I want to insert json array elements like below using MongoDB and CodeIgniter.
"ticket" : [
{
"problem" : "testing",
"time_taken" : "5 hrs",
"number_of_visits" : "7",
"expenses" : {
"amount" : "2000",
"distance_travelled" : "4 km"
}
},
{
"problem" : "testing",
"time_taken" : "5 hrs",
"number_of_visits" : "4",
"expenses" : {
"amount" : "2500",
"distance_travelled" : "5 km"
}
}
]
What is the query I have to use for this?
What you need is the Bulk Write functions (https://docs.mongodb.com/manual/core/bulk-write-operations/)
The PHP depends on which library you are using (old or new, hopefully the new one!) but in plain English you:
create an array of operations (in your case "inserts")
call bulk write.
$this->mongo_db->where('_id', new MongoId($id))->push("ticket",$data)->update('engineer');
This is the query I'm using for inserting multiple array values in mongodb codeigniter.
Related
In our api when we update a group we send the request in following way
{
"title" : "test",
"description" : "test description",
"date_time" : 45525465,
"interest" : "1",
"age_group" : [3],
"capacity" : "10",
"ethnicity" : [],
"privacy_type" : "1",
}
There are several other key-value pairs. By default only those fields are updated which are provided in the request. For example if i want to update only privacy_type I only need to send
{
"privacy_type" : "2"
}
The problem is that when I try to update privacy_type like this, title is also being updated with the value of the route. If I do dd($request->input('title') , I get "api/v1/groups/10024" which is the route for updating a group. Changing title to group_title is an option, but I would like to know if there is any other option. Thanks
You just need a proper ternary:
$group->group_title = $request->has('title') ? $request->get('title') : $group->group_title;
I'm having problems updating a specific field in all the arrays of a subdocument. I have the following structure in MongoDB:
{
"_id" : ObjectId("539c9e97cac5852a1b880397"),
"DocumentoDesgloseER" : [
{
"elemento" : "COSTO VENTA",
"id_rubroer" : "11",
"id_documento" : "45087",
"abreviatura" : "CV",
"orden" : "1",
"formula" : "Cuenta Contable",
"tipo_fila" : "1",
"color" : "#FFD2E9",
"sucursal" : "D",
"documentoID" : "0",
"TOTAL" : "55426.62",
},
{ ... MORE OF THE SAME ... }
],
"id_division" : "2",
"id_empresa" : "9",
"id_sucursal" : "37",
"ejercicio" : "2008",
"lastMonthNumber" : NumberLong(6),
}
I need to update the field "documentoID" to a specific value; like "20" for example, in all the arrays of the subdocument "DocumentoDesgloseER". How I can do this?
I tried the following (with $ operator) and is not working:
$querySearch = array('id_division'=>'2', 'id_empresa'=>'9', 'id_sucursal'=>'37', 'ejercicio'=>'2008');
$queryUpdate = array('$set'=>array('DocumentoDesgloseER.$.documentoID'=>'20'));
Yii::app()->edmsMongoCollection('DocumentosDesgloseER')->update($querySearch,$queryUpdate);
By the way, I'm using Yii Framework to make the connection with Mongo. Any help or advice is welcome.
Thanks ;D!
Unfortunately, you can't currently use a positional operator to update all items in an array. There is a ticket opened in the MongoDB JIRA about this issue.
There a two "solutions":
Change your schema so that your embedded documents are in the separate collection (it's probably not what you want).
The best you can do, if you don't want to change your schema, is to update each subdocument in PHP and then save the whole document.
I’m working on my master thesis where one of my goals is to run tests and experiments against the CouchDB database and tune the performance.
To do that I need some test data. I’ve created a piece of php code to generate some simple relational data for a MySQL database. The tables are:
Customer
Product
Brand
Color
Checkout
I’ve made some relations between for example Product and Colorid and Brandid and in the Checkout table I’ve a relation to Customerid and Productid.
I want to export this entirely data structure and the data with its relations to a JSON format for CouchDB.
So I’ve a JSON string which should contain each customer with its attributes and its purchase with all the parameters for this product, and so on.
I’m thinking that it would look something like:
{
"customer": {
"customerid" : "1",
"firstname" : "somefirstname",
"lastname" : "somelastname",
"email" : "my#mail.com",
"country" : "USA",
"datecreated" : "11111111111111"
}
"purchase" : {
"purchaseid" : "1",
"product": {
"productname" : "mightymouse",
"productcolor": "blue",
"productbrand" : "Apple",
"productprice" : "200",
"checkoutdate" : "1111111111112"
}
"purchaseid" : "2",
"product": {
"productname" : "something nice",
"productcolor": "yellow",
"productbrand" : "Google",
"productprice" : "5000",
"checkoutdate" : "11111111113333"
}
}
}
It’s probably not the right data structure I’ve shown but it something like that.
Can this be done in PHP and if so, how do I create this kind of “CouchDB” ready JSON strings??
If I haven’t explained myself clearly, please let me know.
Thank you
Sincere
- Mestika
PHP has function json_encode, you can use it to convert any PHP object to its JSON representation (and json_decode to convert JSON string to object).
So:
Use PHP database functions to read data from database.
Create object of stdClass:
$data = new stdClass;
Fill this object with properties read from database like this:
$customer = new stdClass;
$customer->customerid = "1"
...
$data->customer = $customer;
Encode generated object with json_encode.
document structure example is:
{
"dob": "12-13-2001",
"name": "Kam",
"visits": {
"0": {
"service_date": "12-5-2011",
"payment": "40",
"chk_number": "1234455",
},
"1": {
"service_date": "12-15-2011",
"payment": "45",
"chk_number": "3461234",
},
"2": {
"service_date": "12-25-2011",
"payment": "25",
"chk_number": "9821234",
}
}
}
{
"dob": "10-01-1998",
"name": "Sam",
"visits": {
"0": {
"service_date": "12-5-2011",
"payment": "30",
"chk_number": "86786464",
},
"1": {
"service_date": "12-15-2011",
"payment": "35",
"chk_number": "45643461234",
},
"2": {
"service_date": "12-25-2011",
"payment": "20",
"chk_number": "4569821234",
}
}
}
In PHP i want to list all those "visits" information (and corresponding "name" ) for which payment is less than "30".
I want to print only the visits with "payment" < "30" not others. Is such query possible, or do i have to get entire document first using search and then use PHP to select such visits??
In the example document, the "payment" values are given as strings which may not work as intended with the $lt command. For this response, I have converted them to integers.
Wildcard queries are not possible with MongoDB, so with the given document structure, the key (0,1,2, etcetera) of the sub-document must be known. For instance, the following query will work:
> db.test.find({"visits.2.payment":{$lt:35}})
However,
> db.test.find({"visits.payment":{$lt:35}})
Will not work in this case, and
> db.test.find({"visits.*.payment":{$lt:35}})
will also not return any results.
In order to be able to query the embedded "visits" documents, you must change your document structure and make "visits" into an array or embedded documents, like so:
> db.test2.find().pretty()
{
"_id" : ObjectId("4f16199d3563af4cb141c547"),
"dob" : "10-01-1998",
"name" : "Sam",
"visits" : [
{
"service_date" : "12-5-2011",
"payment" : 30,
"chk_number" : "86786464"
},
{
"service_date" : "12-15-2011",
"payment" : 35,
"chk_number" : "45643461234"
},
{
"service_date" : "12-25-2011",
"payment" : 20,
"chk_number" : "4569821234"
}
]
}
Now you can query all of the embedded documents in "visits":
> db.test2.find({"visits.payment":{$lt:35}})
For more information, please refer to the Mongo documentation on dot notation:
http://www.mongodb.org/display/DOCS/Dot+Notation+%28Reaching+into+Objects%29
Now on to the second part of your question: it is not possible to return only a conditional sub-set of embedded documents.
With either document format, it is not possible to return a document containing ONLY the sub-documents that match the query. If one of the sub-documents matches the query , then the entire document matches the query, and it will be returned.
As per the Mongo Document "Retrieving a subset of fields"
http://www.mongodb.org/display/DOCS/Retrieving+a+Subset+of+Fields
We can return parts of embedded documents like so:
> db.test2.find({"visits.payment":{$lt:35}},{"visits.service_date":1}).pretty()
{
"_id" : ObjectId("4f16199d3563af4cb141c547"),
"visits" : [
{
"service_date" : "12-5-2011"
},
{
"service_date" : "12-15-2011"
},
{
"service_date" : "12-25-2011"
}
]
}
But we cannot have conditional retrieval of some sub documents. The closest that we can get is the $slice operator, but this is not conditional, and you will have to first know the location of each sub-document in the array:
http://www.mongodb.org/display/DOCS/Retrieving+a+Subset+of+Fields#RetrievingaSubsetofFields-RetrievingaSubrangeofArrayElements
In order for the application to display only the embedded documents that match the query, it will have to be done programmatically.
You may try:
$results = $mongodb->find(array("visits.payment" => array('$lt' => 30)));
But i don't know if it will work since visits is an object. BTW judging from what you posted it could be transfered to array (or should since numerical property names tends to cause confusion)
try - db.test2.find({"visits.payment":"35"})
ok lets say i have this collection in my tv database
"season" : "1", "episodes" : {"code" : ["1x01", "1x02", "1x03"], "title" : ["Dont Look Back", "Genesis", "Third"]},
"season" : "2", "episodes" : {"code" : ["2x01", "2x02", "2x03"], "title" : ["D2ont Look Back", "G2enesis", "T2hird"]},
"season" : "3", "episodes" : {"code" : ["3x01", "3x02", "3x03"], "title" : ["D3ont Look Back", "G3enesis", "T3hird"]},
"season" : "4", "episodes" : {"code" : ["4x01", "4x02", "4x03"], "title" : ["D4ont Look Back", "G4enesis", "T4hird"]}
how do i make it so that only, lets say the episodes from season 2 are shown?
ive been trying using
echo $obj['episodes']['code'][0];
but it only shows episodes from the last row
im pretty sure my nesting is all wrong but im new to mongo and im having trouble trying to map this out
any advice?
You need to use find() first.
$a = $coll->findOne(array('season' => '2');
That will return an array $a which will have keys 'episodes', 'title'
So once you find the document with findOne, you would then just access the data using $a['title'] or whatever fields you need