Skip to content
体验新版
项目
组织
正在加载...
登录
切换导航
打开侧边栏
taosdata
TDengine
提交
b2486b2a
T
TDengine
项目概览
taosdata
/
TDengine
1 年多 前同步成功
通知
1185
Star
22016
Fork
4786
代码
文件
提交
分支
Tags
贡献者
分支图
Diff
Issue
1
列表
看板
标记
里程碑
合并请求
0
Wiki
0
Wiki
分析
仓库
DevOps
项目成员
Pages
T
TDengine
项目概览
项目概览
详情
发布
仓库
仓库
文件
提交
分支
标签
贡献者
分支图
比较
Issue
1
Issue
1
列表
看板
标记
里程碑
合并请求
0
合并请求
0
Pages
分析
分析
仓库分析
DevOps
Wiki
0
Wiki
成员
成员
收起侧边栏
关闭侧边栏
动态
分支图
创建新Issue
提交
Issue看板
提交
b2486b2a
编写于
6月 21, 2022
作者:
W
wenzhouwww@live.cn
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
add case for twa function
上级
4cacb51e
变更
2
隐藏空白更改
内联
并排
Showing
2 changed file
with
171 addition
and
0 deletion
+171
-0
tests/system-test/2-query/twa.py
tests/system-test/2-query/twa.py
+170
-0
tests/system-test/fulltest.sh
tests/system-test/fulltest.sh
+1
-0
未找到文件。
tests/system-test/2-query/twa.py
0 → 100644
浏览文件 @
b2486b2a
from
util.log
import
*
from
util.cases
import
*
from
util.sql
import
*
import
numpy
as
np
import
random
,
os
,
sys
import
platform
import
math
class
TDTestCase
:
updatecfgDict
=
{
'debugFlag'
:
143
,
"cDebugFlag"
:
143
,
"uDebugFlag"
:
143
,
"rpcDebugFlag"
:
143
,
"tmrDebugFlag"
:
143
,
"jniDebugFlag"
:
143
,
"simDebugFlag"
:
143
,
"dDebugFlag"
:
143
,
"dDebugFlag"
:
143
,
"vDebugFlag"
:
143
,
"mDebugFlag"
:
143
,
"qDebugFlag"
:
143
,
"wDebugFlag"
:
143
,
"sDebugFlag"
:
143
,
"tsdbDebugFlag"
:
143
,
"tqDebugFlag"
:
143
,
"fsDebugFlag"
:
143
,
"fnDebugFlag"
:
143
,
"maxTablesPerVnode"
:
2
,
"minTablesPerVnode"
:
2
,
"tableIncStepPerVnode"
:
2
}
def
init
(
self
,
conn
,
logSql
):
tdLog
.
debug
(
"start to execute %s"
%
__file__
)
tdSql
.
init
(
conn
.
cursor
())
self
.
vnode_disbutes
=
None
self
.
ts
=
1537146000000
self
.
tb_nums
=
20
self
.
row_nums
=
100
self
.
time_step
=
1000
def
check_stddev_functions
(
self
,
tbname
,
col_name
):
stddev_sql
=
f
"select stddev(
{
col_name
}
) from
{
tbname
}
;"
same_sql
=
f
"select
{
col_name
}
from
{
tbname
}
where
{
col_name
}
is not null "
tdSql
.
query
(
same_sql
)
pre_data
=
np
.
array
(
tdSql
.
queryResult
)[
np
.
array
(
tdSql
.
queryResult
)
!=
None
]
if
(
platform
.
system
().
lower
()
==
'windows'
and
pre_data
.
dtype
==
'int32'
):
pre_data
=
np
.
array
(
pre_data
,
dtype
=
'int64'
)
pre_avg
=
np
.
sum
(
pre_data
)
/
len
(
pre_data
)
# Calculate variance
stddev_result
=
0
for
num
in
tdSql
.
queryResult
:
stddev_result
+=
(
num
-
pre_avg
)
*
(
num
-
pre_avg
)
/
len
(
tdSql
.
queryResult
)
stddev_result
=
math
.
sqrt
(
stddev_result
)
tdSql
.
query
(
stddev_sql
)
if
-
0.0001
<
tdSql
.
queryResult
[
0
][
0
]
-
stddev_result
<
0.0001
:
tdLog
.
info
(
" sql:%s; row:0 col:0 data:%d , expect:%d"
%
(
stddev_sql
,
tdSql
.
queryResult
[
0
][
0
],
stddev_result
))
else
:
tdLog
.
exit
(
" sql:%s; row:0 col:0 data:%d , expect:%d"
%
(
stddev_sql
,
tdSql
.
queryResult
[
0
][
0
],
stddev_result
))
def
prepare_datas_of_distribute
(
self
):
# prepate datas for 20 tables distributed at different vgroups
tdSql
.
execute
(
"create database if not exists testdb keep 3650 duration 1000 vgroups 5"
)
tdSql
.
execute
(
" use testdb "
)
tdSql
.
execute
(
'''create table stb1
(ts timestamp, c1 int, c2 bigint, c3 smallint, c4 tinyint, c5 float, c6 double, c7 bool, c8 binary(16),c9 nchar(32), c10 timestamp)
tags (t0 timestamp, t1 int, t2 bigint, t3 smallint, t4 tinyint, t5 float, t6 double, t7 bool, t8 binary(16),t9 nchar(32))
'''
)
for
i
in
range
(
self
.
tb_nums
):
tdSql
.
execute
(
f
'create table ct
{
i
+
1
}
using stb1 tags ( now(),
{
1
*
i
}
,
{
11111
*
i
}
,
{
111
*
i
}
,
{
11
*
i
}
,
{
1.11
*
i
}
,
{
11.11
*
i
}
,
{
i
%
2
}
, "binary
{
i
}
", "nchar
{
i
}
" )'
)
ts
=
self
.
ts
for
j
in
range
(
self
.
row_nums
):
ts
+=
j
*
self
.
time_step
tdSql
.
execute
(
f
"insert into ct
{
i
+
1
}
values(
{
ts
}
, 1, 11111, 111, 1, 1.11, 11.11, 2, 'binary
{
j
}
', 'nchar
{
j
}
', now()+
{
1
*
j
}
a )"
)
tdSql
.
execute
(
"insert into ct1 values (now()-810d, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL ) "
)
tdSql
.
execute
(
"insert into ct1 values (now()-400d, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL ) "
)
tdSql
.
execute
(
"insert into ct1 values (now()+90d, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL ) "
)
tdLog
.
info
(
" prepare data for distributed_aggregate done! "
)
def
check_distribute_datas
(
self
):
# get vgroup_ids of all
tdSql
.
query
(
"show vgroups "
)
vgroups
=
tdSql
.
queryResult
vnode_tables
=
{}
for
vgroup_id
in
vgroups
:
vnode_tables
[
vgroup_id
[
0
]]
=
[]
# check sub_table of per vnode ,make sure sub_table has been distributed
tdSql
.
query
(
"show tables like 'ct%'"
)
table_names
=
tdSql
.
queryResult
tablenames
=
[]
for
table_name
in
table_names
:
vnode_tables
[
table_name
[
6
]].
append
(
table_name
[
0
])
self
.
vnode_disbutes
=
vnode_tables
count
=
0
for
k
,
v
in
vnode_tables
.
items
():
if
len
(
v
)
>=
2
:
count
+=
1
if
count
<
2
:
tdLog
.
exit
(
" the datas of all not satisfy sub_table has been distributed "
)
def
distribute_twa_query
(
self
):
# basic filter
tdSql
.
query
(
" select twa(c1) from ct1 "
)
tdSql
.
checkData
(
0
,
0
,
1.000000000
)
tdSql
.
query
(
" select twa(c1) from stb1 partition by tbname "
)
tdSql
.
checkRows
(
self
.
tb_nums
)
tdSql
.
checkData
(
0
,
0
,
1.000000000
)
tdSql
.
query
(
" select twa(c2) from stb1 group by tbname "
)
tdSql
.
checkRows
(
self
.
tb_nums
)
tdSql
.
checkData
(
0
,
0
,
11111.000000000
)
tdSql
.
query
(
"select twa(c1+c2) from stb1 partition by tbname "
)
tdSql
.
checkData
(
0
,
0
,
11112.000000000
)
tdSql
.
query
(
"select twa(c1) from stb1 partition by t1"
)
tdSql
.
checkRows
(
self
.
tb_nums
)
tdSql
.
checkData
(
0
,
0
,
1.000000000
)
# union all
tdSql
.
query
(
" select twa(c1) from stb1 partition by tbname union all select twa(c1) from stb1 partition by tbname "
)
tdSql
.
checkRows
(
40
)
tdSql
.
checkData
(
0
,
0
,
1.000000000
)
# join
tdSql
.
execute
(
" create database if not exists db "
)
tdSql
.
execute
(
" use db "
)
tdSql
.
execute
(
" create stable st (ts timestamp , c1 int ,c2 float) tags(t1 int) "
)
tdSql
.
execute
(
" create table tb1 using st tags(1) "
)
tdSql
.
execute
(
" create table tb2 using st tags(2) "
)
for
i
in
range
(
10
):
ts
=
i
*
10
+
self
.
ts
tdSql
.
execute
(
f
" insert into tb1 values(
{
ts
}
,
{
i
}
,
{
i
}
.0)"
)
tdSql
.
execute
(
f
" insert into tb2 values(
{
ts
}
,
{
i
}
,
{
i
}
.0)"
)
tdSql
.
query
(
" select twa(tb1.c1), twa(tb2.c2) from tb1, tb2 where tb1.ts=tb2.ts "
)
tdSql
.
checkRows
(
1
)
tdSql
.
checkData
(
0
,
0
,
4.500000000
)
tdSql
.
checkData
(
0
,
1
,
4.500000000
)
# group by
tdSql
.
execute
(
" use testdb "
)
# mixup with other functions
tdSql
.
query
(
" select twa(c1),twa(c2),max(c1),elapsed(ts) from stb1 "
)
tdSql
.
checkData
(
0
,
0
,
1.000000000
)
tdSql
.
checkData
(
0
,
1
,
11111.000000000
)
tdSql
.
checkData
(
0
,
2
,
1
)
def
run
(
self
):
self
.
prepare_datas_of_distribute
()
self
.
check_distribute_datas
()
self
.
distribute_twa_query
()
def
stop
(
self
):
tdSql
.
close
()
tdLog
.
success
(
"%s successfully executed"
%
__file__
)
tdCases
.
addWindows
(
__file__
,
TDTestCase
())
tdCases
.
addLinux
(
__file__
,
TDTestCase
())
tests/system-test/fulltest.sh
浏览文件 @
b2486b2a
...
...
@@ -105,6 +105,7 @@ python3 ./test.py -f 2-query/distribute_agg_min.py
python3 ./test.py
-f
2-query/distribute_agg_sum.py
python3 ./test.py
-f
2-query/distribute_agg_spread.py
python3 ./test.py
-f
2-query/distribute_agg_apercentile.py
python3 ./test.py
-f
2-query/twa.py
python3 ./test.py
-f
6-cluster/5dnode1mnode.py
python3 ./test.py
-f
6-cluster/5dnode2mnode.py
...
...
编辑
预览
Markdown
is supported
0%
请重试
或
添加新附件
.
添加附件
取消
You are about to add
0
people
to the discussion. Proceed with caution.
先完成此消息的编辑!
取消
想要评论请
注册
或
登录