JPH04559A - Data retrieving system - Google Patents

Abstract

PURPOSE:To accurately retrieve registered data univocally corresponding to inputted retrieving data by replacing a small index code by a substitutive code having the small number of digits and successively checking whether respective registered data corresponding to the substitutive code correspond to the small index code or not. CONSTITUTION:Registered data 2 are stored in respective lines of a data file 1. An index data part 3 is constituted of a large index code D having l1 digits and a substitutive code P having l3 digits and a data part 4 is constituted of a small index code (d) having l2 digits and real data whose number of digits is not restricted. The code D of inputted retrieving data 5 is retrieved, and when corresponding data 2 are retrieved, respective data 2 are more retrieved while increasing the value of the code P and the data 2 whose small index code (d) coincides with the inputted data 5 is decided as the data 2 univocally corresponding to the finally inputted index data. Thereby, even when the number of digits of the index data exceeds the restricted number of digits, the registered data is accurately retrieved.

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention] (Industrial Application Field) The present invention relates to a data search method for searching registered data stored in a data file using index data input from the outside.

(Prior art) For example, each book stored in a library is registered in a data file with classification numbers such as major classification and minor classification. Search for the target book by specifying a subcategory.In addition to libraries, department stores and other places keep a huge number of customer lists, each of which is classified into various categories and registered in data files. Then, search for customers belonging to the target category and send out related direct mail.

Each registered data stored in such a data file is composed of an index data section and a data section, as shown in FIG. The index data section stores index data composed of, for example, a 92-digit large index code indicating a major category and a small index code of, for example, I12 indicating a small category. In addition, the data section contains actual data consisting of the book base of each book belonging to the classification specified by each sub-index code, the author/publisher, the customer name of each customer belonging to the relevant classification, and various information regarding the relevant customer. is memorized.

Therefore, when searching for this registered data, if you search a data file that stores a large number of registered data using (j2+++N2) digit index data consisting of a large index code and a small index code, you can find the target actual data. Registration data including .

However, even in such a data search method, there are still problems to be improved as described below.

In other words, in a search system made up of computers, etc., there are certain limits to the number of digits of search data used for searches depending on the processing power of the computer, such as 16-bit configuration, 32-bit configuration, etc. There is. However, as the number of registered data stored in a data file increases, there is a strong desire to increase the number of digits in the small index code and classify it in more detail in order to read only the actual data that is needed efficiently. .

However, as described above, there is a certain limit on the total number of digits (jJ+-12) of the major index code and the minor index code, so it is not possible to perform a more detailed classification and search. Therefore, it is necessary for the user to manually classify in more detail a large number of registered data obtained by searching using the conventional method, and finally sort out the necessary registered data.

(Problem to be Solved by the Invention) As described above, in the conventional data search method, there is a certain limit on the number of digits of the index data consisting of the large index code and the small index code, so the data cannot be recorded in the data file. There was a problem that it was not possible to classify and search each registered data in more detail.

The present invention was made in view of these circumstances, and
Replace the minor index code with an alternative code that has fewer digits than this minor index code, and after each registered data corresponding to the alternative code is searched for the missing number of digits, each corresponding registered data will be changed to the minor index. By sequentially checking whether the code corresponds to the code, even if the number of digits in the index data exceeds the limit, it is possible to accurately search for registered data that uniquely corresponds to the input index data. The purpose is to provide a data search method.

[Configuration of the Invention (Means for Solving the Problems) In order to solve the above problems, the present invention provides a system in which a plurality of registered data formed of index data consisting of a large index code and a small index code and actual data is In a data search method that searches a stored data file using index data input from the outside and reads registered data that uniquely corresponds to this index data, A plurality of alternative codes formed with fewer digits than this minor index code indicating the order of appearance of each minor index code within the same major index code, and a major index code included in this index data in response to index data input. a large index code search means for searching data files using the large index code search means; an alternative code search means for sequentially searching each registered data searched by the large index code search means using sequentially increasing alternative codes; a comparison means for comparing the small index code of each registered data sequentially searched with the small index code of the index data; and a comparison means for comparing the small index code of each registered data with the small index code of the index data, and a registration data that uniquely corresponds to the index data by this comparison means. It is equipped with data and a judgment means for making a judgment.

(Function) With the data search method configured in this way, each registered data in the data file has the order of appearance within the same large index code, in addition to the large index code and small index code that make up the index data. An alternative code indicating the

Then, when index data consisting of a large index code and a small index code is input from the outside, the data file is first searched using only the large index code. Then, when each registered data corresponding to the corresponding large index code is searched, the number of alternative codes is sequentially increased, and with this alternative code,
Each of the previously searched registered data items is sequentially searched. If the sub-index code of the searched registration data and the sub-index code of the input index data match, the corresponding registration data uniquely corresponds to the input index data.

Therefore, the actual number of digits to be searched is only the sum of the major index code and the alternative code, so by increasing the number of digits in the minor index code, you can effectively increase the number of digits in the index data beyond the limit. Can be increased.

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

FIG. 1 is a diagram showing the structure of a data file in the data search method of the embodiment. In this data file 1, registered data 2 is stored in each row. Each registered data 2
is composed of an index data section 3 and a data section 4. Further, the index data section 3 is composed of a large index code D having 91 digits and an alternative code P having 93 digits, and the data section 4 consists of a small index code d having 92 digits and an actual index code D having no limit on the number of digits. It consists of data.

The number of digits g3 of the alternative code P is the number of digits 1 of the small index code d.
It is set smaller than 2. This alternative code d is composed of a number indicating the appearance order of the data section 4 that is different from the small index code d and the actual data within the same large index code D.

As shown in FIG. 2, the index data 5 inputted from the outside consists of a large index code D having 91 digits and a small index code d having 112 digits. Therefore, for one index data 5 having L digits, 1
Registered data 2 including actual data is uniquely defined.

Then, when the index data 5 of the column shown in FIG. 2 is inputted from the outside, the data file 1 is inputted according to the flow chart of FIG.
Execute search processing for .

When the flowchart starts, P (program step) 1
At this point, the inputted index data 5 is divided into a large index code D and a small index code d, as shown in FIG. Data file 1 is searched using the large index code D divided at P2.

When the corresponding major index code D is retrieved at P3, the alternative code P to be retrieved next is initialized to 1.

That is, the read line (appearance number) of each registered data 2 within the same large index code D is set to the first line. Then, at P5, the registered data 2 of the row belonging to the appearance number indicated by the alternative code P within the same large index code D is read out. In other words, the actual index code (which is the sum of the major index code D and the alternative code P)
D+P) to search for data file 1.

Then, in P6, the small index code d belonging to the data section 4 of the searched registered data 2 is compared with the small index code d of the index data 5 previously input from the outside. If the two-handed index codes d do not match in P7, it is determined that the registered data 2 corresponding to the appearance number specified by the alternative code P read this time is not the registered data specified by the input index data 5. Then, at P8, the alternative code P is increased by 1.

When it is confirmed at P9 that the alternative code P indicating the increased appearance number does not exceed the limit value PM, the process returns to P5 and the registered data 2 corresponding to the appearance number specified by the increased alternative code P is read out.

If the two-handed index codes d match at P7, it is determined at Plo that the registered data 2 that uniquely corresponds to the index data 5 input from the outside is stored in the data file 1. Then, the corresponding registration data 2 is read out.

In addition, if the large index code D is not found in the data file 1 at P3, or if the alternative code P after increasing exceeds the limit value PM at P9, the index data input from the outside at Pll is It is determined that the registered data 2 corresponding to 5 is not stored in the data file 1. Then, for example, a message saying "no corresponding data" is output.

Further, when registering the registration data 2 input from the outside into the data file 1, the registration process is executed according to the flowchart shown in FIG.

When the flowchart starts, at Ql, as shown in FIG.
The index data of the index data section 3 is taken out, and in Q2, the index data is converted into a 91-digit large index code D and Ω
It is divided into a two-digit small index code d. In Q3, data file 1 is searched using the major index code D.

In Q4, if registered data 2 corresponding to the corresponding large index code D is detected in data file 1, registered data 2 corresponding to the same large index code D has already been registered in data file 1. I judge that.

Then, in Q5, the alternative code P is set to 1, and the successive rows (occurrence number) of each registered data 2 in the same large index code D
Set to the first line. In Q6, the registered data 2 of the row belonging to the appearance number indicated by the alternative code P in the same person index code D is read. That is, the major index code D and the alternative code P
Data file 1 is searched using the actual index code (D + P) obtained by adding the above.

In Ql, if the alternative code P of the corresponding occurrence number is registered in data file 1, proceed to Q8, and enter the sub-index code d of registered data 2 corresponding to this appearance number and the registration data to be registered this time. 2 and the minor index code d. If the two-handed index codes d match in Q9, the registered data 2 having the corresponding index data 5 has already been registered in this data file 1, so equivalence processing is performed in QIO. Specifically, the actual data of the input registered data 2 is additionally registered to the actual data of the data section 4 of the already registered registered data 2. Alternatively, the actual data that has already been registered is rewritten with the actual data of registered data 2 input this time.

If the two-handed index codes d do not match in Q9, Q
The alternative code P is increased by 1 at ll.

After confirming in Q3.2 that the alternative code P indicating the increased appearance number does not exceed the limit value PM, return to Q6 and register the registered data 2 corresponding to the appearance number specified by the increased alternative code P. search for. Note that if the occurrence number specified by the increased alternative code P exceeds the limit value PM in Ql2, it is determined to be an error and an error is displayed.

If registered data 2 corresponding to the occurrence number specified by the increased alternative code P is not found in Ql, Ql
Register the registration data 2 input into the data file 1 in step 3. Specifically, the large index data D input into the index data section 3 of the registration date 2 and the alternative code P indicating the appearance number are incorporated, and furthermore, the small index code d input into the data section 4 and the actual data are incorporated. Incorporate. In this way, the registered data 2 to which the alternative code P is newly added is written into the free area of the data file.

As it is, the input registered data 2 will be stored in the row after the registered data 2 belonging to another large index code D, so in Ql4, the registered data 2 input this time will be stored in the row after the registered data 2 belonging to the same large index code D. The storage position of each registered data 2 in the data file 1 is arranged so that it is located at the end of the registered data 2 belonging to the data file 1.

If the corresponding large index food D is not found in the data file 1 in Q4, the alternative code P indicating the appearance number is set to an initial value of 1 in Q15. Then, the process advances to QlB and the corresponding registration data 2 is registered in the data file 1.

According to the data search method configured in this way, when searching for the required registered data 2 from the data file 1, first, the large index data D of the input index data 5 is searched.
When each corresponding registered data 2 is retrieved, each registered data 2 is further searched while increasing the value of the alternative code P, and finally the registered data 2 with the matching sub-index code d is searched. It is determined that the registered data 2 uniquely corresponds to the index data input in .

In this way, the small index code d forming part of the index data 5 of each registered data 2 is registered in the data section 4 instead of the index data section 3 in the data file 1. The actual actual index code when searching this data file 1 is the sum of the major index code D and the alternative code P, as shown in FIG. Then, the alternative code P is each registered data 2 belonging to the same large index code D.
The number g3 of digits required by this alternative code P can be set smaller than the number 92 digits required by the small index code d.

Therefore, even if the number of digits of the index data 5, which is the sum of the major index code D and the minor index code d, exceeds the limit number of digits determined by the processing capacity of the computer, the minor index code d is not stored in the data file 1. By storing the index code in the data section 4 of the index code and treating it as a type of data, it is possible to reduce the number of digits of the actual index code used in the actual search. Moreover, even if the number of digits of the actual index data is reduced, the registered data 2 that uniquely corresponds to the input original index data can be reliably searched.

Therefore, each registered data 2 can be further classified and registered in the data file 1, and the user can search by specifying the required classification of the registered data 2 more precisely.

Further, since the values of the alternative code P used for the search and the alternative code P registered in the data file 1 are automatically set, there is no need for the user to be equally involved. Therefore, the user can search for the required registered data using the same operation as the conventional search operation.

[Effects of the Invention As explained above, according to the data search method of the present invention,
The minor index code is replaced with an alternative code having fewer digits than this minor index code, and after each registered data corresponding to the alternative code is searched, the corresponding registered data is searched for the number of digits corresponding to the missing number of digits, and then the corresponding registered data is changed to the minor index in order. We are checking whether the code is compatible or not. Therefore, even if the number of digits of the index data exceeds the limited number of digits, it is possible to accurately search for registered data that uniquely corresponds to the input index data. As a result, users can search by specifying more precisely the classification of the registered data they need.

[Brief explanation of the drawing]

1 to 5 show a data search method according to an embodiment of the present invention. FIG. 1 shows the structure of a data file, and FIGS. 2 and 5 show index data and each index. FIG. 3 is a flowchart showing the operation when searching for registered data from a data file, FIG. 4 is a flowchart showing the operation when registering registered data in the data file, and FIG. The figure shows the structure of general registration data. 1...Data file, 2...Registered data, 3...
- Index data section, 4... Data section, 5... Index data 5, D... Large index code, d... Small index code,
P...Alternative code. 1st Ward, 2nd Figure Applicant's Representative Patent Attorney Takehiko Suzue

Claims (1)

[Claims] A data file in which a plurality of registered data formed of index data consisting of a large index code and a small index code and actual data is stored is searched using index data input from the outside, In a data search method for reading registered data that uniquely corresponds to this index data, this small index is attached to each registered data in the data file and indicates the order in which each of the minor index codes appears within the same major index code. a plurality of alternative codes formed with fewer digits than the index code; a large index code search means for searching the data file using the large index code included in the index data in response to the input of the index data; Alternative code search means for sequentially searching each registered data searched by the index code search means with sequentially increasing alternative codes, and a sub-index code of each registered data sequentially searched by the alternative code search means and the index. It is characterized by comprising a comparison means for comparing the data with a minor index code, and a determination means for determining, by the comparison means, registered data whose both minor index codes match as registered data that uniquely corresponds to the index data. Data retrieval method.