Edit Definition Attributes  
All the attributes used to generate the definition documentation are displayed in the edit definition screen. The fields that pertain to the entire dataset are captured within the first row for each dataset.   If the data is changed and therefore the attributes are changed, then a refresh needs to be applied.  The basic steps towards editing the attributes are shown here:

1. Select an attribute by selecting the field to be edited.
2. Edit the attribute directly or enter the value in the "Value Update" text entry underneath and click on the "Apply" button.
3. Optionally, click on the [...] button to select recommended values before applying the update to the values.
4. Alternatively, you can select multiple rows or columns and enter a value in the "Value Update" text entry and click on the apply button to apply to multiple fields at once.

You can choose to commit the changes by clicking on the OK button or click on the "Cancel" button to ignore all the updates to the current session.  

Some of the fields will have recommendations listed if you were to click on the [...] button next to the "Value Update" field.  These include:

1. Dataset Name:
   
   
   
   The list of existing datasets is displayed on the left while the recommended standard CDISC domain dataset names are listed on the right.  Only one value can be selected.
   
2. Variables:
   
   
   
   The list of existing variables is displayed on the left while the recommended standard CDISC variable names are listed on the right.  Only one value can be selected.
   
3. Dataset Label:
   
   
   
   The list of existing dataset labels is displayed on the left while the standard CDISC dataset labels are listed on the right.  Only one value can be selected.
   
4. Variable Labels:
   
   
   
   The list of existing variable labels is listed on the left while the standard recommended CDISC variable labels are listed below.  Only one value can be selected.
   

Editing Value Level Metadata
Value level metadata captures specified metadata that and associated values of the selected variable.  You can select a specified variable to contain value level metadata by scrolling over to the column labeled "Value Level Metadata".  You can have the following options in assignment of this column.

• Missing or No - A missing value or a value of "No" will indicate that there is no value level metadata to be captured and generated in the final documentation.
• Yes - A value of "Yes" indicates that the variable corresponding to the row is going to contain value level metadata.  
• Variable Name - If another variable name is selected, this indicates that a nested of sub variable will be assigned and will therefore have the corresponding values of the nested variables be the value level metadata.  

The assignment of the value level metadata can be specified by first selecting the corresponding cell in the "Edit Definition" screen and then clicking on the "..." button to the right of the "Value Update" entry.  In this case, the "Yes" or list of sub variables can be selected from a list.

Once the specified variable has been assigned to contain a value level metadata, attributes of the value level metadata can be assigned through the "Value Level" button from the "Edit Definition" screen.

You can perform the following tasks from this screen.

• Edit Values - Users can edit by typing the value in the cell entry or through the "Value Update" entry.  The "..." button will also help select default values.
• Export Excel - Export all values of attributes to an Excel spreadsheet.  Note that in this case, you may need to click on the OK button to save changes before exporting all the correct data.
• Import - Import the updated data in the Excel file.
• Refresh - Refresh the latest set of attributes to reflect all variables that has been specified as value level metadata.

Refreshing Definition
When the physical SAS Dataset is not in sync with the attributes captured by the definition, you can apply a "refresh".  This will compare the variable attributes such as variable names and labels against what is currently in the definition.  If there is a mismatch, the refresh will update the definition.  The updates will not update non key fields such as comments and keys.  These "non key" attributes will be edited by the user and therefore not changed by the refresh process except in the event where the variable and the associated attributes are deleted.  

Audit Trail
All edits to the data definition are captured through an audit trail.  This audit trail is updated when the user clicks on the OK button to apply updates to the definition.  A report of the audit trail can be generated by clicking on the "View Audit Trail" button from the edit definition screen.  The report captures all the values that have been updated with a new status column documenting if they were updated or deleted.

Editing Comments 
Comment fields are usually entered as free text from the user.  The comments are usually used to describe how a particular variable is derived.  When a comment is selected and if you click on the [...] button next to the "Value Update" field, it will perform a %getcat which derives categorical values from the selected variable.  This is used as the default value for comments which you can use.  

Capturing Categorical Variables
Categorical variables are usually character variables which contain a small set of distinct values.  It is common to have these values displayed in the comment fields.  You can therefore derive these values in two ways.  

1. Generate Categorical Values - In the edit definition screen, if you were to select the comment field of a categorical variable and select the [...] button, it will automatically generate a default value listing of all the categorical values.  The result in this case begins with the text "VALUES:" followed by a list of values separated by commas.
   
2. Finding Categorical Variables - Clicking on the categorical button on the edit definition screen will present the following:
   
   
   
   This will allow you to choose all the datasets from the source location.  In addition, a report is generated listing all the categorical variables and their distinct values.  A macro version of this %getcat can also be generated through the "Save Code" button.
   

Applying Undo
Each time a report is generated through the "Generate" button or updated through the "OK" button, a backup of the definition is made.  This backup is stored in a user profile so each user will have their own backup version.  You can therefore apply an undo by clicking on the "Undo" button from the main screen.  This will use the latest backup version from your user profile and write it to the _DEFINE dataset which stores the definition.