One of our aims is to provide our users with the ability to maintain and manipulate their data conveniently even outside iteraplan. Among the most accessible and common ways to do this is the representation of the enterprise architecture as a Microsoft Excel file, in which different sheets cover aspects of the architecture such as types or relations. While iteraplan has provided this feature all along, it has now been improved so that is not only possible to obtain the total of the user’s data at once, but also to obtain an extract of the data tailored in accordance with the user’s current needs. These data excerpts can, for example, be sent to responsible persons for data maintenance. Once the responsible persons have edited their corresponding Excel files, these can be imported back into iteraplan by the person in charge, to produce a unified updated picture of the enterprise architecture. This part of the iteraplan user’s guide addresses the steps of this process in detail.
Creating a Partial Export
The first part of the described scenario is the creation of a partial Excel export. This can be accomplished through the iteraplan Export/Import Menu in the Report-Tab.
Once the page is opened, the user can activate the partial functionality by enabling the switch ‘partial’ from 'No' to 'Yes' below the checkboxes for the different export formats.
Note that once the partial configuration is enabled, the only available download is in the XLSX Excel format and the partial export options are presented, as depicted in the figure below.
In the initial configuration, the partial export is based on the building block type "Architectural Domain". The building block around which the export is centered can be changed by selecting a different type from the drop-down menu.
The export can be refined by specifying an iteraQl filter condition in the text box provided below the drop-down box. For example, the configuration
will specify an export in which only the Business Processes for which bob is responsible are contained. Note that while arbitrary iteraQl statements can be used, we recommend to use filters based only on local features of the specified building block type.
After those few steps, the configuration of the partial export is complete and the Excel file can be downloaded.
Excel File Structure - What to Expect
As with the original iteraplan Excel export, the file may contain sheets representing different aspects of the enterprise architecture’s metamodel:
- Type sheets: Sheets containing instances of building block types. Each row contains an instance of a building block type, with its properties.
- Relationship sheets: Sheets representing relationships between types of building block. Each row contains a pair of instances of the respective building block types.
- Enumeration sheets: One of these sheets is provided for each enumeration used by a type sheet. The sheet contains the enumeration values, or literals, of the respective enumeration.
In partial exports, the number of type, relationship and enumeration sheets is greatly reduced. Consider the Excel file obtained in the last section:
Note that the file contains only two type sheets: one for Business Processes and one for Business Mappings, and a single relationship sheet, for the relationship between Business Processes and Business Domains. Recall that the partiality of the export was defined on the basis of the Business Process type of building block.
In general, a partial export is constructed by iteraplan through the following rules:
- One sheet is created for the type of building block selected by the user as the basis of the partial export. This type is denoted as main export type.
- If a filter is provided, the main export type is additionally restricted in accordance with the filter. Note that not specifying a filter is equivalent to specifying a filter which allows all instances through.
- If the main export type has association building block types related to it, all of these also obtain sheets.
- All relationships of the main export type are each provided with a sheet.
- Finally, sheets are created for all enumeration attributes of the exported types.
In the example above, the Business Process (filtered by accountability) is the main export type. The Business Mapping sheet is included because of the Business Mapping being an association type of building block. Finally, a single relationship sheet is included, as the Business Process type has only this one relationship to Business Domains.
Having obtained a partial Excel export, the next step of the example scenario is the actual data maintenance. In this aspect, the partial Excel file is no different than the original full Excel file provided by iteraplan and modifications can be carried out on both type and relationship sheets as described below.
In type sheets (see figure below), the modification of the value of a cell represents the modification of the feature designated by its column for the building block instance represented by the corresponding row. The deletion of a row in a type sheet represents the deletion of the building block instance. Bear in mind that the deletion or renaming of instances can cause a consequent import attempt to fail, if the instance is referenced in another sheet. Finally, new rows in a type sheet are interpreted as newly created instances of the building block type.
In relationship sheets (see figure below), each row represents an association between two building block instances. The deletion of a row in a relationship sheet thus represents the removal of the relation between the two instances. An update of one of the cells of a row represents the update of the association, i.e. the old association is replaced by the new one. Finally, a new row represents a new association between two building block instances.
- The semantics of the modifications to the Excel sheets described here are valid under the assumption that the Excel file is interpreted as an enterprise architecture model. In this sense, while the described operations have the provided meaning in the scope of the Excel file itself, their interpretation when importing them into iteraplan may deviate, depending on the import strategy selected. To illustrate this, consider updating one of the cells of a row in a relationship sheet. The intended meaning of this operation is to replace the old association with the new one. Nonetheless, if the Excel file is imported with an Additive strategy, the old association will not be removed. Instead, both the old and the new associations will be available in iteraplan.
- The partial Excel file contains a number of hidden sheets, rows and cells. These should not be modified, since they contain important context information, necessary for the interpretation of the visible data when importing the Excel file. Modifying this part of the Excel file may corrupt it and render it un-importable, or might have undesirable side-effects.
- With renaming building blocks using the import, we strongly advise to do this in an extra import run with no other changes, using the additive mode. Please also take care to adjust the names in cells referencing the renamed building block, for example in the parent column of the building block sheet or on relationship sheets.
Importing a Partial Excel File in iteraplan
Once an Excel file is modified in accordance with the user’s demands, the data can be re-imported into iteraplan.
The user can select the desired import strategy. Currently, iteraplan supports the Additive, Conservative and Overwrite strategies. These are explained in more detail in the Import Strategies page.
Furthermore, the Import Data section offers a switch-button, which enables the user to also import Metamodel changes from an Excel file. This check-box will not work for partial Excel files. If the check-box is enabled, iteraplan will ignore it with a corresponding message to the user.
The user can start the import process by adding the file through the Select File and Start Import button. iteraplan will automatically detect that the uploaded file contains a partial Excel export.
The import process then navigates the user through a number of steps. As already mentioned, the partial Excel import does not allow for changes to the iteraplan metamodel (creation of attribute types, in particular). Thus, in case the imported file is internally consistent, the process directly proceeds to the data comparison (step 4), while informing the user that a partial Excel import has been initiated (this information is displayed in the box below the Export/Import-boxes). The user can proceed from one step to the next by clicking the Next button.
Only elements which fulfill the filter conditions selected during the partial export will be considered for import. That means if the Excel file was modified sucht that some of the rows contain data contradictory to those filter conditions, those rows will be ignored during import.
From the imported file, iteraplan extracts the date and time at which the export was made. This information is relevant for all import strategies except the additive strategy. Only entities which were not modified since the export are considered when updating the enterprise architecture in iteraplan using one of the other strategies. In other words, the timestamp of the export guarantees that the information being updated is not newer than the information contained in the Excel file.
Running the Validate data step of the import process provides the user with a summary of the changes which will be applied to the enterprise architecture, in accordance with the selected import strategy and the actuallity of the data contained in the Excel file with regard to the current enterprise architecture. If the changes are to the user's satisfaction, the last step of the import process can be triggered. This step performs the actual import of the Excel data into the iteraplan database. Once the step is finished, the user obtains a summarized report of the data written. With this the import process is completed.
Note: If a user has only read permissions for the main export type of a partial Excel file, trying to import the file will result in an error message, due to the user's lack of permissions. The same error message can also be observed, if the user's metamodel at the time of import is missing a feature which was used for the definition of the partiality specification when the Excel file was generated.
Error Messages And Error Locations
Missing building block sheet
Error in step 1: "Import process terminated with errors. No changes were made to the database."
Reason: When the sheet for the main building block type of the partial import is missing, iteraplan can't continue with the import.
User actions: Make sure the building block sheet type is included in the import file.