JPH02138619A - Data retrieving and sorting device for electronic equipment - Google Patents

Abstract

PURPOSE:To shorten the time necessary for sorting by dividing a data storing means into plural blocks and executing the retrieving and sorting with a block unit. CONSTITUTION:When input data is given a data retrieving means C executes a retrieval at every block in order to retrieve data corresponding to this from a data storing means A. That is, the data sorting means B executes the sorting by a unit of a block in accordance with a prescribed rule with respect to data in some block (for instance, a block 2) instead of executing the sorting to all data brought to entry in the data storage means A. Consequently, in accordance with the number of divisions of the block, the time necessary for sorting is shortened.

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Field of Application] The present invention relates to a data search group sorting device for electronic equipment.

[Prior Art] As a conventional data search method, a method of sequentially searching files is known, for example, a successive approximation method as disclosed in Japanese Patent Application Laid-open No. 59-106072. However, in such a system, when the amount of data increases, the search speed slows down and the efficiency deteriorates significantly. Therefore, methods using the hashish method and the two-branch method have been realized.

[Problem to be solved by the invention] As a specific example of using these methods, simple PO3 (
JA N (point of sale information management system), etc.
Japanese Article Number)7
Consider searching and sorting files. In the hash method, as the proportion of valid data in the total file capacity increases, the search operation becomes similar to the successive approximation method.In particular, when the proportion of valid data is 80%
When this level is exceeded, search efficiency becomes significantly worse. Therefore, even if a simple POS is equipped with a memory for 4,000 items, only 3,200 of them can be used as an effective area, that is, only 3,200 items can be entered.

In addition, one of the drawbacks of the hash method is that there are no search results
It is difficult to determine the number of searches for data up to
On the other hand, when using the two-branch method, we are not satisfied with the drawbacks of the hash method, such as the inability to use memory effectively and the difficulty in processing %gj Found. be done. However, on the other hand, in order to perform a search using the two-branch method, it is necessary to sort (classify) the search files in advance according to certain rules (such as numerical order). This sorting is a very time-consuming process, and if it is to be performed for each JAN file setting, for example, the operator will have to wait for a very long time and will be unable to perform other tasks.

The reason for this is that file data must be sorted in order to be able to perform a search using a two-branch method or the like.

In light of this, since it is necessary to sort the file data in order to perform a search using the two-branch method, it would seem that the processing time could be shortened by devising a sorting method.

An object of this invention is to shorten the sorting time.

[Means to solve the problem] The means of this invention are as follows.

Data storage means A (see the functional block diagram of FIG. 1, the same applies hereinafter) is divided into a plurality of blocks, and stores sales data in association with, for example, JAN codes. Then, the data sorting means B sorts the data in the data storage means A in units of blocks, and the data search means C searches for data corresponding to the input data for the sorted data storage means A in units of blocks. .

[Work] The operation of the means of this invention is as follows.

Data search means C is input data (for example, JAN code)
is given, in order to search for the corresponding data from the data storage means A, the search is performed block by block. For example, JAN codes for 1000 items can be recognized in each block, and the data search means C first searches for the corresponding data from the data storage means A. A search is made to see if there is a match in the first block l with the input JAN code. If not, the search within the block is performed by repeating the operation of searching within the next block, for example, using a two-branch method. Data sorting means B is provided to enable this type of intra-block search. In other words, the data sorting means B sorts the data in a certain block (for example, block 2) according to a certain rule on a block-by-block basis instead of sorting all the data entered in the data storage means A. The time required for sorting is reduced depending on the number of divisions. On the other hand, the number of comparisons during a search (the average number of comparisons) increases slightly as the number of blocks increases, but this is not a problem in terms of practical time.

Therefore, sorting time can be shortened.

[Example] Hereinafter, an example will be described in detail with reference to FIGS. 2 to 4.

Figure 2 shows the overall MIJ power diagram when applied to ECR or PO5. Reference numeral 1 denotes a total memory for storing sales data, and within the FLU total memory there is a JAN file used in this embodiment. This JAN file is here divided into four blocks, each block is 1oo
This section stores data for o items. Reference numeral 2 denotes a CPU that controls the entire device, its calculation section 2a executes various calculations, its control section 2b executes various controls, and its transfer section 2c performs data transfer between peripheral devices, etc. . Furthermore, in this embodiment, the CPU 2 uses four block pointers 2d and 2e for specifying each block of the JAN file for searching and sorting.
one array bundle pointer or empty flag, i.e., empty flags 1 to 4 (respectively) supporting an empty position in the corresponding block
Use 2e). Reference numeral 3 denotes a barcode reading unit that reads a product code such as a JAN code from a barcode, and the read JAN code is sent to the CPU 2 for retrieval.

Reference numeral 4 denotes a key input section, which includes various keys such as a mode selection key 4a, an IL number key 4b, an amount setting key 4c, a current/deposit key 4d, and other function keys 40. 5 is a money drawer, 6 is a display section for displaying various data, and 7 is a printing section for printing receipts and the like.

Regarding the operation of the embodiment configured as above, the JAN code and related data (product character, unit price) are
The setting operation for setting files and the registration operation for registering sales data will be explained separately.

9. When the constant operation mode selection key 4a is set to the setting position, the CPU
2 enters the setting process flow shown in FIG. First, the CPU
2 waits for the JAN code to be input in step si, and the operator reads the JAN code scanned by the barcode reading section 3.Next, the CPU 2 initializes the block pointer of the JAN file to "l". In the following step S3, the CPU 2 checks the empty flag of the block indicated by the block pointer to determine whether the designated block is in a full state. When the block is in the full state, the process advances to step S4, where a JAN code that matches the JAN code input in that block is searched using a two-branch method. In other words, read the data at the 500th item position (in the center of the block) of that block and use it as the target JAN.
Compare the relationship between the code and the target JAN.
If the value is smaller than the code, the first half of the 500-item array is further divided into two, and if it is larger, the second half of the 500-item array is further divided into two, and the data at the divided position is compared again with the target JAN code. , continue searching (J
(Assume that AN codes are classified in order from smallest to largest), and if the search is successful in step S5, then
Since the JAN code has been set and the character and unit price are to be changed, take that address and proceed to step S17.
Enter the product character and setting flow below. Step S
If the search cannot be performed in Step 5, the block pointer is updated in Step S6, and if the number exceeds 4 blocks (if the determination result in Step S7 is YES), the process goes to Step S8, and if it is within 4 blocks, the above procedure is performed. Step S of
Return to 3.

When it is detected in step S3 that the block indicated by the pointer is not in a full state, the process proceeds to step S9, and the array of 500 items in the first half of the block is searched using a two-branch method. If the search fails in step S10,
In step 311, it is checked whether up to 500 items have been set in this block by referring to the empty flag (the contents of which indicate the position in the array of the final setting item) corresponding to this block. In this case, step Sll
If 500 items or less are checked, create a new JAN.
In some cases, a code has been set, so proceed to step S14, which will be described later.Also, if 500 items or more is checked in step Sll, in step S12, the array for the latter 500 items of that block is created using the two-branch method. search for.

If the search is not successful in step 313, the process proceeds to step S14, and if the search is successful, the process proceeds to step S17.In addition, in step S14, the empty flag of the designated block is incremented by 1, and in step 315, the empty block indicated by the empty flag is Set the input JAN code in the position, stay, and press S1.
6, the block is stored as a block for quick sort to be performed later. Next, the CPU 2 proceeds to step 317 and waits for the input of a product character. If there is an input, the character is set in the corresponding record (the character setting location of the current JAN code) in step S18, and the unit price is input in the subsequent step S19. If there is an input, step S
In step 20, set the unit price for the corresponding record. Thus, while no end operation is detected in step S21, step S
Repeat the operation from l. When the termination operation is performed,
Proceeding to step S22, the process results in the previous step 316 are referred to and the quick sort block is quickly sorted, and in the final step S23, a receipt is issued and the JAN file setting flow is exited.

Note that although only insertion processing is performed in step S15 in FIG. 3, sorting may be performed here (hereinafter referred to as step 315'). In this case, the array to be sorted will be half of the block ( When the result in step Sll is No, the first 500 items of the specified block are sorted, or when the result is NO in step S13, the latter 500 items of the specified block are sorted.)
, the sorting time is further reduced.

In this case, when sorting within a block is to be performed later (supposed to be performed after step S20), one end of the array of the half of the block sorted in step 515' and the array of the other half of the block Compare the data at 500th and 501st)
, if the sorting conditions are satisfied, the sorting within the block is completed at that point. An arrangement of half blocks can also be considered a type of block, with the original block being the first division block and the subsequent block being the second division block (subblock).

A subblock need not be half of the original block.

Describe the case where sales data is registered next. When the operator scans the barcode of the sold product with the eight-coat reader 3 during a transaction with a customer, step TI in FIG. The input of the JAN code is detected in step T2, and the block pointer of the JAN file is initialized.Next, in step T3, the block pointed to by the pointer is
A search is performed using the branch force formula, and if the search is not successful, the process proceeds from step T4 to step T5, where the block pointer points to the next block, and when all blocks are completed at step T6, error handling (step S7) is performed. If not, return to step T3 and search in the next block. If the desired item has been found, proceed to step T8 and perform normal registration processing (updating the number of sales of the searched product, etc.) In the worst case,
The number of search comparisons is 40 (the total number of items is 4000)
Since the number of blocks is 4 and each block is searched for 2 minutes, the average number of comparisons is 20).
The search is completed within a sufficiently short time for practical purposes.

Note that if the range of values of items that can be entered in each block is determined, specifically, if a range designation pointer is used for each block, the search time is substantially the same as in the case without blocks. In other words, in addition to the number of searches (number of comparisons) in l block, the number of times to search to which block the target item belongs by referring to the range specification pointer (in the case of 4 blocks, the result can be obtained in 4 times at the worst) ) is added.

[Effects of the Invention] According to the present invention, since the data storage means is divided into a plurality of blocks and searching and sorting are performed in block units, the long time required for sorting can be significantly shortened. Therefore, in an electronic device having a limited memory capacity, it is possible to effectively utilize the memory capacity and improve response to the operator.

[Brief explanation of the drawing]

FIG. 1 is a functional block diagram of the present invention, FIG. 2 is a configuration diagram of an embodiment, FIG. 3 is a flowchart at the time of setting in the embodiment, and FIG. 4 is a flowchart at the time of registration in the embodiment. 1...Total memory, 2...CPU, 2
b...control unit, 2d...pointer, 3...barcode reading unit. Patent applicant Casio Computer Co., Ltd. Figure

Claims (1)

[Claims] A data storage means divided into a plurality of blocks, a sorting means for sorting data in the data storage means in block units, and a data search for searching data corresponding to input data in the data storage means in block units. 1. A data search/sorting device for electronic equipment, characterized in that it has means.