I'm charged with porting some old code to a new platform and I'm looking at a bit that has me pretty puzzled. I think I'm familiar with what REPLACE INTO does, but this leaves me feeling unsure. The mySQL query is:
REPLACE INTO theTable SET value1 = ?, value2 = ?
My understanding is this only has an effect if those columns are defined as unique. The table definition for theTable looks like this:
CREATE TABLE `theTable` (
`value1` int(11) unsigned DEFAULT NULL,
`value2` int(11) unsigned DEFAULT NULL,
KEY `value1` (`value2`),
KEY `value2` (`value2`)
) ENGINE=MyISAM DEFAULT CHARSET=latin1
I've inserted a row that's an exact duplicate of an existing row and had nothing happen ( a new row was inserted, the old row still exists ) ... am I missing something or did the original developer just do something that was completely useless?
From MySQL manual:
REPLACE works exactly like INSERT,
except that if an old row in the table
has the same value as a new row for a
PRIMARY KEY or a UNIQUE index, the old
row is deleted before the new row is
inserted...
... Note that unless the table has a
PRIMARY KEY or UNIQUE index, using a
REPLACE statement makes no sense. It
becomes equivalent to INSERT, because
there is no index to be used to
determine whether a new row duplicates
another.
So you need to add UNIQUE KEY in order for it to work.
Like you said: it will only replace (in fact delete + insert) row, if UNIQUE constraint is in place. In your table there is none.
Related
So I just started learning SQL online and while learning about constraints, below example was given for using DEFAULT constraint:
CREATE TABLE persons(
ID INT NULL DEFAULT 100,
f_name VARCHAR(25),
l_name VCARCHAR(25),
UNIQUE(ID)
);
My question is, if ID is defaulted to 100, there can be multiple columns having 100 as ID, so wouldn't that contradict UNIQUE constraint, which ensures all columns to have different values?
Thank you for reading and your inputs!
Rohan
Though it's valid SQL and mysql allows this, it is a bad practice to define DEFAULT value on an column with UNIQUE constraint. This poor schema will lead to inconsistency in your data.
mysql> show create table persons;
+---------+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| Table | Create Table |
+---------+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| persons | CREATE TABLE `persons` (
`id` int(11) DEFAULT '100',
`f_name` varchar(10) COLLATE utf8_unicode_ci DEFAULT NULL,
`l_name` varchar(10) COLLATE utf8_unicode_ci DEFAULT NULL,
UNIQUE KEY `id` (`id`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8 COLLATE=utf8_unicode_ci |
+---------+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
You are right, the combination of DEFAULT 100 and UNIQUE makes no sense.
The column is defined as nullable, so there can be many rows with the value null. Only when a row has a value different from null, must it be unique.
In order to insert nulls, you'd explicitely have this in your INSERT statement. If you don't set null explicitly, the default 100 will be written. This works for the first row treated that way, but the second time the 100 will violate the unique constraint, just as you say.
Well, a nullable ID makes no sense either, and ideally an ID should be auto-incremented, so you don't have to worry about using an unused ID, especially in an environment where multiple processes may try to insert rows at the same time.
So, the given examle is just very bad.
The combination of DEFAULT 100 and UNIQUE makes sense.
This combination means that the newly inserted row should have explicitly specified ID column value primarily.
The scheme allows to insert one row without ID value specified. But only one row. If you need to insert another row with this default/generic ID value then you must edit existing row and alter its ID value previously (or delete it).
In practice - this allows to save raw, incomplete, row, and edit it completely in future. For example, you insert generic row, then calculate needed row parameters and set needed references, and finally you assign some definite ID value to this row. After this you may insert another generic row and work with it.
Of course this situation is rare. But it may be useful in some cases.
hello I have a table as follows:
DROP TABLE IF EXISTS `Master_Product`;
CREATE TABLE IF NOT EXISTS `Master_Product` (
`KeyId_Product` bigint(21) UNSIGNED NOT NULL AUTO_INCREMENT COMMENT 'Id of table',
`ID_OrderInform` bigint(21) UNSIGNED NOT NULL AUTO_INCREMENT COMMENT 'Order desire by user for ouput Inform Printed',
`ID_OrderReport` bigint(21) UNSIGNED NOT NULL AUTO_INCREMENT COMMENT 'Order desire by user for ouput Report View Datatable',
`Name_Product` varchar(200) COLLATE utf8_unicode_ci DEFAULT NULL COMMENT 'PDF Name on disk',
PRIMARY KEY (`KeyId`),
UNIQUE KEY `KeyId` (`KeyId`),
KEY `xID_OrderInform` (`ID_OrderInform`),
KEY `xID_OrderReport` (`ID_OrderReport`)
) ENGINE=InnoDB AUTO_INCREMENT=0 DEFAULT CHARSET=utf8 COLLATE=utf8_unicode_ci ROW_FORMAT=COMPACT;
Each time i insert a new product i need fill this table with the order and name, but i cant use the KeyId_Product to sort in printed Inform or View Datatables, Becouse someone users need use a desirable order.
to get this scalability i need use 2 column aditional to store the desirable order, The problem occurs when a new product must be inserted between 2 existing products, and all products that with a higher index of ordering must be pushed +1 to give space to the new one.
The only solution i find is use 2 query additional to update:
UPDATE
Master_Product
SET
ID_OrderInform = ID_OrderInform + 1
WHERE
ID_OrderInform>$NewitemOrderInform
this other
UPDATE
Master_Product
SET
ID_OrderReport = ID_OrderReport + 1
WHERE
ID_OrderReport>$NewitemOrderReport
how can I do all this in a single query, and if when updating the other products there is an error, apply the rolback that inpides even add the new record.
This does both "at the same time":
UPDATE Master_Product
SET ID_OrderInform = ID_OrderInform + (ID_OrderInform > $NewitemOrderInform),
ID_OrderReport = ID_OrderReport + (ID_OrderReport > $NewitemOrderReport);
To explain,... The (id>$x) is a boolean expression that evaluates to either false or true. false is represented as 0; true as 1. So this adds the 'correct' value (0 or 1) to each of the columns.
Meanwhile, a PRIMARY KEY is a UNIQUE KEY, so get rid of the redundant UNIQUE KEY KeyId (KeyId).
What other queries hit this table? It would probably be better to remove the indexes on Inform and Report. It takes a significant amount of effort to update many of the rows in each for each master UPDATE. And you probably only fetch all the rows when you need the ordered list.
A nitpick: BIGINT is overkill.
I am currently working on a project, which involves altering data stored in a MYSQL database. Since the table that I am working on does not have a key, I add a key with the following command:
ALTER TABLE deCoupledData ADD COLUMN MY_KEY INT NOT NULL AUTO_INCREMENT KEY
Due to the fact that I want to group my records according to selected fields, I try to create an index for the table deCoupledData that consists of MY_KEY, along with the selected fields. For example, If I want to work with the fields STATED_F and NOT_STATED_F, I type:
ALTER TABLE deCoupledData ADD INDEX (MY_KEY, STATED_F, NOT_STATED_F)
The real issue is that the fields that I usually work with are more than 16, so MYSQL does not allow super-keys longer than 16 fields.
In conclusion, Is there another way to do this? Can I make (somehow) MYSQL to order the records according to the desired super-key (something like clustering)? I really need to make my script faster and the main overhead is that each group may contain records which are not stored on the same page of the disk, and I assume that my pc starts random I/Os in order to retrieve records.
Thank you for your time.
Nick Katsipoulakis
CREATE TABLE deCoupledData (
AA double NOT NULL DEFAULT '0',
STATED_F double DEFAULT NULL,
NOT_STATED_F double DEFAULT NULL,
MIN_VALUES varchar(128) NOT NULL DEFAULT '-1,-1',
MY_KEY int(11) NOT NULL AUTO_INCREMENT,
PRIMARY KEY (MY_KEY),
KEY AA (AA) )
ENGINE=InnoDB AUTO_INCREMENT=74358 DEFAULT CHARSET=latin1
Okay, first of all, when you add an index over multiple columns and you don't really use the first column, the index is useless.
Example: You have a query like
SELECT *
FROM deCoupledData
WHERE
stated_f = 5
AND not_stated_f = 10
and an index over (MY_KEY, STATED_F, NOT_STATED_F).
The index can only be used, if you have another AND my_key = 1 or something in the WHERE clause.
Imagine you want to look up every person in a telephone book with first name 'John'. Then the knowledge that the book is sorted by last name is useless, you still have to look up every single name.
Also, the primary key does not have to be a surrogate / artificial one. It's nearly always better to have a primary key which is made up of columns which identify each row uniquely anyway.
Also it's not always good to have many indexes. Not only do indexes slow down INSERTs and UPDATEs, sometimes they just cause an extra lookup, since first a look at the index is taken and a second look to find the actual data.
That's just a few tips. Maybe Jordan's hint is not a bad idea, "You should maybe post a new question that has your actual SQL query, table layout, and performance questions".
UPDATE:
Yes, that is possible. According to manual
If you define a PRIMARY KEY on your table, InnoDB uses it as the clustered index.
which means that the data is practically sorted on disk, yes.
Be aware that it's also possible to define a primary key over multiple columns!
Like
CREATE TABLE deCoupledData (
AA double NOT NULL DEFAULT '0',
STATED_F double DEFAULT NULL,
NOT_STATED_F double DEFAULT NULL,
MIN_VALUES varchar(128) NOT NULL DEFAULT '-1,-1',
MY_KEY int(11) NOT NULL AUTO_INCREMENT,
PRIMARY KEY (NOT_STATED_F, STATED_F, AA),
KEY AA (AA) )
ENGINE=InnoDB AUTO_INCREMENT=74358 DEFAULT CHARSET=latin1
as long as the combination of the columns is unique.
I have a table "Bestelling" with 4 columns: "Id" (PK), "KlantId", "Datum", "BestellingsTypeId", now I want to make the column Id auto_increment, however, when I try to do that, I get this error:
ERROR 1062: ALTER TABLE causes auto_increment resequencing, resulting in duplicate entry '1' for key 'PRIMARY'
SQL Statement:
ALTER TABLE `aafest`.`aafest_bestelling` CHANGE COLUMN `Id` `Id` INT(11) NOT NULL AUTO_INCREMENT
ERROR: Error when running failback script. Details follow.
ERROR 1046: No database selected
SQL Statement:
CREATE TABLE `aafest_bestelling` (
`Id` int(11) NOT NULL,
`KlantId` int(11) DEFAULT NULL,
`Datum` date DEFAULT NULL,
`BestellingstypeId` int(11) DEFAULT NULL,
PRIMARY KEY (`Id`)
) ENGINE=InnoDB DEFAULT CHARSET=latin1
Anyone got an idea?
This will happen if the table contains an existing record with an id of 0 (or negative). Updating all existing records to use positive values will allow auto_increment to be set on that column.
Edit: Some people asked how that 0 got in there. For clarification, the MySQL Reference Manual states that "For numeric types, the default is 0, with the exception that for integer or floating-point types declared with the AUTO_INCREMENT attribute, the default is the next value in the sequence." So, if you performed an insert on a table without providing a value for the numeric column before the auto_increment was enabled, then the default 0 would be used during the insert. More details may be found at https://dev.mysql.com/doc/refman/5.0/en/data-type-defaults.html.
I also had this issue when trying to convert a column to auto_increment where one row had a value of 0. An alternative to changing the 0 value temporarily is via setting:
SET SESSION sql_mode='NO_AUTO_VALUE_ON_ZERO';
for the session.
This allowed the column to be altered to auto_increment with the zero id in place.
The zero isn't ideal - and I also wouldn't recommend it being used in an auto_increment column. Unfortunately it's part of an inherited data set so I'm stuck with it for now.
Best to clear the setting (and any others) afterwards with:
SET SESSION sql_mode='';
although it will be cleared when the current client session clsoes.
Full details on the 'NO_AUTO_VALUE_ON_ZERO' setting here.
This happens when MySQL can not determine a proper auto_increment value. In your case, MySQL choose 1 as next auto_increment value, however there is already row with that value in the table.
One way to resolve the issue is to choose a proper auto_increment value yourself:
ALTER TABLE ... CHANGE COLUMN `Id` `Id` INT(11) NOT NULL AUTO_INCREMENT, AUTO_INCREMENT = 123456;
(Note the AUTO_INCREMENT=123456 at the end.)
The easiest way that I have found to solve this issue is to first set the table's AUTO INCREMENT value before altering the column. Just make sure that you set the auto increment value higher than the largest value currently in that column:
ALTER TABLE `aafest`.`aafest_bestelling`
AUTO_INCREMENT = 100,
CHANGE COLUMN `Id` `Id` INT(11) NOT NULL AUTO_INCREMENT
I tested this on MySQL 5.7 and it worked great for me.
Edit: Don't know exactly how that would be caused, but I do have a workaround.
First, create a new table like the old one:
CREATE TABLE aafest_bestelling_new LIKE aafest_bestelling;
Then change the column
ALTER TABLE `aafest`.`aafest_bestelling_new`
CHANGE COLUMN `Id` `Id` INT(11) NOT NULL AUTO_INCREMENT
Dump in the new data:
INSERT INTO aafest_bestelling_new
(KlantId, Datum, BestellingTypeId)
SELECT
KlantId, Datum, BestellingTypeId
FROM aafest_bestelling;
Move the tables:
RENAME TABLE
aafest_bestelling TO aafest_bestelling_old,
aafest_bestelling_new TO aafest_bestelling;
Maybe there's some corruption going on, and this would fix that as well.
P.S.: As a dutchman, I'd highly recommend coding in english ;)
I had a similar issue. Issue was the table had a record with ID = 0 similar to what SystemParadox pointed out. I handled my issue by the following steps:
Steps:
Update record id 0 to be x where x = MAX(id)+1
Alter table to set primary key and auto increment setting
Set seed value to be x+1
Change record id x back to 0
Code Example:
UPDATE foo SET id = 100 WHERE id = 0;
ALTER TABLE foo MODIFY COLUMN id INT(11) NOT NULL AUTO_INCREMENT;
ALTER TABLE foo AUTO_INCREMENT = 101;
UPDATE foo SET id = 0 WHERE id = 100;
This happens because your primary key column already has values.
As the error says ...
ALTER TABLE causes auto_increment resequencing, resulting in duplicate entry '1' for key 'PRIMARY'
which means that your column already has a primary key value 1 which when you auto_increment that column is reassigned causing duplication and hence this error
the solution to this is to remove the primary constraint and then empty the column. Then alter the table setting the primary key again, this time with auto increment.
This error comes because the any table contains an existing record with an id of 0 (or negative). Update all existing records to use positive values will allow auto_increment to be set on that column.
If this didn't work then export all the data and save it any where in you computer and dont first make foreign key relation then fill data in parent table .
This error will also happen if have a MyISAM table that has a composite AUTO_INCREMENT PRIMARY KEY and are trying to combine the keys
For example
CREATE TABLE test1 (
`id` int(11) NOT NULL,
`ver` int(10) unsigned NOT NULL AUTO_INCREMENT,
PRIMARY KEY (`id`,`ver`)
) ENGINE=MyISAM DEFAULT CHARSET=utf8;
INSERT INTO test1 (`id`, `ver`) VALUES (1,NULL),(1,NULL),(1,NULL), (2,NULL),(2,NULL),(2,NULL);
ALTER TABLE test1 DROP PRIMARY KEY, ADD PRIMARY KEY(`ver`);
Not being able to set an existing column to auto_increment also happens if the column you're trying to modify is included in a foreign key relation in another table (although it won't produce the error message referred to in the question).
(I'm adding this answer even though it doesn't relate to the specific error message in the body of the question because this is the first result that shows up on Google when searching for issues relating to not being able to set an existing MySQL column to auto_increment.)
I need to insert records, but if the records exist do a replace instead. Here is what I am currently using:
$sessionDate = date("Y-m-d H:i:s");
foreach($tmpVP as $sessionVP) {
$res = mysql_query("INSERT INTO sessions
(sessionID,sessionDate,sessionVS,sessionVP)
VALUES('$sessionID','$sessionDate','$sessionVS',
'$sessionVP')") ;
}
What I really need is to update any records matching sessionID, sessionVS, and sessionVP and insert new records that don't match.
EDIT:
Table definition
CREATE TABLE `sessions` (
`ID` bigint(20) NOT NULL auto_increment,
`sessionID` varchar(36) NOT NULL,
`sessionDate` datetime NOT NULL,
`sessionUser` bigint(20) NOT NULL,
`sessionVS` varchar(255) NOT NULL,
`sessionVP` bigint(20) NOT NULL,
`reserved` int(11) NOT NULL,
PRIMARY KEY (`ID`),
KEY `ID` (`ID`)
) ENGINE=MyISAM AUTO_INCREMENT=88 DEFAULT CHARSET=utf8
sessionID, sessionVP, and sessionVS are not unique. Sample:
sessionID sessionDate sessionUser sessionVS sessionVP
0t1m58q9ktejuhqlrjqglcoia0 2010-06-20 09:20:53 0 111111 144268736
0t1m58q9ktejuhqlrjqglcoia0 2010-06-20 09:20:53 0 111111 144268819
0t1m58q9ktejuhqlrjqglcoia0 2010-06-20 09:20:53 0 111111 144268880
So, if I have a new record with 0t1m58q9ktejuhqlrjqglcoia0, 111111, and `144268880' I need to update row #3 instead of adding row #4.
Firstly you should add a unique index on (sessionID, sessionVP, sessionVS). You can do this using the following command:
CREATE UNIQUE INDEX ux_sessions_sessionid_sessionvs_sessionvp
ON sessions (sessionID, sessionVS, sessionVP)
Then there are two relatively simple ways to insert or update in MySQL. The first is to use ON DUPLICATE UPDATE:
INSERT INTO sessions
(sessionID,sessionDate,sessionVS,sessionVP)
VALUES
('$sessionID', '$sessionDate', '$sessionVS', '$sessionVP')
ON DUPLICATE KEY UPDATE sessionDate = '$sessionDate'
There other is to use REPLACE:
REPLACE INTO sessions
(sessionID,sessionDate,sessionVS,sessionVP)
VALUES
('$sessionID', '$sessionDate', '$sessionVS', '$sessionVP')
The second is slightly more concise, but has the disadvantage that it internally causes a delete followed by an insert.
There are also a few other issues:
You don't need both a primary key index and an ordinary index on ID. Remove the ordinary index as it is redundant.
You may have an SQL vulnerability. If you have not already validated the input you might want to consider protecting yourself by using mysql_real_escape_string or intval as appropriate. Alternatively you could look at using query parameters.
You are not checking for error conditions. Consider using trigger_error so that if your query has an error you can see what the error is. Seeing the error message can save you a lot of time debugging.
mysql_query("...") or trigger_error(mysql_error());
You might take a look at INSERT ... ON DUPLICATE KEY UPDATE: http://dev.mysql.com/doc/refman/5.0/en/insert-on-duplicate.html
Add an unique key on (sessionID, sessionVS, sessionVP), then use REPLACE instead of INSERT (just substitute the word, syntax is the same).
Based on your table structure, I mean the primary keys you can use
$res = mysql_query("REPLACE INTO sessions
(sessionID,sessionDate,sessionVS,sessionVP)
VALUES('$sessionID','$sessionDate','$sessionVS',
'$sessionVP')") ;
Are the values part of the primary key? If yes, take a look at: http://dev.mysql.com/doc/refman/5.0/en/insert-on-duplicate.html