Try our conversational search powered by Generative AI!

Jeff Valdez
Jun 14, 2012
  4807
(1 votes)

Entry SortOrder Property, Where Are You?

An EPiServer partner developer asked a question today about the entry sort order property. He could find the interface to affect changes on catalog entry sort order values in Commerce Manager, but couldn’t intuitively find the API hooks within the Mediachase.Commerce.dll assembly. I thought this would be a good topic for an initial blog post.

The sort property for a catalog entry is actually manipulated via a different strongly typed dataset object than the CatalogEntryDto, namely the CatalogRelationDto. In this DTO, we use the NodeEntryRelation datatable and find the desired records based on a unique combination of CatalogId, CatalogNodeId, and CatalogEntryId. Once we have a reference to the row we can affect the SortOrder property and then persist changes as needed.

To get this dataset object, you can use one of the three GetCatalogRelationDto() method overloads from the CatalogContext singleton. The easiest one in this case is the one that takes an integer entry id.

In this method, execution is passed to the CatalogRelationManager, then down to the CatalogRelationAdmin where the stored procedure [ecf_CatalogRelationByChildEntryId] is executed. Query results are mapped to three of CatalogRelationDto's datatables: CatalogNodeRelation, CatalogEntryRelation, NodeEntryRelation. At this point you can iterate through the strongly typed (CatalogRelationDto.NodeEntryRelationRow) datarows of the NodeEntryRelation datatable.

Below is a trivial example. (Apologies if the blog theme horizontally truncates the syntax highlighted portion below.)

   1:  // set parameters
   2:  int catalogEntryId = 7;
   3:  int newSortOrderValue = 10;
   4:   
   5:  CatalogRelationDto catalogRelationDto = 
   6:    CatalogContext.Current.GetCatalogRelationDto();
   7:   
   8:  foreach (CatalogRelationDto.NodeEntryRelationRow row 
   9:    in catalogRelationDto.NodeEntryRelation)
  10:  {
  11:    if (row.RowState != DataRowState.Deleted 
  12:      && row.CatalogEntryId == catalogEntryId)
  13:    {
  14:      // set the value of our NodeEntryRelationRow instance
  15:      row.SortOrder = newSortOrderValue;
  16:    }
  17:  }

Finally, to persist those datarow changes you can use the SaveCatalogRelationDto() method from the CatalogContext singleton. The easiest one in this case is the one that takes an integer entry id.

   1:  // persist changes
   2:  if (catalogRelationDto.HasChanges())
   3:  {
   4:    CatalogContext.Current.SaveCatalogRelationDto(catalogRelationDto);
   5:  }

In this method, execution is again passed to the CatalogRelationManager which raises some pre and post processing pipeline events around the save operation and clears old cached records. The save operation is handled by a CatalogRelationAdmin instance where the DataHelper class is used to persist the dataset.

When looking at the catalog API, you’ll come across this pattern frequently. There is a singleton class funneling calls through a single instance of ICatalogSystem which contains the underlying concrete definition based on configuration. From there methods usually go through a manager class where other framework tasks can be injected, such as caching objects and raising events. Finally, the baton is passed to an admin class where the intended work is done. You may also run head first into the Mediachase.MetaDataPlus.dll assembly, but that is a blog post for another day. Along the way there are strongly typed datasets for ease of programming that seemingly map directly against the database schema and database objects, as well as C# POCO objects to abstract away the details.

We should see these libraries, and the layers that allow presentation, integration, and other functions above improve and innovate in future versions, can’t wait!

Jun 14, 2012

Comments

Jun 15, 2012 02:33 PM

Welcome Jeff and what a great way to start your EPiServer World blogging career!

Please login to comment.
Latest blogs
Optimizely Search and Navigation - Part 2 - Filter Tips

Introduction Continuing from Part 1 – Search Tips , today I will share the next part – filter tips. The platform versions used for this article are...

Binh Nguyen Thi | Jul 1, 2024

Integrating HubSpot CRM without the MA Connector

Have HubSpot CRM? Want to push user data into it from Optimizely? Don’t have any personalisation requirements with that data? Don’t want to pay $80...

Matt Pallatt | Jun 27, 2024

Keeping the website secure by updating external packages

Did you see the latest warning from Optimizely to update this package with a critical security warning? https://world.optimizely.com/documentation/...

Daniel Ovaska | Jun 27, 2024

Optimizely CMS image anonymization now available for Linux!

The famous image anonymization add-on for Optimizely CMS, with at least 5 downloads, is now finally available for use on Linux. Supports simultaneo...

Tomas Hensrud Gulla | Jun 25, 2024 | Syndicated blog