Query plan

The SHOW PLAN SQL statement and the "plan": N JSON interface option display the query execution plan. The plan is generated and stored during the actual execution, so in the case of SQL, profiling must be enabled in the current session before running that statement. This can be done with a SET profiling=1 statement.

Two items are returned in SQL mode:

  • transformed_tree, which displays the full-text query decomposition.
  • enabled_indexes, which provides information about effective secondary indexes.

To view the query execution plan in a JSON query, add "plan": N to the query. The result will appear as a plan property in the result set. N can be one of the following:

  • 1 - Displays only the textual plan of the root node, similar to the one returned in the SHOW PLAN SQL query. This is the most compact form.
  • 2 - Displays only the JSON object plan, useful for processing.
  • 3 - Displays a JSON object with a textual description of every node. Note that the description for child nodes is also present and repeats part of the parent's description, which makes the whole representation quite large.
‹›
  • SQL
  • JSON
📋
set profiling=1;

select * from hn_small where match('dog|cat') limit 0;

show plan;
‹›
Response
*************************** 1. row ***************************
Variable: transformed_tree
   Value: OR(
  AND(KEYWORD(dog, querypos=1)),
  AND(KEYWORD(cat, querypos=2)))
*************************** 2. row ***************************
Variable: enabled_indexes
   Value:
2 rows in set (0.00 sec)

In some cases, the evaluated query tree can be quite different from the original one due to expansions and other transformations.

‹›
  • SQL
  • JSON
📋
SET profiling=1;

SELECT id FROM forum WHERE MATCH('@title way* @content hey') LIMIT 1;

SHOW PLAN;
‹›
Response
Query OK, 0 rows affected (0.00 sec)

+--------+
| id     |
+--------+
| 711651 |
+--------+
1 row in set (0.04 sec)

+------------------+-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| Variable         | Value                                                                                                                                                                                                                                                                                                                                                                                                                   |
+------------------+-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| transformed_tree | AND(
  OR(
    OR(
      AND(fields=(title), KEYWORD(wayne, querypos=1, expanded)),
      OR(
        AND(fields=(title), KEYWORD(ways, querypos=1, expanded)),
        AND(fields=(title), KEYWORD(wayyy, querypos=1, expanded)))),
    AND(fields=(title), KEYWORD(way, querypos=1, expanded)),
    OR(fields=(title), KEYWORD(way*, querypos=1, expanded))),
  AND(fields=(content), KEYWORD(hey, querypos=2))) |
+------------------+-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
1 row in set (0.00 sec)

See also EXPLAIN QUERY. It displays the execution tree of a full-text query without actually executing the query. Note that when using SHOW PLAN after a query to a real-time table, the result will be based on a random disk/RAM chunk. Therefore, if you have recently modified the table's tokenization settings, or if the chunks vary significantly in terms of dictionaries, etc., you might not get the result you are expecting. Take this into account and consider using EXPLAIN QUERY as well.

JSON result set notes

query property contains the transformed full-text query tree. Each node contains:

  • type: node type. Can be AND, OR, PHRASE, KEYWORD, etc.
  • description: query subtree for this node shown as a string (in SHOW PLAN format).
  • children: child nodes, if any.
  • max_field_pos: maximum position within a field.
  • word: transformed keyword. Keyword nodes only.
  • querypos: position of this keyword in a query. Keyword nodes only.
  • excluded: keyword excluded from query. Keyword nodes only.
  • expanded: keyword added by prefix expansion. Keyword nodes only.
  • field_start: keyword must occur at the very start of the field. Keyword nodes only.
  • field_end: keyword must occur at the very end of the field. Keyword nodes only.
  • boost: keyword IDF will be multiplied by this. Keyword nodes only.

Dot format for SHOW PLAN

SHOW PLAN format=dot allows returning the full-text query execution tree in a hierarchical format suitable for visualization by existing tools, such as https://dreampuf.github.io/GraphvizOnline:

MySQL [(none)]> show plan option format=dot\G
*************************** 1. row ***************************
Variable: transformed_tree
   Value: digraph "transformed_tree"
{

0 [shape=record,style=filled,bgcolor="lightgrey" label="AND"]
0 -> 1
1 [shape=record,style=filled,bgcolor="lightgrey" label="AND"]
1 -> 2
2 [shape=record label="i | { querypos=1 }"]
0 -> 3
3 [shape=record,style=filled,bgcolor="lightgrey" label="AND"]
3 -> 4
4 [shape=record label="me | { querypos=2 }"]
}

SHOW PLAN graphviz example

⪢ Table settings and status

SHOW TABLE INDEXES

The SHOW TABLE INDEXES SQL statement displays the secondary indexes available for a specified table, along with their properties. Secondary indexes improve query performance by creating additional data structures that speed up searches on specific columns.

The syntax is:

SHOW TABLE table_name INDEXES

The displayed properties include the following columns:

  • Name: The name of the secondary index. Can be used in query optimizer hints.
  • Type: The type of data stored in the secondary index. For plain attributes, it matches the type of the original attribute. For secondary indexes generated from JSON attributes, the type is deduced by scanning all documents and determining the types of all JSON properties.
  • Enabled: Indicates whether the index is currently enabled and can be used to improve search speed. When an attribute is updated, the secondary index for that attribute is temporarily disabled until the index is rebuilt. You can rebuild disabled indexes using the ALTER TABLE ... REBUILD SECONDARY command.
  • Percent: In an RT table, different disk chunks may contain different secondary indexes especially when JSON attributes are used. This percentage shows how many chunks have an index with the same name, type, and enabled state.
‹›
  • SQL
SQL
📋
SHOW TABLE test INDEXES;
‹›
Response
+------------------------------+--------+---------+---------+
| Name                         | Type   | Enabled | Percent |
+------------------------------+--------+---------+---------+
| j['addresses']               | uint32 | 1       | 100     |
| j['addresses']['a1']         | uint32 | 1       | 100     |
| j['addresses']['a2']         | uint32 | 1       | 100     |
| j['addresses']['a3']         | uint32 | 1       | 100     |
| j['addresses']['a4']         | uint32 | 1       | 100     |
| j['addresses']['a5']         | uint32 | 1       | 100     |
| j['addresses']['a6']         | uint32 | 1       | 100     |
| j['factor']                  | uint32 | 1       | 100     |
| j['int_arr']                 | uint32 | 1       | 100     |
| j['tags']                    | uint32 | 1       | 100     |
| id                           | int64  | 1       | 100     |
| j['price']                   | float  | 1       | 100     |
| j['addresses']['a1']['id']   | string | 1       | 100     |
| j['addresses']['a1']['name'] | string | 1       | 100     |
| j['addresses']['a2']['id']   | string | 1       | 100     |
| j['addresses']['a2']['name'] | string | 1       | 100     |
| j['addresses']['a3']['id']   | string | 1       | 100     |
| j['addresses']['a3']['name'] | string | 1       | 100     |
| j['addresses']['a4']['id']   | string | 1       | 100     |
| j['addresses']['a4']['name'] | string | 1       | 100     |
| j['addresses']['a5']['id']   | string | 1       | 100     |
| j['addresses']['a5']['name'] | string | 1       | 100     |
| j['addresses']['a6']['id']   | string | 1       | 100     |
| j['addresses']['a6']['name'] | string | 1       | 100     |
| j['arr']                     | string | 1       | 100     |
| j['str']                     | string | 1       | 100     |
| j['tags']['1']               | string | 1       | 100     |
| j['tags']['2']               | string | 1       | 100     |
| j['tags']['3']               | string | 1       | 100     |
+------------------------------+--------+---------+---------+
29 rows in set (0.00 sec)