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

Fix normalized quantity basis

Browse Source
This commit is contained in:
ben
2026-03-21 21:50:10 -04:00
parent db761adafc
commit d25448b690
7 changed files with 85 additions and 9 deletions
Download Patch File Download Diff File Expand all files Collapse all files

29
enrich_giant.py
View File

@@ -224,13 +224,17 @@ def normalize_unit(unit):
"OZ": "oz",
"FZ": "fl_oz",
"FL OZ": "fl_oz",
"FLOZ": "fl_oz",
"LB": "lb",
"LBS": "lb",
"ML": "ml",
"L": "l",
"QT": "qt",
"QTS": "qt",
"PT": "pt",
"PTS": "pt",
"GAL": "gal",
"GALS": "gal",
"GA": "gal",
}.get(collapsed, collapsed.lower())
@@ -340,16 +344,24 @@ def derive_prices(item, measure_type, size_value="", size_unit="", pack_qty=""):
return price_per_each, price_per_lb, price_per_oz
def derive_normalized_quantity(size_value, size_unit, pack_qty, measure_type):
def derive_normalized_quantity(qty, size_value, size_unit, pack_qty, measure_type):
parsed_qty = to_decimal(qty)
parsed_size = to_decimal(size_value)
parsed_pack = to_decimal(pack_qty) or Decimal("1")
parsed_pack = to_decimal(pack_qty)
total_multiplier = None
if parsed_qty not in (None, Decimal("0")):
total_multiplier = parsed_qty * (parsed_pack or Decimal("1"))
if parsed_size not in (None, Decimal("0")) and size_unit:
return format_decimal(parsed_size * parsed_pack), size_unit
if parsed_pack not in (None, Decimal("0")) and measure_type == "count":
return format_decimal(parsed_pack), "count"
if measure_type == "each":
return "1", "each"
if (
parsed_size not in (None, Decimal("0"))
and size_unit
and total_multiplier not in (None, Decimal("0"))
):
return format_decimal(parsed_size * total_multiplier), size_unit
if measure_type == "count" and total_multiplier not in (None, Decimal("0")):
return format_decimal(total_multiplier), "count"
if measure_type == "each" and parsed_qty not in (None, Decimal("0")):
return format_decimal(parsed_qty), "each"
return "", ""
@@ -424,6 +436,7 @@ def parse_item(order_id, order_date, raw_path, line_no, item):
normalized_row_id = f"{RETAILER}:{order_id}:{line_no}"
normalized_quantity, normalized_quantity_unit = derive_normalized_quantity(
item.get("shipQy"),
size_value,
size_unit,
pack_qty,