7.3.10. column_copy¶
7.3.10.1. Summary¶
New in version 5.0.7.
column_copy copies all column values to other column.
You can implement the following features with this command:
Changing column configuration
Changing table configuration
You can change column configuration by the following steps:
Create a new column with new configuration
Copy all values from the current column to the new column
Remove the current column
Rename the new column to the current column
You can change table configuration by the following steps:
Create a new table with new configuration
Create all same columns to the new table
Copy all column values from the current table to the new table
Remove the current table
Rename the new table to the current table
Concrete examples are showed later.
You can’t copy column values from a TABLE_NO_KEY table to another
table. And you can’t copy column values to a TABLE_NO_KEY table
from another table. Because Groonga can’t map records without record
key.
You can copy column values from a TABLE_NO_KEY table to the same
TABLE_NO_KEY table.
You can copy column values from a TABLE_HASH_KEY /
TABLE_PAT_KEY / TABLE_DAT_KEY table to the same or another
TABLE_HASH_KEY / TABLE_PAT_KEY / TABLE_DAT_KEY table.
7.3.10.2. Syntax¶
This command takes four parameters.
All parameters are required:
column_copy from_table
from_name
to_table
to_name
7.3.10.3. Usage¶
Here are use cases of this command:
Changing column configuration
Changing table configuration
7.3.10.3.1. How to change column configuration¶
You can change column value type. For example, you can change
UInt32 column value to ShortText column value.
You can change column type. For example, you can change
COLUMN_SCALAR column to COLUMN_VECTOR column.
You can move a column to other table. For example, you can move
high_score column to Users table from Players table.
Here are basic steps to change column configuration:
Create a new column with new configuration
Copy all values from the current column to the new column
Remove the current column
Rename the new column to the current column
Here is an example to change column value type to Int32 from
ShortText.
Here are schema and data:
Execution example:
table_create Logs TABLE_HASH_KEY ShortText
# [[0, 1337566253.89858, 0.000355720520019531], true]
column_create Logs serial COLUMN_SCALAR Int32
# [[0, 1337566253.89858, 0.000355720520019531], true]
load --table Logs
[
{"_key": "log1", "serial": 1}
]
# [[0, 1337566253.89858, 0.000355720520019531], 1]
The following commands change Logs.serial column value type to
ShortText from Int32:
Execution example:
column_create Logs new_serial COLUMN_SCALAR ShortText
# [[0, 1337566253.89858, 0.000355720520019531], true]
column_copy Logs serial Logs new_serial
# [[0, 1337566253.89858, 0.000355720520019531], true]
column_remove Logs serial
# [[0, 1337566253.89858, 0.000355720520019531], true]
column_rename Logs new_serial serial
# [[0, 1337566253.89858, 0.000355720520019531], true]
select Logs
# [
# [
# 0,
# 1337566253.89858,
# 0.000355720520019531
# ],
# [
# [
# [
# 1
# ],
# [
# [
# "_id",
# "UInt32"
# ],
# [
# "_key",
# "ShortText"
# ],
# [
# "serial",
# "ShortText"
# ]
# ],
# [
# 1,
# "log1",
# "1"
# ]
# ]
# ]
# ]
You can find Logs.serial stores ShortText value from the
response of select.
Here is an example to change column type to COLUMN_VECTOR from
COLUMN_SCALAR.
Here are schema and data:
Execution example:
table_create Entries TABLE_HASH_KEY ShortText
# [[0, 1337566253.89858, 0.000355720520019531], true]
column_create Entries tag COLUMN_SCALAR ShortText
# [[0, 1337566253.89858, 0.000355720520019531], true]
load --table Entries
[
{"_key": "entry1", "tag": "Groonga"}
]
# [[0, 1337566253.89858, 0.000355720520019531], 1]
The following commands change Entries.tag column to
COLUMN_VECTOR from COLUMN_SCALAR:
Execution example:
column_create Entries new_tag COLUMN_VECTOR ShortText
# [[0, 1337566253.89858, 0.000355720520019531], true]
column_copy Entries tag Entries new_tag
# [[0, 1337566253.89858, 0.000355720520019531], true]
column_remove Entries tag
# [[0, 1337566253.89858, 0.000355720520019531], true]
column_rename Entries new_tag tag
# [[0, 1337566253.89858, 0.000355720520019531], true]
select Entries
# [
# [
# 0,
# 1337566253.89858,
# 0.000355720520019531
# ],
# [
# [
# [
# 1
# ],
# [
# [
# "_id",
# "UInt32"
# ],
# [
# "_key",
# "ShortText"
# ],
# [
# "tag",
# "ShortText"
# ]
# ],
# [
# 1,
# "entry1",
# [
# "Groonga"
# ]
# ]
# ]
# ]
# ]
You can find Entries.tag stores COLUMN_VECTOR value from the
response of select.
Here is an example to move high_score column to Users table
from Players table.
Here are schema and data:
Execution example:
table_create Players TABLE_HASH_KEY ShortText
# [[0, 1337566253.89858, 0.000355720520019531], true]
column_create Players high_score COLUMN_SCALAR Int32
# [[0, 1337566253.89858, 0.000355720520019531], true]
load --table Players
[
{"_key": "player1", "high_score": 100}
]
# [[0, 1337566253.89858, 0.000355720520019531], 1]
The following commands move high_score column to Users table
from Players table:
Execution example:
table_create Users TABLE_HASH_KEY ShortText
# [[0, 1337566253.89858, 0.000355720520019531], true]
column_create Users high_score COLUMN_SCALAR Int32
# [[0, 1337566253.89858, 0.000355720520019531], true]
column_copy Players high_score Users high_score
# [[0, 1337566253.89858, 0.000355720520019531], true]
column_remove Players high_score
# [[0, 1337566253.89858, 0.000355720520019531], true]
select Users
# [
# [
# 0,
# 1337566253.89858,
# 0.000355720520019531
# ],
# [
# [
# [
# 1
# ],
# [
# [
# "_id",
# "UInt32"
# ],
# [
# "_key",
# "ShortText"
# ],
# [
# "high_score",
# "Int32"
# ]
# ],
# [
# 1,
# "player1",
# 100
# ]
# ]
# ]
# ]
You can find Users.high_score is moved from Players.high_score
from the response of select.
7.3.10.3.2. How to change table configuration¶
You can change table key type. For example, you can change
key type to ShortText from Int32.
You can change table type. For example, you can change
TABLE_HASH_KEY table to TABLE_PAT_KEY table.
You can also change other options such as default tokenizer and
normalizer. For example, you can change default tokenizer to
TokenBigramSplitSymbolAlphaDigit from TokenBigrm.
Note
You can’t change TABLE_NO_KEY table. Because TABLE_NO_KEY
doesn’t have record key. Groonga can’t identify copy destination
record without record key.
Here are basic steps to change table configuration:
Create a new table with new configuration
Create all same columns to the new table
Copy all column values from the current table to the new table
Remove the current table
Rename the new table to the current table
Here is an example to change table key type to ShortText from
Int32.
Here are schema and data:
Execution example:
table_create IDs TABLE_HASH_KEY Int32
# [[0, 1337566253.89858, 0.000355720520019531], true]
column_create IDs label COLUMN_SCALAR ShortText
# [[0, 1337566253.89858, 0.000355720520019531], true]
column_create IDs used COLUMN_SCALAR Bool
# [[0, 1337566253.89858, 0.000355720520019531], true]
load --table IDs
[
{"_key": 100, "label": "ID 100", used: true}
]
# [[0, 1337566253.89858, 0.000355720520019531], 1]
The following commands change IDs table key type to ShortText
from Int32:
Execution example:
table_create NewIDs TABLE_HASH_KEY Int32
# [[0, 1337566253.89858, 0.000355720520019531], true]
column_create NewIDs label COLUMN_SCALAR ShortText
# [[0, 1337566253.89858, 0.000355720520019531], true]
column_create NewIDs used COLUMN_SCALAR Bool
# [[0, 1337566253.89858, 0.000355720520019531], true]
column_copy IDs label NewIDs label
# [[0, 1337566253.89858, 0.000355720520019531], true]
column_copy IDs used NewIDs used
# [[0, 1337566253.89858, 0.000355720520019531], true]
table_remove IDs
# [[0, 1337566253.89858, 0.000355720520019531], true]
table_rename NewIDs IDs
# [[0, 1337566253.89858, 0.000355720520019531], true]
select IDs
# [
# [
# 0,
# 1337566253.89858,
# 0.000355720520019531
# ],
# [
# [
# [
# 1
# ],
# [
# [
# "_id",
# "UInt32"
# ],
# [
# "_key",
# "Int32"
# ],
# [
# "label",
# "ShortText"
# ],
# [
# "used",
# "Bool"
# ]
# ],
# [
# 1,
# 100,
# "ID 100",
# true
# ]
# ]
# ]
# ]
You can find IDs stores ShortText key from the response of
select.
Here is an example to change table type to TABLE_PAT_KEY from
TABLE_HASH_KEY.
Here are schema and data:
Execution example:
table_create Names TABLE_HASH_KEY ShortText
# [[0, 1337566253.89858, 0.000355720520019531], true]
column_create Names used COLUMN_SCALAR Bool
# [[0, 1337566253.89858, 0.000355720520019531], true]
load --table Entries
[
{"_key": "alice", "used": false}
]
# [[0, 1337566253.89858, 0.000355720520019531], 1]
The following commands change Names table to TABLE_PAT_KEY
from TABLE_HASH_KEY:
Execution example:
table_create NewNames TABLE_PAT_KEY ShortText
# [[0, 1337566253.89858, 0.000355720520019531], true]
column_create NewNames used COLUMN_SCALAR Bool
# [[0, 1337566253.89858, 0.000355720520019531], true]
column_copy Names used NewNames used
# [[0, 1337566253.89858, 0.000355720520019531], true]
table_remove Names
# [[0, 1337566253.89858, 0.000355720520019531], true]
table_rename NewNames Names
# [[0, 1337566253.89858, 0.000355720520019531], true]
select Names --filter '_key @^ "ali"'
# [
# [
# 0,
# 1337566253.89858,
# 0.000355720520019531
# ],
# [
# [
# [
# 0
# ],
# [
# [
# "_id",
# "UInt32"
# ],
# [
# "_key",
# "ShortText"
# ],
# [
# "used",
# "Bool"
# ]
# ]
# ]
# ]
# ]
You can find Names is a TABLE_PAT_KEY because select can
use Prefix search operator. You can’t use
Prefix search operator with TABLE_HASH_KEY.
7.3.10.4. Parameters¶
This section describes parameters.
7.3.10.4.1. Required parameters¶
All parameters are required.
7.3.10.4.1.1. from_table¶
Specifies the table name of source column.
You can specify any table including TABLE_NO_KEY table.
If you specify TABLE_NO_KEY table, to_table
must be the same table.
Here is an example to use from_table.
Here are schema and data:
Execution example:
table_create FromTable TABLE_HASH_KEY ShortText
# [[0, 1337566253.89858, 0.000355720520019531], true]
column_create FromTable from_column COLUMN_SCALAR ShortText
# [[0, 1337566253.89858, 0.000355720520019531], true]
column_create FromTable to_column COLUMN_SCALAR ShortText
# [[0, 1337566253.89858, 0.000355720520019531], true]
load --table FromTable
[
{"_key": "key1", "from_column": "value1"}
]
# [[0, 1337566253.89858, 0.000355720520019531], 1]
select FromTable --output_columns _key,from_column,to_column
# [
# [
# 0,
# 1337566253.89858,
# 0.000355720520019531
# ],
# [
# [
# [
# 1
# ],
# [
# [
# "_key",
# "ShortText"
# ],
# [
# "from_column",
# "ShortText"
# ],
# [
# "to_column",
# "ShortText"
# ]
# ],
# [
# "key1",
# "value1",
# ""
# ]
# ]
# ]
# ]
You can copy all values to to_column from from_column:
Execution example:
column_copy FromTable from_column FromTable to_column
# [[0, 1337566253.89858, 0.000355720520019531], true]
select FromTable --output_columns _key,from_column,to_column
# [
# [
# 0,
# 1337566253.89858,
# 0.000355720520019531
# ],
# [
# [
# [
# 1
# ],
# [
# [
# "_key",
# "ShortText"
# ],
# [
# "from_column",
# "ShortText"
# ],
# [
# "to_column",
# "ShortText"
# ]
# ],
# [
# "key1",
# "value1",
# "value1"
# ]
# ]
# ]
# ]
7.3.10.4.1.2. from_name¶
Specifies the column name to be copied values.
See from_table for example.
7.3.10.4.1.3. to_table¶
Specifies the table name of destination column.
You can specify the same table name as from_table when you want to copy column values in the same table.
You can’t specify TABLE_NO_KEY table to to_table because
Groonga can’t identify destination records without record key.
There is one exception. If you specify the same name as from_table
to to_table, you can use TABLE_NO_KEY table as
to_table. Because Groonga can identify destination records when
source table and destination table is the same table.
Here is an example to use to_table.
Here are schema and data:
Execution example:
table_create Table TABLE_HASH_KEY ShortText
# [[0, 1337566253.89858, 0.000355720520019531], true]
column_create Table column COLUMN_SCALAR ShortText
# [[0, 1337566253.89858, 0.000355720520019531], true]
table_create ToTable TABLE_HASH_KEY ShortText
# [[0, 1337566253.89858, 0.000355720520019531], true]
column_create ToTable to_column COLUMN_SCALAR ShortText
# [[0, 1337566253.89858, 0.000355720520019531], true]
load --table Table
[
{"_key": "key1", "column": "value1"}
]
# [[0, 1337566253.89858, 0.000355720520019531], 1]
You can copy all values to ToTable.to_column from
Table.column:
Execution example:
column_copy Table column ToTable to_column
# [[0, 1337566253.89858, 0.000355720520019531], true]
select ToTable --output_columns _key,to_column
# [
# [
# 0,
# 1337566253.89858,
# 0.000355720520019531
# ],
# [
# [
# [
# 1
# ],
# [
# [
# "_key",
# "ShortText"
# ],
# [
# "to_column",
# "ShortText"
# ]
# ],
# [
# "key1",
# "value1"
# ]
# ]
# ]
# ]
7.3.10.4.1.4. to_name¶
Specifies the destination column name.
See to_table for example.
7.3.10.4.2. Optional parameters¶
There is no optional parameter.
7.3.10.5. Return value¶
The command returns true as body on success such as:
[HEADER, true]
If the command fails, error details are in HEADER.
See Output format for HEADER.