Database restructuring system for detecting functionally dependent relations and converting them into third normal form

Claim 2

A database restructuring system comprising: a relational database having at least one relation, each said relation having a plurality of tuples, a set of candidate keys, and a set of non-key attributes; extraction means for extracting a set of candidate keys and non-key attributes from each said relation of said relational database; formation means for forming a combination between a proper subset of said candidate keys extracted by said extraction means and a subset of said non-key attributes; determination means for checking, with reference to values of said tuples which are present in said relation whether said subset of non-key attributes is functionally dependent on said proper subset of candidate keys in each of said combinations, and storing a determination result as dependency information; and division means for dividing said relation into a first relation generated by a projection to an attribute set with non-key attributes having a functional dependency and a second relation generated by a projection to an attribute set without non-key attributes having a functional dependency said division means operating in accordance with said dependency information; and means for checking whether each said relation generated by said division means has at least two non-key attributes, and executing said formation means, said determination means and said division means automatically on each said relation having at least two non-key attributes.

Claim 3

A system according to claim 1, wherein said system further comprises at least one terminal, and a host computer having view means for supplying a table visible image based on relations to said terminal.

Claim 4

A system according to claim 1, further comprising means for calculating a tuple count of each of said relations of said relational database in response to an interruption signal from a CPU and using a relation having said tuple count as a relation to be restructured when the tuple count is equal to or greater than a predetermined value.

Claim 5

A system according to claim 2, wherein said system further comprises means for detecting a predetermined character of attributes of said relation of said relational database, means for informing said terminal by said view means of said detection and inputting an instruction from said terminal, and means for converting a non-normal form into a first normal form in accordance with said input instruction.

Claim 6

A system according to claim 1, wherein said extraction means has means for inputting designation of a relation to be restructured from a terminal, and said division means has means for informing said terminal of attribute information generated by said determination means, inputting an enable/disable division instruction from said terminal, and controlling division processing in accordance with said instruction.

Claim 7

A system according to claim 1, wherein said system further comprises means for converting each of at least one third normal form relations divided by said division means into a relation having a normal form at a level higher than that of said third normal form.

Claim 8

A system according to claim 1, wherein said database management system has detection means for detecting an abnormality occurring when said tuples are updated, and said system further comprises means for using, as a relation to be restructured, a relation related to a tuple in which said abnormality is detected by said abnormality detection means.

Claim 9

A system according to claim 1, wherein said system further comprises means for detecting a null value in values of said tuples and eliminating an attribute corresponding to said null value from a relation to be restructured.

Claim 10

A relational database restructuring system, having at least one relation, each said relation having a plurality of tuples, a set of attributes, a set of candidate keys and a set of non-key attributes, wherein each said tuple is managed as a two-dimensional relation for each said attribute and wherein each said relation has a candidate key serving as a set of attributes which can uniquely identify one of said tuples, comprising: first determination means for checking, with reference to values of tuples which are present in said relations, whether a first set of attributes is functionally dependent on a second set of attributes; division means for dividing a first relation into a second and a third relation by a projecting operation for a designated attribute set; second determination means for determining, from attribute sets A and B which are included in a relation and are mutually primary, whether said attribute sets A and B satisfy at least one of a first condition and a second condition, said first condition being that said attribute set A is a proper subset of said set of candidate keys and said attribute set B is a subset of a non-key attribute set serving as a subset of said set of candidate keys said second condition being that both the attribute sets A and B are proper subsets of said non-key attribute set; and means for checking whether said attribute set B is functionally dependent on said attribute set A once said second determination means has determined that said attribute sets A and B satisfy at least one of said first condition and said second condition, and dividing said relation including said attribute sets A and B, when said attribute set A is functionally dependent on attribute set B, into a relation formed by a projection to a subset of the attribute set B and, a relation formed by a projection to a union between said attribute sets A and B.

Claim 11

A relational database restructuring system, having at least one relation, each said relation having a plurality of tuples, a set of attributes, a set of candidate keys and a set of non-key attributes, wherein each said tuple is managed as a two-dimensional relation for each said attribute and wherein each said relation has a candidate key serving as a set of attributes which can uniquely identify one of said tuples, comprising: first determination means for checking, with reference to values of tuples which are present in said relations, whether a first set of attributes is functionally dependent on a second set of attributes; division means for dividing a first relation into a second and a third relation by a projecting operation for a designated attribute set; second determination means for determining, from attribute sets A and B which are included in a relation and are mutually primary, whether said attribute sets A and B satisfy at least one of a first condition and a second condition, said first condition being that said attribute set A is a proper subset of said set of candidate keys and said attribute set B is a subset of a non-key attribute set serving as a subset of said set of candidate keys said second condition being that both the attribute sets A and B are proper subsets of said non-key attribute set; and means for checking whether said attribute set B is functionally dependent on said attribute set A once said second determination means has determined that said attribute sets A and B satisfy at least one of said first condition and said second condition, and dividing said relation including said attribute sets A and B, when said attribute set A is functionally dependent on attribute set B, into a relation formed by a projection to a subset of the attribute set B and, a relation formed by a projection to a union between said attribute sets A and B.

Claim 12

A system according to claim 9, wherein said system further comprises at least one terminal and a host computer having view means for supplying a table visible image based on relations to said terminal.

Claim 13

A system according to claim 9, wherein said system further comprises means for calculating a tuple count of each relation of said relational database every predetermined time and, if the calculation result is not less than a predetermined value, causing said second determination means to operate a relation having the tuple count.

Claim 14

A system according to claim 9, wherein said system further comprises means for detecting a predetermined character of attributes of said relation of said relational database, means for informing said terminal by said view means of said detection and inputting an instruction from said terminal, means for converting a non-normal relation into first normal form in accordance with said input instruction and means for converting, when a relation of said relational database has a non-normal form, said non-normal form into a first normal form in accordance with said input instruction.

Claim 15

A system according to claim 9, wherein said division means has means for informing an operator of a result determined by said first determination means, inputting an enable/disable a division instruction from said operator, and controlling division processing in accordance with said instruction.

Claim 16

A system according to claim 9, wherein said system further comprises means for converting each of at least one third normal form relations divided by said division means into a relation having a normal form at a level higher than that of said third normal form.

Claim 17

A system according to claim 9, wherein said system further comprises means for detecting an abnormality occurring when said tuples are updated and per forming determination processing in accordance with said updating abnormality.

Claim 18

A system according to claim 9, wherein said system further comprises means for detecting a null value in values of said tuples and eliminating an attribute corresponding to said null value from relations determined by said first determination means and said second determination means.

Claim 21

A method according to claim 17, wherein the method further comprises a step for supplying a table visible image based on relations to said terminal.

Claim 22

A method according to claim 17 further comprising the step of calculating a tuple count of each said relation of said relational database in response to an interruption signal and using a relation having said tuple count as a relation to be structured, when said tuple count is greater than a predetermined value.

Claim 23

A method according to claim 18, wherein the method further comprises a step for detecting a predetermined character of attributes of said relation of said relational database, a step for informing a terminal of said detection results, a step for inputting an instruction from said terminal, a step of converting, when a relation to be restructured of said relational database has a non-normal form, said non-normal form into a first normal form in accordance with said instruction.

Claim 24

A method according to claim 17, wherein step a) comprises a step of inputting designation of a relation to be restructured from a terminal, and step d) further comprises a step of informing said terminal of attribute information generated in step c), inputting an enable/disable division instruction from said terminal, and controlling division processing in accordance with said instruction.

Claim 25

A method according to claim 17, further comprising the step of converting each of at least one third normal form relations divided in step d) into a relation having a form at a level higher than that of third normal form.

Claim 26

A method according to claim 17, wherein said database management system has detection means for detecting an abnormality occurring when tuples are updated, and said method further comprises a step of using, as a relation to be restructured, a relation related to a tuple in which said abnormality is detected in accordance with said abnormality detection performed by said detection means.

Claim 27

A method according to claim 17, wherein the method further comprises a step of detecting a null value in values of said tuples and eliminating an attribute corresponding to said null value from a relation to be restructured.

Claim 30

A method according to claim 25, wherein the method further comprises the step of supplying a table visible image based on relations to a terminal.

Claim 31

A method according to claim 25, wherein the method further comprises the steps of calculating a tuple count for each said relation of said relational database at a predetermined time and, if the calculation result is greater than a predetermined value, performing said step c) and/or said step d) on a relation having said tuple count.

Claim 32

A method according to claim 25, further comprising a step for detecting a predetermined character of attributes of said relation of said relational database,.. a step for informing a terminal of said detection results, a step for inputting an instruction from said terminal, a, step for converting, when a relation to be restructured of said relational database has a non-normal form, said non-normal form into a first normal form in accordance with the instruction.

Claim 33

A method according to claim 25, wherein step b) further comprises the step of informing an operator of a result determined in step a), inputting an enable/disable division instruction from the operator, and controlling division processing in accordance with the instruction.

Claim 34

A system according to claim 25, wherein the method further comprises the step of converting each of at least one third normal form relations divided by the step b) in the step d) into a relation having a normal form at a level higher than that of said third normal form.

Claim 35

A system according to claim 25, wherein the method further comprises the step of detecting an abnormality occurring when tuples are updated and performing the step c) in accordance with the updating abnormality.

Claim 36

A system according to claim 25, wherein the method further comprises the step of detecting a null value in values of said tuples and eliminating an attribute corresponding to said null value from a relation to be restructured.