tpcds: Update query templates to MySQL dialet

tpcds/README.md

@@ -9,6 +9,8 @@ This tpcds-kit is ported from [gregrahn/tpcds-kit](https://github.com/gregrahn/t
   * https://github.com/gregrahn/tpcds-kit/issues/31
   * https://github.com/gregrahn/tpcds-kit/issues/33
 
+* Adjust some query templates to TiDB supported MySQL dialet, adjustments are ported from [FdeFabricio/tpcds-mysql](https://github.com/FdeFabricio/tpcds-mysql), thanks to [FdeFabricio](https://github.com/FdeFabricio/)
+
 ## 1. Setup
 
 ### 1.1 install required development tools
@@ -56,7 +58,7 @@ done
 
 Query generation is done via `dsqgen` with query templetes, here we use a pre-written shell script file [genquery.sh](./genquery.sh), after running this script, queries are located in directory "queries":
 ```sh
-./genquery.sh
+./genquery.sh 1
 ```
 
 All supported TPC-DS queries for TiDB are generated in `tools/queries`

tpcds/genquery.sh

@@ -1,7 +1,7 @@
 #!/bin/bash
 set -eu
 
-SCALE="1"
+SCALE=$1
 TEMPLATE_DIR="../query_templates"
 OUTPUT_DIR="queries"
 QUERY_ID=""
@@ -18,8 +18,23 @@ function generate_query()
     mv "$OUTPUT_DIR/query_0.sql" "$OUTPUT_DIR/query_$QUERY_ID.sql"
 }
 
-#unsupported="87 17 38 8 18 22 27 21 16 32 37 40 82 92 94 12 20 36 49 44 53 63 67 70 86 89 98 1 2 4 5 11 14 23 24 30 31 33 39 47 51 54 56 57 58 59 60 64 74 75 77 78 80 81 83 95 97 13 6"
-unsupported="87 17 38 8 18 22 27 21 16 32 37 40 82 92 94 12 20 36 49 44 53 63 67 70 86 89 98 1 2 4 5 11 14 23 24 30 31 33 39 47 51 54 56 57 58 59 60 64 74 75 77 78 80 81 83 95 97"
+function split_sql()
+{
+    cd $OUTPUT_DIR
+
+    count=`grep ';' query_$QUERY_ID.sql | wc -l`
+    if [[ $count -eq 1 ]]; then
+        cd -
+        return
+    fi
+
+    csplit --quiet --prefix=query_$QUERY_ID -k --suppress-matched --suffix-format="_%d.sql" query_$QUERY_ID.sql "/;/+1"
+    rm query_$QUERY_ID.sql
+    cd -
+}
+
+unsupported="14 18 22 27 36 5 67 70 77 80 86" # rollup function
+unsupported=$unsupported" 2 11 23 39 4 59" # tidb bug
 
 cd tools
 rm -rf $OUTPUT_DIR
@@ -37,6 +52,7 @@ for i in {1..99}; do
     fi
     QUERY_ID="$i"
     generate_query
+    split_sql
 done
 mv $OUTPUT_DIR ..
-cd -
+cd ..

tpcds/query_templates/query12.tpl

@@ -55,7 +55,7 @@ where
   	and i_category in ('[CATEGORY.1]', '[CATEGORY.2]', '[CATEGORY.3]')
   	and ws_sold_date_sk = d_date_sk
 	and d_date between cast('[SDATE]' as date)
-				and (cast('[SDATE]' as date) + 30 days)
+				and date_add(cast('[SDATE]' as date), interval 30 day)
 group by
 	i_item_id
         ,i_item_desc

tpcds/query_templates/query16.tpl

@@ -55,7 +55,7 @@ from
   ,call_center
 where
     d_date between '[YEAR]-[MONTH]-01' and
-           (cast('[YEAR]-[MONTH]-01' as date) + 60 days)
+           date_add(cast('[YEAR]-[MONTH]-01' as date), interval 60 day)
 and cs1.cs_ship_date_sk = d_date_sk
 and cs1.cs_ship_addr_sk = ca_address_sk
 and ca_state = '[STATE]'

tpcds/query_templates/query2.tpl

@@ -37,13 +37,13 @@
  with wscs as
  (select sold_date_sk
         ,sales_price
-  from  select ws_sold_date_sk sold_date_sk
+  from  (select ws_sold_date_sk sold_date_sk
               ,ws_ext_sales_price sales_price
         from web_sales
         union all
         select cs_sold_date_sk sold_date_sk
               ,cs_ext_sales_price sales_price
-        from catalog_sales),
+        from catalog_sales)),
  wswscs as
  (select d_week_seq,
         sum(case when (d_day_name='Sunday') then sales_price else null end) sun_sales,

tpcds/query_templates/query20.tpl

@@ -52,7 +52,7 @@
    and i_category in ('[CATEGORY.1]', '[CATEGORY.2]', '[CATEGORY.3]')
    and cs_sold_date_sk = d_date_sk
  and d_date between cast('[SDATE]' as date)
- 				and (cast('[SDATE]' as date) + 30 days)
+ 				and date_add(cast('[SDATE]' as date), interval 30 day)
  group by i_item_id
          ,i_item_desc
          ,i_category

tpcds/query_templates/query21.tpl

@@ -39,10 +39,10 @@
  [_LIMITA] select [_LIMITB] *
  from(select w_warehouse_name
             ,i_item_id
-            ,sum(case when (cast(d_date as date) < cast ('[SALES_DATE]' as date))
+            ,sum(case when (cast(d_date as date) < cast('[SALES_DATE]' as date))
 	                then inv_quantity_on_hand
                       else 0 end) as inv_before
-            ,sum(case when (cast(d_date as date) >= cast ('[SALES_DATE]' as date))
+            ,sum(case when (cast(d_date as date) >= cast('[SALES_DATE]' as date))
                       then inv_quantity_on_hand
                       else 0 end) as inv_after
    from inventory
@@ -53,8 +53,8 @@
      and i_item_sk          = inv_item_sk
      and inv_warehouse_sk   = w_warehouse_sk
      and inv_date_sk    = d_date_sk
-     and d_date between (cast ('[SALES_DATE]' as date) - 30 days)
-                    and (cast ('[SALES_DATE]' as date) + 30 days)
+     and d_date between date_sub(cast('[SALES_DATE]' as date), interval 30 day)
+                    and date_add(cast('[SALES_DATE]' as date), interval 30 day)
    group by w_warehouse_name, i_item_id) x
  where (case when inv_before > 0
              then inv_after / inv_before

tpcds/query_templates/query23.tpl

@@ -56,7 +56,7 @@
         where ss_customer_sk = c_customer_sk
          and ss_sold_date_sk = d_date_sk
          and d_year in ([YEAR],[YEAR]+1,[YEAR]+2,[YEAR]+3)
-        group by c_customer_sk)),
+        group by c_customer_sk) temp1),
  best_ss_customer as
  (select c_customer_sk,sum(ss_quantity*ss_sales_price) ssales
   from store_sales
@@ -84,7 +84,7 @@ from
          and d_moy = [MONTH]
          and ws_sold_date_sk = d_date_sk
          and ws_item_sk in (select item_sk from frequent_ss_items)
-         and ws_bill_customer_sk in (select c_customer_sk from best_ss_customer))
+         and ws_bill_customer_sk in (select c_customer_sk from best_ss_customer)) temp2
  [_LIMITC];
 
  with frequent_ss_items as
@@ -106,7 +106,7 @@ from
         where ss_customer_sk = c_customer_sk
          and ss_sold_date_sk = d_date_sk
          and d_year in ([YEAR],[YEAR]+1,[YEAR]+2,[YEAR]+3)
-        group by c_customer_sk)),
+        group by c_customer_sk) temp3),
  best_ss_customer as
  (select c_customer_sk,sum(ss_quantity*ss_sales_price) ssales
   from store_sales
@@ -139,6 +139,6 @@ from
          and ws_item_sk in (select item_sk from frequent_ss_items)
          and ws_bill_customer_sk in (select c_customer_sk from best_ss_customer)
          and ws_bill_customer_sk = c_customer_sk
-       group by c_last_name,c_first_name)
+       group by c_last_name,c_first_name) temp4
      order by c_last_name,c_first_name,sales
   [_LIMITC];

tpcds/query_templates/query32.tpl

@@ -46,7 +46,7 @@ where
 i_manufact_id = [IMID]
 and i_item_sk = cs_item_sk
 and d_date between '[CSDATE]' and
-        (cast('[CSDATE]' as date) + 90 days)
+        date_add(cast('[CSDATE]' as date), interval 90 day)
 and d_date_sk = cs_sold_date_sk
 and cs_ext_discount_amt
      > (
@@ -58,7 +58,7 @@ and cs_ext_discount_amt
          where
               cs_item_sk = i_item_sk
           and d_date between '[CSDATE]' and
-                             (cast('[CSDATE]' as date) + 90 days)
+                             date_add(cast('[CSDATE]' as date), interval 90 day)
           and d_date_sk = cs_sold_date_sk
       )
 [_LIMITC];

tpcds/query_templates/query37.tpl

@@ -45,7 +45,7 @@
  where i_current_price between [PRICE] and [PRICE] + 30
  and inv_item_sk = i_item_sk
  and d_date_sk=inv_date_sk
- and d_date between cast('[INVDATE]' as date) and (cast('[INVDATE]' as date) +  60 days)
+ and d_date between cast('[INVDATE]' as date) and date_add(cast('[INVDATE]' as date), interval 60 day)
  and i_manufact_id in ([MANUFACT_ID.1],[MANUFACT_ID.2],[MANUFACT_ID.3],[MANUFACT_ID.4])
  and inv_quantity_on_hand between 100 and 500
  and cs_item_sk = i_item_sk

tpcds/query_templates/query40.tpl

@@ -39,9 +39,9 @@
  [_LIMITA] select [_LIMITB]
    w_state
   ,i_item_id
-  ,sum(case when (cast(d_date as date) < cast ('[SALES_DATE]' as date))
+  ,sum(case when (cast(d_date as date) < cast('[SALES_DATE]' as date))
  		then cs_sales_price - coalesce(cr_refunded_cash,0) else 0 end) as sales_before
-  ,sum(case when (cast(d_date as date) >= cast ('[SALES_DATE]' as date))
+  ,sum(case when (cast(d_date as date) >= cast('[SALES_DATE]' as date))
  		then cs_sales_price - coalesce(cr_refunded_cash,0) else 0 end) as sales_after
  from
    catalog_sales left outer join catalog_returns on
@@ -55,8 +55,8 @@
  and i_item_sk          = cs_item_sk
  and cs_warehouse_sk    = w_warehouse_sk
  and cs_sold_date_sk    = d_date_sk
- and d_date between (cast ('[SALES_DATE]' as date) - 30 days)
-                and (cast ('[SALES_DATE]' as date) + 30 days)
+ and d_date between date_sub(cast('[SALES_DATE]' as date), interval 30 day)
+                and date_add(cast('[SALES_DATE]' as date), interval 30 day)
  group by
     w_state,i_item_id
  order by w_state,i_item_id

tpcds/query_templates/query51.tpl

@@ -71,7 +71,14 @@ from (select item_sk
                  ,case when web.d_date is not null then web.d_date else store.d_date end d_date
                  ,web.cume_sales web_sales
                  ,store.cume_sales store_sales
-           from web_v1 web full outer join store_v1 store on (web.item_sk = store.item_sk
+           from web_v1 web Left join store_v1 store on (web.item_sk = store.item_sk
+                                                          and web.d_date = store.d_date)
+				UNION
+				select case when web.item_sk is not null then web.item_sk else store.item_sk end item_sk
+                 ,case when web.d_date is not null then web.d_date else store.d_date end d_date
+                 ,web.cume_sales web_sales
+                 ,store.cume_sales store_sales
+           from web_v1 web Right join store_v1 store on (web.item_sk = store.item_sk
                                                           and web.d_date = store.d_date)
           )x )y
 where web_cumulative > store_cumulative

tpcds/query_templates/query54.tpl

@@ -84,7 +84,7 @@
  group by c_customer_sk
  )
  , segments as
- (select cast((revenue/50) as int) as segment
+ (select cast((revenue/50) as signed) as segment
   from   my_revenue
  )
  [_LIMITA] select [_LIMITB] segment, count(*) as num_customers, segment*50 as segment_base

tpcds/query_templates/query82.tpl

@@ -45,7 +45,7 @@
  where i_current_price between [PRICE] and [PRICE]+30
  and inv_item_sk = i_item_sk
  and d_date_sk=inv_date_sk
- and d_date between cast('[INVDATE]' as date) and (cast('[INVDATE]' as date) +  60 days)
+ and d_date between cast('[INVDATE]' as date) and date_add(cast('[INVDATE]' as date), interval 60 day)
  and i_manufact_id in ([MANUFACT_ID.1],[MANUFACT_ID.2],[MANUFACT_ID.3],[MANUFACT_ID.4])
  and inv_quantity_on_hand between 100 and 500
  and ss_item_sk = i_item_sk

tpcds/query_templates/query92.tpl

@@ -48,7 +48,7 @@ where
 i_manufact_id = [IMID]
 and i_item_sk = ws_item_sk
 and d_date between '[WSDATE]' and
-        (cast('[WSDATE]' as date) + 90 days)
+        date_add(cast('[WSDATE]' as date), interval 90 day)
 and d_date_sk = ws_sold_date_sk
 and ws_ext_discount_amt
      > (
@@ -60,7 +60,7 @@ and ws_ext_discount_amt
          WHERE
               ws_item_sk = i_item_sk
           and d_date between '[WSDATE]' and
-                             (cast('[WSDATE]' as date) + 90 days)
+                             date_add(cast('[WSDATE]' as date), interval 90 day)
           and d_date_sk = ws_sold_date_sk
       )
 order by sum(ws_ext_discount_amt)

tpcds/query_templates/query94.tpl

@@ -49,7 +49,7 @@ from
   ,web_site
 where
     d_date between '[YEAR]-[MONTH]-01' and
-           (cast('[YEAR]-[MONTH]-01' as date) + 60 days)
+	date_add(cast('[YEAR]-[MONTH]-01' as date), interval 60 day)
 and ws1.ws_ship_date_sk = d_date_sk
 and ws1.ws_ship_addr_sk = ca_address_sk
 and ca_state = '[STATE]'

tpcds/query_templates/query95.tpl

@@ -54,7 +54,7 @@ from
   ,web_site
 where
     d_date between '[YEAR]-[MONTH]-01' and
-           (cast('[YEAR]-[MONTH]-01' as date) + 60 days)
+           date_add(cast('[YEAR]-[MONTH]-01' as date), interval 60 day)
 and ws1.ws_ship_date_sk = d_date_sk
 and ws1.ws_ship_addr_sk = ca_address_sk
 and ca_state = '[STATE]'

tpcds/query_templates/query97.tpl

@@ -57,6 +57,12 @@ group by cs_bill_customer_sk
 [_LIMITA] select [_LIMITB] sum(case when ssci.customer_sk is not null and csci.customer_sk is null then 1 else 0 end) store_only
       ,sum(case when ssci.customer_sk is null and csci.customer_sk is not null then 1 else 0 end) catalog_only
       ,sum(case when ssci.customer_sk is not null and csci.customer_sk is not null then 1 else 0 end) store_and_catalog
-from ssci full outer join csci on (ssci.customer_sk=csci.customer_sk
+from ssci left join csci on (ssci.customer_sk=csci.customer_sk
+                               and ssci.item_sk = csci.item_sk)
+		UNION
+[_LIMITA] select [_LIMITB] sum(case when ssci.customer_sk is not null and csci.customer_sk is null then 1 else 0 end) store_only
+      ,sum(case when ssci.customer_sk is null and csci.customer_sk is not null then 1 else 0 end) catalog_only
+      ,sum(case when ssci.customer_sk is not null and csci.customer_sk is not null then 1 else 0 end) store_and_catalog
+from ssci right join csci on (ssci.customer_sk=csci.customer_sk
                                and ssci.item_sk = csci.item_sk)
 [_LIMITC];

tpcds/query_templates/query98.tpl

@@ -54,7 +54,7 @@ where
   	and i_category in ('[CATEGORY.1]', '[CATEGORY.2]', '[CATEGORY.3]')
   	and ss_sold_date_sk = d_date_sk
 	and d_date between cast('[SDATE]' as date)
-				and (cast('[SDATE]' as date) + 30 days)
+				and date_add(cast('[SDATE]' as date), interval 30 day)
 group by
 	i_item_id
         ,i_item_desc