Yii INSERT ... ON DUPLICATE UPDATE - php

I am working on a Yii project. How can I use the ON DUPLICATE feature of MySQL ( http://dev.mysql.com/doc/refman/5.0/en/insert-on-duplicate.html ) when doing a save() on a Yii model?
My MySQL is as follows:
CREATE TABLE `ck_space_calendar_cache` (
`space_id` int(11) NOT NULL,
`day` date NOT NULL,
`available` tinyint(1) unsigned NOT NULL DEFAULT '0',
`price` decimal(12,2) DEFAULT NULL,
`offer` varchar(45) DEFAULT NULL,
`presale_date` date DEFAULT NULL,
`presale_price` decimal(12,2) DEFAULT NULL,
`value_x` int(11) DEFAULT NULL,
`value_y` int(11) DEFAULT NULL,
PRIMARY KEY (`space_id`,`day`),
KEY `space` (`space_id`),
CONSTRAINT `space` FOREIGN KEY (`space_id`) REFERENCES `ck_space` (`id`) ON DELETE CASCADE ON UPDATE NO ACTION
) ENGINE=InnoDB DEFAULT CHARSET=utf8;
My PHP is a follows:
$cache = new SpaceCalendarCache();
$cache->attributes = $day; //Some array with attributes
$cache->save();
If there is a duplicate in my primary key (sapce_id,day), I don't want it to complain, I just want it to update with the latest data.
I know how to do it in raw SQL, I was just wondering if there is a clean Yii way to do it.

You are using models in Yii, its quite simple .. try to load you model where you suspect to have duplicate entries, if you find the entry the model is loaded else null is return. now if your model is null simply create new model. rest is your normal code to insert a new record.
//try to load model with available id i.e. unique key
$model = someModel::model()->findByPk($id);
//now check if the model is null
if(!$model) $model = new someModel();
//Apply you new changes
$model->attributes = $attributes;
//save
$model->save();
Refer to post controllers update method in sample app Yii blog. I might be wrong with spelling of function names, sorry for that.

I'm repeating two main points from previous answers I think you should keep:
Don't (try to) use "on duplicate key update" since its MySQL-only, as txyoji points out.
Prefer the select->if not found then insert->else insert demonstrated by Uday Sawant.
There's another point here, though: Concurrency. Although for low traffic applications the probability that you'll get in trouble is minimal (still never zero), I think we always be careful about this.
From a transactional point of view, "INSERT .. ON DUPLICATE UPDATE" is not equivalent to selecting into your application's memory and then inserting or updating. The first is a single transaction, then second is not.
Here's a bad scenario:
You do select your record using findByPk() which returns null
Some other transaction (from some other user request) inserts a record with the id you just failed to select
At the next instant you try to insert it again
In this case you'll either get an exception (if you're working with a unique key, as you do here) or a duplicate entry. Duplicate entries are much harder to pick up (usually nothing seems weird until your users see duplicate records).
The solution here is to set a strict isolation level, for example "serializable", and then begin a transaction.
Here's an example for yii:
Yii::app()->db->createCommand('SET TRANSACTION ISOLATION LEVEL SERIALIZABLE');
$trn = Yii::app()->db->beginTransaction();
try {
// Try to load model with available id i.e. unique key
// Since we're in serializable isolation level, even if
// the record does not exist the RDBMS will lock this key
// so nobody can insert it until you commit.
// The same shold for the (most usual) case of findByAttributes()
$model = someModel::model()->findByAttributes(array(
'sapce_id' => $sapceId,
'day' => $day
));
//now check if the model is null
if (!$model) {
$model = new someModel();
}
//Apply you new changes
$model->attributes = $attributes;
//save
$model->save();
// Commit changes
$trn->commit();
} catch (Exception $e) {
// Rollback transaction
$trn->rollback();
echo $e->getMessage();
}
You can see more about isolation levels at least in the following links and see what every isolation level has to offer in data integrity in exchange for concurrency
http://technet.microsoft.com/en-us/library/ms173763.aspx
http://dev.mysql.com/doc/refman/5.0/en/set-transaction.html

I overrode beforeValidate() where I checked if a duplicate exists. If one does, I set $this->setIsNewRecord(false);
Seems to work. Not sure how performant it is.

The "On Duplicate Key Update" feature is specific to MySQL's dialect of SQL. Its unlikely to be implemented in any data abstraction layer. ZendDB and Propel don't have an equivalent.
You can simulate the behavior by attempting an insert in a try/catch and update if insert fails with the proper error code. (duplicate key error).

I agree with #txyoji's analysis of the problem, but I would use a different solution.
You can extend the save() method of the model to look for an existing record, and update it, or insert a new row if it doesn't.

you have to use try catch like that:
try{
$model->save();
}
catch(CDbException $e){
$model->isNewRecord = false;
$model->save();
}

Ooops, sorry.. this answer for yii2
If you dont use yii model, this function generates mysql syntax insert on duplicates key update
static function insertDuplicate($table, $columns, $duplicates, $values="",$ignores=false){
$params=array();
$names=array();
$tipe="VALUES";
$ignore=($ignores===true)?"IGNORE":"";
$placeholders=array();
if(is_array($columns)){
if(!isset($columns[0])){
foreach($columns as $name=>$value)
{
$names[]=$name;
if($value instanceof CDbExpression)
{
$placeholders[] = $value->expression;
foreach($value->params as $n => $v)
$params[$n] = $v;
}
else
{
$placeholders[] = ':' . $name;
$params[':' . $name] = $value;
}
}
}else{
$names=$columns;
}
$myColumn=implode(', ',$names);
if($values!=""){
$myValue=$values;
}else{
$myValue='('.implode(', ', $placeholders).')';
}
}else{
$myColumn=$columns;
$myValue=$values;
}
if($values!=""){
if(substr(strtoupper($values),0,6)=="SELECT"){
$tipe="";
}
}
$d = array();
if(is_array($duplicates)){
if(!isset($duplicates[0])){
foreach($duplicates as $duplicate=>$act)
{
if($act=="increase"){
$dup=$table.".".$duplicate . ' = '.$table.".".$duplicate.' + VALUES('.$duplicate.')';
}elseif($act=="decrease"){
$dup=$table.".".$duplicate . ' = '.$table.".".$duplicate.' - VALUES('.$duplicate.')';
}else{
$dup=$table.".".$duplicate . ' = VALUES('.$duplicate.')';
}
$d[] = $dup;
}
}else{
foreach($duplicates as $duplicate){
$dup=$duplicate . ' = VALUES('.$duplicate.')';
$d[] = $dup;
}
}
$myDuplicate= implode(', ', $d);
}else{
$myDuplicate=$duplicates;
}
$sql='INSERT '.$ignore.' INTO ' . $table
. ' (' . $myColumn . ') '.$tipe.' '
. $myValue . ' ON DUPLICATE KEY UPDATE ' .$myDuplicate;
return Yii::$app->db->createCommand($sql)->bindValues($params)->execute();
}
Place that function into someclass, and dont forget use
use yii\db\Command;
in that class
That function can insert on key update, update increment, update decrement, update multi from a value, and update from select
Usage :
//to update available=1 and price into 100
someclass::insertDuplicate(
'ck_space_calendar_cache',
['sapce_id'=>1,'day'=>'2022-09-01','available'=>1,'price'=>100],
['available','price']
);
//to update price increase by 100, (if price is decrease then change it to decrease)
someclass::insertDuplicate(
'ck_space_calendar_cache',
['sapce_id'=>1,'day'=>'2022-09-01','price'=>100],
['price'=>'increase']
);
//to update mass with a value
someclass::insertDuplicate(
'ck_space_calendar_cache',
['sapce_id','day','price'],
['price'],
'(1,'2022-09-01',100),(2,'2022-09-01',300),(3,'2022-09-01',100)'
);
//to update mass with select from another table
someclass::insertDuplicate(
'ck_space_calendar_cache',
['sapce_id','day','price'],
['price'],
'SELECT otherid as sapce_id, otherday as day, otherprice as price from other WHERE otherprice>100'
);

Related

I am trying to update but it is adding a new row

this is my code for updating:
PS: empid is a foreign key but i think that shouldnt be the reason and the code is in CakePHP
if($this->request->is('post'))
{
$this->request->data["Leave"]["empid"] = $this->request->data["id"];
$this->Leave->empid = $this->request->data["Leave"]["empid"];
$this->request->data["Leave"]["leave_start"] = $this->request->data["start_date"];
$this->request->data["Leave"]["leave_end"] = $this->request->data["end_date"];
$this->request->data["Leave"]["leave_taken"] = $this->request->data["leave_taken"];
if($this->Leave->save($this->request->data['Leave']))
{
return $this->redirect(array('action' => 'manage_leave'));
}
}
// This code is inserting a new row instead of updating and also not adding any value in the new row
May be your trying to update the foreign table data using simple save.
Update multiple records for foreign key
Model::updateAll(array $fields, mixed $conditions)
Example
$this->Ticket->updateAll(
array('Ticket.status' => "'closed'"),
array('Ticket.customer_id' => 453)
);
Simple save for the primary key
Make sure that your HTML has empid
echo $this->Form->input('Leave.empid', array('type' => 'hidden'));
Save Model
$this->Leave->empid = $this->request->data["Leave"]["empid"]; //2
$this->Leave->save($this->request->data);
In between, you can also try to set the model data and check the $this->Leave->validates() and $this->Leave->validationError if they are giving any validation errors.
// Create: id isn't set or is null
$this->Recipe->create();
$this->Recipe->save($this->request->data);
// Update: id is set to a numerical value
$this->Recipe->id = 2;
$this->Recipe->save($this->request->data);
You can find more information about all Saving your data
Hope this helps you :)
And in case if $empid is primary key of corresponding table of Leave model (e.g leaves), Just replace:
$this->Leave->empid = $this->request->data["Leave"]["empid"];
By
$this->Leave->id = $this->request->data["Leave"]["empid"];

Update a row using idiorm and php

I have this function to update a record, but i cannot it fails and send me a "Primary key ID missing from row or is null" message, how can I fix it?
public static function update_child($data)
{
try
{
$update= ORM::for_table("dm_child",DM_TAG)
->where_equal($data["id_child"]);
$update -> set(array(
"gender" => $data["gender"]
"age_year" =>$data["year"]
"age_month" => $data["month"]
));
$update -> save();
}
catch(PDOException $ex)
{
ORM::get_db()->rollBack();
throw $ex;
}
}
Idiorm assumes that the name of the primary key is 'id', which is not that, in your case.
Therefore you have to explicitly specify it to Idiorm:
<?php
ORM::configure('id_column_overrides', array(
'dm_child' => 'id_child',
'other_table' => 'id_table',
));
See Docs>Configuration.
The answer is indeed the one provided by #iNpwd for changing the default 'id' column name for queries on a per table basis:
ORM::configure('id_column_overrides', array(
'table_name' => 'column_name_used_as_id',
'other_table' => array('pk_1', 'pk_2') // a compound primary key
));
The thing that was biting me on getting it to recognize my query was WHERE I was changing the ORM::configure values. I was not in the correct file.
A deeper link to specifically the ID Column configuration: http://idiorm.readthedocs.org/en/latest/configuration.html#id-column
I just met this problem 2 minutes ago. The real reason is, you forgot select id field in querying.
demo:
$demo = ORM::for_table('demo')->select('field_test')->find_one($id);
$demo->field_test = 'do';
$demo->save();
You will get the error.
change to :
$demo = ORM::for_table('demo')->select('field_test')->select('id')->find_one($id);
It will fix the problem.
Some tips in documents:
https://github.com/j4mie/idiorm/blob/master/test/ORMTest.php
/**
* These next two tests are needed because if you have select()ed some fields,
* but not the primary key, then the primary key is not available for the
* update/delete query - see issue #203.
* We need to change the primary key here to something other than id
* becuase MockPDOStatement->fetch() always returns an id.
*/
I've never used idiorm, so cannot guarantee that my answer will work for you, but from this page and under "Updating records", we have an example which is similar but slightly different to yours.
// The 5 means the value of 5 in the primary-key column
$person = ORM::for_table('person')->find_one(5);
// The following two forms are equivalent
$person->set('name', 'Bob Smith');
$person->age = 20;
// This is equivalent to the above two assignments
$person->set(array(
'name' => 'Bob Smith',
'age' => 20
));
// Syncronise the object with the database
$person->save();
I'm sure I'll learn the reason behind this, but let me tell you all I understand at the moment, and how I "fixed" it.
Here is the beginning of idiorm's save function:
public function save() {
$query = array();
// remove any expression fields as they are already baked into the query
$values = array_values(array_diff_key($this->_dirty_fields, $this->_expr_fields));
if (!$this->_is_new) { // UPDATE
// If there are no dirty values, do nothing
if (empty($values) && empty($this->_expr_fields)) {
return true;
}
$query = $this->_build_update();
$id = $this->id(true);
Right there, on that last line, when trying to access the $this->id, you are getting an exception thrown:
throw new Exception('Primary key ID missing from row or is null');
$this does not contain an id property. I'm not really sure how it could. The example given both on their homepage and in the docs doesn't do anything special to address this. In fact I am copying them 1:1 and still yielding the same error as you.
So, all that said, I fixed this error by just adding in my own id:
$crop = ORM::for_table('SCS_Crop')->find_one($id);
$crop->id = $id;
$crop->Name = "Foo";
$crop->save();
This also happens when the id field name is ambiguous, e.g. when joining two tables both having an id column. This is the case with referenced tables
Model::factory('tableOne')
->left_outer_join('tableTwo', array('tableOne.tableTwo_id', '=', 'tableTwo.id'))
->find_one($id);
In these cases set an alias to the ID column of the parent tableOne to later access it while saving. Make sure that you also select other columns you need - e.g. by ->select('*'):
Model::factory('tableOne')
->select('*')
->select('tableOne.id', 'id')
->left_outer_join('tableTwo', array('tableOne.tableTwo_id', '=', 'tableTwo.id'))
->find_one($id);
if in table primary key/ field name not id then following id column overrides required
default id (primary_key) to replace with other id name (primary_key)
ORM::configure('id_column_overrides', array(
'user' => 'user_id',
));
$update = ORM::for_table('user')->find_one(1);
$update->name = "dev";
try{
$update->save();
}catch(Exception $e){
echo $e;
}
print_r($update);

Prepping '0' value to NULL

In CodeIgniter, I want to prep a value returned from a form so that if it is a 0, it will actually be inserted as NULL.
I created a function outside of my controller class:
function prep_zero_to_null($int)
{
if ($int == 0)
{
return NULL;
}
else
{
return $int;
}
}
And at the form validation, I do:
$this->form_validation->set_rules('category_id', 'Category',
'required|integer|prep_zero_to_null');
However, CI still tries to insert zeroes as '0' in the database, which breaks one of my foreign key constraints.
Interestingly enough, if I replace NULL by, say, 25 in the prep_zero_to_null function, CI will indeed recognize it and insert 25 instead of '0'. So my prepping function is indeed getting called, but CI won't allow NULL as a result of it and instead converts it to '0'.
How can I achieve what I want?
Edit: For those wondering, the category_id field does allow null:
`category_id` int(10) unsigned DEFAULT NULL
And the exact error is:
INSERT INTO `articles` (`category_id`, `order`, `title`, `text`)
VALUES ('0', '0', 'test', 'test')
^
Should be NULL
Cannot add or update a child row: a foreign key constraint fails
(`db`.`articles`, CONSTRAINT `articles_ibfk_1` FOREIGN KEY
(`category_id`) REFERENCES `categories` (`id`)
ON DELETE SET NULL ON UPDATE CASCADE)
Just looking at this quickly I think the problem is your $int == 0. Is $int an actual 0 type integer or is it a string? In which case the proper check would be $int == '0'.
Codeigniter validation functions doesn't set the field value based on what you return, your validation function should either return TRUE or FALSE to state that something is valid or not.
If you're after changing the value of something, you'll need to accept variable by reference and modify it in the function then you can return TRUE so it passes the validation.
The best solution would be to make the check before inserting the data into database & not relying on the validation library to do this kind of dirty work.
If you want to insert null to the database you need to return a string with the value "null".
function prep_zero_to_null($int) {
return ($int == 0) ? 'NULL' : $int;
}
Have you tried just unsetting the variable. It's a bit ugly, but it should return NULL for the value then.
Tested here.
I submitted an issue at the GitHub bug tracker for CodeIgniter, as this appears to be a bug.
https://github.com/EllisLab/CodeIgniter/issues/2563
Right now, the workaround at model level is the following:
$category_id = prep_zero_to_null($this->input->post('category_id'));
$data = array
(
'category_id' => $category_id,
'order' => $this->input->post('order'),
'title' => $this->input->post('title'),
'text' => $this->input->post('text')
);
Edit: Apparently, this is the right approach, as there should only be strings at validation/controller level.

PHP RedBean store bean if not exist one

I am a little confused. I actively use PHP RedBean as ORM within my direct mail service and I run into curious situation - I have a table with unique key constraint (i.e. subscriber_id, delivery_id) and two scripts that is writing data into this table.
There is source code that is inserting or updating table:
public static function addOpenPrecedent($nSubscriberId, $nDeliveryId)
{
$oOpenStatBean = \R::findOrDispense('open_stat', 'delivery_id = :did AND subscriber_id = :sid', array(':did' => $nDeliveryId, ':sid' => $nSubscriberId));
$oOpenStatBean = array_values($oOpenStatBean);
if (1 !== count($oOpenStatBean)) {
throw new ModelOpenStatException(
"Ошибка при обновлении статистики открытий: пара (delivery_id,
subscriber_id) не является уникальной: ($nDeliveryId, $nSubscriberId).");
}
$oOpenStatBean = $oOpenStatBean[0];
if (!empty($oOpenStatBean->last_add_dt)) {
$oOpenStatBean->precedent++;
} else {
$oOpenStatBean->delivery_id = $nDeliveryId;
$oOpenStatBean->subscriber_id = $nSubscriberId;
}
$oOpenStatBean->last_add_dt = time('Y-m-d H:i:s');
\R::store($oOpenStatBean);
}
It is called both from two scripts. And I have issues with corruption unique constraint on this table periodically, because race conditions occurs. I know about SQL "INSERT on duplicate key update" feature. But how can I obtain same result purely using my ORM?
Current, that I know if, Redbean will not issue an
INSERT ON DUPLICATE KEY UPDATE
as the discussion of this cited in the comments above indicates that Redbean's developer considers upsert to be a business logic thing that would pollute the ORM's interphase. This being said, it is most likely achievable if one were to extend Redbean with a custom Query Writer or plugin per the Documentation. I haven't tried this because the method below easily achieves this behavior without messing with the internals and plugins of the ORM, however, it does require that you use transactions and models and a couple of extra queries.
Basically, start your transaction with either R::transaction() or R::begin() before your call to R::store(). Then in your "FUSE"d model, use the "update" FUSE method to run a query that checks for duplication and retrieves the existing id while locking the necessary rows (i.e. SELECT FOR UPDATE). If no id is returned, you are good and just let your regular model validation (or lack thereof) continue as usual and return. If an id is found, simply set $this->bean->id to the returned value and Redbean will UPDATE rather than INSERT. So, with a model like this:
class Model_OpenStat extends RedBean_SimpleModel{
function update(){
$sql = 'SELECT * FROM `open_stat` WHERE `delivery_id`=? AND 'subscriber_id'=? LIMIT 1 FOR UPDATE';
$args = array( $this->bean->deliver_id, $this->bean->subscriber_id );
$dupRow = R::getRow( $sql, $args );
if( is_array( $dupRow ) && isset( $dupRow['id'] ) ){
foreach( $this->bean->getProperties() as $property => $value ){
#set your criteria here for which fields
#should be from the one in the database and which should come from this copy
#this version simply takes all unset values in the current and sets them
#from the one in the database
if( !isset( $value ) && isset( $dupRow[$property] ) )
$this->bean->$property = $dupRow[$property];
}
$this->bean->id = $dupId['id']; #set id to the duplicates id
}
return true;
}
}
You would then modify the R::store() call like so:
\R::begin();
\R::store($oOpenStatBean);
\R::commit();
or
\R::transaction( function() use ( $oOpenStatBean ){ R::store( $oOpenStatBean ); } );
The transaction will cause the "FOR UPDATE" clause to lock the found row or, in the event that no row was found, to lock the places in the index where your new row will go so that you don't have concurrency issues.
Now this will not solve one user's update of the record clobbering another, but that is a whole different topic.

Which function architecture is the best for maintainability in this simple case?

I'm adding "promo code" functionality to an online shopping cart written by someone else. This is a pretty micky mouse architecture question, but I would like to hear a couple of opinions. All three of my ideas would work, but which do you think is best for maintainability?
So, here's the basic table structure for the promo codes:
CREATE TABLE `promocodes` (
`promocode_id` INT NOT NULL AUTO_INCREMENT PRIMARY KEY ,
`promocode` VARCHAR( 255 ) NOT NULL ,
`description` VARCHAR( 255 ) NULL ,
`discount_percentage` INT NULL ,
`discount_dollars` DECIMAL( 6, 2 ) NULL ,
`expiration_date` DATE NULL ,
`uses_remaining` INT NULL ,
`is_active` BOOL NOT NULL DEFAULT '1'
)
What's the smartest (without getting overcomplicated):
Check existence with SQL, everything else seperately in code
// LIBRARY
function promoCodeExists($promoCodeString){
// make sql call
return $promoCodeDetailsHash; // or false if no record
}
function isPromoCodeUsable($promoCodeDetailsHash){
// check expiry date and number of uses left and active / inactive
}
function usePromoCode($promoCodeId){
// do order association
// decrement uses left
}
// USAGE
$promoCodeDetailsHash = promoCodeExists($promoCode);
if (is_array($promoCodeDetailsHash) AND isPromoCodeUsable($promoCodeDetailsHash)){
usePromoCode($promoCodeDetailsHash['id'])
} else {
// invalid promo code
}
Or, have a validation function but have it called only by the get function:
// LIBRARY
function validatePromoCode($promoCodeDetailsHash){
// check expiry date and number of uses left and active / inactive
}
function isPromoCodeUsable($promoCodeString){
// make sql call
return validatePromoCode($promoCodeDetailsHash); // or false if no record
}
// USAGE
$promoCodeDetailsHash = promoCodeExists($promoCode);
if (is_array(isPromoCodeUsable($promoCodeDetailsHash))){
usePromoCode($promoCodeDetailsHash['id'])
} else {
// invalid promo code
}
Check everything in SQL with invalid the same as nonexistance:
// LIBRARY
function getDetailsForUsablePromoCode($promoCode){
// use SQL WHERE clauses to only return existence for currently valid promo codes
// or false if no result
}
// USAGE
$promoCodeDetailsHash = getDetailsForUsablePromoCode($promoCode)
if (is_array($promoCodeDetailsHash)){
usePromoCode($promoCodeDetailsHash['id'])
} else {
// error state
}
Feel free to point out any other approaches or wackness here.
In my application, I create it as 2 table. First table just like you, but only hold the usage_limit as integer. In second table, I will hold the usage, one usage per row.
In the promocode_usage table, I will have promocode_id as foreign key, and other required column such as usage datetime, user id, etc. To check if the promo is still available, I will simply count the row in promocode_usage table that have promocode_id I want to check. If the result less than usage_limit value, the promo can be used.

Categories