Introduction

A separate database plugin to support MongoDB-specific features and configurations.

Use-Case

• Users can choose and install MongoDB source and sink plugins.
• Users should see MongoDB logo on plugin configuration page for better experience.
• Users should get relevant information from the tool tip:
  
  • The tool tip should describe accurately what each field is used for.
• Users should not have to specify any redundant configuration
• Users should get field level lineage for the source and sink that is being used.
• Reference documentation should be updated to account for the changes.
• The source code for MongoDB database plugin should be placed in repo under data-integrations org.
• The data pipeline using source and sink plugins should run on both mapreduce and spark engines.

User Stories

• User should be able to install MongoDB specific database source and sink plugins from the Hub
• Users should have each tool tip accurately describe what each field does
• Users should get field level lineage information for the MongoDB source and sink 
• Users should be able to setup a pipeline avoiding specifying redundant information
• Users should get updated reference document for MongoDB source and sink
• Users should be able to read all the DB types

Plugin Type

• Batch Source
• Batch Sink 
• Real-time Source
• Real-time Sink
• Action
• Post-Run Action
• Aggregate
• Join
• Spark Model
• Spark Compute

Design Tips

MongoDB driver reference: http://mongodb.github.io/mongo-java-driver/3.10/driver/

Design

The suggestion is to move existing mongodb-plugins module to the mongodb-plugins repository.

MongoDB Overview

Document database

A record in MongoDB is a document, which is a data structure composed of field and value pairs. MongoDB documents are similar to JSON objects. The values of fields may include other documents, arrays, and arrays of documents.

{
    "_id" : ObjectId("5d3f1c2a2f547625b0bbb397"),
    "string" : "AAPL",
    "int32" : 10,
    "double" : 23.23,
    "array" : [ 
        "a1", 
        "a2"
    ],
    "object" : {
        "inner_field" : "val"
    },
    "binary" : { "$binary" : "YmluYXJ5IGRhdGE=", "$type" : "00" },
    "undefined" : undefined,
    "boolean" : false,
    "date" : ISODate("2019-07-29T16:17:46.109Z"),
    "null" : null,
    "regex" : /./,
    "dbpointer" : DBRef("source", "5d079ee6d078c94008e4bb3a"),
    "javascript" : var l = 1;,
    "javascriptwithscope" : { "$code" : var l = 1; ,  "$scope" : { "scope" : "scope_val" } },
    "symbol" : "a",
    "timestamp" : Timestamp(1564417066, 1),
    "long" : NumberLong(9223372036854775807),
    "decimal" : NumberDecimal("3.100000"),
    "minkey" : { "$minKey" : 1 },
    "maxkey" : { "$maxKey" : 1 }
}

BSON

BSON is a binary serialization format used to store documents and make remote procedure calls in MongoDB. The BSON specification is located at bsonspec.org

Document limitations

• The maximum BSON document size is 16 megabytes.
• In MongoDB, each document stored in a collection requires a unique _id field that acts as a primary key. If an inserted document omits the _id field, the MongoDB driver automatically generates an ObjectId for the _id field.

Flexible schema

Unlike SQL databases, where you must determine and declare a table’s schema before inserting data, MongoDB’s collections, by default, does not require its documents to have the same schema.

• The documents in a single collection do not need to have the same set of fields and the data type for a field can differ across documents within a collection. 
• To change the structure of the documents in a collection, such as add new fields, remove existing fields, or change the field values to a new type, update the documents to the new structure.

Query filter documents

A query filter document and query operators can be used to specify conditions.

The following example uses '{ status: { $in: [ "A", "D" ] } }' query filter document to retrieve all documents from the 'inventory' collection where 'status' equals either "A" or "D":

db.inventory.find( { status: { $in: [ "A", "D" ] } } )

The operation corresponds to the following SQL statement:

SELECT * FROM inventory WHERE status in ("A", "D")

Sink Properties

Sink Data Types Mapping

To support all data types in the Sink we can use MongoDB extended JSON format or/and infer a data type of record field based on its name. 

The table below does not honor non-standard MongoDB data types and lists how CDAP data types are stored.

The following MongoDB data types are missing: Undefined, Regular Expression, DBPointer, JavaScript, Symbol, JavaScript (with scope), Timestamp, Min key, Max key.

["string","int","long"]

Source Properties

Source Data Types Mapping

The source requires Output Schema to be set. Based on the schema source will expect a field in each document to be of a specific Mongo data type.

On Record Error error handling property allows the user to decide whether the pipeline should fail, the record should be skipped, or the record should be sent to the error dataset.

The following table shows what MongoDB data types can be read as CDAP types.

{"type":"record","name":"object","fields":[{"name":"inner_field","type":"string"}]}

{
 "object" : {
        "inner_field" : "val"
    }
}

{
  "type":"record",
  "name":"javascriptwithscope",
  "fields":[
    {"name":"$code","type":"string"},
    {"name":"$scope","type":{"type":"record","name":"scope-object-record","fields"[{"name":"scope","type":"string"}]}}
  ]
}

{
  "javascriptwithscope" : { "$code" : var l = 1; ,  "$scope" : { "scope" : "scope_val" } }
}

{"type":"map","keys":"string","values":"string"}

{
 "object" : {
        "inner_field" : "val"
    }
}

Approach

Move existing mongodb-plugins module to the mongodb-plugins project. Add MongoDB-specific properties to configuration, add support for MongoDB-specific datatypes. Update UI widgets JSON definitions.

Pipeline Samples

Releases

Release X.Y.Z

Related Work

Database plugin enhancements