A mapplet is a reusable data object representing a set of transformations and its logics to transform data among the tablets.
Let's assume there are n number of tables. Our requirement is to select the dimension keys associated to each table. We now create a mapplet with a series of Lookup transformation to find each dimension keys. Now, use these mapplet in each table mapping so that we can find total dimension keys in each table.
The process is defined in 3 steps:
1. You need to add, connect, and configure transformations with a transformation logic.
2. Save the mapplet with a unique name.
3. Use the mapplet in mapping
What are the features of Mapplets?
1. A mapplet can define source definitions of a key data and a source qualifier to provide data for a mapping.
2. A mapplet can have no source definitions on data . Maplet can accept those data also from a mapping process through the mapplet input ports.
3. A mapplet can have multiple transformations.
4. Data can be transformed to multiple pipelines. A mapplet can contain multiple groups of output ports/groups that can be connected with each other through a different pipelines in the mapping.
Mapplets are created for 2 main purposes. They are:
1. Every time while loading a new data into a table, there occur some space in source system and every time a new transformation has to set on every mapping. Mapplets are created to avoid these repetition of creating transformation expression and to terminate the unwanted space formed in the source system.
2. On every workflow running in the system, a batch id is formed based on the session timestamp by using a mapplet.
Both have one similarity that they work on data. But Mapping deals with those data on which a modification need to be made whereas Maplets deals with multiple mappings. Other major differences are as follows:
|It is a collection of source data objects linked by set of transactions, targets on which the data has to move.||It is a collection of transactions or set of rules applied on data.|
|The transformations cannot be reused.||The transformations can be reused.|
|It uses data sources and a transformation logic to transform the data to the target||One can create a series of transformation logics.|
|Mappings are applied on small amount of data||Mapplets are applied on Bulk data or Big data|
|Mapplet is a part of Mapping.|
|Basic components of Mapping are
1. Source tables
2. Parameters and variables
3. Target objects
|Basic Components of Mapplets are
As shown in the below figure, Mapping is all about connecting one database to another. These connections are made by defining a set of roles called as transformation. Each transformation has a definite validating elements defined called as parameters and variables. To define a transformation, there should be a source and destination objects which are called as Source Tables and Target objects respectively.
Mapplet applies on Mapped data. As shown in the figure, the source and target objects are defined and stored in the database. A logic to transform is defined by providing an Mapplet Input transformation that passes the data from Mapping to Mapplet and then a desired Mapplet output transformation that passes data from Mapplet to Mapping. . Thus Input- Output transformations are revolved in a cycle making Mapping, transaction logic and data reused several times without affecting the data.
Reusable transformation means any transition logic or rule defined on a data to transform from source to destination through mapping can be provided for multiple times with different methods and the logic can be used on other transformations.
1. You cannot connect a single Input port to multiple transformations.
2. An Input transformation must get data from single active user only.
3. Depending on the source qualifiers- a set of rules defined on source data to extract data, Mapplet output ports use pipeline method for single qualifier on source data or Join method for more than one qualifier
4. PowerMart 3.5-style LOOKUP functions are not supported in a Mapplet.
5. If Mapplets are changed from passive to active, the mapping defined on data is invalid.
1. Commenting on each Input and output transformations made in Mapplets is the best way to avoid mistakes. They will trigger us why we made them.
2. Do not provide any alterations or changes on the source data datatype, data precision, or select the connected ports in I/O transformation, from passive to an active mapplet.
Mindmajix offers training for many of other informatica courses depends on your requirement:
|Informatica Analyst||Informatica PIM|
|Informatica SRM||Informatica MDM|
|Informatica Data Quality||Informatica ILM|
|Informatica Big Data Edition||Informatica Multi Domain MDM|
Free Demo for Corporate & Online Trainings.