JP6483630B2 - Test content generation apparatus and operation method thereof - Google Patents

Description

  The present invention relates to a test content generation apparatus that automatically generates test content and an operation method thereof.

  One of the software development methods is the development with the waterfall model. In the waterfall model, each process of “method study / basic design”, “functional design”, “detailed design”, “manufacturing”, “unit test”, “combination test”, “multiple combined test / stabilization test” is performed. Implement sequentially.

  Conventionally, the verifier manually extracts the test contents of the “multiple compound test / stabilization test” based on the specifications generated in “method study / basic design”. In this method, the contents of the test greatly depend on individual know-how, and a long time is required for extraction.

Kazuo Murakami, “Software Quality Knowledge System Guide -SQuBOK Guide- (2nd Edition) 2. 2.3.1 Waterfall Model (P95), 2.18.1.4 V-Shaped Model (P169)”, SQuBOK Development Committee, Ohm, 2014 / 11/28

  The present invention has been made in view of the above circumstances, and an object thereof is to provide a test content generation apparatus that automatically generates test content and an operation method thereof.

  In order to solve the above problems, a test content generation apparatus according to the present invention includes a natural language analysis unit that analyzes operation specification data described in a natural language, and a tag based on the analysis result for the operation specification data. A tag assigning unit to be assigned, a tag extracting unit for extracting a tag and a portion to which the tag is assigned from the operation specification data after the grant, and the extracted tag and the portion to which the tag is assigned to the operation specification data. And a test content conversion unit for converting the test content into a test content indicating a result obtained by the method.

  The operation method of the test content generation device of the present invention includes a step in which the test content generation device analyzes the operation specification data described in a natural language, and the test content generation device analyzes the operation specification data. A step of assigning a tag based on a result, a step of extracting the portion to which the tag and the tag are attached from the operation specification data after the grant, and the test content generating device are extracted. The method includes a step of converting a tag and a portion to which the tag is attached into a test method for the target of the operation specification data and a test content indicating a result obtained by the method.

  According to the test content generation apparatus and the operation method thereof of the present invention, the test content can be automatically generated.

It is a block diagram which shows schematic structure of the test content production | generation apparatus which concerns on 1st Embodiment. It is the figure which illustrated a part of operation specification data. It is the figure which illustrated a part of structured operation specification data. It is the figure which illustrated test contents. It is explanatory drawing of the learning operation | movement by the tag provision part. It is the figure which illustrated the test content of the modification. It is a block diagram which shows schematic structure of the test content production | generation apparatus which concerns on 2nd Embodiment. It is the figure which illustrated a part of operation specification data. It is the figure which illustrated a part of structured operation specification data.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

[First Embodiment]
FIG. 1 is a block diagram showing a schematic configuration of a test content generation apparatus according to the first embodiment.

  The test content generation apparatus includes a structured function unit 1 and a test content generation unit 2.

  The structured function unit 1 includes a natural language analysis unit 11 that analyzes operation specification data written in a natural language, and a tag addition unit 12 that adds a tag based on the analysis result to the operation specification data.

  The test content generation unit 2 extracts a tag and a portion to which a tag is attached from the operation specification data after the assignment, and extracts the extracted tag and the portion to which the tag is attached in the operation specification data. A test content conversion unit 22 that converts a test method for a subject and a test content indicating a result obtained by the method is provided.

  The operation specification data is intended for software and devices, for example.

  FIG. 2 is a diagram illustrating a part of the operation specification data.

  The operation specification data includes, for example, a natural language sentence in which a character or a word is applied to a portion such as XX in “XX, when XX, results in △ Δ”.

  The natural language analysis unit 11 in FIG. 1 analyzes such a sentence and determines that “when XX” is a condition, “XX” is an input value, and “ΔΔ” is an output value (result). To do.

  The tag assigning unit 12 assigns a tag <cond> </ cond> to the condition “in the case of OO”, and assigns a tag <input> </ input> to the input value “XX”. Then, a tag <output> </ output> is assigned to the result “ΔΔ”. The operation specification data to which the tag is assigned in this way is referred to as “structured operation specification data”.

  FIG. 3 is a diagram illustrating a part of the structured operation specification data.

  In the structured operation specification data, a tag <cond> </ cond> is added to the condition “when OO”, and <input> </ input> is input to the input value “XX”. A tag is assigned, and the tag “<output> </ output>” is assigned to the result “ΔΔ”. The tags are not limited to these three types, and other tags may be used.

  The tag extraction unit 21 in FIG. 1 extracts, from the structured operation specification data, for example, the one shown in FIG. 3, that is, the tag and the part to which the tag is assigned.

  The test content conversion unit 22 converts the extracted tag and the portion to which the tag is assigned into a test content indicating a test method for the target software of the operation specification data and a result obtained by the method.

  FIG. 4 is a diagram illustrating test contents.

  The test content is, for example, “condition: XX” converted from </ cond> when <cond> XX, “input value: XX” converted from <input> xxx </ input>, < “Output value: ΔΔ” converted from output> ΔΔ </ output> is included.

  The test content conversion unit 22 outputs test specification data including the test content generated in this way. The test specification data may include a plurality of test contents.

  Therefore, according to the test content generation apparatus of the first embodiment, the test content can be automatically generated.

  Here, it supplements about the tag provision part 12. FIG.

  FIG. 5 is an explanatory diagram of a learning operation by the tag assigning unit 12.

  First, the test content creator U creates structured operation specification data for each of a plurality of operation specification data on the basis of the operation specification data and the provision criteria prepared in advance. In the assignment standard, for each tag such as <cond> </ cond>, a sentence expression to which the tag is to be attached, a location to which the tag is attached, and the like are described.

  The test content creator U searches for the corresponding sentence expression from the operation specification data and attaches a tag to the designated location.

  The tag assigning unit 12 learns the relationship between the operation specification data (structured operation specification data) pre-assigned in this way and the operation specification data before assignment in advance by using a plurality of sets. Remember.

  And when attaching a tag to the operation specification data which has not been learned, a tag is provided based on the learning result and the analysis result by the natural language analysis unit 11.

  Therefore, it is possible to generate test contents regarding operation specification data that has not been learned.

(Modification)
FIG. 6 is a diagram illustrating test content generated by a test content generation apparatus according to a modification.

  The test content conversion unit 22 of the modified example is a portion to which a tag and a tag are attached when the extracted tag and the portion to which the tag is attached indicate normal operating conditions such as target software and a normal result. Is converted into a method for producing an abnormal result different from the normal result and a test content indicating the abnormal result.

  In many cases, the operation specification data includes the normal operating conditions (XX for XX) and conditions, such as “XX: XX results in △△”. The result when it is satisfied, that is, a normal result (result of ΔΔ) is shown.

  For example, the test content conversion unit 22 of the modification may include “condition: XX converted from </ cond> when <cond> XX”, “input” converted from <input> xxx </ input> A test content including “value: other than xx” and “output value: does not become ΔΔ” converted from <output> ΔΔ </ output> is generated.

  Therefore, it is possible to generate a test content for a test that causes an abnormal operation as well as a normal operation, and to perform the test. This modification can also be applied to a second embodiment described later.

[Second Embodiment]
Next, a second embodiment will be described. Here, it demonstrates centering on the difference with 1st Embodiment, and duplication description is abbreviate | omitted.

  FIG. 7 is a block diagram illustrating a schematic configuration of a test content generation apparatus according to the second embodiment.

  In addition to the configuration of the test content generation device of the first embodiment, the test content generation device of the second embodiment includes a database 3 that assigns and stores a label to each of a plurality of test content.

  The test content conversion unit 22 gives a label to the generated test content and stores it in the database 3. Thus, the database 3 assigns and stores a label to each of the plurality of test contents.

  The label indicates the function corresponding to the test content, such as “restart (restart condition for construction / failure etc.)” and “basic call processing service (minimum communication condition)”. is there. Hereinafter, the description will be continued by taking “restart” as an example.

  FIG. 8 is a diagram illustrating a part of the operation specification data.

  The label is disclosed to the creator of the operation specification data, and if the creator does not change the function corresponding to the label as in the past, as shown in FIG. Include the text “The function is as it is now.”

  The natural language analysis unit 11 analyzes such a sentence and determines that the “restart function” is “as is”.

  FIG. 9 is a diagram illustrating a part of the structured operation specification data.

  As shown in FIG. 9, the tag assigning unit 12 reads a test content stored in the database 3 in response to the “restart function” determined to be “as is”, for example, <ref> A tag </ ref> is assigned. The tag may be different.

  The tag extraction unit 21 extracts such a tag <ref> </ ref> and the part “restart function” to which the tag is attached from the structured operation specification data.

  The test content conversion unit 22 stores the tag <ref> </ ref> as a tag for reading the test content stored in the database 3, and the tag extracted in this way is used for reading the test content. In the case of a tag, the label “restart” corresponding to the part “restart function” to which the tag is attached is searched from the database 3, and the test content to which the label is attached is read.

  Then, the test content conversion unit 22 outputs test specification data including the test content read in this way. In the test specification data, the generated test content and the read test content may be mixed.

  Therefore, according to the test content generation apparatus of the second embodiment, the test content generated in the past can be read.

  The computer program for causing the computer to function as the test content generation apparatus of each embodiment can be recorded on a computer-readable recording medium such as a semiconductor memory, a magnetic disk, an optical disk, a magneto-optical disk, or a magnetic tape. It can be widely distributed by being transmitted via a communication network such as the Internet.

DESCRIPTION OF SYMBOLS 1 Structured function part 2 Test content generation part 3 Database 11 Natural language analysis part 12 Tag assignment part 21 Tag extraction part 22 Test content conversion part

Claims (6)

A natural language analysis unit that analyzes operation specification data written in natural language;
A tag attaching unit for attaching a tag based on an analysis result to the operation specification data;
A tag extraction unit that extracts a tag and a portion to which the tag is attached from the operation specification data after the grant;
A test content conversion unit that converts the extracted tag and the portion to which the tag is attached into a test method for the target of the operation specification data and a test content indicating a result obtained by the method, and
The tag giving unit is
Learning and storing the relationship between pre-granting operation specification data and post-granting operation specification data using multiple sets of pre-granting operation specification data and pre-granting operation specification data. and, learning result and the analysis result test content generating apparatus characterized by imparting tags based on. A natural language analysis unit that analyzes operation specification data written in natural language;
A tag attaching unit for attaching a tag based on an analysis result to the operation specification data;
A tag extraction unit that extracts a tag and a portion to which the tag is attached from the operation specification data after the grant;
A test content conversion unit that converts the extracted tag and the portion to which the tag is assigned into a test method for the target of the operation specification data and a test content indicating a result obtained by the method;
And a database for imparting and storing a label to each of a plurality of said test content,
The test content conversion unit
When the extracted tag is a tag for reading the test content stored in the database, the label corresponding to the portion to which the tag is assigned is searched from the database, and the test content to which the label is assigned test content generation apparatus characterized by reading out. The test content conversion unit
When the extracted tag and the part to which the tag is assigned indicate the normal operating condition and the normal result of the target, the tag and the part to which the tag is assigned are different from the normal result. The test content generation apparatus according to claim 1, wherein the test content generation apparatus converts the method for generating an abnormal result and the test content indicating the abnormal result. A test content generation device analyzing operation specification data written in a natural language;
The test content generation device gives a tag based on an analysis result to the operation specification data;
The test content generation device extracts a tag and a portion to which the tag is attached from the operation specification data after the grant;
The test content generation device converts the extracted tag and the portion to which the tag is attached into a test method for the target of the operation specification data and a test content indicating a result obtained by the method;
The test content generation device uses a plurality of sets of operation specification data to which a tag is attached in advance and the operation specification data before the application, and the operation specification data before the application and the operation specification data after the application are provided. storing the relationship learned in advance, the learning result and method of operation of the test content generation apparatus characterized by performing the step of applying a tag on the basis of the analysis result. A test content generation device analyzing operation specification data written in a natural language;
The test content generation device gives a tag based on an analysis result to the operation specification data;
The test content generation device extracts a tag and a portion to which the tag is attached from the operation specification data after the grant;
The test content generation device converts the extracted tag and the part to which the tag is assigned into a test method for the target of the operation specification data and a test content indicating a result obtained by the method. Done
The test content generation device includes:
A database for storing a label for each of the plurality of test contents;
In the step of converting into the test content,
When the extracted tag is a tag for reading the test content stored in the database, the label corresponding to the portion to which the tag is assigned is searched from the database, and the test content to which the label is assigned operation method of testing the contents generation apparatus characterized by reading out. In the step of converting into the test content,
When the extracted tag and the part to which the tag is assigned indicate the normal operating condition and the normal result of the target, the tag and the part to which the tag is assigned are different from the normal result. 6. The method for operating a test content generation apparatus according to claim 4 or 5 , wherein the method generates an abnormal result and converts it into a test content indicating the abnormal result.

Images