testdata imgs

testdata/generate_testdata.py

@@ -8,6 +8,7 @@ who are friends with each other.
 import csv
 import io
 import json
+import mimetypes
 import os
 import random
 import zipfile
@@ -15,6 +16,9 @@ from hashlib import sha256
 from nacl.signing import SigningKey
 from nacl.encoding import HexEncoder
 
+IMGS_DIR = os.path.join(os.path.dirname(__file__), 'imgs')
+IMAGE_ATTACHMENT_PROBABILITY = 0.9
+
 
 # Realistic item database with shared_data category/tag references using fully qualified handles
 # Format: (name, category_handle, tags_handles, policy, description, qty, properties_list)
@@ -250,6 +254,21 @@ def load_shared_data():
     return all_tags, all_categories, all_properties
 
 
+def load_available_images():
+    """Read every file in imgs/ and return a list of (data, mime_type) tuples."""
+    images = []
+    for filename in sorted(os.listdir(IMGS_DIR)):
+        filepath = os.path.join(IMGS_DIR, filename)
+        if not os.path.isfile(filepath):
+            continue
+        mime_type, _ = mimetypes.guess_type(filename)
+        if not mime_type:
+            continue
+        with open(filepath, 'rb') as f:
+            images.append((f.read(), mime_type))
+    return images
+
+
 def generate_keypair():
     """Generate a signing key pair and return (hex_private_key, hex_public_key)."""
     signing_key = SigningKey.generate()
@@ -341,7 +360,7 @@ def _quote_value_if_needed(value):
     return value
 
 
-def generate_inventory_csv(count, num_locations, location_id_map, item_range=None):
+def generate_inventory_csv(count, num_locations, location_id_map, item_range=None, available_images=None):
     """Generate a realistic inventory.csv with varied items using shared_data references and properties.
 
     Args:
@@ -349,6 +368,12 @@ def generate_inventory_csv(count, num_locations, location_id_map, item_range=Non
         num_locations: Number of locations
         location_id_map: Map of location names to IDs
         item_range: Tuple of (start_idx, end_idx) to partition the ITEMS_DATABASE, or None for all items
+        available_images: List of (data, mime_type) tuples to randomly attach to items, or None to skip
+
+    Returns:
+        Tuple of (inventory_csv_bytes, files) where files maps a zip arcname (matching
+        `toolshed.offlinedata.inventory_files()`'s 'files/<hash><ext>' convention) to file bytes,
+        for every image actually referenced by an item.
     """
     # Get location names from the map
     location_names = list(location_id_map.keys())
@@ -365,6 +390,7 @@ def generate_inventory_csv(count, num_locations, location_id_map, item_range=Non
     selected_items = random.sample(available_items, min(count, len(available_items)))
 
     rows = []
+    files = {}
     for i, item_tuple in enumerate(selected_items, 1):
         # Handle both old format (6 elements) and new format (7 elements with properties)
         if len(item_tuple) == 7:
@@ -382,6 +408,16 @@ def generate_inventory_csv(count, num_locations, location_id_map, item_range=Non
         properties_str = ', '.join(
             f"{prop_handle}={_quote_value_if_needed(value)}" for prop_handle, value in properties_list)
 
+        # Attach a random image to most items, matching `toolshed.offlinedata.inventory_files()`'s
+        # 'files/<hash><ext>' arcname convention so the import side resolves it as the same file.
+        files_str = ''
+        if available_images and random.random() < IMAGE_ATTACHMENT_PROBABILITY:
+            data, mime_type = random.choice(available_images)
+            extension = mimetypes.guess_extension(mime_type) or ''
+            arcname = f'files/{sha256(data).hexdigest()}{extension}'
+            files[arcname] = data
+            files_str = arcname
+
         rows.append({
             'id': str(i),
             'name': name,
@@ -393,7 +429,7 @@ def generate_inventory_csv(count, num_locations, location_id_map, item_range=Non
             'storage_location': location,
             'tags': ', '.join(tags_for_item),
             'properties': properties_str,
-            'files': '',
+            'files': files_str,
             'created_at': '2026-08-01T00:00:00+00:00',
         })
 
@@ -403,7 +439,7 @@ def generate_inventory_csv(count, num_locations, location_id_map, item_range=Non
     writer = csv.DictWriter(output, fieldnames=fieldnames)
     writer.writeheader()
     writer.writerows(rows)
-    return output.getvalue().encode('utf-8')
+    return output.getvalue().encode('utf-8'), files
 
 
 def generate_sample_files():
@@ -447,9 +483,14 @@ def generate_export_zip(username, domain, friend_username, friend_domain, friend
         friends_csv = generate_friends_csv(friend_username, friend_domain, friend_public_key)
         zip_file.writestr('friends.csv', friends_csv)
 
-        inventory_csv = generate_inventory_csv(num_items, num_locations, location_id_map, item_range)
+        available_images = load_available_images()
+        inventory_csv, item_files = generate_inventory_csv(
+            num_items, num_locations, location_id_map, item_range, available_images)
         zip_file.writestr('inventory.csv', inventory_csv)
 
+        for arcname, content in item_files.items():
+            zip_file.writestr(arcname, content)
+
         sample_files = generate_sample_files()
         for arcname, content in sample_files.items():
             zip_file.writestr(arcname, content)