Loading the return item into a transactional fact table is a process that requires a number of stages to obtain and work with the data in different ways. To clarify this process, it is presented as a data flow in which data flows from one processor to the next. Along the way the data is transformed until the final processor records the data into the transactional fact table.
Data loading starts with the implementation of the Loader
. The atg.reporting.datawarehouse.
component uses the
loader.Loader.LoaderqueueName
property to point to a log file. The Loader
runs using a scheduler, creates a log file reader for returned items and invokes the processReader
method of the PipelineDriver
component.
The PipelineDriver
reads delimited lines from the log file then parses them and populates a pipeline parameter. The parameter is then sent down a pipeline chain. The atg.reporting.datawarehouse.
component uses the properties
loader.FilePipelineDriver.PipelineDriverparamPropertyNames
and paramClasses
properties to specify the names and types of parameters that are read.
The PipelineManager
runs the pipeline chain described in the pipeline.xml
file. Pipelines consist of several processors. These processors collect data, prepare line items and insert them in into the data warehouse.
Returns and Exchanges Pipeline Processors
The following processors are used in the Submit Return pipeline:
Pipeline Chain | Pipeline Link | Description |
---|---|---|
| Uses the return ID to look up the return/exchange in the CSR repository. | |
| This processor fetches all return items in the warehouse for the current return ID. If there are return items the parameter map entry is created for the | |
| This processor allows a switch to be implemented in the pipeline by determining if a warehouse item exists in the properties. If a | |
| This processor creates an array or the returned items map. | |
| This processor gets refund information from | |
| This processor gets the return fee from | |
| This process sums up all adjustments for each returned item, such as shipping share, tax share, and other return fee allocation amounts. The suggested shares are used to calculate the actual share. If the suggested share is zero, the share will be calculated based on the return item’s quantity. | |
| This processor sums up all total adjustments and item refund items. | |
| This processor converts local currency to standard currency values. | |
| This processor converts shipping, tax and other refunds, the return fee, total adjustments and refunds. | |
| This processor looks for the ID of the day for a given time stamp. This processor uses the return request | |
| This processor looks for the ID of the time for a given time stamp. This processor uses the return request | |
| This processor gets the customer ID from the | |
| This processor gets the agent ID from the | |
| This processor looks up the return channel ID in the data warehouse. | |
| This processor runs the return item pipeline for each element of the | |
|
| This processor looks up the SKU ID for each return item in the data warehouse. |
| This processor looks up the product ID for each return item in the data warehouse. | |
| This processor looks up the return reason ID for each return item in the data warehouse. | |
| This processor creates a repository item for the logged data that is based upon the lookup properties. |
Calls Pipeline Processors
The following processors are used in the call pipeline:
Pipeline Chain | Pipeline Link | Description |
---|---|---|
|
| This processor obtains the call item for the call ID. If the call item exists, the parameter map entry is created for the |
| This processor implements a switch in the pipeline by determining if a warehouse item exists in the properties. If a | |
| This processor gets the customer ID from the parameter map. If the customer is not found in the data warehouse, it will look in the production schema. If found in neither schemas, the processor will return | |
| This processor gets the agent ID from the parameter map. If the agent is not found in the data warehouse, it will look in the production schema. If found in neither schemas, the processor will return | |
| This processor looks for the start time timestamp from the parameter map as well as the lookup for the day of the time stamp. | |
| This processor looks for the start time timestamp from the parameter map as well as the lookup for the time of the time stamp. | |
| This processor looks for the end time timestamp from the parameter map as well as the lookup for the end day of the time stamp. | |
| This processor looks for the end time timestamp from the parameter map as well as the lookup for the end time of the time stamp. | |
| This processor calculates the total amount of call time in seconds. | |
| This processor creates a repository item for the logged data based on the lookup properties. |