feat(products): diff resolves hosted image URLs to existing records — INV-12 over images (SD-0002 §6.3)

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

backend/app/domains/products/diff.py

@@ -21,10 +21,11 @@ from __future__ import annotations
 
 import hashlib
 import json
-from dataclasses import dataclass, field
+from dataclasses import dataclass, field, replace
 from decimal import Decimal
 
-from .models import CLEAR_DEFAULTS, CanonicalProduct, CanonicalVariant
+from . import hosted
+from .models import CLEAR_DEFAULTS, CanonicalProduct, CanonicalVariant, RowError
 
 
 @dataclass
@@ -119,6 +120,7 @@ def compute_diff(catalog: dict[str, CatalogProduct], products: list[CanonicalPro
     summary = {"adds": 0, "updates": 0, "unchanged": 0, "errors": 0}
     for product in products:
         current = catalog.get(product.handle)
+        _resolve_hosted_images(product, current)
         if product.errors:
             product_plan = ProductPlan(kind="error", canonical=product, catalog=current)
             detail: dict = {"errors": [e.as_json() for e in product.errors]}
@@ -145,6 +147,31 @@ def compute_diff(catalog: dict[str, CatalogProduct], products: list[CanonicalPro
     return DiffResult(records=records, summary=summary, fingerprint=fingerprint, plan=plan)
 
 
+def _resolve_hosted_images(product: CanonicalProduct, current: CatalogProduct | None) -> None:
+    """Recognize re-imported hosted image URLs (INV-12). A hosted URL is resolved
+    to this product's existing image with that id — re-keyed onto that image's
+    source_url so the by_src match treats it as the existing image (never an add,
+    never a fetch). A hosted URL whose id is not one of this product's images
+    (cross-storefront, deleted, or a brand-new product) is a row error."""
+    current_by_id = {i.id: i for i in current.images} if current else {}
+    resolved: list = []
+    for image in product.images:
+        hosted_id = hosted.parse_image_id(image.source_url)
+        if hosted_id is None:
+            resolved.append(image)
+            continue
+        match = current_by_id.get(hosted_id)
+        if match is None:
+            product.errors.append(
+                RowError(image.line_number, "Image Src",
+                         f"image '{image.source_url}' refers to an unknown hosted id")
+            )
+            resolved.append(image)
+            continue
+        resolved.append(replace(image, source_url=match.source_url))
+    product.images = resolved
+
+
 def _add_plan(product: CanonicalProduct) -> ProductPlan:
     return ProductPlan(
         kind="add",