We have previously talked about what does it mean to manually manipulate the Dataverse data model, and the benefits of data model scripting.

We'll now deep dive on what it means to create tables, columns and relations via PACX.

Creating a table with PACX is quite easy.

As a prerequisite, you must connect to a Dataverse environment via pacx auth create or pacx auth select, and optionally you can select a default solution via pacx solution setDefault.

Then you can just type

pacx table create --name "My Table"

to create a new table named "My Table" in the context of the solution set as default via pacx solution setDefault. If no solution is set as default, the --solution argument is mandatory.

The table schema name will be automatically generated extrapolating only chars, numbers and underscores from the display name, setting them lowercase, prefixed with the solution's publisher prefix.

In this case, if the publisher prefix is greg, the generated schema name will be greg_mytable.

The command can be also run using the alias pacx create table

The following default conventions apply automatically:

• DisplayCollectionName (aka the plural name of the entity) can be specified using --plural argument. If not specified, is inferred automatically pluralizing the display name (automatic pluralization is done via Pluralize.NET, and is currently supported only if base language is English - 1033).
• Description can be specified using --description argument, otherwise is left empty.
• SchemaName can be specified using --schemaName argument, otherwise is inferred from the display name as described above.
• OwnershipType can be specified using --ownership argument, otherwise is set as UserOwned.
• IsActivity by default is false, can be changed via --isActivity argument
• IsAvailableOffline by default is false unless the table is an activity table, or the --offline argument is specified
• IsValidForQueue by default is false unless the table is an activity table, or the --queue argument is specified
• IsConnectionsEnabled by default is false unless the table is an activity table, or the --connection argument is specified
• HasNotes by default is false unless the table is an activity table, or the --notes argument is specified
• HasFeedback by default is false unless the table is an activity table, or the --feedback argument is specified
• IsAuditEnabled by default is true, but can be overridden via --audit false argument

About the table primary attribute:

• You can set it as an Autonumber field via --primaryAttributeAutoNumberFormat argument (-paan). If not specified, it is assumed to be plain text.
• The display name
  • if the table is an activity, is fixed to Subject
  • otherwise, it can be specified via --primaryAttributeName (-pan) argument.
  • if not specified
  • if the the primary attribute is an autonumber, it's set by default to Code
  • otherwise it's set to Name
• The requirement level
  • can be specified via --primaryAttributeRequiredLevel (-par).
  • if not specified
  • if it's an autonumber, it's set by default to None
  • otherwise it's set by default to ApplicationRequired
• The max length is 100 by default, unless specified via --primaryAttributeMaxLength (-palen) argument
• The description can be set via --primaryAttributeDescription argument (-pad). If not specified, is left empty

Take a look to the official documentation to get the list of all the command arguments.

Once created, you can use the following command

pacx table exportMetadata --table greg_mytable -r

To generate and open a JSON representation of the table metadata.

Hope this can be helpful to start using PACX in your daily activities. In the next articles we'll deep dive on how to work with columns.