JPH09330333A - Automatic verifying method for inspection result - Google Patents

Abstract

PROBLEM TO BE SOLVED: To inspect whether or not a retrieved document meets retrieval conditions and whether or not documents in a data base, which meet the retrieval conditions, are all retrieved, by placing an inspected system actually in operation and obtaining a retrieval results, comparing the retrieval result with its predicted value, and outputting the comparison result. SOLUTION: A document generating program 13 generates more than one kind of new documents from a ciphered basic document and an amplification table 11 and stores them in the data base 19. A retrieval condition generating program 14 generates retrieval conditions by combining amplification tables 20 and 21 with each other. A retrieval result predicted value calculating program 15 calculates a predicted value for the retrieval result corresponding to the retrieval conditions generated by the retrieval condition generating program 14 to generate a retrieval result predicted value table. A retrieval result comparing program 16 inputs the retrieval conditions of a retrieval result predicted value table to a full-text retrieval system 17, compares an acquired retrieval result with the retrieval result predicted value table, and outputs the actual retrieval result to comparison result information 18.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method of verifying an inspection result in software inspection, and more particularly to a method of verifying an inspection result in inspection of a full-text search system.

[0002]

2. Description of the Related Art In developing a full-text search system for searching a document stored in a database for a document that matches a search condition, it is necessary to check whether or not the full-text search system operates normally. In the full-text search system, the documents stored in the database are numbered, and when a search condition is input to the full-text search system, the documents stored in the database are searched and the document number matching the search condition is returned. There is no known method for checking this full-text search system. As an undisclosed method that is not publicly known, enter search conditions in the system under test,
The searched system is actually operated and the search result is acquired.
Then, verify the document with the number obtained as the search result,
Check if the search conditions are met.

[0003]

The conventional method has the following problems.

That is, the search condition is input to the system to be inspected, the number of the document that matches the search condition is obtained as the search result, and it is verified whether this document matches the search condition to obtain the search result. Although it is possible to confirm whether or not the specified document matches the search condition, it is not possible to verify whether or not all the documents stored in the database that match the search condition are obtained by the search result.

As described above, according to the conventional method, it is possible to check whether the document searched by the system to be inspected matches the search condition given, but whether all the documents matching the search condition in the database have been searched. It cannot be inspected.

An object of the present invention is to check whether or not the documents searched by the system to be inspected match the search condition given and to check whether all the documents matching the search condition in the database are searched. To provide a way to do.

[0007]

According to the present invention, in the inspection of a full-text search system, a new document is created by encrypting a basic document and replacing a part of the encrypted basic document with an amplification table. By using multiple amplification tables,
Create multiple new documents. Here, the basic document is a document prepared in advance as a base of a newly created document, and the amplification table is a table storing a plurality of types of character strings that replace the encrypted basic document. One amplification table is provided for each replacement part in the encrypted basic document.
Use one. The reason why the basic document is encrypted is to prevent the same character as the character string in the amplification table from being included in the basic document. These documents are stored in the database as a search target. A condition for searching a document in the database is created by the amplification table, an expected value of the search result corresponding to the search condition is calculated, and an expected search result value table is created. Here, the search result expected value table is a table in which the search condition and the expected value of the search result are associated with each other. The system under test is actually operated to obtain the search result, compared with the expected value of the search result, and the comparison result is output.
By calculating the expected value of the search result in advance, it is checked whether the system to be inspected is correctly searching for the document matching the given search condition, and all the documents matching the search condition in the database are searched. It is possible to reliably carry out an inspection as to whether or not In addition, since the information of the comparison result is output, when the results do not match, it becomes a material for searching for defects in the full-text inspection system.

[0008]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described in detail below.

FIG. 1 is a block diagram showing the configuration of a full-text search system and a full-text search system inspection program according to the present invention. FIG. 2 is a diagram showing the configurations of the amplification table and the search result expected value table. FIG. 3 is a flowchart showing an embodiment of a processing procedure when the present invention is applied to the inspection of the full-text search system. FIG. 4 is a diagram showing a procedure for creating a document to be searched by using the encrypted basic document and the amplification table.

In FIG. 1, a basic document 10 is a document which is a base of a newly created document prepared in advance for creating a new document to be searched. Amplification table 11
Is a table in which a plurality of character strings for replacing a part of the encrypted basic document are stored, and one table is prepared for each place where the encrypted basic document is replaced. The first character string in the amplification table represents a character string to be replaced in the basic document, and the second and subsequent character strings are character strings that replace the character string in the basic document to be replaced. The encryption program 12 encrypts the basic document. The document creation program 13 uses the encrypted basic document and amplification table (amplification table 120 and amplification table 2 shown in FIG. 2).
21) newer document (search target document 29 shown in FIG. 2)
A plurality of types are created, and these newly created documents are stored in the database 19 as search targets of the inspected system. The search condition creation program 14 uses each amplification table (amplification table 120 and amplification table 2 2 shown in FIG. 2).
From the combination of 1), search conditions (search condition 3 shown in FIG.
2) is created. Expected search result calculation program 1
Reference numeral 5 calculates the expected value of the search result (the expected value 33 of the search result shown in FIG. 2) for the search condition (search condition 32 shown in FIG. 2) created by the search condition creation program 14.
A search result predicted value table (search result predicted value table 31 shown in FIG. 2) is created. Search result comparison program 16
Is the search condition (search condition 32 shown in FIG. 2) of the search result predicted value table (search result predicted value table 31 shown in FIG. 2)
Is input to the full-text search system 17, the system to be inspected is operated, and the acquired search result and the search result predicted value table (search result predicted value table 33 shown in FIG. 2) are compared.
If they match, the information corresponding to the match is output to the comparison result information 18 with the search condition, the calculated search result, and the actual search result if they do not match. The full-text search system 17 is a system to be inspected, searches the database 19 according to search conditions, and outputs the search result to the search result comparison program 16. The comparison result information 18 is information of a result obtained by comparing the expected value of the search result with the actual search result, which is output from the search result comparison program 16. The database 19 is a database that stores documents to be searched by the full-text search system 17.

Next, the operation of each part shown in FIGS. 1, 2 and 4 will be described with reference to the flow chart of FIG.

First, the basic document 10 is encrypted by the encryption program 1
It is encrypted by 2 (step 100). Next, the document creation program 13 replaces the encrypted basic document (encrypted basic document 22 shown in FIG. 4) with the amplification table 11, creates a plurality of documents to be searched, and stores them in the database 19. (Step 101). To create a document, for example, the encrypted basic document (encrypted basic document 22 shown in FIG. 4) says "My name is A. Age is B.
I'm old. ], Amplification table 1 (amplification table 120 shown in FIG. 4) is “A, Tanaka, Sato”, amplification table 2 (amplification table 222 shown in FIG. 4) is “B, 25, 30”.
Suppose First, the encrypted basic document (encrypted basic document 22 shown in FIG. 3) is replaced by the amplification table 1 (amplification table 120 shown in FIG. 4). Amplification table 1 (amplification table 120 shown in FIG. 4) is an encrypted basic document (encrypted basic document 22 shown in FIG. 4).
Since the part of "A" is replaced, as a result, the intermediate document (Fig.
In the interim documents 23 and 24) shown in ", my name is Tanaka.
The age is B. , "My name is Sato. Age is B
I'm old. , Two documents are created. Next, the amplification table 2 (search condition 21 shown in FIG. 4) replaces the intermediate document (intermediate documents 23 and 24 shown in FIG. 4). Since the amplification table 2 (amplification table 223 shown in FIG. 4) replaces the part “B” of the intermediate document (the intermediate documents 23 and 24 shown in FIG. 4), as a result, a new document (search shown in FIG. 4) is obtained. In the target documents 25, 26, 27, 28), "My name is Tanaka. Age is 25.", "My name is Tanaka. Age is 30.", "My name is I'm Sato. I'm 25 years old. "," My name is Sato. I'm 3 years old.
I am 0 years old. Four documents are created. That is,
Each amplification table (amplification tables 20 and 21 shown in FIG. 4)
A new document is created by multiplying the number of character strings in. In this example, the amplification table 1 (amplification table 20 shown in FIG. 4) has two character strings, and the amplification table 2 (amplification table 21 shown in FIG. 4) has two character strings.
× 2 = 4 new documents are created. Next, search conditions are created by the search condition creation program 14 using the amplification table 11 (step 102). In creating search conditions, the conditions are created using all combinations of the amplification tables (amplification tables 20 and 21 shown in FIG. 4). Using the above example, by combining each amplification table (amplification tables 20, 21 shown in FIG. 4),
"Tanaka, 25""Tanaka,30""Tanaka, no condition""Sato,25""Sato,30""Sato, no condition""No condition, 25""No condition, 30""No condition, no condition""
9 types of search conditions (search condition 32 shown in FIG. 4) are created. “No condition” means that no character string is selected from the amplification table (amplification tables 20 and 21 shown in FIG. 4). For example, the search condition “Tanaka, no condition” is changed to the search condition “Tanaka”. Have the same meaning. Next, the search result predicted value calculation program 15 calculates a search result predicted value for the search condition (search result predicted value 33 shown in FIG. 2), and a search result predicted value table (search result predicted shown in FIG. 2). A value table 31) is created (step 103). Next, the search result comparison program 16 issues a search request to the full-text search system 17, and the full-text search system 17 searches the database 19.
The search result is transmitted to the search result comparison program 16 (step 104). Next, the search result comparison program 16
Compares the predicted value of the search result from the search result predicted value calculation program 15 with the actual search result from the full-text search system 17 (step 105). If the comparison result is a match, a match is output to the comparison result information 18 (step 10
6) If they do not match, the search condition, the expected value of the search result, and the actual search result are output (step 107). Next,
If the search condition is completed, the inspection is completed, and if there is any, the process returns to step 104 (step 108).

[0013]

As described above, according to the present invention, since the retrieval result can be calculated in advance, it is possible to inspect whether the system to be inspected is correctly retrieving the document which matches the given retrieval condition. With this, it is possible to surely perform an inspection as to whether or not all documents matching the search condition in the database are searched. In addition, since the information of the comparison result is output, when the results do not match, it becomes a material for searching for defects in the full-text inspection system.

[Brief description of drawings]

FIG. 1 is a block diagram of an inspection system of a full-text search system according to the present invention.

FIG. 2 is a diagram showing configurations of an amplification table and a search result expected value table.

FIG. 3 is a flowchart showing an embodiment of a processing procedure of the present invention.

FIG. 4 is a diagram showing a method of creating a plurality of new documents by using an encrypted basic document and an amplification table.

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Shogo Saito 5030 Totsuka-cho, Totsuka-ku, Yokohama, Kanagawa Prefecture Software Development Division, Hitachi, Ltd. (72) Inventor Jukugawa ▲ Hiro ▼ Yuko Chuo-ku, Osaka-shi, Osaka Kitahama 3-chome No. 529 Nissei Seibu Software Co., Ltd. (72) Inventor Hiroaki Morita No. 3-529 Kitahama Kitahama, Chuo-ku, Osaka City, Osaka (72) Inventor Kazuya Ueoka Osaka 3-529 Kitahama, Chuo-ku, Osaka-shi, Niigata Nishi Software Co., Ltd. (72) Inventor Bungo Shichinohe 5030 Totsuka-cho, Totsuka-ku, Yokohama-shi, Kanagawa Hitachi, Ltd. Software Development Headquarters

Claims (1)

[Claims] 1. A new document is created by replacing a part of a prepared document with a character string, and a plurality of new documents are created by using a plurality of types of this character string. Store the created document in the database as a search target, create a condition to search the document stored in the database by the above character string, calculate the expected value of the search result corresponding to the search condition, and An automatic verification method of an inspection result characterized by comparing an expected value and a search result obtained by actually executing the inspected system and outputting the comparison result.