JPH02239370A - Data retrieve system - Google Patents

Abstract

PURPOSE:To reduce the number of times of record reading at the time of reading a column value by making a flag indicating the soleness of a component name into a column as a retrieve key, and directly retrieving the flag as one of the keys at the time of retrieving. CONSTITUTION:An attribute column 1 is the combined column of an attribute value column 2 and a duplicate flag 3, and >= one column 1 exists in one record. Whereas, the column 1 without the duplicate flag 3 exists as well. When many data having the same attribute value exist in the attribute value column 2, the duplicate flag 3 sets the flag to indicate the existence of duplication at the only one duplicate flag 3 in the data group. Further the flag 3 sets the non-existence of the duplication at the all corresponding duplicate flags in the data group at the time of registering the data. As a result, when the data group corresponding to the column value are retrieved, the number of times of record reading can be suppressed to a minimum when the column value necessary for the generation of the retrieve conditions of answer aggregation is read from the condition aggregation.

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention] (Field of Industrial Application) The present invention relates to a data search method for efficiently searching for desired data in a database.

(Prior art) In this type of conventional technology, parts with the same name are managed as separate records using a search key such as a record number (hereinafter referred to as a unique key) that can be uniquely identified in the system. Therefore, when searching for attribute elements (hereinafter referred to as columns) that constitute records other than unique keys in a data group searched with arbitrary search conditions, the data group is read out one by one and the columns are searched. I needed to check for duplicate values.

(Problems to be Solved by the Invention) As described above, in the conventional technology, there is a problem in that reading out duplicate records requires a lot of search time.

The present invention has been made in view of the above, and its purpose is to columnize a flag indicating the uniqueness of a part name as a search key, and configure it so that it can be directly searched as one of the keys at the time of search. An object of the present invention is to provide a data search method that reduces the number of times records are read.

[Structure of the invention]

(Means for Solving the Problems) In order to solve the above problems, the data search method of the present invention is as follows:
A data search method that searches for child information whose record format, which is a combination of items, is different from that of the child information for records that store child information that constitutes multiple items, and where duplication indicates duplication of identification information that identifies child information. When registering child information by setting a flag as a search key, if the identification information is the same in multiple child information of the same record format, only the duplicate flag of any one child information is set to be duplicated, When searching for child information by setting the duplication flag of child information of the same identification other than the child information to "no duplication", a condition in which the setting of "duplication exists" to the duplication flag is added to the given search condition by logical AND. The purpose of the search is to perform a search using .

(Operation) In the data search method of the present invention, when multiple pieces of child information in the same record format have the same identification information, only the duplication flag of any one child information is set to “duplicate” and the child information is searched. , the search is performed using a condition in which the duplicate flag's duplicate setting is added to the logical product.

(Example) Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a diagram showing part of a data structure used in a data search method according to an embodiment of the present invention. In the figure, 1 is an attribute column, 2 is an attribute value column, and 3 is a duplication flag. Attribute column 1 is a composite column that is a combination of attribute value column 2 and duplication flag 3, and one or more exists in one record, but there are also columns that do not have a duplication flag. Duplicate flag 3 is an attribute value. If there is a large number of data with the same attribute value in the RAM 2, the only duplicate flag 3 of the data group is set to "Duplicate", and the corresponding duplicate of the data group is set. "No duplication" is set for all flags 3 at the time of data registration.

An example of specifying a search condition to check for duplicates can be written as follows.

Here, 4 is a search condition for performing a search using the attribute value column 2, and 5 is a search condition that indicates whether there is a duplicate test. A conditional search using only attribute value column 2 can be performed by combining search condition 4 with a logical expression.

Attribute 1 having a duplicate flag can specify the presence or absence of a test using search condition 5, which specifies the presence or absence of a duplicate test. If duplicates are to be excluded, search condition 5 is designated as "checked"; if duplicates are allowed, search condition 5 is designated as "no search" or search condition 5 is not designated.

FIG. 2 is a flowchart showing a search operation using a duplicate flag in a data search method according to an embodiment of the present invention. In the figure, first, a search is performed using a logical expression composed of only the search condition 4 based on the value of the attribute value input column 2 as a search condition, and a corresponding data group 11 is generated (step 1
10). Copy this data group 11 to data group 12 (
Step 120). Then, the presence or absence of a duplicate check is checked using the search condition 5 for the attribute power column 1 having the duplicate flag (step 130). For items that require duplicate inspection, corresponding data 13 is generated using the logical product of search conditions 4 and 5 as a search condition (step 1).
40). Next, take the AND of data group 12 and data group 13 and replace it with data group 12 (step 150)
. It is determined whether the duplicate inspection has been performed for all of the attribute power RAMs 1 having the duplicate flag, and if the duplicate inspection has not been completed, the process returns to step 130 for determining whether or not there is a duplicate inspection (step 16
0).

〔Effect of the invention〕

As explained above, according to the present invention, when identification information is the same in multiple pieces of child information in the same record format, only the duplication flag of any one child information is set to "duplicate", and the child information is searched. In this case, since the search is performed using the condition in which the duplicate flag's duplicate setting is added to the logical AND, it is possible to generate a data group (condition set) in which there is only one matching record for the value of a specific power ram in the search condition. , when searching for a data group (answer set) corresponding to the column value, it is possible to minimize the number of record reads when reading the column values necessary for generating the search conditions of the answer set from the condition set.

[Brief explanation of drawings]

FIG. 1 is a diagram showing a part of the data structure used in a data search method according to an embodiment of the present invention, and FIG. 2 is a flowchart showing the operation of the data search method. 1・●・Attribute power ram, 2...Attribute value power ram, 3...Duplicate flag. Agent Patent Attorney Hidekazu Miyoshi Figure 1 Figure 2

Claims (3)

[Claims] (1) A data search method that searches for child information whose record format, which is a combination of items, is different from that of the child information for a record that stores child information that constitutes multiple items, and where identification information that identifies the child information overlaps. When registering child information, if the identification information is the same in multiple child information in the same record format, only the duplicate flag for any one child information can be set as a search key. Set,
When searching for child information by setting the duplication flag of child information of the same identification other than the child information to "no duplication", a condition in which the setting of "duplication exists" to the duplication flag is added to the given search condition by logical AND. A data search method characterized by performing a search using . (2) A record that matches the search condition is a set of child information that is a set of child information for which the duplicate flag is set to “duplicate” and a set of child information that falls under the search condition where the duplicate flag is not included in the condition. 2. The data retrieval method according to claim 1, wherein the data retrieval method is generated as a set of data retrieval methods. (3) As a set of child information that corresponds to the search conditions,
3. The data search method according to claim 1, wherein one of two sets of child information can be selectively generated.