ben/scrape-giant
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Refine product review prompt flow

Browse Source
This commit is contained in:
ben
2026-03-17 13:25:12 -04:00
parent 7b8141cd42
commit d39497c298
3 changed files with 288 additions and 53 deletions
Download Patch File Download Diff File Expand all files Collapse all files

184
tests/test_review_workflow.py
View File

@@ -158,16 +158,186 @@ class ReviewWorkflowTests(unittest.TestCase):
)
self.assertEqual(0, result.exit_code)
self.assertIn("Review observed product (1/1)", result.output)
self.assertIn("Resolve this observed product group", result.output)
self.assertIn("actions: [l]ink existing [n]ew canonical e[x]clude [s]kip [q]uit", result.output)
first_item = result.output.index("2026-03-14 | 7.49")
second_item = result.output.index("2026-03-12 | 6.99")
self.assertIn("Review 1/1: Resolve observed_product MIXED PEPPER to canonical_name [__]?", result.output)
self.assertIn("2 matched items:", result.output)
self.assertIn("[l]ink existing [n]ew canonical e[x]clude [s]kip [q]uit:", result.output)
first_item = result.output.index("[1] 2026-03-14 | 7.49")
second_item = result.output.index("[2] 2026-03-12 | 6.99")
self.assertLess(first_item, second_item)
self.assertIn("image: https://example.test/mixed-pepper.jpg", result.output)
self.assertIn("gcan_mix | MIXED PEPPER (exact normalized name)", result.output)
self.assertIn("https://example.test/mixed-pepper.jpg", result.output)
self.assertIn("1 canonical suggestions found:", result.output)
self.assertIn("[1] MIXED PEPPER", result.output)
self.assertIn("\x1b[", result.output)
def test_review_products_no_suggestions_is_informational(self):
with tempfile.TemporaryDirectory() as tmpdir:
purchases_csv = Path(tmpdir) / "purchases.csv"
queue_csv = Path(tmpdir) / "review_queue.csv"
resolutions_csv = Path(tmpdir) / "review_resolutions.csv"
catalog_csv = Path(tmpdir) / "canonical_catalog.csv"
with purchases_csv.open("w", newline="", encoding="utf-8") as handle:
writer = csv.DictWriter(
handle,
fieldnames=[
"purchase_date",
"retailer",
"order_id",
"line_no",
"observed_product_id",
"canonical_product_id",
"raw_item_name",
"normalized_item_name",
"image_url",
"upc",
"line_total",
],
)
writer.writeheader()
writer.writerow(
{
"purchase_date": "2026-03-14",
"retailer": "giant",
"order_id": "g1",
"line_no": "1",
"observed_product_id": "gobs_ice",
"canonical_product_id": "",
"raw_item_name": "SB BAGGED ICE 20LB",
"normalized_item_name": "BAGGED ICE",
"image_url": "",
"upc": "",
"line_total": "3.50",
}
)
with catalog_csv.open("w", newline="", encoding="utf-8") as handle:
writer = csv.DictWriter(handle, fieldnames=review_products.build_purchases.CATALOG_FIELDS)
writer.writeheader()
result = CliRunner().invoke(
review_products.main,
[
"--purchases-csv",
str(purchases_csv),
"--queue-csv",
str(queue_csv),
"--resolutions-csv",
str(resolutions_csv),
"--catalog-csv",
str(catalog_csv),
],
input="q\n",
color=True,
)
self.assertEqual(0, result.exit_code)
self.assertIn("no canonical_name suggestions found", result.output)
def test_link_existing_uses_numbered_selection_and_confirmation(self):
with tempfile.TemporaryDirectory() as tmpdir:
purchases_csv = Path(tmpdir) / "purchases.csv"
queue_csv = Path(tmpdir) / "review_queue.csv"
resolutions_csv = Path(tmpdir) / "review_resolutions.csv"
catalog_csv = Path(tmpdir) / "canonical_catalog.csv"
with purchases_csv.open("w", newline="", encoding="utf-8") as handle:
writer = csv.DictWriter(
handle,
fieldnames=[
"purchase_date",
"retailer",
"order_id",
"line_no",
"observed_product_id",
"canonical_product_id",
"raw_item_name",
"normalized_item_name",
"image_url",
"upc",
"line_total",
],
)
writer.writeheader()
writer.writerows(
[
{
"purchase_date": "2026-03-14",
"retailer": "costco",
"order_id": "c2",
"line_no": "2",
"observed_product_id": "gobs_mix",
"canonical_product_id": "",
"raw_item_name": "MIXED PEPPER 6-PACK",
"normalized_item_name": "MIXED PEPPER",
"image_url": "",
"upc": "",
"line_total": "7.49",
},
{
"purchase_date": "2026-03-12",
"retailer": "costco",
"order_id": "c1",
"line_no": "1",
"observed_product_id": "gobs_mix",
"canonical_product_id": "",
"raw_item_name": "MIXED PEPPER 6-PACK",
"normalized_item_name": "MIXED PEPPER",
"image_url": "",
"upc": "",
"line_total": "6.99",
},
]
)
with catalog_csv.open("w", newline="", encoding="utf-8") as handle:
writer = csv.DictWriter(handle, fieldnames=review_products.build_purchases.CATALOG_FIELDS)
writer.writeheader()
writer.writerow(
{
"canonical_product_id": "gcan_mix",
"canonical_name": "MIXED PEPPER",
"category": "",
"product_type": "",
"brand": "",
"variant": "",
"size_value": "",
"size_unit": "",
"pack_qty": "",
"measure_type": "",
"notes": "",
"created_at": "",
"updated_at": "",
}
)
result = CliRunner().invoke(
review_products.main,
[
"--purchases-csv",
str(purchases_csv),
"--queue-csv",
str(queue_csv),
"--resolutions-csv",
str(resolutions_csv),
"--catalog-csv",
str(catalog_csv),
"--limit",
"1",
],
input="l\n1\ny\nlinked by test\n",
color=True,
)
self.assertEqual(0, result.exit_code)
self.assertIn("Select the canonical_name to associate 2 items with:", result.output)
self.assertIn('[1] MIXED PEPPER | gcan_mix', result.output)
self.assertIn('2 "MIXED PEPPER" items and future matches will be associated with "MIXED PEPPER".', result.output)
self.assertIn("actions: [y]es [n]o [b]ack [s]kip [q]uit", result.output)
with resolutions_csv.open(newline="", encoding="utf-8") as handle:
rows = list(csv.DictReader(handle))
self.assertEqual("gcan_mix", rows[0]["canonical_product_id"])
self.assertEqual("link", rows[0]["resolution_action"])
def test_review_products_creates_canonical_and_resolution(self):
with tempfile.TemporaryDirectory() as tmpdir:
purchases_csv = Path(tmpdir) / "purchases.csv"