Browse Prior Art Database

Inter and Intra Related Organized Multiple File Technique

IP.com Disclosure Number: IPCOM000074673D
Original Publication Date: 1971-May-01
Included in the Prior Art Database: 2005-Feb-23
Document File: 3 page(s) / 47K

Publishing Venue

IBM

Related People

Moore, AC: AUTHOR

Abstract

Figure 1 shows a typical file, as it appears to the user, and Fig. 2, an organization of such a file which facilitates rapid search and retrieval, with provisions for updating as required. The organization shown in Fig. 2 is an example of the "inverted file" organization. The notation used to describe that organization is the IDS notation. (IDS stands for Integrated Data Store, a product of the General Electric Company). In the IDS notation, the rectangles stand for records; the directed arrows stand for closed chains. In Fig. 2, there is a single record, the Unique Master, which is the master of a chain "A". On chain A there is a set of "File Name" records containing the names of all files in the data base.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 52% of the total text.

Page 1 of 3

Inter and Intra Related Organized Multiple File Technique

Figure 1 shows a typical file, as it appears to the user, and Fig. 2, an organization of such a file which facilitates rapid search and retrieval, with provisions for updating as required. The organization shown in Fig. 2 is an example of the "inverted file" organization. The notation used to describe that organization is the IDS notation. (IDS stands for Integrated Data Store, a product of the General Electric Company). In the IDS notation, the rectangles stand for records; the directed arrows stand for closed chains. In Fig. 2, there is a single record, the Unique Master, which is the master of a chain "A". On chain A there is a set of "File Name" records containing the names of all files in the data base. The Unique Master record points to the first File Name record on chain A; the first File Name record points to the second; the second to the third; and so on. The last File Name record points to the Unique Master record, thus forming a closed chain. Similarly, each File Name record on chain A is the master of a chain B whose records contain the Attribute names applicable to the file named in the B chain master. Each Attribute Name record is the master of a chain C, whose detail records contain the unique values of the attributes in the file named by the B chain master. Each unique value record is the master of a chain D, whose detail records contain the RID's (Record Identifier) and AID's (Attribute Identifier), i.e., the row and column coordinates, of each occurrence of that value in the file. Chain D links all records having a common RID. Finally, each Attribute Name record is also the master of a chain E, which links all of the chain D details having a common AID.

In Fig. 2 (dashed lines) and Fig. 3, the "inverted file" organization is modified and generalized so that both intrafile and interfile relationships may be represented by a single file organization.

The dashed lines of Fig. 2 show the modifications required to establish intrafile relationships, of indefinite number, among the rows of a single file. The basic "inverted file" organization is left undisturbed, i.e., chains A through E. Each File Name record is the master of a chain F, whose detail records contain the record identifiers used by the file (RID). Each F chain detail record is the master of a G chain which links all of the details on the D chain which have the RID of the G chain master; thus, the G chain links together the fields in a row of the "typical" file. Now, if there are no intrafile relationships among the rows of the "typical" file, chains H through M are not required. If there are one or more relationships defined on the rows of the file, chains H through M provide for sorting, searching, retrieving and updating an arbitrary number of these relationships. Each File Name record is the master of an H chain which links records containing the names of the relationships. Each relation name re...