity with gallery-dl's per-file stdout), so the per-post handling shows in
   the panel the operator already uses.
2. DRY: stop re-reading post['attributes'] inline in ingest_core. write_post_record
   now returns a PostRecordOutcome (path, post_type, title, body_chars) — mirroring
   the download_post -> MediaOutcome contract — and the downloader owns the read;
   ingest_core only formats the outcome into the log line.

Reverts the post_diagnostics metadata field + DownloadDetailModal 'Post capture'
section added earlier. Per-post line: 'post <id> [<post_type>] body: N chars' (+
' — EMPTY' when 0), so an empty body is self-explanatory by post_type.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

tests/test_download_service.py

@@ -60,11 +60,10 @@ def _make_fake_dl_result(
     *, success=True, written_paths=None, quarantined_paths=None,
     files_downloaded=0, error_type=None, error_message=None,
     stdout="", stderr="", cursor=None, run_stats=None, posts_processed=0,
-    relink_source_paths=None, post_diagnostics=None,
+    relink_source_paths=None,
 ):
     return SimpleNamespace(
         relink_source_paths=relink_source_paths or [],
-        post_diagnostics=post_diagnostics or [],
         success=success,
         url="https://patreon.com/alice",
         artist_slug="alice",
@@ -821,28 +820,6 @@ async def test_recapture_mode_selected_and_flag_cleared_on_complete(
     assert "_backfill_recapture" not in co
 
 
-@pytest.mark.asyncio
-async def test_post_diagnostics_written_to_event_metadata(
-    db, db_sync, tmp_path, seed_artist_and_source,
-):
-    """#842: per-post body diagnostics land on the DownloadEvent.metadata_ so the
-    UI can show how each post was handled (post_type + body length)."""
-    _artist, source = seed_artist_and_source
-    diag = [
-        {"post_id": "p1", "title": "Has body", "post_type": "image_file", "body_chars": 42},
-        {"post_id": "p2", "title": "Empty poll", "post_type": "poll", "body_chars": 0},
-    ]
-    svc, _ = _backfill_svc(db, db_sync, tmp_path, _make_fake_dl_result(
-        success=True, post_diagnostics=diag,
-    ))
-    await svc.download_source(source.id)
-
-    ev = (await db.execute(
-        select(DownloadEvent).order_by(DownloadEvent.id.desc()).limit(1)
-    )).scalar_one()
-    assert ev.metadata_["post_diagnostics"] == diag
-
-
 @pytest.mark.asyncio
 async def test_partial_error_type_maps_to_ok_status(
     db, db_sync, tmp_path, seed_artist_and_source,

tests/test_patreon_downloader.py

@@ -18,6 +18,7 @@ from backend.app.services.patreon_client import MediaItem
 from backend.app.services.patreon_downloader import (
     MediaOutcome,
     PatreonDownloader,
+    PostRecordOutcome,
     _sanitize,
 )
 from backend.app.utils.sidecar import find_sidecar
@@ -646,14 +647,17 @@ def test_write_post_record_writes_enriched_post_only_sidecar(tmp_path):
     )
     post = _post()
     post["attributes"]["content"] = ""  # media-less post, empty feed body
-    sc = dl.write_post_record(post, "artist-x")
+    rec = dl.write_post_record(post, "artist-x")
 
-    assert sc is not None
-    assert sc.name == "_post.json"  # can't collide with a media `NN_*.json`
-    data = json.loads(sc.read_text())
+    assert isinstance(rec, PostRecordOutcome)
+    assert rec.path is not None
+    assert rec.path.name == "_post.json"  # can't collide with a media `NN_*.json`
+    data = json.loads(rec.path.read_text())
     assert data["id"] == "1001"
     assert data["content"] == "<p>text post body</p>"
     assert calls == ["1001"]
+    # Outcome reports the captured body's shape (detail-fetched here).
+    assert rec.body_chars == len("<p>text post body</p>")
 
 
 def test_write_post_record_none_without_post_id(tmp_path):
@@ -661,4 +665,6 @@ def test_write_post_record_none_without_post_id(tmp_path):
         images_root=tmp_path, cookies_path=None, validate=False,
         session=_FakeSession(),
     )
-    assert dl.write_post_record({"id": "", "attributes": {}}, "artist-x") is None
+    rec = dl.write_post_record({"id": "", "attributes": {}}, "artist-x")
+    assert rec.path is None
+    assert rec.body_chars == 0

tests/test_patreon_ingester.py

@@ -24,7 +24,7 @@ from backend.app.services.patreon_client import (
     PatreonAuthError,
     PatreonDriftError,
 )
-from backend.app.services.patreon_downloader import MediaOutcome
+from backend.app.services.patreon_downloader import MediaOutcome, PostRecordOutcome
 from backend.app.services.patreon_ingester import PatreonIngester, _ledger_key
 
 pytestmark = pytest.mark.integration
@@ -137,7 +137,14 @@ class _FakeDownloader:
         self.post_records += 1
         p = self.tmp_path / f"{post.get('id')}__post.json"
         p.write_text("{}")
-        return p
+        attrs = post.get("attributes") or {}
+        body = attrs.get("content")
+        return PostRecordOutcome(
+            path=p,
+            post_type=attrs.get("post_type"),
+            title=attrs.get("title"),
+            body_chars=len(body) if isinstance(body, str) else 0,
+        )
 
 
 @pytest.fixture
@@ -654,13 +661,10 @@ async def test_backfill_skips_already_captured_post_but_recapture_forces_it(
     # Diagnostic summary surfaces the post-record + relink counts (operator reads
     # this off the event stdout to see what a recapture actually did).
     assert "1 post-record(s), 1 relinked" in res_recap.stdout
-    # #842: per-post body diagnostic carries post_type + body length so the UI can
-    # show how each post was handled (and explain empties by post_type).
-    assert len(res_recap.post_diagnostics) == 1
-    diag = res_recap.post_diagnostics[0]
-    assert diag["post_id"] == "p1"
-    assert diag["post_type"] == "image_file"
-    assert diag["body_chars"] == len("<p>body p1</p>")
+    # #842: per-post handling is written into the SAME run stdout (reuses the
+    # existing "Raw stdout" panel) — post_type + body length, so an empty body is
+    # self-explanatory by its post_type.
+    assert f"post p1 [image_file] body: {len('<p>body p1</p>')} chars" in res_recap.stdout
 
 
 @pytest.mark.asyncio