Advanced Queries

Introduction

MongoDB offers a rich query environment with lots of features. This page lists some of those features.

Queries in MongoDB are represented as JSON-style objects, very much like the documents we actually store in the database.  For example:

// i.e., select * from things where x=3 and y="foo"
db.things.find( { x : 3, y : "foo" } );

Note that any of the operators on this page can be combined in the same query document. For example, to find all document where j is not equal to 3 and k is greater than 10, you'd query like so:

db.things.find({j: {$ne: 3}, k: {$gt: 10} });

Retrieving a Subset of Fields

By default on a find operation, the entire document/object is returned. However we may also request that only certain fields are returned. Note that the _id field is always returned automatically.

// select z from things where x=3
db.things.find( { x : 3 }, { z : 1 } );

You can also remove specific fields that you know will be large:

// get all posts about mongodb without comments
db.posts.find( { tags : 'mongodb' }, { comments : 0 } );

$slice operator

New in MongoDB 1.5.1

You can use the $slice operator to retrieve a subset of elements in an array.

db.posts.find({}, {comments:{$slice: 5}}) // first 5 comments
db.posts.find({}, {comments:{$slice: -5}}) // last 5 comments
db.posts.find({}, {comments:{$slice: [20, 10]}}) // skip 20, limit 10
db.posts.find({}, {comments:{$slice: [-20, 10]}}) // 20 from end, limit 10

More examples at example slice1

Conditional Operators : <, <=, >, >=

Use these special forms for greater than and less than comparisons in queries, since they have to be represented in the query document:

db.collection.find({ "field" : { $gt: value } } );   // greater than  : field > value
db.collection.find({ "field" : { $lt: value } } );   // less than  :  field < value
db.collection.find({ "field" : { $gte: value } } );  // greater than or equal to : field >= value
db.collection.find({ "field" : { $lte: value } } );  // less than or equal to : field <= value

For example:

db.things.find({j : {$lt: 3}});
db.things.find({j : {$gte: 4}});

You can also combine these operators to specify ranges:

db.collection.find({ "field" : { $gt: value1, $lt: value2 } } );    // value1 < field < value

Conditional Operator : $ne

Use $ne for "not equals".

db.things.find( { x : { $ne : 3 } } );

Conditional Operator : $in

The $in operator is analogous to the SQL IN modifier, allowing you to specify an array of possible matches.

db.collection.find( { "field" : { $in : array } } );

Let's consider a couple of examples. From our things collection, we could choose to get a subset of documents based upon the value of the 'j' key:

db.things.find({j:{$in: [2,4,6]}});

Suppose the collection updates is a list of social network style news items; we want to see the 10 most recent updates from our friends. We might invoke:

db.updates.ensureIndex( { ts : 1 } ); // ts == timestamp
var myFriends = myUserObject.friends; // let's assume this gives us an array of DBRef's of my friends
var latestUpdatesForMe = db.updates.find( { user : { $in : myFriends } } ).sort( { ts : -1 } ).limit(10);

Conditional Operator : $nin

The $nin operator is similar to $in except that it selects objects for which the specified field does not have any value in the specified array. For example

db.things.find({j:{$nin: [2,4,6]}});

would match {j:1,b:2} but not {j:2,c:9}.

Conditional Operator : $mod

The $mod operator allows you to do fast modulo queries to replace a common case for where clauses. For example, the following $where query:

db.things.find( "this.a % 10 == 1")

can be replaced by:

db.things.find( { a : { $mod : [ 10 , 1 ] } } )

Conditional Operator: $all

The $all operator is similar to $in, but instead of matching any value in the specified array all values in the array must be matched. For example, the object

{ a: [ 1, 2, 3 ] }

would be matched by

db.things.find( { a: { $all: [ 2, 3 ] } } );

but not

db.things.find( { a: { $all: [ 2, 3, 4 ] } } );

An array can have more elements than those specified by the $all criteria. $all specifies a minimum set of elements that must be matched.

Conditional Operator : $size

The $size operator matches any array with the specified number of elements. The following example would match the object {a:["foo"]}, since that array has just one element:

db.things.find( { a : { $size: 1 } } );

You cannot use $size to find a range of sizes (for example: arrays with more than 1 element). If you need to query for a range, create an extra size field that you increment when you add elements.

Conditional Operator: $exists

Check for existence (or lack thereof) of a field.

db.things.find( { a : { $exists : true } } ); // return object if a is present
db.things.find( { a : { $exists : false } } ); // return if a is missing

Currently $exists is not able to use an index. Indexes on other fields are still used.

Conditional Operator: $type

The $type operator matches values based on their BSON type.

db.things.find( { a : { $type : 2 } } ); // matches if a is a string
db.things.find( { a : { $type : 16 } } ); // matches if a is an int

Possible types are:

For more information on types and BSON in general, see http://www.bsonspec.org.

$or

The $or operator lets you use a boolean or expression to do queries. You give $or a list of expressions, any of which can satisfy the query.

New in MongoDB 1.5.3

Simple:

db.foo.find( { $or : [ { a : 1 } , { b : 2 } ] } )

With another field

db.foo.find( { name : "bob" , $or : [ { a : 1 } , { b : 2 } ] } )

The $or operator retrieves matches for each or clause individually and eliminates duplicates when returning results. A number of $or optimizations are planned for 1.8. See this thread for details.

Regular Expressions

You may use regexes in database query expressions:

db.customers.find( { name : /acme.*corp/i } );

For simple prefix queries (also called rooted regexps) like /^prefix/, the database will use an index when available and appropriate (much like most SQL databases that use indexes for a LIKE 'prefix%' expression). This only works if you don't have i (case-insensitivity) in the flags.

While /^a/, /^a.*/, and /^a.*$/ are equivalent and will all use an index in the same way, the later two require scanning the whole string so they will be slower. The first format can stop scanning after the prefix is matched.

MongoDB uses PCRE for regular expressions. Valid flags are:

• i - Case insensitive. Letters in the pattern match both upper and lower
  case letters.
• m - Multiline. By default, Mongo treats the subject string as consisting of a single line of characters (even if it actually contains newlines). The "start of line" metacharacter (^) matches only at the start of the string, while the "end of line" metacharacter ($) matches only at the end of the string, or before a terminating newline.
  When m it is set, the "start of line" and "end of line" constructs match immediately following or immediately before internal newlines in the subject string, respectively, as well as at the very start and end. If there are no newlines in a subject string, or no occurrences of ^ or $ in a pattern, setting m has no effect.
• x - Extended. If set, whitespace data characters in the pattern are totally ignored except when escaped or inside a character class. Whitespace does not include the VT character (code 11). In addition, characters between an unescaped # outside a character class and the next newline, inclusive, are also ignored.
  This option makes it possible to include comments inside complicated patterns. Note, however, that this applies only to data characters. Whitespace characters may never appear within special character sequences in a pattern, for example within the sequence (?( which introduces a conditional subpattern.

Value in an Array

To look for the value "red" in an array field colors:

db.things.find( { colors : "red" } );

That is, when "colors" is inspected, if it is an array, each value in the array is checked.  This technique may be mixed with the embedded object technique below.

Conditional Operator: $elemMatch

Version 1.3.1 and higher.

Use $elemMatch to check if an element in an array matches the specified match expression.

> t.find( { x : { $elemMatch : { a : 1, b : { $gt : 1 } } } } ) 
{ "_id" : ObjectId("4b5783300334000000000aa9"),
  "x" : [ { "a" : 1, "b" : 3 }, 7, { "b" : 99 }, { "a" : 11 } ]
}

Note that a single array element must match all the criteria specified; thus, the following query is semantically different in that each criteria can match a different element in the x array:

> t.find( { "x.a" : 1, "x.b" : { $gt : 1 } } )

See the dot notation page for more.

Value in an Embedded Object

For example, to look author.name=="joe" in a postings collection with embedded author objects:

db.postings.find( { "author.name" : "joe" } );

See the dot notation page for more.

Meta operator: $not

Version 1.3.3 and higher.

The $not meta operator can be used to negate the check performed by a standard operator. For example:

db.customers.find( { name : { $not : /acme.*corp/i } } );

db.things.find( { a : { $not : { $mod : [ 10 , 1 ] } } } );

Javascript Expressions and $where

In addition to the structured query syntax shown so far, you may specify query expressions as Javascript. To do so, pass a string containing a Javascript expression to find(), or assign such a string to the query object member $where. The database will evaluate this expression for each object scanned. When the result is true, the object is returned in the query results.

For example, the following statements all do the same thing:

db.myCollection.find( { a : { $gt: 3 } } );
db.myCollection.find( { $where: "this.a > 3" } );
db.myCollection.find("this.a > 3");
f = function() { return this.a > 3; } db.myCollection.find(f);

Javascript executes more slowly than the native operators listed on this page, but is very flexible. See the server-side processing page for more information.

sort()

sort() is analogous to the ORDER BY statement in SQL - it requests that items be returned in a particular order. We pass sort() a key pattern which indicates the desired order for the result.

db.myCollection.find().sort( { ts : -1 } ); // sort by ts, descending order

sort() may be combined with the limit() function. In fact, if you do not have a relevant index for the specified key pattern, limit() is recommended as there is a limit on the size of sorted results when an index is not used. Without a limit(), or index, a full in-memory sort must be done but by using a limit() it reduces the memory and increases the speed of the operation by using an optimized sorting algorithm.

limit()

limit() is analogous to the LIMIT statement in MySQL: it specifies a maximum number of results to return. For best performance, use limit() whenever possible. Otherwise, the database may return more objects than are required for processing.

db.students.find().limit(10).forEach( function(student) { print(student.name + "<p>"); } );

In the shell (and most drivers), a limit of 0 is equivalent to setting no limit at all.

skip()

The skip() expression allows one to specify at which object the database should begin returning results. This is often useful for implementing "paging". Here's an example of how it might be used in a JavaScript application:

function printStudents(pageNumber, nPerPage) {
   print("Page: " + pageNumber);
   db.students.find().skip((pageNumber-1)*nPerPage).limit(nPerPage).forEach( function(student) { print(student.name + "<p>"); } );
}

snapshot()

Indicates use of snapshot mode for the query. Snapshot mode assures no duplicates are returned, or objects missed, which were present at both the start and end of the query's execution (even if the object were updated). If an object is new during the query, or deleted during the query, it may or may not be returned, even with snapshot mode.

Note that short query responses (less than 1MB) are always effectively snapshotted.

Currently, snapshot mode may not be used with sorting or explicit hints.

count()

The count() method returns the number of objects matching the query specified. It is specially optimized to perform the count in the MongoDB server, rather than on the client side for speed and efficiency:

nstudents = db.students.find({'address.state' : 'CA'}).count();

Note that you can achieve the same result with the following, but the following is slow and inefficient as it requires all documents to be put into memory on the client, and then counted. Don't do this:

nstudents = db.students.find({'address.state' : 'CA'}).toArray().length; // VERY BAD: slow and uses excess memory

On a query using skip() and limit(), count ignores these parameters by default.  Use count(true) to have it consider the skip and limit values in the calculation.

n = db.students.find().skip(20).limit(10).count(true);

group()

The group() method is analogous to GROUP BY in SQL. group() is more flexible, actually, allowing the specification of arbitrary reduction operations. See the Aggregation section of the Mongo Developers' Guide for more information.

Special operators

only return the index key

db.foo.find()._addSpecial("$returnKey" , true )

limit the number of items to scan

t.find()._addSpecial( "$maxScan" , 50 )

See Also

* Optimizing Queries (including explain() and hint())