This function is used to describe the facts table which are to be used as base for the reports of the driven version of the software. A facts table is a data table loaded by one or several tables coming from one or several live folders and present in a datamart. It integrates:

  • fields (which can be used in the management reports)
  • links (to dimension tables used as analysis criteria)
  • aggregates (in order to store the totaled amounts on criteria combinations for optimization purposes)
  • indexes (also for optimization purposes).

The facts tables are not created in the folder itself but in a data structure called datawarehouse, which may be common to several folders.

The facts tables are loaded via a function integrated to the software (it is thus not necessary to use an external software of type ETL). Its parameters are defined by a dedicated function.

ABITUNNEL entry point

The ABITUNNEL entry point is used to modify the value of the OKK variable.

If [L]OKK = 0 then no tunnel can be defined on the dimension.

In any other case, it is possible to define a tunnel on the dimension.

Prerequisites

SEEREFERTTO Refer to documentation Implementation

Screen management

Header

Is used to define the code and the title of the table but also an abbreviation which will be used to access the facts table via the update processings.

Tab Description

General information on the facts tables can be found here, notably on:

  • the authorization conditions by site, if a field of this type is present. In the description of each report based on the corresponding data, it is possible to define if the restrictions by site must be applied to the users by specifying what is the function whose authorizations by site are under consideration.
  • A date that ca be used for the purge of fact table data.
  • A type of update used by the loading functions.

Tab Fields

Here is the detailed definition of the fields which can be:

  • dimensions (that is analysis criteria which are as a general rule linked to an annex table).
  • measures (values that can be displayed in detail or aggregated according to the dimension values).
  • information (additional information fields).

This function is used to add to the field grid fields coming from a table entered previously in the header of the window being opened from the current line. If the facts table has the name of one the tables of the software, this name is proposed by default.

Default values are applied to the grid containing the lines that are thus recovered:

  • The main characteristics of the field are recovered as is (title, type, length, local menu, miscellaneous table, etc.)
  • Once the field is characterized by a type linked to an object (or if it is of the type Date or Miscellaneous table, it is defined as a Dimension.
  • If it is of type DCB, it is defined as a Measurement.
  • Otherwise, it is defined as an Information.

If fields having the same title as a field to insert already exist, they are not inserted.

Tab Links

This tab is used to define the various links to the dimension tables managed by the software.

A link is characterized by the code of the dimension table and by a link expression. When there are several links to a same dimension table, it is necessary to give a distinct alias number.

Note that the automatic links do not need to be described here. Automatic links are the ones generated for the following fields:

  • The fields of type date (D), which have an implicit join with an alias of the AX3DAT table.
  • The fields of type Local menu (M), which have an implicit join with an alias of the AX3MNLxxx table.
  • The fields of type Miscellaneous table (ADI), which have an implicit join with an alias of the AX3ADIyyy table.
  • The fields of type Folder code (ADS), which have an implicit join with an alias of AX3DOS.
  • The fields of type Access code (ACS), which have an implicit join with an alias of AX3ACC.

Tab Index

This type is used to define the indexes which are created in the facts table in order to optimize the access times.

Tab Aggregates

In this tab the aggregates to be managed are to be defined. An aggregate is a table storing the intermediate totals based on the various values that a field or combination of fields can take. The advantage of these aggregates is that they help optimize the access times when a report concerns only the measures totaled on one of the aggregates: the access is direct since the data are not recalculated.

In this tab, one or several aggregates can be defined:

  • a line of the upper grid represents an aggregate
  • when the cursor is positioned on a grid line, the bottom grids (Dimensions and Index) are loaded with the characteristics of each aggregate.

The table Dimensions displays the list of dimensions whose crossing defines an aggregate. These dimensions can have a predefined type (date, local menu, miscellaneous table), or have type Other; this type is used to enter one of the dimensions listed in the corresponding tab.

Note that for each aggregate defined in the facts table:

  • the site dimension is mandatory if an authorization by site is managed in the facts table.
  • the dimensions having an access code are also mandatory.

The table Index is used to define the indexes in order to make faster the access to the aggregate data.

Specific Buttons

Validation

Is used to create the processing which loads the created aggregates.

The name of the processing is WMF followed by the facts table code. The labels OUVRE, FERME, INSERT, ERASE, ALIM_i (i=aggregate number) can be found, which are called by the processings allowing the data update.