Browse Prior Art Database

Universal Data Record

IP.com Disclosure Number: IPCOM000083588D
Original Publication Date: 1975-Jun-01
Included in the Prior Art Database: 2005-Mar-01
Document File: 6 page(s) / 48K

Publishing Venue

IBM

Related People

Gorrick, JJ: AUTHOR [+2]

Abstract

A data collection and retrieval system is described which allows data to be processed without knowing before hand what data, data format, or type of data is to be processed.

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 27% of the total text.

Page 1 of 6

Universal Data Record

A data collection and retrieval system is described which allows data to be processed without knowing before hand what data, data format, or type of data is to be processed.

Data to be collected and retrieved will be self-describing data. Data will be initially gathered by data dependent code and formatted into a data item as follows:

(Image Omitted)

Where: Data item code is a value in the range X'0001' to X'7FFF'. This value is arbitrarily assigned and is used in determining how the data should be formatted. It is used as an index into the Definition Pointer Table which is later discussed. Length is a value in the range X'00000000' to X'FFFFFFFF'. This is the length of the data. Data group is the data being gathered.

This data item is then passed as input to a record creation routine. This record creation routine is data independent and table driven and has two tables defined for it, a Definition Pointer Table (DPT) and a Definition Table (DT).

The DPT is a variable-length table containing from 1 to 32,767 four-byte entries. Each entry in the DPT points to an entry in the DT. The data item code contained in a data item is used as an index into the DPT. Each entry in the DT, called a DTE, defines how a data item is to be formatted. Each DTE has the following format:

(Image Omitted)

where the parentheses are metalingual symbols used to indicate a grouping and: Data group code is a two-byte field whose value is in the range X'0000'- X'7FFF'. This value is arbitrarily assigned to represent a data group which is any piece of data one byte or greater in length. The value `0000' is reserved to indicate a null data group. The values from X'0001' - X'7FFF' are used as indices into a Data Group Name Pointer Table which is later discussed. The value X'FFFF' is reserved to be used as an escape, i.e., a technique to invoke a presently undefined algorithm which will be used when the data group code must exceed the X'7FFF' limit. Length1 is a four-byte length field in the range X'0000000' - X'FFFFFFFF', and it represents the total length (in bytes) of the information following it. Keyword code is a two-byte value in the range X'0000' - X'7FFF'. This value is arbitrarily assigned to represent a data group field in a data group. The value X'0000' is reserved to indicate unencoded data, i.e., data which is treated as hexadecimal data. The values from X'0001' - X'7FFF' are used as indices into a Keyword Name Pointer Table which is later discussed. The value X'FFFF' is reserved to invoke a presently undefined algorithm which will be used when the keyword code must exceed the X'7FFF' limit. Length2 is a two-byte length field in the range X'0000' - X'7FFF' and it represents the length (in bytes) of the data following it. Data is a definition of a field within a data group. Its format is as follows:

1

Page 2 of 6

(Image Omitted)

Where: Byte offset is a four-byte offset into a data group and is used in locating the first byte assoc...