[TD-3984] <docs>: describe "union all" clause.

documentation20/cn/08.connector/01.java/docs.md

@@ -16,7 +16,6 @@ TDengine 的 JDBC 驱动实现尽可能与关系型数据库驱动保持一致
 
 * TDengine 目前不支持针对单条数据记录的删除操作。
 * 目前不支持事务操作。
-* 目前不支持表间的 union 操作。
 * 目前不支持嵌套查询（nested query）。
 * 对每个 Connection 的实例，至多只能有一个打开的 ResultSet 实例；如果在 ResultSet 还没关闭的情况下执行了新的查询，taos-jdbcdriver 会自动关闭上一个 ResultSet。
 
@@ -447,7 +446,7 @@ Query OK, 1 row(s) in set (0.000141s)
 
 
 
-## TAOS-JDBCDriver 版本以及支持的 TDengine 版本和 JDK 版本
+## <a class="anchor" id="version"></a>TAOS-JDBCDriver 版本以及支持的 TDengine 版本和 JDK 版本
 
 | taos-jdbcdriver 版本 | TDengine 版本     | JDK 版本 |
 | -------------------- | ----------------- | -------- |

documentation20/cn/11.administrator/docs.md

@@ -462,31 +462,31 @@ TDengine的所有可执行文件默认存放在 _/usr/local/taos/bin_ 目录下
 
 | 关键字列表 |             |              |            |           |
 | ---------- | ----------- | ------------ | ---------- | --------- |
-| ABLOCKS    | CONNECTIONS | HAVING       | MODULES    | SLIMIT    |
-| ABORT      | COPY        | ID           | NCHAR      | SMALLINT  |
-| ACCOUNT    | COUNT       | IF           | NE         | SPREAD    |
-| ACCOUNTS   | CREATE      | IGNORE       | NONE       | STABLE    |
-| ADD        | CTIME       | IMMEDIATE    | NOT        | STABLES   |
-| AFTER      | DATABASE    | IMPORT       | NOTNULL    | STAR      |
-| ALL        | DATABASES   | IN           | NOW        | STATEMENT |
-| ALTER      | DAYS        | INITIALLY    | OF         | STDDEV    |
-| AND        | DEFERRED    | INSERT       | OFFSET     | STREAM    |
-| AS         | DELIMITERS  | INSTEAD      | OR         | STREAMS   |
-| ASC        | DESC        | INTEGER      | ORDER      | STRING    |
-| ATTACH     | DESCRIBE    | INTERVAL     | PASS       | SUM       |
-| AVG        | DETACH      | INTO         | PERCENTILE | TABLE     |
-| BEFORE     | DIFF        | IP           | PLUS       | TABLES    |
-| BEGIN      | DISTINCT    | IS           | PRAGMA     | TAG       |
-| BETWEEN    | DIVIDE      | ISNULL       | PREV       | TAGS      |
-| BIGINT     | DNODE       | JOIN         | PRIVILEGE  | TBLOCKS   |
-| BINARY     | DNODES      | KEEP         | QUERIES    | TBNAME    |
-| BITAND     | DOT         | KEY          | QUERY      | TIMES     |
-| BITNOT     | DOUBLE      | KILL         | RAISE      | TIMESTAMP |
-| BITOR      | DROP        | LAST         | REM        | TINYINT   |
-| BOOL       | EACH        | LE           | REPLACE    | TOP       |
-| BOTTOM     | END         | LEASTSQUARES | REPLICA    | TOPIC     |
-| BY         | EQ          | LIKE         | RESET      | TRIGGER   |
-| CACHE      | EXISTS      | LIMIT        | RESTRICT   | UMINUS    |
+| ABLOCKS    | CONNECTIONS | HAVING       | MODULES    | SMALLINT  |
+| ABORT      | COPY        | ID           | NCHAR      | SPREAD    |
+| ACCOUNT    | COUNT       | IF           | NE         | STABLE    |
+| ACCOUNTS   | CREATE      | IGNORE       | NONE       | STABLES   |
+| ADD        | CTIME       | IMMEDIATE    | NOT        | STAR      |
+| AFTER      | DATABASE    | IMPORT       | NOTNULL    | STATEMENT |
+| ALL        | DATABASES   | IN           | NOW        | STDDEV    |
+| ALTER      | DAYS        | INITIALLY    | OF         | STREAM    |
+| AND        | DEFERRED    | INSERT       | OFFSET     | STREAMS   |
+| AS         | DELIMITERS  | INSTEAD      | OR         | STRING    |
+| ASC        | DESC        | INTEGER      | ORDER      | SUM       |
+| ATTACH     | DESCRIBE    | INTERVAL     | PASS       | TABLE     |
+| AVG        | DETACH      | INTO         | PERCENTILE | TABLES    |
+| BEFORE     | DIFF        | IP           | PLUS       | TAG       |
+| BEGIN      | DISTINCT    | IS           | PRAGMA     | TAGS      |
+| BETWEEN    | DIVIDE      | ISNULL       | PREV       | TBLOCKS   |
+| BIGINT     | DNODE       | JOIN         | PRIVILEGE  | TBNAME    |
+| BINARY     | DNODES      | KEEP         | QUERIES    | TIMES     |
+| BITAND     | DOT         | KEY          | QUERY      | TIMESTAMP |
+| BITNOT     | DOUBLE      | KILL         | RAISE      | TINYINT   |
+| BITOR      | DROP        | LAST         | REM        | TOP       |
+| BOOL       | EACH        | LE           | REPLACE    | TOPIC     |
+| BOTTOM     | END         | LEASTSQUARES | REPLICA    | TRIGGER   |
+| BY         | EQ          | LIKE         | RESET      | UMINUS    |
+| CACHE      | EXISTS      | LIMIT        | RESTRICT   | UNION     |
 | CASCADE    | EXPLAIN     | LINEAR       | ROW        | UPLUS     |
 | CHANGE     | FAIL        | LOCAL        | ROWS       | USE       |
 | CLOG       | FILL        | LP           | RP         | USER      |
@@ -498,5 +498,5 @@ TDengine的所有可执行文件默认存放在 _/usr/local/taos/bin_ 目录下
 | CONCAT     | GLOB        | METRICS      | SHOW       | VIEW      |
 | CONFIGS    | GRANTS      | MIN          | SLASH      | WAVG      |
 | CONFLICT   | GROUP       | MINUS        | SLIDING    | WHERE     |
-| CONNECTION | GT          | MNODES       |            |           |
+| CONNECTION | GT          | MNODES       | SLIMIT     |           |
 

documentation20/cn/12.taos-sql/docs.md

@@ -407,7 +407,7 @@ SELECT select_expr [, select_expr ...]
     [INTERVAL (interval_val [, interval_offset])]
     [SLIDING sliding_val]
     [FILL fill_val]
-    [GROUP BY col_list <!-- [HAVING having_condition] -->]
+    [GROUP BY col_list]
     [ORDER BY col_list { DESC | ASC }]
     [SLIMIT limit_val [SOFFSET offset_val]]
     [LIMIT limit_val [OFFSET offset_val]]
@@ -647,7 +647,7 @@ Query OK, 1 row(s) in set (0.001091s)
 3. 从 2.0.17 版本开始，条件过滤开始支持 BETWEEN AND 语法，例如 `WHERE col2 BETWEEN 1.5 AND 3.25` 表示查询条件为“1.5 ≤ col2 ≤ 3.25”。
 
 <!-- 
-### GROUP BY 之后的 HAVING 过滤
+### <a class="anchor" id="having"></a>GROUP BY 之后的 HAVING 过滤
 
 从 2.0.20 版本开始，GROUP BY 之后允许再跟一个 HAVING 子句，对成组后的各组数据再做筛选。HAVING 子句可以使用聚合函数和选择函数作为过滤条件（但暂时不支持 LEASTSQUARES、TOP、BOTTOM、LAST_ROW）。
 
@@ -657,6 +657,16 @@ SELECT AVG(f1), SPREAD(f1, f2, st2.f1) FROM st2 WHERE f1 > 0 GROUP BY f1 HAVING
 ```
 -->
 
+### <a class="anchor" id="union"></a>UNION ALL 操作符
+
+```mysql
+SELECT ...
+UNION ALL SELECT ...
+[UNION ALL SELECT ...]
+```
+
+TDengine 支持 UNION ALL 操作符。也就是说，如果多个 SELECT 子句返回结果集的结构完全相同（列名、列类型、列数、顺序），那么可以通过 UNION ALL 把这些结果集合并到一起。目前只支持 UNION ALL 模式，也即在结果集的合并过程中是不去重的。
+
 ### SQL 示例 
 
 - 对于下面的例子，表tb1用以下语句创建