chore: update tpch q7/q14 (#1128)

Author: Genesis929

third_party/bigframes_vendored/tpch/queries/q14.py

@@ -26,9 +26,20 @@ def q(project_id: str, dataset_id: str, session: bigframes.Session):
  filtered["L_EXTENDEDPRICE"] * (1 - filtered["L_DISCOUNT"])
  ) * filtered["P_TYPE"].str.contains("PROMO").astype("Int64")
 
- total_revenue = (filtered["L_EXTENDEDPRICE"] * (1 - filtered["L_DISCOUNT"])).sum()
- promo_revenue = filtered["CONDI_REVENUE"].sum()
+ total_revenue = (
+ (filtered["L_EXTENDEDPRICE"] * (1 - filtered["L_DISCOUNT"]))
+ .to_frame(name="TEMP")
+ .sum()
+ )
+
+ promo_revenue = filtered["CONDI_REVENUE"].to_frame(name="TEMP").sum()
 
- promo_revenue_percent = 100.00 * promo_revenue / total_revenue
+ promo_revenue_percent = (
+ (100.00 * promo_revenue / total_revenue)
+ .sort_index()
+ .reset_index(drop=True)
+ .round(2)
+ .to_frame(name="PROMO_REVENUE")
+ )
 
- _ = round(promo_revenue_percent, 2)
+ promo_revenue_percent.to_gbq()

third_party/bigframes_vendored/tpch/queries/q7.py

@@ -35,28 +35,20 @@ def q(project_id: str, dataset_id: str, session: bigframes.Session):
  var3 = date(1995, 1, 1)
  var4 = date(1996, 12, 31)
 
- n1 = nation[(nation["N_NAME"] == var1)]
- n2 = nation[(nation["N_NAME"] == var2)]
+ nation = nation[nation["N_NAME"].isin([var1, var2])]
+ lineitem = lineitem[
+ (lineitem["L_SHIPDATE"] >= var3) & (lineitem["L_SHIPDATE"] <= var4)
+ ]
 
- jn1 = customer.merge(n1, left_on="C_NATIONKEY", right_on="N_NATIONKEY")
+ jn1 = customer.merge(nation, left_on="C_NATIONKEY", right_on="N_NATIONKEY")
  jn2 = jn1.merge(orders, left_on="C_CUSTKEY", right_on="O_CUSTKEY")
  jn2 = jn2.rename(columns={"N_NAME": "CUST_NATION"})
  jn3 = jn2.merge(lineitem, left_on="O_ORDERKEY", right_on="L_ORDERKEY")
  jn4 = jn3.merge(supplier, left_on="L_SUPPKEY", right_on="S_SUPPKEY")
- jn5 = jn4.merge(n2, left_on="S_NATIONKEY", right_on="N_NATIONKEY")
+ jn5 = jn4.merge(nation, left_on="S_NATIONKEY", right_on="N_NATIONKEY")
  df1 = jn5.rename(columns={"N_NAME": "SUPP_NATION"})
+ total = df1[df1["CUST_NATION"] != df1["SUPP_NATION"]]
 
- jn1 = customer.merge(n2, left_on="C_NATIONKEY", right_on="N_NATIONKEY")
- jn2 = jn1.merge(orders, left_on="C_CUSTKEY", right_on="O_CUSTKEY")
- jn2 = jn2.rename(columns={"N_NAME": "CUST_NATION"})
- jn3 = jn2.merge(lineitem, left_on="O_ORDERKEY", right_on="L_ORDERKEY")
- jn4 = jn3.merge(supplier, left_on="L_SUPPKEY", right_on="S_SUPPKEY")
- jn5 = jn4.merge(n1, left_on="S_NATIONKEY", right_on="N_NATIONKEY")
- df2 = jn5.rename(columns={"N_NAME": "SUPP_NATION"})
-
- total = bpd.concat([df1, df2])
-
- total = total[(total["L_SHIPDATE"] >= var3) & (total["L_SHIPDATE"] <= var4)]
  total["VOLUME"] = total["L_EXTENDEDPRICE"] * (1.0 - total["L_DISCOUNT"])
  total["L_YEAR"] = total["L_SHIPDATE"].dt.year