Spider2lite local273 gold solution is incorrect.

[ritahuja]

local273 asks: Calculate the average pick percentage for each product name, using a first-in-first-out approach that selects from inventory locations based on the earliest purchase date and smallest available quantity, ensuring that the picked quantity reflects only the overlapping range between each order’s required quantity and the inventory’s available quantity, and then grouping and ordering the results by product name?

Problem: The gold answer is wrong.
https://github.com/xlang-ai/Spider2/blob/main/spider2-lite/evaluation_suite/gold/exec_result/local273_a.csv

PRODUCT_NAME,AVG_PICK_PERCENTAGE
4280,0.0
6520,0.0

It shows an average pick percentage of 0, which is concretely incorrect because in the data we have both
we have inventory and we have order, which means we MUST have a percentage of >0.