Work with Generated Repositories and NoetixAnswers
This chapter provides information about the Oracle Business Intelligence (Oracle BI) repository objects, how Noetix Generator for Oracle Business Intelligence (Noetix Generator) interacts with these objects, and tells you how to use the Oracle BI Presentation Services to view the generated content. In addition, this chapter provides information about NoetixAnswers, a prebuilt set of report and dashboard templates.
About Generated Objects
The Noetix Generator generates the Magnitude NoetixViews (NoetixViews) and Noetix Analytics for Oracle E-Business Suite (Noetix Analytics) metadata into the following Oracle BI repository layers:
Physical
Business Model and Mapping
Presentation
Physical
The Physical layer of the Administration Tool defines the data sources to which the Oracle BI Server sends queries and the relationships between physical databases and other data sources that are used to process multiple database queries. In this layer, Noetix Generator does the following:
For NoetixViews:
Creates a data source that points to the TNS name associated with the NOETIX_SYS schema.
Creates a connection pool in the data source. The following properties are set for the connection pool, based on the values given by you as an administrator:
Name
Connection pool type
Timeout
Maximum number of connections
Creates a schema folder based on the NOETIX_SYS schema. A SYS schema is also created to house the DUAL table. This is necessary to support multiple-view reports. See “Writing Multi-view Reports and the JOIN ASSIST Column” for more information.
Creates a physical table in the generated schema folder for each view in the NOETIX_SYS schema, based on the roles you select to generate.
Creates foreign key relationships between the physical tables. These relationships are based on the Z$ columns found in the NoetixViews.
Creates a physical table for each List of Values (LoV) view.
Creates the necessary joins between the LoV view and the views referencing them.
For Noetix Analytics:
Creates a data source that points to the TNS name associated with the schema used for data connections to the Operational Data Store and data marts.
-
Creates a connection pool in the data source. The following properties are set for the connection pool, based on values given in Noetix Generator:
Name
Connection pool type
Timeout
Maximum number of connections
Creates a schema for the Operational Data Store (ODS) and one for the data marts.
Creates a physical table in the appropriate schema folder for each ODS and data mart table found in the Active Metadata Repository (AMDR).
Creates alias tables for each table alias defined in the AMDR.
Creates foreign key relationships between the physical tables. These relationships are based on the relationship metadata found in the AMDR.
For both NoetixViews and Noetix Analytics:
Creates a “Noetix-Custom” physical display folder for custom physical objects added to the generated database. See Adding Custom Objects to the Physical Layer in Customizing Generated Content for more information on adding custom objects.
Business Model and Mapping
The Business Model and Mapping layer of the Administration Tool defines the business model of the data and the mapping between the business model and the physical layer schemas. In this layer, Noetix Generator does the following:
For NoetixViews (Standard and Cross Operations):
Generates a business model for each organizational unit (set of books, inventory organization, etc.) selected for generation.
Generates logical tables in the business layer based on each generated physical table.
-
Creates logical dimension tables for each physical table with the following attributes:
Have the same name as the physical table.
Contain all the columns that are present in the physical table, except columns used in other dimension tables based on the physical table.
-
Creates logical fact tables based on each physical table with the following attributes:
Have the same name as the physical table with a suffix “_Fact.”
Contain aggregate columns for each measure column in the view. These columns are defined in the fact table with a default aggregation function, such as SUM or COUNT. The SUM function is used by default, but can be modified in the Noetix Generator metadata.
-
Creates a logical dimension table for each key flexfield and key flexfield description hierarchy found in the underlying view. These tables have the following attributes:
Are named using the convention: <View Name>_Dimension_<Key flexfield Name>, for example, AR2_Cash_Receipts_Dimension_Account.
Contain all the columns from the view that are associated with the segments of the associated flexfield.
Example:
The HR_Applicant_Hist view has four key flexfields defined in it: Jobd, Posd, Vacancy_Job_Name, and Vacancy_Position_Name. Noetix Generator creates the following dimension tables for this view:
HR_Applicant_Hist: This table is named after the physical table and contains all of the columns in the physical table, except columns referenced in the other dimension tables.
HR_Applicant_Hist_Dimension_Jobd: This table describes the Jobd key flexfield and contains all columns related to this flexfield.
HR_Appliant_Hist_Dimension_Posd: This table describes the Posd key flexfield and contains all columns related to this flexfield.
HR_Applicant_Hist_Dimension_Vacancy_Job_Name: This table describes the Vacancy_Job_Name key flexfield and contains all columns related to this flexfield.
HR_Applicant_Hist_Dimension_Vacancy_Position_Name: This table describes the Vacancy_Position_Name key flexfield and contains all columns related to this flexfield.
Generates logical joins between the logical dimension tables and the logical fact table generated for the view. The resulting effect creates a star schema for each view.
-
Creates dimension objects for each dimension table generated with the following attributes:
Each dimension object has a total logical level, set as a grand total level.
Key flexfield based dimensions have each KFF segment added in a hierarchy, the first segment as the parent and each subsequent section as a child of the segment before it.
The key flexfield segment columns from the associated dimension table are added to each associated level of the dimension object.
Non-key flexfield based dimensions have a single Detail level under the Total level. The columns from the associated dimension table are added to the single Detail level.
Logical levels are added for each associated LoV column in the underlying view, if a foreign key join exists between the Noetix view and LoV view.
Each logical table source linked to a dimension table has aggregate content settings configured to map the table to the lowest level of the dimension objects associated with it. For key flex-field-based dimensions, the logical table is mapped to the last segment in the hierarchy. For non-key flex-field-based dimensions, the logical table is mapped to the Detail level of the dimension object because that is the lowest level.
-
Each logical table source linked to a fact table has aggregate content settings configured to map the table to:
The lowest level of each dimension object associated with the dimension tables joined to the fact table.
The lowest level of each “Detail” dimension object associated with dimension tables that can be joined to the view using Z$ relationships found in Noetix Generator metadata layer.
-
Generates logical display folders based on the NoetixViews roles with the following attributes:
Adds the logical dimension tables for views associated with the role to the appropriate folders.
Adds the logical fact tables for views associated with the role to the appropriate folders.
Adds the dimension objects for views associated with the role to the appropriate folders.
For NoetixViews (Global Extension):
Generates a business model for each global subject area (Financials, Human Resources, etc) selected for generation.
-
Generates logical dimension tables for each physical table based on the global transaction view. These logical dimension tables have the following attributes:
Have the same name as the physical table
Contain all columns present in the physical table, except any single structure key flexfield columns contained in the view. They will be modeled in separate logical dimension tables.
-
Generates logical dimension tables for each single structure key flexfield and key flexfield description hierarchy defined in the transaction views. These logical dimension tables have the following attributes:
Are named using the convention: <View Name>_Dimension_<Key Flexfield Name>, for example, FAG0_Asset_Assignment_Hist_Dimension_AssetLoc.
-
Contain all of the columns from the view that are associated with the segments of the associated flexfield.
Example:
The FAG0_Asset_Assignment_Hist view has three single structure key flexfields defined in it: AssetCat, AssetKey, and AssetLoc. Noetix Generator will create the following dimension tables for this view:
FAG0_Asset_Assignment_Hist: This table is named after the physical table it is based on and contains all of the columns in the physical table, except columns referenced in the other dimension tables listed below.
FAG0_Asset_Assignment_Hist_Dimension_AssetCat: This table describes the AssetCat key flexfield and contains all segment columns related to this flexfield.
FAG0_Asset_Assignment_Hist_Dimension_AssetKey: This table describes the AssetKey key flexfield and contains all segment columns related to this flexfield.
FAG0_Asset_Assignment_Hist_Dimension_AssetLoc: This table describes the AssetLoc key flexfield and contains all segment columns related to this flexfield.
-
Creates logical fact tables for each physical table based on the global transaction views. These logical fact tables have the following attributes:
Have the same name as the physical table they are based on, with a suffix of “_Fact”
Contain aggregate columns for each measure column in the view. These columns are defined in the fact table with a default aggregation function, such as SUM or COUNT. The SUM function is used by default, but can be modified in the Noetix Generator metadata.
-
Creates logical dimension tables for each key flexfield (XXK) view. These logical dimension tables have the following attributes:
-
One dimension table is created for each instance of the key flexfield in the related transaction views. The column label associated with the instance of the key flexfield will be appended to the end of the logical table’s name.
For example, an XXK_GL_Acct_AcctLiab dimension table will be created for the liability account defined in the general ledger accounting key flexfield.
These dimension tables will contain all of the columns found in the associated physical table.
-
One dimension table is created for each key flexfield structure associated with the key flexfield. This logical dimension table will have multiple instances in the business model, one per instance of the key flexfield in the related transaction views.
For example, an XXK_GL_Acct_AcctLiab (Operations Accounting Flex) dimension table will be created for the Operations Accounting Flex structure and liability account.
These dimension tables will contain just the columns from the underlying physical table that are associated with the structure in question. The logical table source for this dimension table will also have a WHERE clause defined that filters the results of the underlying physical table to just those rows that pertain to the associated key flexfield structure.
The key flexfield logical dimension tables will have multiple logical table sources defined, linking the logical table to each duplicated copy of the physical key flexfield table.
Each key flexfield dimension table will have multiple logical table sources. Each key flexfield physical table is duplicated for each transaction view it is associated with to make it possible to select different key flexfield values when joining multiple transaction views together. A logical table source is created for each duplicate physical table.
-
-
Creates logical dimension tables for each parent-child hierarchy view, when available. These logical dimension tables have the following attributes:
-
Dimension tables are created for each “Essentials” and “Hierarchies” parent-child view created for each segment in the appropriate key flexfields.
These dimension tables will contain all of the columns found in the associated physical table. The Code_Combination_Id column will be defined as the logical key for these tables.
-
The parent-child logical tables will have multiple logical table sources defined, linking the logical table to each duplicated copy of the physical parent-child table.
Parent-child physical tables will be duplicated to make it possible to select different values when joining multiple transaction views together.
-
Generates logical joins between the logical dimension tables and the logical fact table generated for the view. The resulting effect creates a star schema for each view.
Generates logical joins between the logical dimension tables of one view to the logical fact table of another view, based on the Z$ columns defined in the views. This enables report authors to create reports that join multiple views together.
Generates logical joins between the key flexfield (XXK) logical dimension tables and the transaction views they are associated with, based on the Z$ columns defined in the key flexfield and transaction views.
Generates logical joins between the parent-child logical dimension tables and the transaction views they are associated with, based on the Code_Combination_Id columns defined in the parent-child and transaction views.
-
Creates dimension objects for each dimension table generated against the transaction views. These dimension objects have the following attributes:
Each dimension object has a total logical level, set as a grand total level.
Key flexfield based dimensions have each KFF segment added in a hierarchy, the first segment as the parent and each subsequent section as a child of the segment before it.
The key flexfield segment columns from the associated dimension table are added to each associated level of the dimension object.
Non-key flexfield based dimensions have a single Detail level under the Total level. The columns from the associated dimension table are added to the single Detail level.
Logical levels are added for each associated LoV column in the underlying view, if a foreign key join exists between the transaction view and LoV view.
-
Creates dimension objects for each dimension table generated against the key flexfield (XXK) views. These dimension objects are similar to the ones generated for the transaction view-based dimension tables, with the following differences:
Dimension objects created for the multi-structure key flexfield dimension tables, XXK_GL_Acct_AcctLiab for example, will have a grand total level and a detail level. A hierarchy based on the key flexfield segment columns contained in the table will not be created because these views define multiple hierarchies that may share segments. Multiple hierarchies that share level columns are not supported in Oracle BI.
Dimension objects created for the single-structure key flexfield dimension tables, XXK_GL_Acct_AcctLiab (Operations Accounting Flex) for example, will have a grand total level and then levels describing the segment hierarchy defined for the structure. This makes it possible for report authors to create reports that provide drilling capability based on the key flexfield structure hierarchy.
-
Creates dimension objects for each dimension table generated against the parent-child views. For the “Essentials” parent-child logical tables, these dimension objects will be similar to the “Detail” dimension objects created for the transaction view-based logical tables.
For the ‘Hierarchies” parent-child logical tables, these dimension objects will be defined as ragged hierarchies based on the different hierarchy level columns defined in the associated logical dimension table.
Each logical table source linked to a dimension table has aggregate content settings configured to map the table to the lowest level of the dimension objects associated with it. For key flexfield-based dimensions, the logical table is mapped to the last segment in the hierarchy. For non-key flexfield-based dimensions, the logical table is mapped to the Detail level of the dimension object because it is the lowest level.
-
Each logical table source linked to a fact table has aggregate content settings configured to map the table to:
The lowest level of each dimension object associated with the dimension tables joined to the fact table.
The lowest level of each “Detail” dimension object associated with dimension tables that can be joined to the view using Z$ relationships found in Noetix Generator metadata layer.
-
Generates logical display folders based on the global roles with the following attributes:
Adds the logical dimension tables for views associated with the role to the appropriate folders.
Adds the logical fact tables for views associated with the role to the appropriate folders.
Adds the dimension objects for views associated with the role to the appropriate folders.
For Noetix Analytics:
Generates a business model for each module (account payables, etc.) and data source (ODS and data marts) selected for generation.
Creates a logical dimension table for each dimension table defined in the physical layer.
Generates consolidated logical dimension tables, if the dimension tables in a given relationship set result in a snowflake schema pattern. This is done to ensure that each relationship set is described as a logical star schema, based on Oracle BI’s modeling best practices. Logical table sources referencing the individual physical dimension tables are generated as well.
Creates a logical fact table for each fact table that is defined in the physical layer. These logical fact tables contain all the measures defined in the corresponding physical fact table, including measure columns for the calculated fields, if available. These measure columns are defined with a default aggregation rule, such as SUM.
Creates logical columns for columns of tables used in the expressions of the calculated fields.
Generates consolidated logical fact tables for relationship sets that include more than one fact table. This is done to ensure that each relationship set is described as a logical star schema, based on Oracle BI’s modeling best practices. Logical table sources referencing the individual physical fact tables are generated as well.
-
Breaks denormalized fact tables into separate logical dimension and fact tables. These tables will have the following attributes:
The dimension table will have the same name as the underlying physical table.
The fact table will have the same name as the underlying physical table, with a suffix of “_Fact”.
The logical dimension table will contain all of the dimension columns found in the underlying physical table.
The logical fact table will contain all of the measures found in there. The measure columns will be defined with a default aggregation rule, such as SUM.
Generates logical joins between the logical dimension tables and the logical fact tables, creating star schemas.
Creates dimension objects for each logical dimension table. These dimension objects will have grand total and detail levels. Hierarchical dimension objects are also generated for date dimension and parent-child logical tables. The dimension objects related to the date dimension include hierarchies for both Gregorian and fiscal calendars.
Date dimensions are defined as “time dimensions” with a chronological key set at the lowest level in the hierarchy.
Each logical table source linked to a logical dimension and fact tables will have aggregate content settings configured to map the table to the lowest level of the dimension objects associated with it.
Generates logical display folders based on the relationship sets defined for the business area. The dimensions, logical dimension tables and logical fact tables associated with the relationship set will be added to the appropriate logical display folders.
For both NoetixViews and Noetix Analytics:
Creates a “Noetix-Custom” logical display folder for custom logical objects added to the generated business model. See Adding Custom Objects to the Business Layer in Customizing Generated Content for more information on adding custom objects to the business model.
Presentation
The Presentation layer provides an interface to present customized views of a business model to users. Presentation Catalogs in the presentation layer are visible as business models to users of Oracle BI Presentation Services. In this layer, Noetix Generator does the following:
For NoetixViews (Standard and Cross Operations editions):
Generates presentation catalogs based on the organizational units, for example, sets of books, business groups, and so on, associated with the roles that are selected to be generated.
Each presentation catalog contains presentation tables based on the views associated with the roles that are selected to be generated.
-
The presentation tables are grouped by the role that each one belongs to, using a hidden feature of Oracle BI that lets you create the appearance of a hierarchy in Presentation Services.
An empty presentation table is created based on the name of each role.
Presentation tables based on each view are sorted under the appropriate role-based presentation table.
Presentation tables based on each view are also suffixed with the name of the role to which they belong, to ensure unique names within the presentation catalog.
The presentation tables based on each view are given a more user friendly display name based on the view name.
-
Generates presentation tables based on the logical tables in the business layer with the following attributes:
Adds columns from each dimension table to the appropriate presentation table.
Adds the aggregated columns from each fact table to the appropriate presentation table.
Sorts columns in the presentation table alphabetically, with the key flexfield based hierarchical columns at the end of the list.
Distinguishes between LoV columns and non-LoV columns using column description provided as a ToolTip in Oracle BI Answers.
Removes underscores in column names, if that option was selected by the administrator during generation.
For NoetixViews (Global edition):
Generates presentation catalogs based on the global subject areas, for example, NoetixViews for Oracle Financials and NoetixViews for Oracle Supply Chain, based on the global roles selected to be generated.
Each presentation catalog contains presentation tables based on the views associated with the global roles that were selected for generation into the presentation catalogs.
-
The presentation tables are grouped by the functional area each one belongs to, using a hidden feature of Oracle BI that allows you to create the appearance of a hierarchy in Presentation Services.
An empty presentation table is created based on the name of the functional area, for example General Ledger Views.
Presentation tables based on each transaction view are sorted under the appropriate functional area-based presentation table.
An empty presentation table is also created to group the multi-structure key flexfield views together. Presentation tables are created for each multi-structure key flexfield view and are sorted below this empty presentation table.
-
An empty presentation table is also created for each key flexfield structure. Presentation tables for each associated single-structure logical dimension table are created and sorted under this empty presentation table, grouping them together.
In addition, presentation tables based on the parent-child views are sorted in these groups. Presentation hierarchies defined against the parent-child “Hierarchies” logical tables are also included.
Presentation tables based on each transaction view are also suffixed with the name of the functional area or key flexfield structure to which they belong, to ensure unique names within the presentation catalog.
The presentation tables based on each view are given a more user friendly display name, based on the name of the underlying view.
-
The presentation tables generated have the following attributes:
Presentation columns will be created based on the columns found in each logical dimension tables associated with the transaction view.
Presentation columns will be created for each aggregated measure column found in the logical fact tables associated with the transaction views.
Columns will be sorted alphabetically, with the single structure key flexfield segment columns sorted to the bottom of the list in hierarchy order.
Removes underscores in column names, if that option was selected by the administrator during generation.
-
The presentation hierarchies will have the following attributes:
They will have the same name as the corresponding parent-child “Hierarchies” presentation table, with a suffix of “$DIM”.
They will have multiple hierarchies defined, based on new hierarchy metadata available in NoetixViews Global Extension.
For Noetix Analytics:
Generates presentation catalogs based on the business areas that were selected for generation.
Each presentation catalog contains presentation tables based on the logical dimension and fact tables associated with the business area.
-
The presentation tables are grouped by the relationship set to which they belong using a hidden feature of Oracle BI that allows you to create the appearance of a hierarchy in Presentation Services.
An empty presentation table is created based on the name of each relationship set.
Presentation tables associated with the relationship set are prefixed with a minus sign and sorted under the empty presentation table.
Presentation tables are also suffixed with the name of the relationship set to which they belong to ensure name uniqueness within the presentation catalog.
The presentation tables based on each relationship set are given a more user friendly display name based on the information group names defined in the AMDR.
Presentation columns are added for each of the logical columns in the corresponding logical tables in the business layer.
Presentation columns are added for each logical column corresponding to the calculated fields.
Separator columns are created to make the distinction between types of columns (metrics, attributes, etc.) more obvious. These columns are meant for display purposes only and are not meant to be used in any report.
How Noetix Generator Interacts with Non-Noetix Objects in the Repository
The Noetix Generator interacts with non-Noetix objects in the following manner:
Does not change objects in the Physical, Business Model and Mapping and Presentation layers that are outside the Noetix generated top level physical, business, and presentation catalogs.
Deletes custom objects that the administrator manually adds to the Noetix generated business model or physical database, unless they have been included in the “Noetix-Custom” physical and logical display folders.
Deletes any custom objects in the Noetix-generated presentation catalogs that do not conform to the naming or content requirements listed in Adding Custom Objects to the Physical Layer in Customizing Generated Content.
After the initial generation, Noetix Generator does not modify a Noetix generated connection pool in the physical layer. Only administrators can modify settings for the connection pool.
-
Noetix Generator overwrites permission and any other changes made to Noetix generated security groups, with the exception of:
Group membership; Noetix Generator retains the membership of a group after regeneration.
Permission grants to custom presentation tables and connection pools. See Adding Custom Objects to the Physical Layer in Customizing Generated Content. for more information on adding custom objects.
About Generated Report and Dashboard Templates
The Noetix Generator generates NoetixAnswers, a set of prebuilt report and dashboard templates based on NoetixViews into the Shared Folders area of the Oracle BI Presentation Catalog. These templates will be generated into a folder based on the Top Level Name specified before the generation started.
NOTE: NoetixAnswers is not yet available for Noetix Analytics.
Report Templates in Oracle BI Answers
The report templates are generated as Oracle BI Answers reports. They are organized into a folder hierarchy based on the associated organizational units (sets of books, inventory organizations, etc.) and NoetixViews roles. The report templates can be found in folders named after the NoetixViews role associated with each template. This folder hierarchy follows the organizational structure of the NoetixViews Help File.
The generated report templates will be named after a question that the template is meant to answer. Users with appropriate privileges can view report templates. They can also create their own custom reports based on the report templates.
Each report template will contain the following:
A list of the display columns defined in the underlying Noetix answer. These columns will have the appropriate sort order defined. Some columns may be based on complex expressions.
A set of filters based on the filters and parameter prompts defined in the underlying answer. Filters meant to be used as parameter prompts will be defined with Oracle BI’s is prompted operator, enabling it to be used in conjunction with a global prompt.
A pivot table view that includes subtotals and grand totals, if the underlying Noetix answer has totals defined.
WARNING: Report templates with filters that use the is prompted operator should not be executed directly in Oracle BI Answers. Reports with is prompted filters may perform poorly when executed in OBI Answers. These report templates are meant to be used in Oracle BI Dashboards where global prompts have been defined to supply values for the is prompted filters.
Using Generated Report Templates
The reports generated into Oracle BI Answers are meant to be templates to start from when seeking the answer to a business question. Some may answer the business question right out of the box, but most will have to be copied and modified to match your company’s specific reporting requirements.
If a report template does not meet the business need out of the box, a copy should be saved to the My Folders area or a folder outside of the Top Level Name folder under Shared Folders. Once copied, it can then be modified to meet the business need.
NOTE: The “Save Content with HTML Markup” privilege must be granted to Oracle BI users that need the ability to save copies of the dashboard templates created by Noetix Generator. See the Oracle BI administration documentation for instructions on how to do this.
CAUTION: The generated report templates should never be modified directly in Oracle BI Answers. Any changes made to the templates will be overridden during the next regeneration. NoetixAnswers is meant to be either run as generated, or used as a starting point to make a new report. For information about modifying how the templates are generated, see “Customizing NoetixAnswers” in Customizing Generated Content.
Prompts in Generated Report Templates
The report templates generated into Oracle BI Answers are meant to be used in conjunction with the dashboard templates generated into Oracle BI Dashboards. Many of the report templates contain filters that use Oracle BI’s is prompted operator, which offload the specification of a filter value to a global prompt.
Those global prompts are included in the dashboard page associated with the report template and will supply the values necessary to ensure that the report template is filtered appropriately.
If end users desire to use the generated report templates outside of the generated dashboard templates, then they must modify the is prompted filters to use a different operator and then specify a filter value, or remove the filters and create prompts in Oracle BI Answers instead.
WARNING: The report templates that contain is prompted filters should never be executed directly in Oracle BI Answers. Doing so may take up more resources on the Business Intelligence or database server than is desired and may produce a long running query. Many of the underlying Noetix views must have filters applied in order to be queried successfully.
Dashboard Templates in Oracle BI Dashboards
The dashboard templates are generated as Oracle BI Dashboards pages. Noetix Generator creates one dashboard for each NoetixViews role selected prior to the start of the generation.
Each dashboard contains an Overview page that lists all of the Noetix answers associated with the NoetixViews role. The Overview page displays the title and description for the answer, along with a link to view the answer.
A dashboard page is generated for each Noetix answer. These dashboard pages display the following:
A list of prompts for the dashboard. If the associated Noetix answer has parameter prompts defined, they will be displayed across the top of the dashboard page. Required prompts will specify a default value of --Select actual values--. These values should be replaced by user-selected values prior to using the dashboard.
An option that lets users switch between viewing the detailed data from the report or a list of summary tables and charts for the report. This option is only displayed when the underlying Noetix answer contains totals or aggregated columns.
A table displaying the detail data from the report. This table will list all of the dimension columns, measure columns, subtotals and grand totals found in the underlying Noetix answer.
-
A summary table and chart for each aggregated measure column, subtotal column or grand total column found in the underlying Noetix answer. The summary table will display the left-most dimension column or left-most break column associated with the aggregated measure or subtotal columns.
A chart will be displayed to the right of the summary table to show a graphical representation of the data in the summary table.
Some answers will contain additional table and chart views which are not displayed in dashboards by default. Specifically, answers with multiple aggregate measures will each have a table and chart that displays all measures. These table and chart views may be exposed as indicated in “Customizing NoetixAnswers” in Customizing Generated Content or by manually editing copies of generated dashboards.
A Back to Overview link to take the user back to the Overview page.
Using Generated Dashboard Templates
The dashboards generated into Oracle BI Dashboards are meant to be templates to start from instead of fully finished dashboards. Some may meet your company’s needs out of the box, but many may need to be copied and customized to provide all of the information required.
If a dashboard template does not meet the business need out of the box, a copy should be saved to the My Dashboard area or a folder outside of the Top Level Name folder under Shared Folders. If the underlying report template in Oracle BI Answers needs to be modified as well, a copy of the report should be created as well. Once copies of the dashboard and underlying report template have been made, they can be modified to meet the business need.
NOTE: The Change or Full Control access level should be selected in the Dashboard Access Level option in Noetix Generator to provide users with the ability to make a copy of the generated dashboard templates that they have access to.
While this will give end users access to modify the dashboard templates in place, they should not do so. Any changes to the templates directly in Oracle BI Dashboards will be overwritten during the next regeneration. NoetixAnswers is meant to be either run as generated, or used as a starting point to make a new dashboard. For information about modifying how the dashboard templates are generated, see Customizing NoetixAnswers in Customizing Generated Content.
Writing Multi-view Reports and the JOIN ASSIST Column
When creating a report from two or more views, the special JOIN ASSIST column must be included to ensure that OBI properly processes the request. This column (based on SYS.DUAL.DUMMY) assures that OBI will construct the proper joins between the views, thus ensuring the returned results are correct.
This column typically only needs to be used in reports that meet both of the following criteria:
The report uses 2 or more Noetix views.
The report does not include a measure column. Measure columns have a suffix like “_SUM” or “_COUNT”.
Each report should include only one JOIN ASSIST column. Reports with three or more views should choose the JOIN ASSIST column from the central view in the report.
This column should be included in reports as a JOIN ASSIST = ‘X’ filter. This ensures that the “Limited Values” prompt list of values option will work as expected.
NOTE: If you fail to include the JOIN ASSIST column in reports that meet the above criteria, you may receive incorrect results. On the other hand, including the JOIN ASSIST column in reports that do not meet these criteria will not affect the accuracy of the results.
Reports against 1 view
There are certain instances when it will be necessary to include the JOIN ASSIST column in reports that utilize just one view. This column should be included when key flexfield columns from the selected view are included in the report.
Reports with 2 views
When joining only two Noetix views together, you can choose the JOIN ASSIST column from either view.
Reports with 3 or more views
When writing a report with 3 or more views, you must choose at least one view that can join to all the other views in your report. You should choose the JOIN ASSIST column from this view.
Below are examples from the NoetixViews for Oracle Purchasing and Assets modules to help you identify the center view of a report from the Z$ columns. In each example, it is assumed that we have already identified our reporting requirements and that the report requires columns from all views in the example. For the sake of brevity, only the Z$ columns between the views in the example are shown.
Example 1
-
PO_Buyers
Z$PO_Buyers
-
PO_Vendors
Z$PO_Vendors
-
PO_Purchase_Orders
Z$PO_Buyers
Z$PO_Purchase_Orders
Z$PO_Vendors
From the Z$ columns, we can see that PO_Purchase_Orders has a join to both PO_Buyers and PO_Vendors. A diagram of the relationships between these views looks like this:
PO_Buyers
PO_Vendors
PO_Purchase_Orders
Since PO_Purchase_Orders is involved with joins to both PO_Buyers and PO_Vendors, choose the JOIN ASSIST column from PO_Purchase_Orders.
Example 2
-
PO_Buyers
Z$PO_Buyers
-
PO_Purchase_Orders
Z$PO_Buyers
Z$PO_Purchase_Orders
-
PO_Invoices
Z$PO_Purchase_Orders
In this example, we see that PO_Invoices has a join to PO_Purchase_Orders and PO_Purchase_Orders has a join to PO_Buyers. A diagram of the relationships between these views looks like this:
PO_Buyers
PO_Purchase_Orders
PO_Invoices
Again, since PO_Purchase_Orders is involved with joins to both PO_Buyers and PO_Invoices, choose the JOIN ASSIST column from PO_Purchase_Orders.
Example 3
-
FA_Assets
Z$FA_Assets
-
FA_Asset_Assignments
Z$FA_Assets
-
FA_Asset_Financials
Z$FA_Assets
From the Z$ columns, we can see that FA_Asset_Assignments has a join to FA_Assets, as does FA_Asset_Financials. A diagram of the relationships between these views looks like this:
FA_Assets
FA_Asset_Assignments
FA_Asset_Financials
Since FA_Assets is involved with joins to both FA_Asset_Assignments and FA_Asset_Financials, choose the JOIN ASSIST column from FA_Assets.