ChatGPT解决这个技术问题 Extra ChatGPT

How to query MongoDB with "like"

I want to query something with SQL's like query:

SELECT * FROM users  WHERE name LIKE '%m%'

How can I achieve the same in MongoDB? I can't find an operator for like in the documentation.

see mongodb's docs: Advanced Queries -- Regular expressions mongodb.org/display/DOCS/…
I seriously suggest taking a look at MongoDB Atlas Search, as it is much more resource efficient and feature rich for "like"-like queries that $text or $regex

P
Peter Mortensen

That would have to be:

db.users.find({"name": /.*m.*/})

Or, similar:

db.users.find({"name": /m/})

You're looking for something that contains "m" somewhere (SQL's '%' operator is equivalent to regular expressions' '.*'), not something that has "m" anchored to the beginning of the string.

Note: MongoDB uses regular expressions which are more powerful than "LIKE" in SQL. With regular expressions you can create any pattern that you imagine.

For more information on regular expressions, refer to Regular expressions (MDN).


is searching by regex expensive?
Actually, it depends. If the query doesn't use an index, and must do a table scan, then it can certainly be expensive. If you're doing a 'starts with' regex query, then that can use an index. Best to run an explain() to see what's happening.
When not anchored to the beginning of the string, it is somewhat expensive. But then again, so is a LIKE query in SQL.
I would add regex i javascript db.users.find({ "name": { $regex: /m/i } })
It is worth mentioning that if you want to use it from Node app and you want a dynamic search, you can use: users.find({"name": new RegExp('.*' + searchVariable + '.*')}) So, this way you can use it with other operators like $in, $nin, etc.
M
Manuel Jordan
db.users.insert({name: 'paulo'})
db.users.insert({name: 'patric'})
db.users.insert({name: 'pedro'})

Therefore:

For:

db.users.find({name: /a/})  // Like '%a%'

Output: paulo, patric

For:

db.users.find({name: /^pa/}) // Like 'pa%'

Output: paulo, patric

For:

db.users.find({name: /ro$/}) //like '%ro'

Output: pedro


You know why you have so many upvotes? Because the docs deserve equal many downvotes! Thank you
K
Kirill Husiatyn

In

PyMongo using Python

Mongoose using Node.js

Jongo, using Java

mgo, using Go

you can do:

db.users.find({'name': {'$regex': 'sometext'}})

@TahirYasin if you're still wondering, case-insensitive search would be done like this: db.users.find({'name': {'$regex': 'sometext', '$options': 'i'}})
this applies to the whole word, not to a part of the word.
P
Peter Mortensen

In PHP, you could use the following code:

$collection->find(array('name'=> array('$regex' => 'm'));

P
Peter Mortensen

Here are different types of requirements and solutions for string search with regular expressions.

You can do with a regular expression which contains a word, i.e., like. Also you can use $options => i for a case insensitive search.

Contains string

db.collection.find({name:{'$regex' : 'string', '$options' : 'i'}})

Doesn't contain string, only with a regular expression

db.collection.find({name:{'$regex' : '^((?!string).)*$', '$options' : 'i'}})

Exact case insensitive string

db.collection.find({name:{'$regex' : '^string$', '$options' : 'i'}})

Start with string

db.collection.find({name:{'$regex' : '^string', '$options' : 'i'}})

End with string

db.collection.find({name:{'$regex' : 'string$', '$options' : 'i'}})

Keep Regular Expressions Cheat Sheet as a bookmark, and a reference for any other alterations you may need.


this is a later comment but, how can I use a variable in above example? like let name = 'john doe' . how can I implement name variable in regex? thanks
P
Peter Mortensen

You would use a regular expression for that in MongoDB.

For example,

db.users.find({"name": /^m/})

I think this only shows documents with a name value that starts with "m"
Important to remove quotes when you use regex, i.e. don't do "/^m/", just do /^m/
P
Peter Mortensen

You have two choices:

db.users.find({"name": /string/})

or

db.users.find({"name": {"$regex": "string", "$options": "i"}})

For the second one, you have more options, like "i" in options to find using case insensitive.

And about the "string", you can use like ".string." (%string%), or "string.*" (string%) and ".*string) (%string) for example. You can use a regular expression as you want.


P
Peter Mortensen

If using Node.js, it says that you can write this:

db.collection.find( { field: /acme.*corp/i } );

// Or
db.collection.find( { field: { $regex: 'acme.*corp', $options: 'i' } } );

Also, you can write this:

db.collection.find( { field: new RegExp('acme.*corp', 'i') } );

Thanks @Eddy. Providing '$options: 'i'' with regex made refactoring process to easy for me.
P
Peter Mortensen

Already you got the answers, but to match with a regular expression with case insensitivity, you could use the following query:

db.users.find ({ "name" : /m/i } ).pretty()

The i in the /m/i indicates case insensitivity and .pretty() provides a prettier output.


P
Peter Mortensen

For Mongoose in Node.js:

db.users.find({'name': {'$regex': '.*sometext.*'}})

d
damd

With MongoDB Compass, you need to use the strict mode syntax, as such:

{ "text": { "$regex": "^Foo.*", "$options": "i" } }

(In MongoDB Compass, it's important that you use " instead of ')


P
Peter Mortensen

You can use the new feature of MongoDB 2.6:

db.foo.insert({desc: "This is a string with text"});
db.foo.insert({desc:"This is a another string with Text"});
db.foo.ensureIndex({"desc":"text"});
db.foo.find({
    $text:{
        $search:"text"
    }
});

Note, AFAIK Mongodb's text searching works on whole words only by default, so this will match values like "This is a string with text", but not "This is a string with subtext". So it's not quite like sql's "LIKE" operator.
P
Peter Mortensen

In a Node.js project and using Mongoose, use a like query:

var User = mongoose.model('User');

var searchQuery = {};
searchQuery.email = req.query.email;
searchQuery.name = {$regex: req.query.name, $options: 'i'};
User.find(searchQuery, function(error, user) {
                if(error || user === null) {
                    return res.status(500).send(error);
                }
                return res.status(200).send(user);
            });

P
Peter Mortensen

You can use a where statement to build any JavaScript script:

db.myCollection.find( { $where: "this.name.toLowerCase().indexOf('m') >= 0" } );

Reference: $where


$where is highly inefficient. Do full collection scan :(
b
besthost

Using template literals with variables also works:

{"firstname": {$regex : `^${req.body.firstname}.*` , $options: 'si' }}


P
Peter Mortensen

In Go and the mgo driver:

Collection.Find(bson.M{"name": bson.RegEx{"m", ""}}).All(&result)

where the result is the struct instance of the sought-after type.


pls avoid unkeyed fields in literals, do bson:RegEx{Pattern:"m", Options:"i"} instead
P
Peter Mortensen

For PHP mongo Like.

I had several issues with PHP mongo like. I found that concatenating the regular expression parameters helps in some situations - PHP mongo find field starts with.

For example,

db()->users->insert(['name' => 'john']);
db()->users->insert(['name' => 'joe']);
db()->users->insert(['name' => 'jason']);

// starts with
$like_var = 'jo';
$prefix = '/^';
$suffix = '/';
$name = $prefix . $like_var . $suffix;
db()->users->find(['name' => array('$regex'=>new MongoRegex($name))]);
output: (joe, john)

// contains
$like_var = 'j';
$prefix = '/';
$suffix = '/';
$name = $prefix . $like_var . $suffix;
db()->users->find(['name' => array('$regex'=>new MongoRegex($name))]);

output: (joe, john, jason)

P
Peter Mortensen

In SQL, the ‘like’ query looks like this:

select * from users where name like '%m%'

In the MongoDB console, it looks like this:

db.users.find({"name": /m/})     // Not JSON formatted

db.users.find({"name": /m/}).pretty()  // JSON formatted

In addition, the pretty() method will produce a formatted JSON structure in all the places which is more readable.


D
Deepak parmar

String yourdb={deepakparmar, dipak, parmar}

db.getCollection('yourdb').find({"name":/^dee/})

ans deepakparmar

db.getCollection('yourdb').find({"name":/d/})

ans deepakparmar, dipak

db.getCollection('yourdb').find({"name":/mar$/})

ans deepakparmar, parmar


P
Peter Mortensen

Regular expressions are expensive to process.

Another way is to create an index of text and then search it using $search.

Create a text index of fields you want to make searchable:

db.collection.createIndex({name: 'text', otherField: 'text'});

Search for a string in the text index:

db.collection.find({
  '$text'=>{'$search': "The string"}
})

This doesn't work with partials.
S
Sahil Thummar

In MongoDb, can use like using MongoDb reference operator regular expression(regex).

For Same Ex.

MySQL - SELECT * FROM users  WHERE name LIKE '%m%'

MongoDb

    1) db.users.find({ "name": { "$regex": "m", "$options": "i" } })

    2) db.users.find({ "name": { $regex: new RegExp("m", 'i') } })

    3) db.users.find({ "name": { $regex:/m/i } })

    4) db.users.find({ "name": /mail/ })

    5) db.users.find({ "name": /.*m.*/ })

MySQL - SELECT * FROM users  WHERE name LIKE 'm%'

MongoDb Any of Above with /^String/

    6) db.users.find({ "name": /^m/ })

MySQL - SELECT * FROM users  WHERE name LIKE '%m'

MongoDb Any of Above with /String$/

    7) db.users.find({ "name": /m$/ })

It becomes extremely slow on large datasets just FYI. And there's no way to know what the score is.
A
Abhijit Bashetti

Use regular expressions matching as below. The 'i' shows case insensitivity.

var collections = mongoDatabase.GetCollection("Abcd");

var queryA = Query.And(
         Query.Matches("strName", new BsonRegularExpression("ABCD", "i")), 
         Query.Matches("strVal", new BsonRegularExpression("4121", "i")));

var queryB = Query.Or(
       Query.Matches("strName", new BsonRegularExpression("ABCD","i")),
       Query.Matches("strVal", new BsonRegularExpression("33156", "i")));

var getA = collections.Find(queryA);
var getB = collections.Find(queryB);

P
Peter Mortensen

It seems that there are reasons for using both the JavaScript /regex_pattern/ pattern as well as the MongoDB {'$regex': 'regex_pattern'} pattern. See: MongoDB RegEx Syntax Restrictions

This is not a complete regular expression tutorial, but I was inspired to run these tests after seeing a highly voted ambiguous post above.

> ['abbbb','bbabb','bbbba'].forEach(function(v){db.test_collection.insert({val: v})})

> db.test_collection.find({val: /a/})
{ "val" : "abbbb" }
{ "val" : "bbabb" }
{ "val" : "bbbba" }

> db.test_collection.find({val: /.*a.*/})
{ "val" : "abbbb" }
{ "val" : "bbabb" }
{ "val" : "bbbba" }

> db.test_collection.find({val: /.+a.+/})
{ "val" : "bbabb" }

> db.test_collection.find({val: /^a/})
{ "val" : "abbbb" }

> db.test_collection.find({val: /a$/})
{ "val" : "bbbba" }

> db.test_collection.find({val: {'$regex': 'a$'}})
{ "val" : "bbbba" }

P
Peter Mortensen

A like query would be as shown below:

db.movies.find({title: /.*Twelve Monkeys.*/}).sort({regularizedCorRelation : 1}).limit(10);

For the Scala ReactiveMongo API,

val query = BSONDocument("title" -> BSONRegex(".*" + name + ".*", "")) // like
val sortQ = BSONDocument("regularizedCorRelation" -> BSONInteger(1))
val cursor = collection.find(query).sort(sortQ).options(QueryOpts().batchSize(10)).cursor[BSONDocument]

P
Peter Mortensen

If you are using Spring-Data MongoDB, you can do it in this way:

String tagName = "m";
Query query = new Query();
query.limit(10);
query.addCriteria(Criteria.where("tagName").regex(tagName));

P
Peter Mortensen

If you have a string variable, you must convert it to a regex, so MongoDB will use a like statement on it.

const name = req.query.title; //John
db.users.find({ "name": new Regex(name) });

Is the same result as:

db.users.find({"name": /John/})

k
kz_sergey

Use aggregation substring search (with index!!!):

db.collection.aggregate([{
        $project : {
            fieldExists : {
                $indexOfBytes : ['$field', 'string']
            }
        }
    }, {
        $match : {
            fieldExists : {
                $gt : -1
            }
        }
    }, {
        $limit : 5
    }
]);

This might still be better than all these regex suggestions - but it still does a colscan.
E
Ezequias Dinella

You can query with a regular expression:

db.users.find({"name": /m/});

If the string is coming from the user, maybe you want to escape the string before using it. This will prevent literal chars from the user to be interpreted as regex tokens.

For example, searching the string "A." will also match "AB" if not escaped. You can use a simple replace to escape your string before using it. I made it a function for reusing:

function textLike(str) {
  var escaped = str.replace(/[\-\[\]\/\{\}\(\)\*\+\?\.\\\^\$\|]/g, '\\$&');
  return new RegExp(escaped, 'i');
}

So now, the string becomes a case-insensitive pattern matching also the literal dot. Example:

>  textLike('A.');
<  /A\./i

Now we are ready to generate the regular expression on the go:

db.users.find({ "name": textLike("m") });

P
Peter Mortensen

As the MongoDB shell supports regular expressions, that's completely possible.

db.users.findOne({"name" : /.*sometext.*/});

If we want the query to be case-insensitive, we can use the "i" option, like shown below:

db.users.findOne({"name" : /.*sometext.*/i});

P
Peter Mortensen

If you want a 'like' search in MongoDB then you should go with $regex. By using it, the query will be:

db.product.find({name:{$regex:/m/i}})

For more, you can read the documentation as well - $regex