MongoDB Plugin

We provide precompiled dependency libraries (libmongoc, libbson, libicudata, libicuuc) at the /mongodb/bin directory. Optionally, you can manually compile them by following the steps described in this section.

Syntax

mongodb::connect(host, port, user, password, [db])

Arguments

• host: a string indicating the MongoDB server address
• port: a integer indicating the port number of the MongoDB server
• user: a string indicating the username to the MongoDB server. If the MongoDB authentication service is disabled, enter an empty string "".
• password: a string indicating the password to the MongoDB server. If the MongoDB authentication service is disabled, enter an empty string "".
• db: a string indicating the MongoDB database for user authentication. The database must have the credential information on the specified user. If this parameter is not specified, the "admin" database from the MongoDB server (specified by host) is used.

Details

Set up connection with the MongoDB server. Return a connection handle that can be used when calling mongodb::load and other methods.

Examples

conn = mongodb::connect(`localhost, 27017, `root, `root, `DolphinDB)
conn2 = mongodb::connect(`localhost, 27017, `root, `root)

Syntax

mongodb::load(connection, collcetionName, query, option, [schema])

Arguments

• connection: the MongoDB connection handle returned by mongodb::connect
• collectionName: the name of a MongoDB collection. There are two options to specify this parameter:
  • specify only the `collection name - the system searches for the collection in the database (db) specified in mongodb::connect
  • specify "databaseName:collectionName" - the system searches for the collection in the specified database
• query: a JSON STRING indicating the query conditions in MongoDB. For example: { "aa" : { "$numberInt" : "13232" } }, { "datetime" : { "$gt" : {"$date":"2019-02-28T00:00:00.000Z" }} }
• option: a JSON STRING indicating MongoDB query options. For example: {"limit":123} limits the number of records or documents returned in the query.
• schema: a table containing column names and columns types. You can optionally change the default column types through the schema table and pass it to mongodb::load as an argument.

Details

Import the query result in MongoDB into a DolphinDB in-memory table. For information on data type support, see the Data Type section in the DolphinDB User Manual.

Examples

conn = mongodb::connect(`localhost, 27017, `root, `root, `DolphinDB)
query='{ "datetime" : { "$gt" : {"$date":"2019-02-28T00:00:00.000Z" }} }'
option='{"limit":1234}'
tb=mongodb::load(conn, `US,query,option)
select count(*) from tb
tb2 = mongodb::load(conn, 'dolphindb:US',query,option)
select count(*) from tb
schema=table(`item`type`qty as name,`STRING`STRING`INT as type)
tb2 = mongodb::load(conn, 'dolphindb:US',query,option,schema)

Syntax

mongodb::aggregate(connection, collcetionName, pipeline, option, [schema])

Arguments

• connection: the MongoDB connection handle returned by mongodb::connect
• collectionName: the name of a MongoDB collection. There are two options to specify this parameter:
  • specify only the `collection name - the system searches for the collection in the database (db) specified in mongodb::connect
  • specify "databaseName:collectionName" - the system searches for the collection in the specified database
• pipeline: a STRING indicating the MongoDB aggregation pipeline. For example, {$group : {_id : "$by_user", num_tutorial : {$sum : 1}}}.
• option: a STRING indicating MongoDB query options. It is JSON document in BSON format. For example: {"limit":123} limits the number of records or documents returned in the query.
• schema: a table containing column names and columns types. You can optionally change the default column types through the schema table and pass it to mongodb::load as an argument.

Details

Import the result of an aggregate operation in MongoDB into a DolphinDB in-memory table. For information on data type support, see the Data Type section in the DolphinDB User Manual.

Examples

conn = mongodb::connect(`localhost, 27017, "", "", `DolphinDB)
pipeline = "{ \"pipeline\" : [ { \"$project\" : { \"str\" : \"$obj1.str\" } } ] }"
option="{}"
mongodb::aggregate(conn, "test1:collnetion1",pipeline,option)

Syntax

mongodb::close(connection)

Arguments

• the MongoDB connection handle returned by mongodb::connect

Details

Close a MongoDB server connection.

Examples

conn = mongodb::connect(`localhost, 27017, `root, `root, `DolphinDB)
query=`{ "datetime" : { "$gt" : {"$date":"2019-02-28T00:00:00.000Z" }} }
option=`{"limit":1234}
tb = mongodb::load(conn, `US,query,option)
select count(*) from tb
mongodb::close(conn)

Syntax

mongodb::parseJson(str, keys, colnames, colTypes)

Details

Parse a JSON string and convert it into a DolphinDB in-memory table. Return the in-memory table.

Arguments

• str: a STRING vector indicating the JSON string to be converted
• keys: a STRING vector indicating the keys of str
• colnames: a STRING vector indicating the column names in the result table. The column names correspond to the keys of str.
• colTypes: a vector indicating the column types in the result table. Supported DolphinDB data types: BOOL, INT, FLOAT, DOUBLE, STRING and array vector of BOOL[], INT[] and DOUBLE[] type. You can convert the int, float and double types in JSON into any of the INT, FLOAT or DOUBLE type in DolphinDB.

Example

data = ['{"a": 1, "b": 2}', '{"a": 2, "b": 3}']
 mongodb::parseJson(data, 
`a`b, 
`col1`col2,
[INT, INT] )

Syntax

mongodb::getCollections([databaseName])

Arguments

• databaseName: a STRING indicating a MongoDB database. If this parameter is not specified, the database specified in mongodb::connect is used.

Details

Get the names of all the collections in the specified database.

Examples

conn = mongodb::connect("192.168.1.38", 27017, "", "")
mongodb::getCollections(conn, "dolphindb")

In DolphinDB, the smallest value of each integral type (e.g. -2,147,483,648 for INT and -9,223,372,036,854,775,808 for LONG) is a NULL value.