Upgrading to Jet 2019: What’s New In Jet Analytics 2019

This release entirely supports Azure SQL Database Managed Instance, that represents fully managed SQL Server Instance hosted in Azure cloud – allowing customers to easily take advantage of new technology when released by Microsoft. In fact, 18.10 supports any of Microsoft SQL Servers, be it Microsoft Azure SQL DB Managed Instance, Single or Elastic Pool and of course the on-prem SQL database.

Row Level Security
In version 18.10, you can set up row-level security on models in the shared semantic layer. The security setup can be deployed to SSAS Tabular Endpoints.
Controlling access to data on the row level can be a very useful addition to controlling access on the table or model level. One common example is a sales department where each salesperson needs access to data from their own sales region, but not everyone else’s.

In Jet Data Manager 18.10, you can order fields on semantic layer tables in hierarchies. All endpoint types support hierarchies.

Ordering fields in hierarchies can be a good way of making a model easier to grasp and use in PowerBI. An example is a table with products that often have multiple levels of categories.

You can now configure data formats and categories on fields which will be applied to SSAS Tabular endpoints and used in front-ends to decide how to display data. For instance, text categorized as “Web URL” can be displayed as a link.

“Display folders” is a feature of SSAS Tabular which allows you to organize the fields, measures and hierarchies of a semantic table by adding them to folders. This makes the model easier to browse, especially if it has a lot of fields. With display folders, you can (for example) bucket all address information on a customer table in a “Geography” display folder.

In version 18.10, Semantic Models can now be prioritized in the execution engine, which gives the user an option to influence the execution order to get certain items finished earlier than others.

Jet Analytics 18.10 includes new context menu command “Clone field” on all fields, as well as a new dialog for cloning fields (incl. Conditional Lookups, Supernatural Keys, Custom Hash Fields, transformations, standard data fields).

Two options appear in the dialog: Cloning structure (incl. transformations) or data (only adding transformations for cloned data).

When you extract data from a data source that is in use, you can end up with inconsistent data where a foreign key in a record points to a record in another table that does not exist.

Jet Data Manager 18.10 can use the fact that Dynamics NAV has a timestamp on every table to minimize this problem. If you enable the new consistent read setting, the JDM will only extract records older than the start of the data extraction. This will prevent inconsistencies caused by data added while extraction is running.

One of the challenges of using incremental load is that the value of a lookup field on an incrementally loaded table does not change if the value being looked up changes. The lookup field value is calculated when the row is loaded so the usual solution to this problem is to schedule regular full loads of the table.
Jet Data Manager 18.10 includes a new setting on incrementally loaded tables: “Keep lookup values up-to-date”.

When this setting is enabled, the JDM will do comparisons to see if any values being looked up have changed and will update the lookup fields accordingly.

Version 18.10 contains improvements to the Oracle data source, which makes it easier to set up the connection and makes the connection Oracle-version independent.

The data source connector now supports EZConnect which means that you can simply enter the connection information (server, port and service name).

New features:

• When dragging and dropping fields, a new easy-to-use dialog for adding relations is shown, displaying existing relations with a tooltip, possibility of extending an existing relation, or adding a new default relation or non-default relation.
• When adding a new conditional lookup field or a new lookup field, a new simplified, comprehensible dialog for adding lookup joins is shown, with a possibility of copying joins not only from relations but also from existing lookup fields. This dialog also shows details of all joins in tooltips. Only different joins from existing lookups are shown (duplicates are removed from the list).
• Dragging and dropping of fields in another table with the right mouse button creates a new conditional lookup field instead of a new relationship and allows for easy positioning of the conditional lookup field before or after the dropped field.
• In the Add/Edit Lookup Field dialog, you can now select which Operator you want to “Set as Default”.
• New lookup operator ‘None’ greatly improves performance in some scenarios + automatic changing (after confirmation) of the lookup operator to ‘None’ when adding joins different than ‘Equal’.
• In the “Add Join” dialog, new lookup join fixed value operators have been added: ‘In list’, ‘Not in list’, ‘Like’, ‘Not like’.
• Copying Lookup Fields within the same table via drag and drop (the right mouse button also offers two options: add or replace)
• Copying Lookup Joins within the same table via drag and drop (the right mouse button also offers two options: add or replace)
• Copying Lookup Sortings within the same table via drag and drop (the right mouse button also offers two options: add or replace)
• Copying Lookup Conditions within the same table via drag and drop, also between lookups and transformations, and vice versa (the right mouse button also offers two options: add or replace)
• Automatic renaming (after confirmation) of lookup fields and conditional lookup fields when the lookup data field is changed

Performance Improvements:

• Batch data cleansing now works for all conditional lookups and supernatural keys.
• History tables have been optimized and work significantly faster
• Conditional lookups can be merged, so all lookups are performed at one time.

Fixes:

• The first relation is automatically created as default, and it is no longer possible to remove default from a relation other than by adding default to another relation or deleting the relation. If there are several relations, one is always default.
• Creating relations by dragging and dropping fields is now possible also if a relation was already created in the opposite table. Relations now work regardless of in which table they were created, i.e. they can be used in both tables.
• There are now consistent dialogs for adding/modifying conditional lookup fields, lookup fields, lookup joins, and lookup sortings
• Lookups now allow MAX and MIN with the bit data type