Browse Prior Art Database

Entity-Relationship Data Model Reporting

IP.com Disclosure Number: IPCOM000115706D
Original Publication Date: 1995-Jun-01
Included in the Prior Art Database: 2005-Mar-30
Document File: 4 page(s) / 139K

Publishing Venue

IBM

Related People

Eisenberg, NR: AUTHOR [+4]

Abstract

Entity-Relationship (ER) data is essentially organized into a network of data nodes ("entities") and connections ("relationships"). A specific request for data (a "query") from an ER repository manager organizes the data into a logical "tree", and using a "flattening algorithm," returns the tree to the caller in a series of data records. Due to the hierarchical nature of a tree, this data is "sparse", that is, only data for an individual leg of the tree is filled into each record in its appropriate position; the rest of the record is essentially blank.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 31% of the total text.

Entity-Relationship Data Model Reporting

      Entity-Relationship (ER) data is essentially organized into a
network of data nodes ("entities") and connections ("relationships").
A specific request for data (a "query") from an ER repository manager
organizes the data into a logical "tree", and using a "flattening
algorithm," returns the tree to the caller in a series of data
records.  Due to the hierarchical nature of a tree, this data is
"sparse", that is, only data for an individual leg of the tree is
filled into each record in its appropriate position; the rest of the
record is essentially blank.

      The invention involves receiving these sparse data records and
automatically producing a tabular report of the data, where each line
of the report represents data from a single leg of the tree.
Further, the user can lay out distinct free-format panels for each
leg in the tree.  This offers the user the function of condensing the
sparse (and wide) data into a more compact and visually
understandable report.  Distinct report formats for each leg in the
tree reflect the conceptually different groupings of data defined in
each leg.

      Once the ER data is converted to the Relational data model, it
can be displayed as a structured report.  The structure of the report
is oriented to the contents of an individual row or groupings of rows
in the data.  The format of this structured can be either a direct
mapping of the underlying data to a tabular report, or can be a
repeating patterned arrangement of the data intermixed with text.

      In the discussion which follows, a simple ER data example is
used for a personnel data base to illustrate the significant point.
The RM tree is shown in Fig. 1.
              LOCATION
               /   \
              /     \
             /       \
            /         \
          NAME       DEPT
          JOB        MANAGER
         SALARY

Figure 1.  RM Tree with 2 legs for simple Personnel data base.

      This tree indicates that there are two groupings of information
associated  with LOCATION, one is a set of employees, the other a set
of departments.

      The data rows are encoded.  Each actual data value is  preceded
by  a  special  indicator  field  which describes that content of the
field as one of the following:
  1.  field contains real data;
  2.  field has no assigned value ("null");
  3.  field is related to but does not occur for given leg ("no
       instance");
  4.  field is not related to given leg ("no  relationship"); this is
       the case where we say "data does not exist" for the given
column.

      Data rows are shown in Fig. 2.  For simplicity, the special
indicators are shown (in brackets) in place of the actual data.
  (location)  (name)    (job)         (salary)   (dept)  (man...