From 797a8e040406eaf9b8bc1d4cf615b3a71d412065 Mon Sep 17 00:00:00 2001
From: Patrick Buckley <eous@users.noreply.github.com>
Date: Mon, 15 Jun 2026 20:40:01 -0700
Subject: [PATCH] fix(attachments): preserve pdf/audio kind when reloading
 attachments from the DB

_reconstruct_attachment_refs collapsed every non-image attachment to the 'document' placeholder kind, so a reloaded session's pdf/audio placeholder type ({type:document}) mismatched the live-injection type ({type:pdf}/{type:audio}). Harmless today (resolution keys on attachment_id + blob kind) but a latent footgun for any consumer branching on the pre-resolution placeholder type. Preserve image/pdf/audio verbatim; only a stored 'text' blob collapses to 'document'.
---
 tests/test_attachments_pdf_audio.py | 39 +++++++++++++++++++++++++++++
 turnstone/core/storage/_utils.py    | 14 +++++++----
 2 files changed, 48 insertions(+), 5 deletions(-)

diff --git a/tests/test_attachments_pdf_audio.py b/tests/test_attachments_pdf_audio.py
index 069cd478..cebdcdf2 100644
--- a/tests/test_attachments_pdf_audio.py
+++ b/tests/test_attachments_pdf_audio.py
@@ -75,6 +75,45 @@ class TestSniffAudio:
         assert sniff_audio_mime(b"") is None
 
 
+class TestReconstructAttachmentRefs:
+    def test_preserves_pdf_audio_kind_on_reload(self) -> None:
+        from turnstone.core.storage._utils import _reconstruct_attachment_refs
+
+        atts = {
+            1: [
+                {
+                    "attachment_id": "a1",
+                    "kind": "pdf",
+                    "filename": "r.pdf",
+                    "mime_type": "application/pdf",
+                },
+                {
+                    "attachment_id": "a2",
+                    "kind": "audio",
+                    "filename": "a.wav",
+                    "mime_type": "audio/wav",
+                },
+                {
+                    "attachment_id": "a3",
+                    "kind": "image",
+                    "filename": "i.png",
+                    "mime_type": "image/png",
+                },
+                {
+                    "attachment_id": "a4",
+                    "kind": "text",
+                    "filename": "t.txt",
+                    "mime_type": "text/plain",
+                },
+            ]
+        }
+        refs, meta = _reconstruct_attachment_refs(atts, 1)
+        # pdf/audio/image kept verbatim; only 'text' collapses to 'document'
+        # (so the placeholder type can't collide with a real text content part).
+        assert [r.kind for r in refs] == ["pdf", "audio", "image", "document"]
+        assert [m["kind"] for m in meta] == ["pdf", "audio", "image", "text"]
+
+
 class TestSafeAttachmentLabel:
     def test_strips_frame_breakers(self) -> None:
         from turnstone.core.attachments import safe_attachment_label
diff --git a/turnstone/core/storage/_utils.py b/turnstone/core/storage/_utils.py
index d9aeafb3..b321103a 100644
--- a/turnstone/core/storage/_utils.py
+++ b/turnstone/core/storage/_utils.py
@@ -410,11 +410,15 @@ def _reconstruct_attachment_refs(
     if not attachments_by_msg or row_id is None:
         return refs, meta
     for att in attachments_by_msg.get(row_id, []):
-        # AttachmentRef.kind is the by-reference content kind ('image' |
-        # 'document'); the stored blob kind ('image' | 'text') drives the actual
-        # resolution.  'document' (not 'text') keeps the placeholder type from
-        # colliding with a real text content part on the dict round-trip.
-        ref_kind = "image" if str(att.get("kind") or "") == "image" else "document"
+        # AttachmentRef.kind is the by-reference placeholder kind: the stored
+        # blob kind verbatim for image / pdf / audio, else 'document' for a
+        # stored 'text' blob (so the placeholder type can't collide with a real
+        # text content part on the dict round-trip).  The blob kind drives the
+        # actual resolution; preserving pdf / audio here keeps the reloaded
+        # placeholder type ({type:pdf} / {type:audio}) consistent with the live
+        # injection path, which already emits those.
+        kind_str = str(att.get("kind") or "")
+        ref_kind = kind_str if kind_str in ("image", "pdf", "audio") else "document"
         refs.append(
             AttachmentRef(
                 attachment_id=str(att.get("attachment_id") or ""),