JPH04328680A - Data storing method - Google Patents

Abstract

PURPOSE:To store the massive quantity of data constituted of the data and the identification code into the comparatively small capacity of the storing area. CONSTITUTION:The identification code of the data constituted of the data and plural digits is made into one record, the identification code is dissolved into digits, respective digits are assigned into each node 311 of the tree structure, a digit value is stored into a list 312 designated from each node 311, the node address of the next digit is stored, and into the list 312 of the prescribed digit, out of the data elements identified by the identification code from the most significant digit up to the digit, the address of the data elements excluding the identified data elements from the digit up to the higher-order digit than the relevant digit is stored. On the other hand, the data elements of respective records, when they are the same as other records, are stored into the same address as the data element of the record.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for storing data in which a plurality of digit identification codes are attached to each data and common data is included in the plurality of data.

0002

2. Description of the Related Art FIG. 8 is a diagram showing an example of JAN code master data that compares product classification codes and product names, and classification codes are assigned to each product by each sales company.
Used for product barcodes, etc.

FIGS. 9 to 11 are conceptual diagrams showing the storage state of the above-mentioned JAN code master data according to the conventional storage method. Conventionally, data in which one record consists of a fixed-length numeric data section and a variable-length character string data section is stored in a binary tree structure. That is, as shown in FIG. 9, a pointer indicating the storage position of character string data is stored in association with the numerical data section, and branch destinations from the numerical data are stored as left and right branch pointers. The string data is shown in Figure 1.
0, they are stored at the addresses indicated by the pointers. FIG. 11 is a conceptual diagram showing the storage state of tree-structured data as described above.

0004

[Problem to be Solved by the Invention] In the conventional data storage method as described above, character string codes such as sales company names that are common to a large number of data, such as JAN code master data, and corresponding numerical data are stored. When storing data, all data for each record is stored in a different storage area, so as the number of records increases, more of the same data will be stored, reducing the efficiency of storage area usage. there were.

[0005] The present invention has been made to solve these problems, and provides a data storage method with high storage area utilization efficiency by storing common data with other records at the same address. For the purpose of providing.

[0006]

[Means for Solving the Problem] In the data storage method of the present invention, data and an identification code of the data consisting of a plurality of digits are set as one record, and some elements of the data of one record are some elements of other records. If they are the same, in a method for storing data in which the same value is set in the digit part of each identification code corresponding to the same data element, each digit value of the identification code is stored in each node of a tree structure that becomes deeper in digit order. The data elements identified by the identification code from the most significant digit to the relevant digit are stored in a list of each digit value specified by , and the node address of the next digit is stored in each list. It stores the storage addresses of data elements other than the data elements already identified in the digits higher than the relevant digit, and stores the data elements of each record as the data element of that record if the data element is the same as that of another record. It is characterized by being stored at the same address.

[0007]

[Operation] The data storage method of the present invention decomposes the identification code into digits, assigns each digit to each node of a tree structure that becomes deeper in order of digits, and stores the next digit in a list by digit value specified from each node. In addition to storing the address, data elements identified by the identification code from the most significant digit to the relevant digit are stored in a list of predetermined digits, excluding data elements that have already been identified by the digits higher than the relevant digit. The address is stored, while the data element of each record is stored at the same address of the data element of the record if the data element is the same as that of another record. When searching for data stored in the above manner, the node address of the next digit is obtained from the list specified by the node for each digit of the identification code, and the identification code is followed in digit order by a match search. If the storage address of the data element is stored, read the data element from this storage address, add it to the identified data element from the most significant digit to the relevant digit, and when all digit values have been searched for a match, Output all data elements as search results.

[0008]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be explained below based on drawings showing embodiments thereof. FIG. 1 is a block diagram showing the configuration of a data processing device that implements a data storage method according to the present invention. In the figure, reference numeral 1 denotes an input unit consisting of a keyboard for inputting data such as numerical values and characters, and various commands. The data and commands input from the input unit 1 are input to a data processing unit 2. The data processing section 2 has a numerical data section and a character string data section.
Data is read from the database 3 that stores data to be records and processed, and the processing results are transferred to C
The information is displayed on the display section 5 consisting of an RT and printed on the printer 6. The data processing unit 2 searches the database 3 and stores the search results in the search result character string buffer 4.

The database 3 includes a tree-structured data storage section 31 that stores a numerical data section as the identification code in a tree structure described later, and a string data section as the data that stores data elements common to a plurality of records. It consists of an aggregated data storage unit 32 that aggregates and stores data at the same address.

FIG. 2 is a conceptual diagram showing the data structure,
FIG. 2(a) shows the structure of one record consisting of a fixed-length numerical data section and a variable-length character string data section, with the numerical data section serving as a search key. FIG. 2(b) shows the structure of each node data 311 in the tree structure, and the data is a pointer to the first element of the bidirectional list.

FIG. 2(c) is a diagram showing the element configuration of the interactive list 312, in which each element, starting from the beginning, contains numerical data (numb), a pointer to aggregated data (comp),
These are a pointer to the previous element (prev), a pointer to the next element (next), and a pointer to the next digit of node data (nxcl).

FIG. 3 is a conceptual diagram showing a data storage state according to the data storage method of the present invention, in which numerical data is stored in ascending order in a bidirectional list 312 pointed to by each node data 311. Aggregated data 321 is stored at the address pointed to by the pointer to aggregated data in the interactive list of numerical data.

FIGS. 4 and 5 show the procedure for searching data stored by the data storage method of the present invention as described above.
The explanation will be based on the flowchart shown in . First, each digit of the search key is set to S[n] (S1). next,
Set "A" to an empty string, "i" to "0", and "N" to "root node data", that is, a pointer to the bidirectional list in the first digit of the numerical data section (S2).

The element data of the bidirectional list pointed to by "N" is set in "M" (S3). That is, “Mnu
mb” is the numerical data for one digit, “Mcomp” is the pointer to the aggregated data, “Mprev” is the pointer to the forward element, “Mnext” is the pointer to the backward element,
A pointer to the next digit of node data is set in "Mnxcl".

"S[i]" and "Mnumb" are compared, and if they are not equal, it is determined whether "Mnext" is "0" (S5). If "Mnext" is "0", the search target is not found (S7), and the process ends with an error. If "Mnext" is not "0", "Mnext" is set to "N" (S6), and the process returns to step 3 to consider the next numerical data.

[0016] If "S[i]" and "Mnumb" are equal in S4, since this is the desired numerical data, "Mco
mp” is “0” (S8).
If "comp" is not "0", aggregate data is stored, so the aggregate data character string pointed to by "Mcomp" is read out and added to the end of "A" (S9).

If "Mcomp" is "0" and no aggregated data is stored, it is determined whether "Mnxcl" is "0" (S10). “Mnxcl” is not “0”,
That is, if there is the next digit, set the node data pointed to by "Mnxcl" to "N" (S11), increment "i" (S12), return to step 3, and set the above to S[0]. ~S[12] Repeat for all digits. When the final digit is reached and “Mnxcl” becomes “0”, “
The character string "A" is output as a search result to the display section 5, printer 6, etc. (S13), and the search for one record is completed.

Next, a procedure for searching for data using "4971710007213" as a key will be explained based on conceptual diagrams of the storage state of tree-structured data shown in FIGS. 6 and 7. In the figure, 61, 64, 66, 68, 70, 77 and 80
respectively indicate node data, 62, 65, 67, 6
9, 71, 72, 74, 75, 78 and 81 indicate element data of the interactive list. Note that, for example, 68 and 69 are linearly connected data. Further, 63, 76 and 79 are aggregated data.

[0019] First, S[0] ←4, S[1] ←9,
S[2] ←7,S[3] ←1,S[4] ←7,S
[5] ←1,S[6] ←0,S[7] ←0,S[
8] ←0,S[9] ←7,S[10]←2,S[1
Set 1]←1, S[12]←3. “Mnumb
” contains the numerical data “4” for one digit, “Mcomp” contains the pointer to the aggregated data “Kose Kobayashi”, “Mprev
” sets the pointer to the forward element, “Mnext” sets the pointer to the backward element, and “Mnxcl” sets the pointer to the node data of the next digit “9”.

A tree structure that becomes deeper by one digit is followed, and as a result of a match search for "4" in the first digit, aggregated data "Kobayashi Kose" 63 is stored in the search result character string buffer 4. “i”
is incremented 9 times and the match search is completed to "497171000", the node data 70 of the 10th digit S[9] is reached, and the digit value "7" is moved to the top of the bidirectional list pointed to by the node data 70. "M" of element data 71 of
As a result of comparison with "number", it is larger, so branching is performed to the subsequent element data 72.

Next, as a result of comparing the numerical data "0" of the first element data of the two-way list pointed to by the node data 73 of the 11th digit S[10] and the digit value "2" of the 11th digit, it is found that Branching to the backward element data 75, and by matching the digit value with the numerical data "2" of the element data 75, the aggregated data "S.B.S Cake (Quick T) R" 76 is obtained, and the search result string Buffer 4's "Kose Kobayashi"
S. B. S Cake (Quick T) R'' is added. Numerical data and digit value “1” of the first element data 78 of the bidirectional list pointing to the node data 77 of S[11] in the 12th digit
As a result of comparing, they match, so the aggregated data “11
0"79 is obtained, and the search result character string buffer 4 "Kobayashi Kose S. B. Add “110” to “S Cake (Quick T) R”.

[0022] As a result of comparing the numerical data of the element data 80 of the two-way list pointed to by the node data 80 of the last digit S[12] with the digit value "3" of the last digit, they match, and the node data of the next digit is Since “Mnxcl” of the element data 81 of the two-way list pointed to is “0”, it is determined that the match search has been completed up to the last digit, and “Kobayashi Kose S.B.S Cake (Quick T) R110” is searched. Output as result. As described above, the nodes of the tree structure are traced from the root to the leaves, and when the end of the leaf is reached, the “Kobayashi Kose
．． B. S Cake (Quick T) R110” is output as a search result.

[0023] In this embodiment, a case has been described in which JAN code master data is stored, but the present invention is not limited to this, and can also be applied to, for example, postal codes.

[0024]

Effects of the Invention As described above, the data storage method of the present invention aggregates the common data parts of multiple records and stores them at a common address, so it is possible to save the storage area and reduce the storage capacity. It has the excellent effect of being able to store a huge amount of data in a storage area of .

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the configuration of an apparatus that implements the data storage method of the present invention.

FIG. 2 is a conceptual diagram showing the structure of data stored by the data storage method of the present invention.

FIG. 3 is a conceptual diagram showing a data storage state according to the data storage method of the present invention.

FIG. 4 is a flowchart showing a procedure for searching data stored by the data storage method of the present invention.

FIG. 5 is a flowchart showing a procedure for searching data stored by the data storage method of the present invention.

FIG. 6 is a conceptual diagram showing the relationship between data storage status and search procedure according to the data storage method of the present invention.

FIG. 7 is a conceptual diagram showing the relationship between data storage status and search procedure according to the data storage method of the present invention.

FIG. 8 is a diagram showing an example of JAN code master data.

FIG. 9 is a conceptual diagram showing a storage state of a numerical data section according to a conventional data storage method.

FIG. 10 is a conceptual diagram showing a storage state of a character string data section according to a conventional data storage method.

FIG. 11 is a conceptual diagram showing a data storage state according to a conventional data storage method.

[Explanation of symbols]

1 Input section 2 Data processing section 3 Database 4 Search result string buffer 31 Tree structure data storage unit 32 Aggregated data storage unit

Claims (1)

[Claims] Claim 1: If the data and the identification code of the data consisting of multiple digits are one record, and some elements of the data in one record are the same as some elements in other records, each corresponding to the same data element In a method of storing data in which the same value is set in the digit part of the identification code, each digit value of the identification code is stored in a list of digit values specified from each node of a tree structure that deepens in digit order. The node address of the next digit is also stored in each list, and among the data elements identified by the identification code from the most significant digit to the relevant digit, the data already identified by the digit higher than the relevant digit is stored in the list of predetermined digits. Data that is characterized by storing storage addresses of data elements other than Element, and storing the data element of each record at the same address as the data element of the record if the data element is the same as that of another record. Memorization method.