From 2fd2b714c1fe64db68ecad4efd2661ddf122f51a Mon Sep 17 00:00:00 2001
From: HangyuanLiu <460660596@qq.com>
Date: Thu, 11 Jul 2019 16:45:45 +0800
Subject: [PATCH] Add aggregate function doc (#1434)

---
 .../sql-functions/aggregate-functions/avg.md  | 30 ++++++++++++++++
 .../aggregate-functions/count.md              | 34 +++++++++++++++++++
 .../aggregate-functions/count_distinct.md     | 20 +++++++++++
 .../aggregate-functions/group_concat.md       | 21 ++++++++++++
 .../aggregate-functions/hll_union_agg.md      | 25 ++++++++++++++
 .../sql-functions/aggregate-functions/max.md  | 19 +++++++++++
 .../sql-functions/aggregate-functions/min.md  | 19 +++++++++++
 .../sql-functions/aggregate-functions/ndv.md  | 21 ++++++++++++
 .../aggregate-functions/percentile_approx.md  | 21 ++++++++++++
 .../aggregate-functions/stddev.md             | 26 ++++++++++++++
 .../aggregate-functions/stddev_samp.md        | 19 +++++++++++
 .../sql-functions/aggregate-functions/sum.md  | 19 +++++++++++
 .../aggregate-functions/var_samp.md           | 19 +++++++++++
 .../aggregate-functions/variance.md           | 26 ++++++++++++++
 14 files changed, 319 insertions(+)
 create mode 100644 docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/avg.md
 create mode 100755 docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/count.md
 create mode 100755 docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/count_distinct.md
 create mode 100755 docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/group_concat.md
 create mode 100644 docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/hll_union_agg.md
 create mode 100755 docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/max.md
 create mode 100755 docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/min.md
 create mode 100644 docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/ndv.md
 create mode 100755 docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/percentile_approx.md
 create mode 100755 docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/stddev.md
 create mode 100755 docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/stddev_samp.md
 create mode 100755 docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/sum.md
 create mode 100755 docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/var_samp.md
 create mode 100755 docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/variance.md

diff --git a/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/avg.md b/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/avg.md
new file mode 100644
index 0000000000..d73f2a9a8c
--- /dev/null
+++ b/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/avg.md
@@ -0,0 +1,30 @@
+# AVG
+
+## Syntax
+
+`AVG([DISTINCT] expr)`
+
+## Description
+
+用于返回选中字段的平均值
+
+可选字段DISTINCT参数可以用来返回去重平均值
+
+## Examples
+
+```
+mysql> SELECT datetime, AVG(cost_time) FROM log_statis group by datetime;
++---------------------+--------------------+
+| datetime            | avg(`cost_time`)   |
++---------------------+--------------------+
+| 2019-07-03 21:01:20 | 25.827794561933533 |
++---------------------+--------------------+
+
+mysql> SELECT datetime, AVG(distinct cost_time) FROM log_statis group by datetime;
++---------------------+---------------------------+
+| datetime            | avg(DISTINCT `cost_time`) |
++---------------------+---------------------------+
+| 2019-07-04 02:23:24 |        20.666666666666668 |
++---------------------+---------------------------+
+
+```
diff --git a/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/count.md b/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/count.md
new file mode 100755
index 0000000000..507ac1c649
--- /dev/null
+++ b/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/count.md
@@ -0,0 +1,34 @@
+# COUNT
+
+## Syntax
+
+`COUNT([DISTINCT] expr)`
+
+## Description
+
+用于返回满足要求的行的数目
+
+## Examples
+
+```
+MySQL > select count(*) from log_statis group by datetime;
++----------+
+| count(*) |
++----------+
+| 28515903 |
++----------+
+
+MySQL > select count(datetime) from log_statis group by datetime;
++-------------------+
+| count(`datetime`) |
++-------------------+
+|         28521682  |
++-------------------+
+
+MySQL > select count(distinct datetime) from log_statis group by datetime;
++-------------------------------+
+| count(DISTINCT `datetime`)    |
++-------------------------------+
+|                       71045   |
++-------------------------------+
+```
diff --git a/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/count_distinct.md b/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/count_distinct.md
new file mode 100755
index 0000000000..ecaee4400e
--- /dev/null
+++ b/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/count_distinct.md
@@ -0,0 +1,20 @@
+# COUNT_DISTINCT
+
+## Syntax
+
+`COUNT_DISTINCT(expr)`
+
+## Description
+
+用于返回满足要求的行的数目，或者非NULL行的数目
+
+## Examples
+
+```
+MySQL > select count_distinct(query_id) from log_statis group by datetime;
++----------------------------+
+| count_distinct(`query_id`) |
++----------------------------+
+|                        577 |
++----------------------------+
+```
diff --git a/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/group_concat.md b/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/group_concat.md
new file mode 100755
index 0000000000..c326dbb10a
--- /dev/null
+++ b/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/group_concat.md
@@ -0,0 +1,21 @@
+# GROUP_CONCAT	
+
+## Syntax
+
+`GROUP_CONCAT(expr)`
+
+## Description
+
+用于返回选中字段字符串连接起来的新字符串
+
+使用逗号连接
+
+## Examples
+```
+MySQL> select group_concat(`query_id`) from log_statis group by datetime;
++---------------------------------------------------------------------------------------------------------+
+| group_concat(`query_id`)                                                                                |
++---------------------------------------------------------------------------------------------------------+
+| 445f5875f8854dfa:b9358d5cd86488a2, d5055534e6c04fa0:9b3f645913c4973c, a55801d0764d47fe:a98f3710649dc558 |
++------------------------------------------------------------------------------------------------ --------+
+```
diff --git a/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/hll_union_agg.md b/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/hll_union_agg.md
new file mode 100644
index 0000000000..69253c8c72
--- /dev/null
+++ b/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/hll_union_agg.md
@@ -0,0 +1,25 @@
+# HLL_UNION_AGG
+
+## Syntax
+
+`HLL_UNION_AGG(hll)`
+
+## Description
+
+HLL是基于HyperLogLog算法的工程实现，用于保存HyperLogLog计算过程的中间结果
+
+它只能作为表的value列类型、通过聚合来不断的减少数据量，以此来实现加快查询的目的
+
+基于它得到的是一个估算结果，误差大概在1%左右，hll列是通过其它列或者导入数据里面的数据生成的
+
+导入的时候通过hll_hash函数来指定数据中哪一列用于生成hll列，它常用于替代count distinct，通过结合rollup在业务上用于快速计算uv等
+
+## Examples
+```
+MySQL > select HLL_UNION_AGG(uv_set) from test_uv;;
++-------------------------+
+| HLL_UNION_AGG(`uv_set`) |
++-------------------------+
+| 17721                   |
++-------------------------+
+```
diff --git a/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/max.md b/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/max.md
new file mode 100755
index 0000000000..d4ca250647
--- /dev/null
+++ b/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/max.md
@@ -0,0 +1,19 @@
+# MAX
+
+## Syntax
+
+`MAX(expr)`
+
+## Description
+
+返回expr表达式的最大值
+
+## Examples
+```
+MySQL > select max(scan_rows) from log_statis group by datetime;
++------------------+
+| max(`scan_rows`) |
++------------------+
+|          4671587 |
++------------------+
+```
diff --git a/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/min.md b/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/min.md
new file mode 100755
index 0000000000..e6c8f18516
--- /dev/null
+++ b/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/min.md
@@ -0,0 +1,19 @@
+# MIN
+
+## Syntax
+
+`MIN(expr)`
+
+## Description
+
+返回expr表达式的最小值
+
+## Examples
+```
+MySQL > select min(scan_rows) from log_statis group by datetime;
++------------------+
+| min(`scan_rows`) |
++------------------+
+|                0 |
++------------------+
+```
diff --git a/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/ndv.md b/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/ndv.md
new file mode 100644
index 0000000000..ff3a087534
--- /dev/null
+++ b/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/ndv.md
@@ -0,0 +1,21 @@
+# NDV
+
+## Syntax
+
+`NDV(expr)`
+
+## Description
+
+返回类似于 COUNT(DISTINCT col) 结果的近似值聚合函数。
+
+它比 COUNT 和 DISTINCT 组合的速度更快，并使用固定大小的内存，因此对于高基数的列可以使用更少的内存。
+
+## Examples
+```
+MySQL > select ndv(query_id) from log_statis group by datetime;
++-----------------+
+| ndv(`query_id`) |
++-----------------+
+| 17721           |
++-----------------+
+```
diff --git a/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/percentile_approx.md b/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/percentile_approx.md
new file mode 100755
index 0000000000..329f6b104a
--- /dev/null
+++ b/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/percentile_approx.md
@@ -0,0 +1,21 @@
+# PERCENTILE_APPROX
+
+## Syntax
+
+`PERCENTILE_APPROX(expr, DOUBLE p)`
+
+## Description
+
+返回第p个百分位点的近似值，p的值介于0到1之间
+
+该函数使用固定大小的内存，因此对于高基数的列可以使用更少的内存，可用于计算tp99等统计值
+
+## Examples
+```
+MySQL > select `table`, percentile_approx(cost_time,0.99) from log_statis group by `table`;
++---------------------+---------------------------+
+| table    | percentile_approx(`cost_time`, 0.99) |
++----------+--------------------------------------+
+| test     |                                54.22 |
++----------+--------------------------------------+
+```
\ No newline at end of file
diff --git a/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/stddev.md b/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/stddev.md
new file mode 100755
index 0000000000..393ffdf141
--- /dev/null
+++ b/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/stddev.md
@@ -0,0 +1,26 @@
+# STDDEV,STDDEV_POP
+
+## Syntax
+
+`STDDEV(expr)`
+
+## Description
+
+返回expr表达式的标准差
+
+## Examples
+```
+MySQL > select stddev(scan_rows) from log_statis group by datetime;
++---------------------+
+| stddev(`scan_rows`) |
++---------------------+
+|  2.3736656687790934 |
++---------------------+
+
+MySQL > select stddev_pop(scan_rows) from log_statis group by datetime;
++-------------------------+
+| stddev_pop(`scan_rows`) |
++-------------------------+
+|      2.3722760595994914 |
++-------------------------+
+```
diff --git a/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/stddev_samp.md b/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/stddev_samp.md
new file mode 100755
index 0000000000..5243a99b90
--- /dev/null
+++ b/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/stddev_samp.md
@@ -0,0 +1,19 @@
+# STDDEV_SAMP
+
+## Syntax
+
+`STDDEV_SAMP(expr)`
+
+## Description
+
+返回expr表达式的样本标准差
+
+## Examples
+```
+MySQL > select stddev_samp(scan_rows) from log_statis group by datetime;
++--------------------------+
+| stddev_samp(`scan_rows`) |
++--------------------------+
+|        2.372044195280762 |
++--------------------------+
+```
diff --git a/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/sum.md b/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/sum.md
new file mode 100755
index 0000000000..ad4fe7c993
--- /dev/null
+++ b/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/sum.md
@@ -0,0 +1,19 @@
+# SUM
+
+## Syntax
+
+`SUM(expr)`
+
+## Description
+
+用于返回选中字段所有值的和
+
+## Examples
+```
+MySQL > select sum(scan_rows) from log_statis group by datetime;
++------------------+
+| sum(`scan_rows`) |
++------------------+
+|       8217360135 |
++------------------+
+```
diff --git a/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/var_samp.md b/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/var_samp.md
new file mode 100755
index 0000000000..5da00ad73d
--- /dev/null
+++ b/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/var_samp.md
@@ -0,0 +1,19 @@
+# VAR_SAMP,VARIANCE_SAMP
+
+## Syntax
+
+`VAR_SAMP(expr)`
+
+## Description
+
+返回expr表达式的样本方差
+
+## Examples
+```
+MySQL > select var_samp(scan_rows) from log_statis group by datetime;
++-----------------------+
+| var_samp(`scan_rows`) |
++-----------------------+
+|    5.6227132145741789 |
++-----------------------+
+```
\ No newline at end of file
diff --git a/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/variance.md b/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/variance.md
new file mode 100755
index 0000000000..27e948f3ae
--- /dev/null
+++ b/docs/documentation/cn/sql-reference/sql-functions/aggregate-functions/variance.md
@@ -0,0 +1,26 @@
+# VARIANCE,VAR_POP,VARIANCE_POP
+
+## Syntax
+
+`VARIANCE(expr)`
+
+## Description
+
+返回expr表达式的方差
+
+## Examples
+```
+MySQL > select variance(scan_rows) from log_statis group by datetime;
++-----------------------+
+| variance(`scan_rows`) |
++-----------------------+
+|    5.6183332881176211 |
++-----------------------+
+
+MySQL > select var_pop(scan_rows) from log_statis group by datetime;
++----------------------+
+| var_pop(`scan_rows`) |
++----------------------+
+|   5.6230744719006163 |
++----------------------+
+```
\ No newline at end of file