I have a simple table visitor and another table visitor_tokens.
SQL creation script of visitor:
CREATE TABLE `visitor` (
`id` int(7) UNSIGNED NOT NULL,
`phone` varchar(15) CHARACTER SET utf8mb4 COLLATE utf8mb4_0900_ai_ci DEFAULT NULL,
`created_at` datetime NOT NULL DEFAULT CURRENT_TIMESTAMP,
`updated_at` datetime NOT NULL DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP
) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_0900_ai_ci;
ALTER TABLE `visitor`
ADD PRIMARY KEY (`id`),
ADD UNIQUE KEY `phone_index` (`phone`);
ALTER TABLE `visitor`
MODIFY `id` int UNSIGNED NOT NULL AUTO_INCREMENT, AUTO_INCREMENT=1;
INSERT INTO `visitor` (`phone`) VALUES
('111111111');
SQL creation script of visitor_tokens:
CREATE TABLE `visitor_tokens` (
`id` int(7) UNSIGNED NOT NULL,
`visitor` int(7) UNSIGNED NOT NULL,
`token` varchar(100) COLLATE utf8mb4_general_ci NOT NULL,
`created_at` datetime NOT NULL DEFAULT CURRENT_TIMESTAMP,
`updated_at` datetime NOT NULL DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP
) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_general_ci;
ALTER TABLE `visitor_tokens`
ADD PRIMARY KEY (`id`);
ALTER TABLE `visitor_tokens`
MODIFY `id` int UNSIGNED NOT NULL AUTO_INCREMENT;
INSERT INTO `visitor_tokens` (`visitor`, `token`) VALUES
(1, 'abc_token'),
(1, 'xyz_token');
I want to get some data of visitor by id:
I want to get the phone number (phone column) and the visitor tokens (visitor_tokens.token). All - according to given id.
My current SQL script is: SELECT visitor.phone, visitor_tokens.token FROM visitor JOIN visitor_tokens ON visitor_tokens.visitor=visitor.id WHERE id=1. This gives me only the phone and the first token: abc. But I also want to get the tokens of the visitor from the second table. To get something like [abc_token, xyz_token]. How can I do it?
So this is my solution (also recommend on reading the main comments):
SELECT visitor.phone, GROUP_CONCAT(visitor_tokens.token) AS tokens FROM visitor LEFT JOIN visitor_tokens ON visitor_tokens.visitor=visitor.id WHERE visitor.id=1 LIMIT 1;
It returns result regardless of missing visitor tokens (LEFT JOIN), and also if there are tokens - it returns the tokens separated by comma in one row.
Related
I have the following table structure (example)
CREATE TABLE `Test` (
`id` int(11) NOT NULL,
`order_id` int(11) NOT NULL,
`position_id` int(11) NOT NULL,
`name` varchar(255) COLLATE utf8mb4_unicode_ci NOT NULL,
`price` decimal(10,2) NOT NULL
) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_unicode_ci;
ALTER TABLE `Test` ADD PRIMARY KEY (`id`);
ALTER TABLE `Test` MODIFY `id` int(11) NOT NULL AUTO_INCREMENT;
This table contains data that is constantly in need of updating. There is also new data that needs to be entered. Since there is a lot of data, it will take quite a long time to check each record to make it insert or update.
After studying the question, I realized that I need to use batch insert/update with:
INSERT on DUPLICATE KEY UPDATE
But the documentation says that the fields must have a unique index. But I don't have any unique fields, I can't use the ID field. The uniqueness of the record can only be in a combination of two fields order_id and position_id.
Is it possible to make a batch insert/update if the uniqueness of the record is a bundle of two fields?
You need a composite primary-key. You also don't need your AUTO_INCREMENT id column, so you can drop it.
Like so:
CREATE TABLE `Test` (
`order_id` int NOT NULL,
`position_id` int NOT NULL,
`name` varchar(255) NOT NULL COLLATE utf8mb4_unicode_ci,
`price` decimal(10,2) NOT NULL,
CONSTRAINT PK_Test PRIMARY KEY ( `order_id`, `position_id` )
) ENGINE=InnoDB
Then you can use INSERT ON DUPLICATE KEY UPDATE.
I am trying to set the Default value as UUID() in MySQL version 8.0.13. But upon successful execution, the default value resets to NOT NULL.
MySQL version:
Here is my CREATE TABLE script
CREATE TABLE `session` (
`id` binary(16) NOT NULL DEFAULT (UUID_TO_BIN(UUID(), TRUE)),
`start_timestamp` timestamp NOT NULL,
`end_timestamp` timestamp NULL DEFAULT NULL,
`created_at` timestamp NULL DEFAULT CURRENT_TIMESTAMP,
`updated_at` timestamp NULL DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP,
`status` varchar(20) DEFAULT NULL,
PRIMARY KEY (`id`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_0900_ai_ci;
the log output on table generation,
SQL script was successfully applied to the database.
The TABLE definition post execution:
CREATE TABLE `session` (
`id` binary(16) NOT NULL,
`start_timestamp` timestamp NOT NULL,
`end_timestamp` timestamp NULL DEFAULT NULL,
`created_at` timestamp NULL DEFAULT CURRENT_TIMESTAMP,
`updated_at` timestamp NULL DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP,
`status` varchar(20) DEFAULT NULL,
PRIMARY KEY (`id`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_0900_ai_ci
I am not able to figure out why this could happen when documentation clearly mentions that parenthesis enclosed functions are allowed.
This is unfortunately a bug with default expressions for primary key columns, Expression Default is made NULL during CREATE TABLE query, if field is made PK.
It is fixed in MySQL 8.0.19:
For a column defined as a PRIMARY KEY in a CREATE TABLE statement, a default value given as an expression was ignored. (Bug #29596969, Bug #94668)
As a workaround (if you cannot upgrade), you can add the primary key afterwards with an ALTER TABLE-statement:
CREATE TABLE `session` (
`id` binary(16) NOT NULL DEFAULT (UUID_TO_BIN(UUID(), TRUE)),
`start_timestamp` timestamp NOT NULL,
`end_timestamp` timestamp NULL DEFAULT NULL,
`created_at` timestamp NULL DEFAULT CURRENT_TIMESTAMP,
`updated_at` timestamp NULL DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP,
`status` varchar(20) DEFAULT NULL
) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_0900_ai_ci;
ALTER TABLE `session` ADD PRIMARY KEY(`id`);
I needed the column to not be a binary one. So, in my case, I declared it like this:
`id` char(36) CHARACTER SET utf8mb4 COLLATE utf8mb4_bin NOT NULL DEFAULT (UUID()),
For anyone who needs it to store UUIDs with default values in a char column.
I have implemented custom url shortener in my app and I have one table for that. table structure looks like this:
CREATE TABLE `urls` (
`id` int(11) NOT NULL,
`url_id` varchar(10) DEFAULT NULL,
`long_url` varchar(255) DEFAULT NULL,
`clicked` mediumint(5) NOT NULL DEFAULT 0,
`user_id` varchar(7) DEFAULT NULL,
`type` varchar(15) DEFAULT NULL,
`ad_id` int(11) DEFAULT NULL,
`campaign` int(11) DEFAULT,
`increment` tinyint(1) NOT NULL DEFAULT 0,
`date` date DEFAULT NULL,
`del` enum('1','0') NOT NULL DEFAULT '0'
) ENGINE=InnoDB DEFAULT CHARSET=utf8 ROW_FORMAT=COMPACT
ALTER TABLE `urls`
ADD PRIMARY KEY (`id`),
ADD KEY `url_id` (`url_id`),
ADD KEY `type` (`type`),
ADD KEY `campaign` (`campaign`),
ADD KEY `ad_id` (`ad_id`),
ADD KEY `date` (`date`),
ADD KEY `user_id` (`user_id`);
The table now has 20.000.000 records and currently growing by 300k-400k records per day.
url_id column is unique varchar(10) and url looks like that: http://example.com/asdfghjklu
Now i have partitioned this table into 10 partitions by HASH(id):
PARTITION BY HASH (`id`)
PARTITIONS 10;
When I try to generate reports and join this table on others query is getting really slow, so slow even can't get 1 week report.
When I try to make big query in this table I filter almost every query with dates and I think it will be much better if I partition this table by date column.
Is it good idea?
As I read if I want to partition this table by date I need to add date in composite primary key: PRIMARY KEY(id, date)
What do you think about this? How do I improve my query performance?
I wold recommend use hash partition using date or month or YEAR
CREATE TABLE `urls` (
`id` int(11) NOT NULL,
`url_id` varchar(10) DEFAULT NULL,
`long_url` varchar(255) DEFAULT NULL,
`clicked` mediumint(5) NOT NULL DEFAULT 0,
`user_id` varchar(7) DEFAULT NULL,
`type` varchar(15) DEFAULT NULL,
`ad_id` int(11) DEFAULT NULL,
`campaign` int(11) DEFAULT,
`increment` tinyint(1) NOT NULL DEFAULT 0,
`date` date DEFAULT NULL,
`del` enum('1','0') NOT NULL DEFAULT '0',
PartitionsID int(4) unsigned NOT NULL,
KEY PartitionsID (PartitionsID)
) ENGINE=InnoDB DEFAULT CHARSET=latin1
PARTITION BY HASH (PartitionsID)
PARTITIONS 366;
IN PARTITION ID you just need to insert TO_DAYS(date) so you have only one value for entire day .
SOURCE
and it will make easy for partition for each day or you can do with month wise also depending on your data size .
for select
you can use below query as example
SELECT *
FROM TT ACT
WHERE ACT.CustomerID = vCustomerID
AND ACT.TransactionTime BETWEEN vInvoiceEndDate AND vPaymentDueDate
AND ACT.TrxnInfoTypeID IN (19, 23)
AND ACT.PaymentType = '1'
AND ACT.PartitionsID BETWEEN TO_DAYS(vInvoiceEndDate) AND TO_DAYS(vPaymentDueDate);
I have archived some old line items for invoices that are no longer current but still need to reference them. I think I need to create a VIEW but not really understanding it. Can someone help so I can run a query to pull the invoice and then the total of all the line items assigned (no matter what table the items are in)?
CREATE TABLE `Invoice` (
`Invoice_ID` INT(11) UNSIGNED NOT NULL AUTO_INCREMENT,
`Invoice_CreatedDateTime` DATETIME DEFAULT NULL,
`Invoice_Status` ENUM('Paid','Sent','Unsent','Hold') DEFAULT NULL,
`LastUpdatedAt` TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP,
PRIMARY KEY (`ID`),
KEY `LastUpdatedAt` (`LastUpdatedAt`)
) ENGINE=MYISAM DEFAULT CHARSET=latin1
CREATE TABLE `Invoice_LineItem` (
`LineItem_ID` INT(11) UNSIGNED NOT NULL AUTO_INCREMENT,
`LineItem_ChargeType` VARCHAR(64) NOT NULL DEFAULT '',
`LineItem_InvoiceID` INT(11) UNSIGNED DEFAULT NULL,
`LineItem_Amount` DECIMAL(11,4) DEFAULT NULL,
`LastUpdatedAt` TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP,
PRIMARY KEY (`LineItem_ID`),
KEY `LastUpdatedAt` (`LastUpdatedAt`),
KEY `LineItem_InvoiceID` (`LineItem_InvoiceID`)
) ENGINE=MYISAM AUTO_INCREMENT=1 DEFAULT CHARSET=latin1
CREATE TABLE `Invoice_LineItem_Archived` (
`LineItem_ID` INT(11) UNSIGNED NOT NULL AUTO_INCREMENT,
`LineItem_ChargeType` VARCHAR(64) NOT NULL DEFAULT '',
`LineItem_InvoiceID` INT(11) UNSIGNED DEFAULT NULL,
`LineItem_Amount` DECIMAL(11,4) DEFAULT NULL,
`LastUpdatedAt` TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP,
PRIMARY KEY (`LineItem_ID`),
KEY `LastUpdatedAt` (`LastUpdatedAt`),
KEY `LineItem_InvoiceID` (`LineItem_InvoiceID`)
) ENGINE=MYISAM AUTO_INCREMENT=1 DEFAULT CHARSET=latin1
Typically I would just run the following query to get the amount due on the invoices
SELECT
Invoice_ID,
Invoice_CreatedDateTime,
Invoice_Status,
(SELECT SUM(LineItem_Amount) AS totAmt FROM Invoice_LineItem WHERE LineItem_InvoiceID=Invoice_ID) AS Invoice_Total
FROM
Invoice
WHERE
Invoice_Status='Sent'
Also how can I select all the line items from both tables in one query?
SELECT
LineItem_ID,
LineItem_ChargeType,
LineItem_Amount
FROM
Invoice_LineItem
WHERE
LineItem_InvoiceID='1234'
You can use the MERGE Storage Engine to create a virtual table that's the union of two real tables:
CREATE TABLE Invoice_LineItem_All
(
`LineItem_ID` INT(11) UNSIGNED NOT NULL AUTO_INCREMENT,
`LineItem_ChargeType` VARCHAR(64) NOT NULL DEFAULT '',
`LineItem_InvoiceID` INT(11) UNSIGNED DEFAULT NULL,
`LineItem_Amount` DECIMAL(11,4) DEFAULT NULL,
`LastUpdatedAt` TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP,
KEY (`LineItem_ID`),
KEY `LastUpdatedAt` (`LastUpdatedAt`),
KEY `LineItem_InvoiceID` (`LineItem_InvoiceID`)
) ENGINE=MERGE UNION=(Invoice_LineItem_Archived, Invoice_LineItem);
You can use UNION :
SELECT a.* FROM a
UNION
SELECT b.* FROM b;
You just need to have the same number and type of column in your different queries.
As far as I remember, you can add test in sub-queries, but I'm not sure you can order on the global result.
http://dev.mysql.com/doc/refman/4.1/en/union.html
I like to append a table to add column but without using alert table command
e.g.
This is the table which is missing some columns.
CREATE TABLE IF NOT EXISTS `admin` (
`id` int(11) NOT NULL auto_increment,
`username` varchar(20) NOT NULL,
`passwd` varchar(40) NOT NULL,
`isActive` tinyint(1) NOT NULL default '1',
`lastVisit` datetime NOT NULL default '0000-00-00 00:00:00',
`modifyAt` datetime NOT NULL,
`createdAt` datetime NOT NULL,
PRIMARY KEY (`id`)
) ENGINE=MyISAM DEFAULT CHARSET=utf8;
So if i run this query then it should automatically add missing columns into my tables
CREATE TABLE IF NOT EXISTS `admin` (
`id` int(11) NOT NULL auto_increment,
`username` varchar(20) NOT NULL,
`passwd` varchar(40) NOT NULL,
`name` varchar(100) NOT NULL,
`originalUser` tinyint(1) NOT NULL default '0',
`isActive` tinyint(1) NOT NULL default '1',
`lastVisit` datetime NOT NULL default '0000-00-00 00:00:00',
`modifyAt` datetime NOT NULL,
`createdAt` datetime NOT NULL,
PRIMARY KEY (`id`)
) ENGINE=MyISAM DEFAULT CHARSET=utf8;
Can this be possible to do without using alert table command ?
I understand your question as you want to add some columns to your table. Please be informed that the term row is usually related to the actual data in your table, not the columns itself. If my assumption is wrong, please clarify your question.
You cannot use CREATE TABLE for altering a table. It is there to create table,
and if it cannot create it, it will in most cases throw an error like you described. Another command exists for that reason: ALTER TABLE.
You might do it something like this.
(1) Create your table with your CREATE TABLE syntax above:
CREATE TABLE IF NOT EXISTS `admin` (
`id` int(11) NOT NULL auto_increment,
`username` varchar(20) NOT NULL,
`passwd` varchar(40) NOT NULL,
`isActive` tinyint(1) NOT NULL default '1',
`lastVisit` datetime NOT NULL default '0000-00-00 00:00:00',
`modifyAt` datetime NOT NULL,
`createdAt` datetime NOT NULL,
PRIMARY KEY (`id`)
) ENGINE=MyISAM DEFAULT CHARSET=utf8;
(2) Use ALTER TABLE like this to make the modifications I think you want to have in your second statement (two more columns):
ALTER TABLE
ADD COLUMN `name` varchar(100) NOT NULL AFTER `passwd`,
ADD COLUMN `originalUser` tinyint(1) NOT NULL default '0' AFTER `name`;
Not related to your question, but I'd avoid column names like name, because if you don't escape them properly it'll throw you other errors (see reserved words).