Community Blog Get Started with MongoDB BI Connector in 4 Steps

Get Started with MongoDB BI Connector in 4 Steps

This article is a quick beginner's guide that demonstrates how to implement MongoDB BI connector in four easy steps.

By Zhang Youdong (Linqing)

MongoDB uses BI Connector to enable BI components for directly accessing MongoDB using SQL or ODBC data sources. At first, MongoDB directly used PostgreSQL Foreign Data Wrapper (FDW) to convert SQL statements to MQL. Later, the lighter-weighted mongosqld was implemented to support connections with BI tools.


Install BI Connector

Refer to this link to install BI Connector.

wget https://info-mongodb-com.s3.amazonaws.com/mongodb-bi/v2/mongodb-bi-linux-x86_64-rhel70-v2.12.0.tgz

$tar xvf mongodb-bi-linux-x86_64-rhel70-v2.12.0.tgz
  • The mongosqld tool receives SQL query requests and forwards the requests to MongoDB Server, which is the core of BI Connector.
  • The mongodrdl tool generates database schema information that is used for BI SQL queries.
  • The mongotranslate tool converts SQL queries to the MongoDB Aggregation Pipeline format.

Start mongosqld

Refer to this link to launch BI Connector

mongodb-bi-linux-x86_64-rhel70-v2.12.0/bin/mongosqld --addr --mongo-uri
  • addr: specifies the address of the mongosqld listener.
  • mongo-uri: specifies the address of the connected MongoDB Server.

By default, mongosqld automatically analyzes the schema of the data in the destination MongoDB Server and caches it to the memory. Specify the schema mapping during startup. Use the mongodrdl tool to generate a schema, specify a collection, and export the schema information for the fields in the collection.

$./bin/mongodrdl --uri=mongodb:// -c coll01
- db: test
  - table: coll01
    collection: coll01
    pipeline: []
    - Name: _id
      MongoType: float64
      SqlName: _id
      SqlType: float
    - Name: qty
      MongoType: float64
      SqlName: qty
      SqlType: float
    - Name: type
      MongoType: string
      SqlName: type
      SqlType: varchar

Use a MySQL Client to Connect to mongosqld

The mongosqld tool is directly accessed from a MySQL client. Connect to the tool by using a BI tool such as Excel, Access, or Tableau. Refer to this page for more information.

mysql --protocol=tcp --port=3307

mysql> use test
Database changed
mysql> show tables;
| Tables_in_test |
| coll           |
| coll01         |
| coll02         |
| inventory      |
| myCollection   |
| yourCollection |
6 rows in set (0.00 sec)

mysql> select * from coll01;
| _id  | qty  | type   |
|    1 |    5 | apple  |
|    2 |   10 | orange |
|    3 |   15 | banana |
3 rows in set (0.00 sec)

// 对照 MongoDB 数据库里的原始数据

mongo --port
mymongo:PRIMARY> use test
switched to db test
mymongo:PRIMARY> show tables;
mymongo:PRIMARY> db.coll01.find()
{ "_id" : 1, "type" : "apple", "qty" : 5 }
{ "_id" : 2, "type" : "orange", "qty" : 10 }
{ "_id" : 3, "type" : "banana", "qty" : 15 }

Convert SQL to Aggregation

For example, to convert an SQL query for test.coll01 to the MongoDB Aggregation Pipeline format, analyze the schema by using mongodrdl and then use the mongotranslate tool to convert the schema.

// 导出分析的 shema 文件
$./bin/mongodrdl --uri=mongodb:// -c coll01 > coll01.schema  

// SQL 转换为 Aggregation
$./bin/mongotranslate --query "select * from test.coll01" --schema coll01.schema
    {"$project": {"test_DOT_coll01_DOT__id": "$_id","test_DOT_coll01_DOT_qty": "$qty","test_DOT_coll01_DOT_type": "$type","_id": NumberInt("0")}},
0 0 0
Share on


97 posts | 7 followers

You may also like