JPH03288226A - Software evaluating method - Google Patents

Abstract

PURPOSE:To enable a user to concretely grasp the problems and to easily work out countermeasures by analyzing a program to detect the description information included in the program and having the defective maintenance property based on the analysis information and outputting the relative information showing the factor of the defective maintenance property. CONSTITUTION:A program is analyzed and the description information included in the program and having the defective maintenance property is detected based on the analysis information. Then this program description information is outputted with relation secured to the information showing the problem of the maintenance property. For instance, the check is carried out based on the basis of standardization for each word/phrase. Then the corresponding line No and message are stored in a memory and outputted when a word or phase opposed to the standardization reference rule is detected. Thus a user can concretely grasp a specific problem produced in a specific area of a program in regard of the maintenance property and can work out the countermeasure.

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention relates to a method for improving software productivity, and in particular, the present invention relates to a method for improving the productivity of software, and in particular, to evaluating the maintainability of a developed program, and detecting and changing parts with poor maintainability. This invention relates to a software evaluation method suitable for

[Prior Art] One of the approaches to make software maintenance work more efficient is to consider in advance the efficiency of future maintenance work and create a program with good maintainability.

For this purpose, if there is a means to automatically inspect and evaluate the ease of maintenance of created programs, it will be easier to create programs with good maintainability, and it will also be possible to easily determine the direction of creating programs with good maintainability. It can also be used for inspection.

One of the conventional techniques related to this purpose is the development of a tool that analyzes a program and measures and outputs information that is an index of maintainability, such as the ratio of Go TO statements in the program and the program size.

For example, [Nakamura et al.: IMAP system (9)]
- Software quality assurance technology - Information Processing Society of Japan 31st Annual Conference Proceedings, p505-506 (September 1985)
month)”.

[Problem to be solved by the invention]

In the above-mentioned conventional technology, it is not possible to associate a specific part of a program with a maintainability problem with the problem and present it to the user. Therefore, it is difficult for the user to know specifically what kind of problem exists in which part of the program.

Further, the above-mentioned conventional technology has a problem in that it does not take into consideration a method of creating a program using a semi-finished program, that is, a program pattern/component.

That is, in recent years, a method is often used to create a program by using previously created program patterns and parts and performing additional coding on these to complete the program. In the above-mentioned conventional technology, it is not possible to identify whether the target program uses patterns or parts, and it is not possible to distinguish between the parts of the target program that are based on program patterns and parts and the parts that are additionally coded. It is not possible to obtain the size and ratio of the data or to evaluate maintainability. Therefore, it is not possible to evaluate maintainability only for the parts that the program creator has actually added coding, or to evaluate maintainability based on the usage rate of patterns and parts. It was difficult to come up with countermeasures.

Further, in the above-mentioned conventional technology, it is not possible to visually display the ease of maintenance of each internal procedure constituting a program on a diagram showing the structure of the program. Therefore, it is difficult for users to know which internal procedures have particularly poor maintainability and to rewrite those procedures to improve maintainability.

Further, in the above-mentioned conventional technology, it is not possible to visually display the ease of maintenance of each program on a program relationship diagram showing the structure of the entire software. Therefore, it is difficult for users to know which programs have particularly poor maintainability and to rewrite those programs to improve maintainability.

Furthermore, in the above-mentioned conventional technology, the statements in the program and their description format are automatically rewritten to the statements and format for structured programs, for example, in order to improve maintainability without changing the description of the entire program. , or could not be automatically rewritten to the standard sentences and formats in that department. Furthermore, it was not possible to automatically evaluate the maintainability of rewritten programs. Therefore, it has been difficult to improve the maintainability of the created programs.

Furthermore, with the above-mentioned conventional technology, it is not possible to aggregate the maintainability evaluation information of each developed program by person or group, or in chronological order, and output the results for comparison. Therefore, it has been difficult to formulate measures to improve the maintainability of the programs being developed for each person or group.

The first object of the present invention is to provide a Rutame software evaluation method that assists the user in knowing specifically where in a program there is a problem regarding maintainability and taking countermeasures. It is in.

A second object of the present invention is to provide a software evaluation method that evaluates the maintainability of a program and supports the formulation of measures to improve the maintainability of the program.

A third object of the present invention is to provide a software evaluation method for supporting the evaluation of program maintainability when developing programs using program patterns and components, and for developing measures to improve program maintainability. It is about providing.

A fourth object of the present invention is to provide a software evaluation method that makes it easy for a user to know internal procedures in a program that have particularly poor maintainability and to take countermeasures.

A fifth object of the present invention is to provide a software evaluation method that makes it easy for users to know which programs have particularly poor maintainability and to take countermeasures.

A sixth object of the present invention is to provide a software evaluation method that facilitates improving the maintainability of a created program without changing the structure of the entire program.

A seventh object of the present invention is to provide a software evaluation method that facilitates formulating measures to improve the maintainability of programs to be developed for each person or group.

[Means to solve the problem]

In order to achieve the above first objective, the software evaluation method of the present invention analyzes a program, uses the analysis information to detect program descriptions with maintainability problems in the program, and solves maintainability problems. It is distinctive in that it is output in association with the information shown.

Another feature is that lines of the source program that include program descriptions with problems in maintainability are output in association with information indicating problems in maintainability.

Additionally, information indicating the maintainability problem is associated with the line containing the program description with the maintainability problem, and
The feature is that the source program is output.

In order to achieve the above second objective, the software evaluation method of the present invention analyzes a program and uses the analysis information to detect program description information in the program that has maintainability problems or It is characterized by measuring the presence or absence and number of program description information that serves as an index, calculating evaluation points based on the measured information according to established evaluation criteria, and outputting the results in a chart format.

Another feature is that the user sets and changes the program description information type or evaluation criteria to be detected and measured.

In addition, the program description information to be detected and measured is a statement that has a maintainability problem, that is, GOTo
It is characterized by including at least one of the following: a sentence, an ALTER sentence, or a sentence that is determined to be contrary to #A$.

In addition, the program description information to be detected and measured above is an IF statement without END-IF at the end, TH
It is characterized by including at least one of the sentence notations that are determined to have problems in maintainability, such as an IF statement without both EN and ELSE.

In addition, the program description information to be detected and measured above includes at least one of the executable statements determined to have maintainability problems, such as Go To statements and PERFORM statements located above the program. It is characterized by its inclusion.

Further, the program description information to be detected and measured is characterized in that it includes at least one of lexical delimiter information determined to have problems in maintainability, such as rtJtr;J.

In addition, the program description information to be detected and measured above includes locations with multiple statements on the same line, IF statements, and PERFOR statements.
In the program, there are sentences that are not indented among the indented sentences included in M sentences, etc., or sentences that are out of the one indentation target range but are not returned from the indented position. It is characterized by including at least one of the information determined to have maintainability problems regarding the description position.

Furthermore, the above information to be measured includes at least one of the number of lines and sentences of the program internal procedure, and information on the size of the program internal procedure, such as the average value or maximum value of at least one of them. There is.

Further, the program description information to be measured is characterized in that it includes at least one of the number of comment lines or the ratio of comment lines in the program.

Also, the information to be measured above is the number of lines of the program 1
It is characterized by containing at least one of information related to the program size, such as the number of lines in the procedure division, the number of statements in the program, and the number of executable statements in the procedure division.

In order to achieve the above third objective, the software evaluation method of the present invention analyzes a program and detects whether a program pattern/part created in advance is used in the program. Identification of parts and additional coding parts by program pattern/parts is done from the identification information number in the comment, and output of whether or not the program pattern/parts are used, or additional coding part or 1 is the number of lines of the program pattern/parts. , outputting size information such as the number of executable statements or their ratio, or evaluating and outputting the maintainability of the program in the detected additional coding part using the method described above. be.

In addition, to identify the program pattern/component part and the additional coding part for the above program analysis information, input the identification information of the program pattern/component part and the additional coding part corresponding to the corresponding pattern, and use that information. It is characterized by the fact that it is done by identifying.

In addition, to obtain information regarding the size of the above program pattern/part part/additional coding part, or to evaluate the maintainability of the additional coding part, input measurement information that is an index of maintainability regarding the corresponding pattern/part, It is characterized in that it is performed by calculating using that information.

In order to achieve the fourth objective, the software evaluation method of the present invention detects at least start information of each internal procedure and call statements between internal procedures in analyzing the program. Identify the internal structure of the internal procedure, measure the information of the part that is an index of program maintainability using analysis information, and assign evaluation points based on the measurement information to each internal procedure unit using the method described above. It is unique in that it creates internal program structure diagram information based on structural information and visually outputs the evaluation points for each internal procedure, distinguishing them by color or shape.

In order to achieve the above-mentioned fifth objective, the software evaluation method of the present invention measures program analysis/maintainability information and assigns evaluation points to each program constituting a specified program group using the above-described method. At the same time, it detects calling relationships between programs using other program calling statements in each program, creates relationship diagram information between programs based on the calling relationship information, and displays evaluation points for each program in color and shape. It is characterized by distinguishing and outputting it visually.

In order to achieve the above-mentioned sixth object, the software conversion method of the present invention analyzes a program, and uses the analysis information to convert a program description with a maintainability problem in the program and information on the problem to the above-mentioned method. The feature is that the analysis information is detected using a predetermined method, and the program is reproduced and output from the changed analysis information.

Also, I analyzed the program and found that END-IF was added at the end.1. N Detects at least one of the IF statements that do not include an IF statement, THEN, or ELSE, and if they are not attached, END-IF, THEN,
ELSE (7) The feature is that at least one of the pieces of information is inserted into the analysis information, and a program is regenerated and output from the changed analysis information.

Also, from the program analysis information above, ",".

Detect the use of at least one of the lexical delimiter program description information that is determined to have maintainability problems, such as ";", and if it is used, delete it from the analysis information and change it. It is characterized by regenerating and outputting a program from analysis information.

In addition, from the above program analysis information, we have found that there are multiple statements on the same line, sentences that are not indented among the sentences that are included in IF and PERFORM statements, or sentences that are targeted for indentation. Detects program description information that determines that there is a maintainability problem regarding the program description position, such as a sentence that does not return the indentation in a sentence that has come out of the range, and detects the program description information that determines that there is a problem with maintainability regarding the position of the program description, and detects the second and subsequent sentences in the position where there are multiple sentences on the same line. Converting a sentence to a separate line, changing the first column position of a non-indented sentence within the indentation target sentence to the indentation column position, returning from indentation within a sentence that has come out of the indentation target range Make at least one of the following changes, such as changing the position of the first column of the sentence that has not been changed to the column position of the indentation return.

The feature is that the program is regenerated and output from this changed analysis information.

In addition, for programs that have been automatically modified for the purpose of improving maintainability as described above, the above method can be used to (1) detect program description information that has maintainability problems in the program, or use it as an index of maintainability; (2) Maintainability within the program A feature of the present invention is that it detects problematic program description information and its causes, and outputs them in association with each other.

In order to achieve the seventh objective, the software evaluation method of the present invention analyzes maintainability of each developed program by performing program analysis, measuring maintainability information, and calculating evaluation points using the methods described above. It is characterized by obtaining information, aggregating the evaluation information by person or group, or by time series, and comparing it.

[Effect]

The present invention analyzes a program, uses the analysis information to detect a program description in the program that has a maintainability problem, and outputs the detected program description in association with information indicating the maintainability problem. Therefore, the user can easily and concretely know the parts of the target program that have maintainability problems and the problems, and can take countermeasures.

The present invention analyzes a program, uses the analysis information to detect program description information in the program that has problems with maintainability, or program descriptions that are indicators of maintainability, and detects the presence or absence of such information and the number of such information. is measured, and based on the measurement information, evaluation points are calculated according to established evaluation criteria and output in a single chart format. Therefore, the user can automatically obtain the evaluation value of the maintainability of the program in an easy-to-understand chart format, so the user can evaluate the maintainability of the program,
Measures to improve maintainability can be easily taken.

The present invention analyzes a program, detects whether a program pattern/component is used in the program, or identifies a part based on a program pattern/component and a part with additional coding using identification information in comments. Based on the information, it outputs whether or not a program pattern/component is used, or outputs size information such as the number of lines, the number of executable statements, or their ratio of the additional coding part or the program pattern/component part, or the detected It is possible to output the maintainability evaluation of the additional coding part. Therefore, the user can easily evaluate program maintainability and take measures to improve maintainability in program development using program patterns and components.

Furthermore, the present invention can input identification information of the program pattern/component part and the additional coding part corresponding to the corresponding pattern, and use that information to identify the program pattern/component part and the additional coding part. I can do it.

In addition, the present invention inputs measurement information that is an index of maintainability regarding a corresponding pattern/component, and uses that information to obtain information regarding the size of a program pattern/component part/additional coding part, or to obtain additional coding It is also possible to evaluate the maintainability of parts.

The present invention identifies the internal structure within a program by detecting the start information of each internal procedure and the call statements between internal procedures, evaluates the maintainability of the program for each internal procedure, and In-program structure diagram information is created based on the structure information, and the evaluation points for each internal procedure are differentiated by color and shape and output visually.

Therefore, the user can easily learn about particularly poorly maintainable internal procedures in the target program and take countermeasures.

The present invention evaluates the maintainability of each program constituting a specified program group, detects calling relationships between programs using other program call statements in each program, and uses this calling relationship information to The system creates related diagram information and visually outputs the evaluation points for each program, distinguishing them by color and shape. Therefore, the user can easily know which programs have particularly poor maintainability and can take countermeasures.

°The present invention analyzes a program, uses the analysis information to detect a program description with a maintainability problem in the program and information about the problem, and changes the analysis information in a predetermined manner. Then, the program is regenerated and output from the changed analysis information. In addition, for programs that have been automatically modified for the purpose of improving maintainability as described above, (1) detection of program description information that has maintainability problems in the program, or program description information that is an index of maintainability, Detect and measure that information, give evaluation points based on the measurement information according to established evaluation criteria, and output in chart format. (2) Program description information that has maintainability problems in the program and its cause, and output the associated information.

Therefore, the program can be automatically converted to improve maintainability without changing the structure of the entire program, and the maintainability of the program can be automatically evaluated after conversion.

The present invention obtains maintainability evaluation information by performing program analysis, measuring maintainability information, and calculating evaluation points for each developed program using the method described above, and the evaluation information is divided by person, group, or time. Summarize each series and output for comparison.

Therefore, it becomes easy to formulate measures to improve the maintainability of the program being developed for each person or group.

〔Example〕

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

FIG. 1 is a diagram showing a processing procedure in a first embodiment of the present invention, and FIG. 2 is a diagram showing a hardware environment of this embodiment.

In FIG. 2, 201 is a CPU, 202 is a memory, 2
03 is an input device, and 204 is an output device.

205 is an external storage device, and 206 is a display device.

The software evaluation method of this embodiment will be explained with reference to FIG. First, in step 1, input one line from the program to be evaluated stored in the external storage device 205.
01), analyze the input line (step 103
).

FIG. 3 shows an example of a source program written in the COBOL language, FIG. 4 shows an example in which the analysis results obtained when line 301 is input are stored in memory, and FIG. 5 shows an example of a source program written in the COBOL language.

FIG. 6, an example of the standardization standard rules to be checked at the time of analysis, shows an example in which messages indicating non-standards on a line-by-line basis are stored in memory.

Check the input line sequentially from left to right for groups of characters other than spaces, including space symbols and ",".

A collection of characters separated by punctuation symbols such as ";" (
For each (hereinafter referred to as a token), the column position of the start of the token is stored in memory together with the token and the line number, which is counted by 1 for each input line.

For example, in row 301 of FIG. 3, rA L T E R
J is the first non-space token from the beginning of the line,
It starts in column 15 and is terminated by a space symbol, so the lexical rALTER) (401
) is code number ``82'' (402), column r15'' (4
03), it is stored in memory as lexical information 404.

In addition, if the symbol "umbrella" in the C0BOL language, which indicates that the entire line is an annotation (comment), is in column 7, lexical information that indicates that the symbol "umbrella" is a lexical and the column is "7", Create lexical information in which the entire line after column "8" is one lexical phrase. for example.

In line 302 of FIG. 3, since there is a symbol "*" in column 7, the entire line including spaces after garum 8 is treated as one token, and the starting column is stored in the memory as column "8".

Next, a check is performed based on the standardization criteria for each lexical unit, and if the lexical phrase violates the standardization standard rules, the corresponding code number and message are stored in the memory, and at the same time, the corresponding lexical information is deleted from the memory.

For example, rule 501 shown in the example of FIG. 5 is a rule regarding "words" and indicates that the use of words "," is prohibited. The lexical information 405 shown in the example of FIG. 4 is the rule 50 shown in FIG.
1, the line message 601 consisting of the line number and message shown in the example of FIG. 6 is stored in the memory. Thereafter, the lexical information 405 shown in the example of FIG. 4 is deleted from the memory.

Lexical information 406 is similarly deleted from memory.

When the appendix analysis is completed for one line, in step 104 shown in FIG. 1, if the lexical information stored in the memory has a column of "7" and a lexical of "*", the lexical information creation light is The line marked with the number is determined to be a comment line, the lexical information having the number is deleted from the memory, and the process returns to step 101 shown in FIG. 1 to read the next line. If the line is not a note line, sentence information based on language syntax is extracted based on the lexical information stored in the memory, and a sentence is formed (step 105).

FIG. 7 shows an example in which the source program shown in the example of FIG. 3 is additionally analyzed and the results of sentence formation are stored in memory.

In the sentence formation (step 105) shown in the example of Figure 1, the lexical information obtained by the appendix analysis (step 103) is
Sentence information is created by parsing based on the L language syntax and stored in memory.

For example, if the lexical information shown in the example of FIG. 4 (however, lexical information 405 and 406 are deleted) is
By performing syntax analysis based on the language syntax of , sentence information 701 shown in the example of FIG. 7 is formed. Each piece of information includes an item rALTERJ (702) that identifies sentence information from the lexical rALTERJ (401) shown in the example of FIG.
) from column r15J (403) to the starting column "
15'' (704), and also the lexical rSW by syntactic information.
J (407), parameter from the lexical “Pl” (4o8)
The data rSW and PIJ (705) are respectively created and stored in memory. Item number r70J (706) is a number for identifying sentence information that is incremented by one each time sentence information is created and stored in the memory. Also, parent item r52J
(707) stores in memory the item number of the sentence information corresponding to the upper layer when indicating the hierarchical relationship in the C0BOL syntax, for example. After storing this sentence information in memory, the lexical information is deleted from memory. In the sentence information 701 shown in the example of FIG. 7, the lexical information group 406 shown in the example of FIG. 4 is deleted from the memory.

When one piece of sentence information can be formed in the sentence formation (step 1o5) shown in FIG. 1 (step 106), sentence analysis (step 1o7) is performed based on the sentence information. If the sentence formation (step 105) has not resulted in one sentence information (if one sentence spans multiple lines) (step 106), the process returns to step 101 and reads the next line.

In sentence analysis (step 1o7), sentence formation (step 1o
Targeting the latest sentence information created in step 5), apply the standardization standard rules with the classification "sentence" and check the sentence. If the sentence violates the standardization standard rules, the "Start" Store the code number J and the message in memory.

For example, sentence analysis targeting the sentence information 701 shown in the example of FIG.
Apply 02. Here, "apply" indicates that a rule to be checked is found among the rules classified as "sentence" for the sentence information. Rule 502 indicates that the use of the sentence "rule related to sentence J, item name is rALTER" is "prohibited". According to this rule 502, it is determined that the sentence information 701 shown in the example of FIG. 7 is a "prohibited" command, and the message is stored in the memory along with the starting line "82") (Line message 602 in FIG. 6) ,.

After the sentence analysis (step 107) shown in FIG. 106), sentence analysis (step 1
07) - Repeat sentence formation (step 105).

For example, in the example shown in FIG. 4, the lexical information group 409 is deleted from the memory during the first sentence formation, and when forming the second sentence, the lexical information group 410 is used as the target for sentence formation, and the lexical information group 409 is Text information 708 shown in the example shown in the figure is created. In the sentence analysis in the sentence information 708, the rule 50 shown in FIG.
3 is applied, and the start line "82" in the sentence information 708 in FIG.
Since the values of the start line "82" in the previous sentence information 701 are the same, the line message 603 shown in FIG. 6 is stored in the memory. In addition, the sentence information 70 shown in the example of FIG.
In the sentence analysis in 9, the above-mentioned rule 503 shown in FIG.
In addition, rule 504 is applicable.

Rule 504 is a rule regarding r-sentence J, and indicates that if the nesting level is deeper than the previous sentence information, "indentation" description is to be performed. For example, the sentence information 709 shown in the example of FIG.
The parent item number "87" is the parent item number "87" of the sentence information 710 mentioned above.
Since the value is larger than "2", it can be seen that the nesting level is deep. It is the intent of the above-mentioned "indentation" rule to have a starting column position that is larger in value than the starting column position of the previous sentence when the nesting level becomes deep.

Therefore, the start column "15" of the sentence information 709 is the sentence information 7
It must be larger than the starting column "15" of 01, but since it is against the rules, line message 6 shown in Figure 6
04 in memory.

On the other hand, if the nesting level is shallower than the previous sentence information, the "one character down" description means one character back J, and the start column value of the sentence is smaller than the start column value of the previous sentence. shows. For example, in the example shown in FIG.
2" has a smaller value than the parent item number "89" of the sentence information 712, which indicates that the nesting level is shallower. From this, starting column r17 of sentence information 711
J must be smaller than the starting column "17" of sentence information 712, but since this is against the rule, line message 605 shown in FIG. 6 is stored in memory.

Next, when the reading of the source program (step 101 shown in FIG. 1) is completed (step 102), the entire program is analyzed based on the sentence formation results stored in the memory (step 108).

In the overall analysis (step 108), sentence formation (step 1
05) and stored in memory, apply the standardization standard rules whose category is "whole", check the entire program, and check for sentences that violate the standardization standard rules. The message is stored in memory along with the code number of this sentence.

For example, in the standardization standard rule shown in the example in Figure 5, rule 505
is a rule regarding "whole", and at least one of the contents indicated by the parameters of the sentence information with the item name rGOTOJ has a smaller item number than this sentence information. Indicates that matching with the content shown in is prohibited. By applying the rule 505, first, among the sentence information shown in the example of FIG. 7, sentence information with the item name "rGOT○" is detected. Statement information 713 is applicable, but the parameter does not have a value, so the next item name is r.
Detect the sentence information of GOTOJ. The sentence information 714 is applicable, and the parameter "Pl" is obtained.

Next, sentence information with a section name "procedure section" or "procedure paragraph" having "Pl" as a parameter is detected. Sentence information 7
15 is applicable, and the item number “82” is the item number “82” of the sentence information 714.
Items smaller than ``86''.

The sentence information 714 is found to be against the rules, and the message is stored in the memory along with the number "95" (line message 606 shown in the example of FIG. 6). Also, the fifth
Rule 506 in the standardization standard shown in the example in the figure is a rule regarding "whole", and the term names are rIFJ, rTHENJ, rE
LsEJ.

This shows that the sentence information of rENDIFJ has a structure in which the item numbers increase in this order and the parent item numbers of each piece of sentence information are the same. By applying this rule 506, first, among the sentence information shown in the example of FIG. 7, sentence information with the item name rIFJ is detected. Sentence information 716 corresponds to this. Next, it has the same parent item number as the parent item number “82” of this sentence information 716, and has the item number “82”.
Sentence information having an item number greater than "5" is detected. In the example of FIG. 7, sentence information 710 corresponds, but the item name is rE L S
In EJ, since it is not rTHENJ as specified in the rule, it is found that the structure is contrary to the rule, and by applying this rule, the code number "95" of the sentence information 716 detected first and the message are stored in the memory (Fig. 6 line message 607).

In the standardization standard shown in the example in Figure 5, rule 507 is "overall"
rules, the number of lines from the statement information whose section name is "Procedure Section" to the section immediately before the statement information whose section name is "Procedure Section" and whose section number is greater than the section number of this statement information is "30" or less. This indicates that the limit is set to . By applying this rule 507, in the sentence information shown in the example of FIG.
Detects sentence information of "procedure clause". Sentence information 717 corresponds to this. Next, the sentence information having the smallest item number is detected from among the sentence information 717 whose item name is ``procedure clause'' and has an item number greater than ``51''. Sentence information 718 corresponds to this. The numerical value rlO5J is obtained by subtracting the start line “61” of the sentence information 717 from the start line “166” of the sentence information 718.
It is found that the number of lines belongs to the "procedure clause" of the sentence information 717, and it is larger than the number of lines "30" in the rule 507 shown in FIG. 5, so the line message 608 shown in FIG. 6 is against the rule. is stored in memory.

Next, various numerical measurements in step 109 in FIG. 1 are performed.

FIG. 8 shows an example in which various numerical information collection results regarding the entire program are stored in the memory.

The value "480" for the numerical name "number of rows" shown in Figure 8 is the value stored in memory of the "number" that is counted sequentially in the additional analysis (step 103) shown in Figure 1, and the numerical name The value r31oJ of ``number of sentences'' is the ``item number'' that is sequentially counted in the sentence analysis (step 105) shown in FIG. 1, stored in memory, and the value E35 of the numerical name ``number of comments.''
is a memory that stores the values sequentially counted in the comment judgment (step 104) shown in FIG.
The value "11" of the numerical name "rGOTO number" is the value counted in the memory for the sentence information with the item name "rGOT○" among the sentence information of the sentence formation results stored in the memory,
The value "40" of the numerical name "average number of lines in the procedure cylinder" and the value "52" of the "maximum number of lines in the procedure lung" are based on the values in Figure 5 applied in the overall analysis (step 108) shown in Figure 1. The number of rows belonging to the "procedure clause" is calculated for each "procedure clause" using the same method as rule 507 shown in the example, and the average value and maximum value are calculated, and the value stored in the memory and the numerical name "procedure clause" are calculated. The value "38" for "average number of sentences in lungs" and the value "49" for "maximum number of human sentences in procedure clause" are based on the rules shown in the example in Figure 5 that were applied in the overall analysis (step 108) shown in Figure 1. 5
The method is similar to 07, but by setting the calculation target to 1 item number, the number of sentences belonging to the "procedure clause" is calculated for each "procedure clause", and the average value and maximum value are calculated, and the retrieval memory is The value ``22'' of the numerical name ``number of defective prohibition commands'' stored in the memory is calculated by counting the number of line messages in which the line message shown in Figure 6 has ``the prohibition command ``-'' is used.'' If the value is stored in memory, the value ``8'' with the numerical name [same line number system defective number] has the line message shown in Figure 6 ``Multiple sentences are used on the same line.'' The value of the counted number of line messages is stored in memory, and the value ``5'' of the numerical name ``Number of lines using unnecessary words'' indicates that the line message shown in Figure 6 is ``-'' is used.'' The value of the number of different code numbers among the line messages with the ``numeric name r'' value ``5'' is stored in memory.
1" is the value stored in memory, which is the count of the number of line messages in which the line message shown in Figure 6 has "no indentation" or "no indentation." The value "32" for the name "Number of structured defects" is the value stored in memory by counting the number of row messages whose row messages have "unstructured" as shown in Figure 6. The value ``3'' of the line number limit defective number is the value obtained by counting the number of line messages in which the line message shown in Figure 6 has ``the number of lines exceeds the limit value for ``-''lines.'' This is what I remembered.

Next, output the analysis results shown in FIG. 1 (step 110)
outputs each line of the source program and the corresponding line message as a pair to the output device.

FIG. 9 shows an example in which the analysis results of the source program shown in the example of FIG. 3 are output to an output device.

For example, the source program shown in the example of FIG. 3 stores the input lines in the memory in units of units in the program input (step 101) of FIG. Output at the same time. At this time, if there is a line message with the same number as the line to be output in the line message shown in the example of FIG. 6, the message is added and output. The row 301 shown in the example 3- is the row 9 shown in FIG.
The line message group (6
09) are simultaneously output (903).

When the output of the analysis results is finished due to the termination of the source program, the analysis results shown in FIG. end.

FIG. 10 shows an example of evaluation standard rules, and FIGS. 11 and 12 show examples in which evaluation results are output to an output device. Evaluation standard rule A shown in FIG. 1o is a rule for obtaining various evaluation points based on the various numerical information collection results shown in the example of FIG. , indicates the type of evaluation item, and if the condition 1002 is correct, the evaluation point r
5J (1003) is calculated. Condition 1002
Now, since the numerical value r480J of the numerical name "number of lines" shown in FIG. 8 is greater than r300J, the condition 1004 of the next item "number of lines" is checked. Numerical name shown in Figure 8 “
Because the number r480J of "Number of rows" is smaller than r500J,
It is determined that the condition is correct, and an evaluation score of "4" (1005) is obtained and stored in the memory. Similarly, conditions are checked for each item name, and evaluation points are obtained and stored in the memory. The obtained evaluation points are added up sequentially and stored in the memory as a "total score." Each "evaluation score" and "overall score" stored in the memory are associated with each numerical value of the various numerical information collection results shown in the example of FIG. 8 together with the item name, and the evaluation results are output (step 112), the data is sequentially edited and output (FIG. 11). Furthermore, the evaluation standard rule B shown in FIG. 10 is based on the various numerical information collection results shown in the example of FIG.
Item name [Easy to see shape J (1006) is formula 100
By calculating 7, the value can be calculated as a percentage.

"Number of defective sentences in the same line", "Number of lines using unnecessary words", "Number of defective indentations", and "Number of lines" in formula 1007 are replaced by the numerical name "Number of defective sentences in the same line" shown in the example in Figure 8. ``Number'' value ``8'', ``Number of unnecessary lines used'' value ``5'', ``Number of defective font indentations'' value ``5''.
1", the value of "number of rows" is r480J, and the formula is rx-(
8+5+51)/480J, and the calculation result is ro, 86
7J is stored in memory as "R-shaped visibility". Similarly, by calculating formulas for each item, various standardization rates are calculated and stored in the memory. The various standardization rates stored in the memory are output to the output device in the form of a radar chart, for example, as shown in FIG. 11, together with the item names (step 112).

According to this embodiment, it is possible to see to what extent the source program adheres to the standard coding standards, and therefore the user can make fair and accurate judgments regarding ease of maintenance. be.

FIG. 13 shows the processing procedure in the second embodiment of the present invention. In this embodiment, in addition to the embodiment shown in FIG. 1, standardization standard rules and evaluation standard rules are input from an external storage device.

The steps other than the rule input step (step 1301) shown in the example of FIG. 13 are the same as the embodiment shown in FIG.

Rule input (step 1301) includes standardization standard rules to be used (an example is shown in Figure 5) for additional analysis (step 1o3), sentence analysis (step 107), and overall analysis (step 1o8) stored in the external storage device. ) and the evaluation criteria rules (first
(Example shown in Figure 0) and store it in memory.

According to this embodiment, the standardization standards or evaluation standards stored in the external storage device are set in advance to standards suitable for software development projects, departments, offices, etc. By keeping it in place, changes can be made, and it has the effect of being universally usable even when standardization standards or evaluation standards are different.

FIG. 14 shows the processing procedure of the third embodiment of the software evaluation method of the present invention. In addition to the embodiment shown in FIG. 1, this embodiment provides maintainability only for the part of the standard pattern that is own-coded when the input program is a program created using a standard pattern. This made it possible to evaluate the Here, the standard pattern is a standard semi-completed program that has created the main parts of the program's processing procedure, and the user codes individual parts by adding them to the standard pattern (hereinafter referred to as own coding). This allows you to create standard programs. Figure 15 is an example of a source program written in C0BOL language created using standard patterns, and Figure 16 is a table that describes comments for identifying own coding parts and pattern parts for standard patterns. Example, 17th
The figure shows an example of a table in which the standard pattern part and the own coding part are identified and the starting numbers of each part are stored, and FIG. 18 shows an example of outputting the maintainability evaluation results for the own coding part. When one line of the input program is determined to be a comment in the processing procedure of FIG. 14 (step 1o4), it is determined whether the comment includes a pattern. For example, as in the comment 2101 in Fig. 15, rPATTERN-ID: PATT
If the program starts with ERNIJ, then the program contains a standard pattern and rPATTERNIJ wo pattern TD.
After judgment L (step 2001), pattern information corresponding to the pattern is input (step 2002). Inputting pattern information (step 2002) involves inputting and storing pattern information identified by the extracted pattern among the pattern information stored in the external storage device using the pattern as an identifier into the memory (step 2002). 2002). FIG. 16 shows an example of pattern information stored in the memory, and shows the pattern I in the comment line 21o1.
Standard pattern rPATTERNIJ identified by D
(2201) (7) Indicates pattern information. The pattern information indicates that the comment in the comment line in the program is
What you remember, for example, ``*G Main Shori J
(2202), the following line of the source program is a line of "Pattern J (2203)", indicating that it is a part coded in advance in the standard pattern. Also, if the comment is, for example, "*G If the source program starts with ``Own Coding J'' (2204), the following lines of the source program are ``Own Coding J''.
This line (2205) indicates that the user has added additional coding to the standard pattern. If the comment does not include a pattern ID, in identification determination step 2003, pattern information input step 2 is
From the pattern information obtained in 0o2, it is determined whether the comment line is the start line of own coding, the start line of a pattern, or something else. If it is an own coding start line, the code number of the line is used as own coding start information in the own coding start information storage step 2004, and if it is a pattern start line, the line is used as pattern start information in the pattern start information storage step 2005. The indicated own coding and the starting line of the pattern are stored in a table, and the process returns to the program input step 101 to read the next line of the program. After all lines of the program have been input and the overall analysis step 108 has been performed in the same manner as in the embodiment of FIG. It is obtained from the own coding start number information obtained in the pattern start information storage step 2005 and the pattern start number assignment information obtained in the pattern start information storage step 2005. Based on this, the own coding part is analyzed in the own coding range measurement step 2008, and the own coding analysis result output step 2009 outputs a program source list with messages for the own coding part, similar to that shown in FIG. Further, in own coding part analysis result evaluation step 2010, the own coding part measured according to the coding standard is evaluated, and the own coding part evaluation result is output step 2o1.
1, the own coding unit evaluation result list shown in FIG. 18 is obtained.

For example, when the source program shown in Figure 15 is input to this system, r * PATTERN -I on the second line
D PATTERNl'' (2101) is recognized as a pattern ID in the pattern ID determination step 2o01, and information regarding the pattern PATTERNI shown in FIG. 16 is obtained in the pattern information input unit 2002. When "Ning G Kaishi Shori" on the 88th line of FIG. It is stored in the number 2 column of the table in FIG. 17 that 88 is the start line of ○ (own coding). The own coding range creation unit 2006 uses 88 in FIG. 15 from the table.
The range from line 92 to line 92 (the line immediately before the starting numbering of the next number) is defined as one own-coding range, and the other own-coding ranges are calculated in the same manner. Own coding range numerical value measurement step 2008 is row step 1
03 and the line message information obtained in the overall analysis step 108, those within this range are extracted, and various numerical measurements within this range are performed in the same manner as in FIG. The step 2009 for outputting the own coding part analysis results outputs the own coding line message information extracted in 2008 corresponding to the target source program in a format similar to that shown in FIG. The own coding analysis result evaluation step 2010 evaluates the own coding part in the same way as 110 in FIG. 1, and the own coding part evaluation result output step 2011 outputs the evaluation result of the own coding part in the same way as in FIG. . FIG. 18 shows an example of the output.

According to this embodiment, the maintainability of a program using standard patterns and parts can be efficiently evaluated.

FIG. 19 shows the processing procedure of the fourth embodiment of the software evaluation method of the present invention. The source program in Figure 20 is either own coding (0) or pattern (P) for each block.
) or the identifier is attached to the comment, which makes the first
By the processing procedure shown in FIG. 9, the 16th
Pattern portions and own-coding portions can be identified without the pattern information shown in the figure.

For example, if you input the source program shown in Figure 20, 2
Row r r * PATTERN - ID PAT
TERN2J is recognized as a pattern ID in pattern ID identification step 2001, and pattern ID storage step 25
It is remembered that the program input in step 01 uses the pattern rPATTERN2J. Figure 20 88
When the line "Kushi○ Kaishi Shori" is input, the identifier determination step 2502 determines that it is own coding start line information based on the leading "umbrella○", and the own coding start information storage unit 2004 stores the information in the table shown in FIG. number 2
The number in the column indicates that line 88 is the start line of ○ (own coding). Thereafter, the evaluation result list of the own coding unit shown in FIG. 18 is generated in the same manner as in the third embodiment. The following is the same as the third embodiment. In this embodiment, even when the program includes other parts of the pattern (described later), the program can be analyzed and evaluated by distinguishing between the two parts of the pattern part and the own coding part.

FIG. 21 shows the processing procedure of the fifth embodiment of the software evaluation method of the present invention, in which a program including parts is input,
It outputs the usage rate of parts along with the evaluation of the own coding section, etc. Here, a component is a standard internal module created in advance for each function, and by using the component as part of newly developed software, the development process can be reduced. FIG. 22 shows an example of a C0BOL source program input in this embodiment. FIG. 23 shows an own coding section and a table showing the start numbering of patterns or parts. FIG. 24 is a parts information table in which a combination of part name and step number is registered. FIG. 25 is an output list output in this embodiment, and shows the results of analyzing the parts included in the input program.

In the processing procedure shown in FIG. 21, the steps from the program input section 101 to the pattern ID identification section 2001 are the same as in the embodiment shown in FIG. It is determined whether it is a coding start, a pattern start, or a part start, and if it is identified as a part start, in a part information input step 2704, information on the already registered part shown in FIG. 24 is input. Furthermore, set the part start number to 2nd in 27Q5.
Store it in the table shown in Figure 3. Thus step 27o4
Based on the parts information obtained in step 27, a list showing the usage rate of parts as shown in FIG.
Output from 08. For example, C0B shown in FIG.
Line 109 of the OL source program is determined to have an identifier in the identifier determination step 2701, determined to be the start of a component in the identifier determination step 2702, and one of the component information shown in FIG. 24 is input in the component information input step 2704. , pattern/component start information storage step 2
In step 705, it is stored as pattern/component start line information as shown in FIG. As a result of the above, the parts utilization rate and the like are calculated in the parts analysis step 27o6, and the parts analysis list shown in FIG. 25 is outputted in the analysis result output step 2707 together with the own coding section evaluation result list shown in FIG. 18. In this embodiment, a list with a message from the own coding section and an own coding evaluation list are output together with the parts analysis list, but an alternative solution in which only the parts analysis list is output can be easily realized.

FIG. 26 shows the processing procedure of the sixth embodiment of the software evaluation method of the present invention. FIG. 27 is an output list of this embodiment. This is a diagram that shows the structure of blocks in a program, with upper blocks calling lower blocks.

In addition, evaluation points are set for each block, and the evaluation points are color-coded based on the evaluation score.

Up to the overall analysis step 108, the same processing as in the first embodiment is performed. In the block range structure analysis step 3011, the range of the block is identified by the rsEcTIONJ statement indicating the start of the block, and the calling relationship for each block is determined by rPE.
The block structure information is created by grasping the information using the RFORMJ command. For example, in the program in Figure 15, "PERF
ORMKAISH (21o3) If found, program block MA I N4 block KA
It is possible to understand that ISHI is subordinate.

Next, based on the information for grasping the block range obtained in the block range structure analysis step 3011, various numerical value measurement step 3014 describes various numerical values for each block, and in the measurement result evaluation step 3012, the blocks are Evaluate units. Then, in a structure step 3013, a program internal structure diagram shown in FIG. 27 is output based on the block structure information within the program obtained in step 3011 and the evaluation result obtained in step 3012.

For example, the source program shown in FIG. 22 is analyzed in accordance with the processing procedure shown in FIG. 26 from the program input step 101 to the overall analysis step 108 in the same manner as in the first embodiment. Then block range structure analysis step 30
11 performs block-by-block analysis and rPERFORM
Understand the structure of calling relationships between blocks from the J instruction. In this way, program MAIN has program KAI.
SHI, Program NAKA, Program 5tJRYO
, to which program TSUIKA is dependent, program 5UR
It is understood that the program TSUIKA is dependent on YO. In the various numerical value measurement step 3014, measurements are performed based on the information obtained in the block range structure analysis step 3011, in the measurement result evaluation step 3012, the measurement results are evaluated, and in the structural figure output step 3013, between the programs shown in FIG. A structure diagram is output.

FIG. 28 shows the processing procedure of the seventh embodiment of the software evaluation method of the present invention. Figure 29 shows the four C0s to be evaluated.
The BOL source program is shown below.

rCALL 5HOUKEIJ (2901) and rCA
By LLURIAGEJ (2902), program CHOUHYO calls program 5HOUKEI and program URIAGE. Also rCALL
Program 5HOUK by 5ORTJ (2903)
EI is calling program 5ORT. FIG. 30 is a diagram showing the calling relationship of programs, and shows the relationship in which a higher-level program calls a lower-level program. Evaluation points are also provided for each program and are color coded.

In this embodiment, a diagram showing the calling relationships between multiple programs as shown in FIG. 30 is output by inputting a plurality of programs having calling relationships as shown in FIG. 29 and performing the processing procedure shown in FIG. 28. be done. In step 3201 of FIG. 28, it is determined whether there are any remaining programs to be input, and if there are any remaining programs, the program input step 101 and subsequent steps are performed as one of the remaining programs.

Hereinafter, the same processing as in the first embodiment is performed up to the analysis evaluation output step 111, and then the call relationship analysis step 32
In step 03, the call relationship between programs is grasped from the "rCALL" command, and call/output relationship information is stored. For example, in Figure 29, program C) IOHYO's rCA
It can be understood from LL 5HOUKEIJ (2901) that program 5HOUKEI is subordinate to program CHOUHYO. In the evaluation result accumulation step 3204, evaluation results of this program are accumulated. When the analysis and evaluation of all programs is completed in this way, it is determined in 3201 that there is no program to be input, and the evaluation result output step 32o5 is performed based on the information accumulated in the accumulation step 3204 and the call related information accumulated in step 3202. The diagram shown in Figure 30 is output.

FIG. 31 shows the processing procedure in the eighth embodiment of the present invention. In addition to the example shown in Figure 1, this example improves the overall evaluation score by automatically correcting some parts of the target source program that violate the standardization standard rules. It is something.

The steps shown in the example of FIG. 31 are the same as the embodiment shown in FIG. 1 except for the steps of sentence correction (step 4001) and output of the analysis result (step 4002).

FIG. 3 shows an example of a source program written in the COBOL language, and FIG. 32 shows the additional analysis (step 103) and sentence analysis (step 107) shown in FIG. 31.

FIG. 33 shows an example of a line message indicating outside the standardization standard for each line in this embodiment, which was stored in the memory by the overall analysis (step 108). The source program shown in the example of FIG. Figure 34 is an example of adding a new start line, new start column, and adding a new sentence to the result. 32 shows an example of the line message shown in FIG.

In FIG. 32, item information of "new code number" is added to the line message example of FIG. 6, and a new code number created by dividing a line and adding a sentence is stored as a standard. Figure 33 shows an example of the sentence formation result in Figure 7 with "Used or not" and "New start line".
, is a table to which each item information of "new start column" is added. "Used or not" indicates information that identifies whether sentence information is obtained from the sentence formation result information or from line information when generating a new source program (step 4002 shown in FIG. 31). "New start row" and "New start column" are
Memorizes new numbers created by splitting lines and adding sentences, and new starting columns created by "indentation".

In the sentence modification (step 4001) shown in the example of Figure 31, among the line descriptions of the source program that violate the standardization standard rules, line descriptions that can be automatically corrected or added are modified to conform to the pre-standardization standard rules. Modify and add written content;
Messages for lines that violate the standardization standard rules are stored in memory as line messages, so line messages are sequentially examined, line messages that can be automatically corrected or added are detected, and sentence information stored in memory is stored. Make corrections and additions.

For example, the line message 4101°41o shown in FIG.
2 indicates that there is an unnecessary token in the line numbered "82". In the sentence information shown in the example in Figure 33, the starting line is "
82" is detected.

Statement information 4201 and 4202 correspond to this, and by generating new source program lines from this statement information,
Since it is possible to create a line that does not include the unnecessary words and phrases mentioned above, the "Usage status" of 4201 and 4202 of this text information
"Y" is set to indicate that the statement information is to be used for line generation of the new source program. Further, a line message 4103 shown in FIG. 32 indicates that there are multiple sentences in the line numbered "82". Among the sentence information shown in the example of FIG. 33, sentence information whose starting line is "82" is detected. Sentence information 420
1.4202 is applicable. In order to output multiple pieces of sentence information each on one line, increase the new start line of the second and subsequent sentence information by 1 in the order of increasing item number among the sentence information with the same number. At the same time, among sentence information other than sentence information whose parent item number has the same value as the parent item number of the sentence information, the start column of the sentence information with the largest item number value is set as the new start column of the sentence information.

For example, in text information 4201.4202, text information 420
The new start column of sentence information 4202 is incremented by 1 for all new and start lines of sentence information after 2, and the new start column of sentence information 4202 is
Set it to "15", which is the same as the start column of 01. In addition, since a new source program is generated with multiple pieces of statement information each on a separate line, rYJ
Set.

Next, the line message 41o4 shown in FIG.
This indicates that there is a sentence in the line ``97'' that is not inverted by one character. Among the sentence information shown in the example of FIG. 33, sentence information whose starting line is "97" is detected. Sentence information 4203 corresponds, and a new start column indicating "character inscription" is set. The new start column “19” of this sentence information 4203 is
2o3's parent item number "87" is detected as the item number value, and the detected sentence information of 4204 is the start column of 42o4 "
17'', for example, calculate and set ``19'' from r17+2=19J, which indicates that the ``2'' column is ``indented''.

Furthermore, until sentence information with a parent item number smaller than the parent item number "87" of sentence information 42o3 is detected, this sentence information 420
From 3 onwards, the item number increases by 1 to the new start column for each sentence information.

For example, set the value by adding "2" columns to the start column.

Next, the line message 4105 shown in FIG.
This indicates that there is a sentence in the line loOJ that has not been "back-lined." Among the sentence information shown in the example of FIG. 33, sentence information whose starting line is rloOJ is detected. Sentence information 42
o5 corresponds to this, and a new start column indicating "backward" is set. The new start column “15” of the sentence information 42o5 is
Among the sentence information other than sentence information 42o5 that has the parent item number "82" of this sentence information 42o5 as the value of the parent item number, the sentence information with the largest item number value is detected. (Sentence information 4206 item 1 is applicable) , the start column of the detected sentence information is set.

Next, the line message 4106 shown in FIG.
This indicates that the sequence of statement information related to structured programming starting with the rIFJ statement included in the line ``95'' is unstructured, and indicates that the item name of the corresponding statement information is rIFJ. Among the sentence information shown in the example of FIG. 33, sentence information whose starting line is "95" is detected. Sentence information 4206.
4207 is applicable, but since the item name is rIFJ, text information for structured programming is added to text information 4206.

First, the sentence information having the smallest item number among the sentence information whose parent item number is the item number of this sentence information is detected. If the item name of the detected sentence information is not rTHENJ, the item name r is next to this sentence information.
new sentence information with "THEN" is added (4208 in FIG. 33). Add 1 to the item number of this sentence information to the added sentence information item number, and add the value of this sentence information item number to the parent item number.
Set "Y" to the usage status, set the new start line to the value of the start line of this statement information plus 1, and set the new start column to the value of the start column of this statement information plus, for example, "2". Set.

In conjunction with the addition of text information, the following modifications will be made. That is,
Except for the added sentence information, 1 is added to the item number for all sentence information whose item number has a value larger than the item number of the added sentence information. All sentence information whose item number has a value larger than the item number of the added sentence information, except for the added sentence information and the sentence information with item name rE L S EJ whose parent item number is the item number of this sentence information. For all sentence information whose parent item number is equal to or larger than the item number of this sentence information, 1 is added to the parent item number. Furthermore, 1 is added to the new start for all sentence information whose item numbers have values larger than the added sentence information. And, for all statement information whose parent item number is the item number of the added statement information, the new start column is changed from the new start column of the added statement information to "
Check whether the text is set to "Inset" (a larger value), and if it is not set to "Inset", add, for example, "2" to the new start column.

Next, among the statement information with an item number larger than this statement information item number, the statement information whose parent item number has the same value as the parent item number of this statement information, and whose item number has the smallest value. Detect. If the item name of the detected sentence information is not rENDIFJ, add new sentence information with the item name rENDIFJ before the detected sentence information (4209 in Figure 33), and add the detected sentence to the item number of the added sentence information. The value of the item number of the information is set to the parent item number, the value of the parent item number of the detected sentence information is set to "Y" for whether to use it, and the value of the new start line of the detected sentence information is set to the new start line. The value of the new start column of the sentence information is set in each column.

Along with the addition of text information, the item numbers and parent item numbers of other text information are corrected in the same manner as described above.

By modifying the text information for the line messages mentioned above,
Since the corresponding line message is no longer needed, the corresponding line message is deleted from the memory after the processing for each line message is completed.

FIG. 34 shows a line message after the sentence information has been corrected based on the line message of FIG. 32.

The output of the analysis results shown in FIG. 31 (step 4002) is almost the same as the embodiment shown in FIG.

Each line of the source program to be output is the program input (
The difference is that the line input in step 101) is created from the sentence information stored in the memory, and the created new source program is also output to an external storage device.

Each line of the source program to be output has the output number in the new start line of the statement information, and if there is statement information whose use status is rYJ, from the items, parameters, and new start columns of the statement information, for example, the C0BOL language Generates a source line according to the syntax and outputs it to an output device and an external storage device. Whether or not to use the statement information that has the output number in the new start line of the statement information is "
If not, the source line is shifted to the left by the number of columns of "Start Column - New Start Column" of this statement information, and outputs the corresponding line input in the program input (step 101) to the output device and external storage device. do.

Figure 35 shows the source program shown in Figure 3 and the source program shown in Figure 33.
An example of a source program list generated from the sentence information shown in the figure and output to an output device is shown.

According to this example, coding that violates standardization standards can be avoided, especially in terms of maintainability, such as "easy to see shape" and "easy to see structure".
It is possible to automatically correct points related to standardization standards, which has the effect of improving maintainability. In addition, by modifying only parts that are problematic in terms of legibility, rather than modifying the entire source program, it is possible to preserve the original source program as much as possible, which is difficult to do with automatic modification. It has the effect of preserving.

FIG. 36 shows the processing procedure in the ninth embodiment of the present invention. In addition to the embodiment shown in FIG. 1, this embodiment extracts the creator of the target source program, compares it with the evaluation history information of this creator stored in an external storage device in advance, and extracts the creator of the target source program. The comparison result is output to an output device.

Input of evaluation history information shown in the example of FIG. 36 (step 50)
01), comparison of evaluation results (step 5002).

Other than outputting the evaluation comparison results (step 5003),
This is the same as the embodiment shown in FIG.

Figure 37 shows an example of evaluation history information for each creator, and Figure 38 shows an example of an evaluation comparison result list that compares the evaluation of a target source program with the evaluation history information of the creator of the source program. show.

Input of evaluation history information shown in the example of FIG. 36 (step 50)
01) is the name of the creator of the source program (for example, AUTHOR of the COBOL language
The evaluation history information stored in the external storage device is input into the memory using a sentence (or a comment using a specific identification symbol, etc.) as an identifier.

For example, the value of the item name "Creator name" shown in FIG. 37 [Hitachi Taro J (5101) is a value that identifies evaluation history information, and the value r5 of the item name "Number of samples"]
(5102) indicates that the number of evaluations performed by the creator of the source program so far is "5", and the evaluation point group 5103 indicates that the number of evaluations performed by the creator of the source program is "5". ” indicates the total score obtained from the evaluation.

Next, compare the evaluation results shown in the example of Fig. 36 (step 5).
002) calculates the difference between the average of the evaluation points obtained in the evaluation of the analysis results (step 111) and the evaluation points obtained in the input of evaluation history information (step 5001).

First, the evaluation score for each item of the evaluation history information stored in the memory is divided by the number of samples, and the difference between this value and the evaluation score for each item of the evaluation result information stored in the memory is determined.

Next, the evaluation result of the source program, the average evaluation score of the history information, and the evaluation comparison result of the difference in the second term are output for each evaluation item to the output device (step 5003), and at the same time, The evaluation points for each evaluation item are added to the evaluation points for each evaluation item in the history information, 1 is added to the value of the item name "number of samples," and the result is output to an external storage device.

According to this embodiment, evaluation results can be accumulated for each source program creator, and quality improvement for each program creator or standardized coding for each creator can be managed in chronological order. effective. In addition, by changing the value that identifies the evaluation history information in this example to "creator group" or "creator company" instead of "creator," the above-mentioned quality improvement and management can be achieved by "creator group". , can be done for each "creating company".

Furthermore, the user can input and estimate the "creator", "creator group", or "creator company" instead of extracting them from the program.

Furthermore, it is also possible to easily compare and output the evaluation information in the evaluation history information between "creators", "creator groups", or "creator companies".

FIG. 39 shows a processing procedure in a tenth embodiment of the present invention. This embodiment is almost the same as the embodiment shown in FIG. 1, but in outputting the analysis results, only the lines in the source program that have corresponding line messages are output in pairs with the corresponding line messages. The only difference is that

Output of analysis results shown in the example of FIG. 39 (step 6001
) is the same as the embodiment shown in FIG.

FIG. 39 shows an example in which the result of dividing the source program shown in the example of FIG. 3 is output to an output device.

For example, the source program shown in the example of FIG.
In the program input (step 101) shown in the figure, the input line is stored in the memory in units of units. The step of outputting the analysis results (step 6001) in FIG.
If there is a line message with the same number as this line, the number and message are added and output in order. The row 301 shown in the example of FIG. 3 is the row 901 shown in FIG.
2" (902), and the line message group (609) having the item number "82" shown in the example of FIG. 6 is output at the same time (903). Unlike the output example shown in FIG. 9, source lines without corresponding line messages are not output.

Since the embodiment of the present invention has the processing procedure described above, it has the effects described below.

By analyzing the program, it is possible to detect program descriptions with maintainability problems in the program and output them in association with information indicating the maintainability problems.
Users can easily learn about areas with poor maintainability and their problems, and take countermeasures.

Furthermore, lines in the source program that include program descriptions with problems in maintainability can be output in association with information indicating the problems in maintainability.

Further, it is possible to output the source program by associating the line containing the program description with the problem with maintainability with information indicating the problem with maintainability in the source program.

Furthermore, since the maintainability evaluation score can be automatically calculated and output in a chart format, the user can easily evaluate the maintainability of the program and formulate measures to improve maintainability.

Furthermore, since the user can set and change the type and evaluation criteria of the program description information to be detected and evaluated, it is possible to easily evaluate maintainability using evaluation criteria suitable for the department in which it is used.

In addition, it is possible to detect the usage of statements in a program that have problems with maintainability, such as Go TO statements + ALTER statements, or statements that violate standards, and to analyze the number and usage rate of these statements. Maintainability can be evaluated.

Also, an IF statement without END-IF at the end, T
IF statements without both HEN and ELSE, etc.
It is possible to detect the usage of grammar notation that has problems with maintainability, and to evaluate maintainability based on the number and usage rate of the usage.

Also, the going rabbit is G at the top of the program.

Maintainability can be evaluated based on the detection of usage locations of executable statements determined to have maintainability problems, such as TO statements and PERFORM statements, and the number of uses and usage rates.

Furthermore, the maintainability can be evaluated based on the detection of the usage of lexical delimiter information such as ",", ";", etc. that is determined to have a maintainability problem, and the number and usage rate of the information.

Also, places with multiple statements on the same line, IF statements and PERFO
In the program, such as a sentence that is not indented among the indented sentences included in the RM statement, etc., or a sentence that has come out of the indented range and has not been returned from the indented position. Maintainability can be evaluated based on the detection of the usage location of descriptive information determined to have a maintainability problem regarding the description position, and the number and usage rate of the usage.

Also, based on the size information of program internal procedures such as the number of lines or sentences of program internal procedures, or the average value or maximum value of at least one of them,
Maintainability can be evaluated.

Furthermore, maintainability can be evaluated based on the number or ratio of comment lines in a program.

Furthermore, maintainability can be evaluated based on information regarding the program size, such as the number of lines in the program, the number of lines in the procedure division, the number of statements in the program, and the number of executable statements in the procedure division.

In addition, by analyzing the program, output the presence or absence of use of program patterns and parts based on comment information in the program, or the number of lines and executable statements of additional coding parts, program patterns, and parts, or the like. It is possible to output size information such as the ratio of , or to evaluate and output the maintainability of the program in the additional coding part. Therefore, it is possible to easily formulate measures to improve maintainability.

In addition, the above-mentioned program pattern/part part and additional coding part are identified according to the corresponding pattern. Enter the identification information of the program pattern/component part and the additional coding part, and use that information to
By identifying and performing this, even if the program/part part cannot be identified from comment information alone, it is possible to output whether or not the above program pattern/part is used, or to output size information of the additional coding/program pattern/part part. Alternatively, it is possible to evaluate and output the maintainability of the program in the additional coding part, and to evaluate the maintainability of the program when developing a program using program patterns and parts, and to formulate measures to improve maintainability. It can be done easily.

In addition, you can input measurement information that is an index of maintainability for the corresponding pattern/part, and use that information to obtain information about the size of the program pattern/part part/additional coding part, or to evaluate the maintainability of the additional coding part. By evaluating program maintainability when developing a program using program patterns and parts, without measuring information that is an index of the maintainability of parts created from patterns and parts in the program. Furthermore, it is possible to easily formulate measures to improve maintainability.

In addition, the program can be analyzed to create internal program structure diagram information, and the evaluation points for each internal procedure can be visually output by distinguishing them by color and shape, allowing users to check the maintainability of the program. Be aware of particularly bad internal procedures;
Countermeasures can be easily taken.

In addition, for each program that makes up a specified program group, it is possible to create relationship diagram information between programs, and visually output the evaluation points for each program, distinguishing them by color and shape. Users can easily learn about internal procedures that are particularly poorly maintained and take countermeasures.

In addition, it is possible to obtain maintainability evaluation information for each program developed, aggregate the evaluation information by person or group, or by time series, and output it for comparison.
This makes it easier to devise measures to improve the maintainability of programs being developed.

〔Effect of the invention〕

As described above, according to the present invention, it is possible to efficiently detect parts of a program with poor maintainability and to evaluate the ease of maintenance of the program, and furthermore, it is possible to take measures to improve the maintainability of the software. It has excellent effects, such as making it easier to stand up.

[Brief explanation of the drawing]

FIG. 1 is a diagram showing the processing procedure in the first embodiment of the present invention, FIG. 2 is a diagram showing the hardware environment of the present invention, and FIG.
The figure shows an example of a C0BOL source program to be evaluated.
Figure 5 shows an example of lexical information obtained through program analysis, Figure 5 shows an example of standardization standard rules used to check a program, Figure 6 shows an example of memory information for line messages of non-standard lines, and Figure 7 The figure shows an example of sentence information obtained by analyzing a program, Fig. 8 shows an example of information stored in memory as a result of collecting various numerical information of a program, Fig. 9 shows an example of the output of the analysis result of a source program, and Fig. 10 shows an example of the output of the analysis result of a source program. Example of evaluation criteria rules, Figure 11, Part 1
2 shows an example of output of the evaluation results, FIG. 13 shows the processing procedure in the second embodiment of the present invention, FIG. 14 shows the processing procedure in the third embodiment of the invention, and FIG. 15 shows the processing procedure in the third embodiment of the invention. The figure shows an example of a COBOL source program to be evaluated, FIG. 16 shows an example of an own coding section identification table corresponding to a standard pattern, FIG. 17 shows an example of a starting number storage table for standard pattern sections/own coding sections, and FIG. The figure shows an example of the output of the own coding evaluation result, FIG. 19 shows the processing procedure in the fourth embodiment of the present invention, FIG. 20 shows an example of the C0BOL source program to be evaluated, and FIG. Diagram showing the processing procedure in Example 5, No. 22
The figure shows an example of a C0BOL source program to be evaluated.
Figure 3 is an example of a standard pattern/component part/own coding part start number storage table, Figure 24 is an example of a part name/step information table, Figure 25 is an example of a parts analysis list, and Figure 26 is an example of the present invention. FIG. 27 is an output example of a program internal structure diagram, FIG. 28 is a diagram showing the processing procedure in the seventh embodiment of the present invention, and FIG. 29 is an evaluation target An example of multiple C0BOL source programs, FIG. 30 is an output example of a call relationship diagram between multiple programs, FIG. 31 is a diagram showing the processing procedure in the eighth embodiment of the present invention, and FIG. 32 is a diagram showing non-standard lines. An example of message memory information, Figure 33 is an example of corrected sentence information, Figure 34 is an example of memory information of a line message from which unnecessary line messages have been deleted, and Figure 35 is an output example of the analysis result of a revised source program. , FIG. 36 is a diagram showing the processing procedure in the ninth embodiment of the present invention, FIG. 37 is an example of evaluation history information for each creator, and FIG. 38 is a comparison result list of the target source program and evaluation history information. For example, FIG. 39 is a diagram showing the processing procedure in the tenth embodiment of the present invention, and FIG. 40 is an example of output of the analysis result of the source program. 101... Program one line input step, 102...
- Program input end determination step, 103...Additional analysis step, 104...Comment line determination step, 1
05... Sentence formation step, 106... 1 sentence formation judgment step, 107... Sentence analysis step, 108...
Overall analysis step, 109...Various numerical value judgment step, 110...Analysis result output step, 111...Analysis result evaluation step, 112...Evaluation result output step, 301...Program line, 302 ...Program comment line, 401... Lexical, 402... Line number, 403... Start column, 404-406... Lexical information, 407-408... Lexical information, 409-410.
... Lexical information group, 501-507... Standardization standard rules, 601-608... Line message, 609... Line message group, 701... Sentence information, 702... Vertex angle, 703...・Start row, 704...Start column, 70
5...Parameter, 706...Item number, 707...
Parent item number, 708-718...Text information, 901...Source line, 902...Line number, 903...Line message group, 1001...Evaluation criteria rule item name, 1002...
・Condition, 1003...Evaluation point, 10o4...Condition,
1005... Evaluation score, 1006... Evaluation standard item name, 1007... Calculation formula, 1301... Rule input step, 2o01... Pattern ID identification step, 20
02...Pattern information input step, 2003...
Identifier determination step, 2004... Own coding start information storage step, 2005... Pattern start information storage step, 2006... Own coding range determination step, 2008... Own coding range measurement step, 2009... Own Coding part analysis result output step, 2010... Own coding part result evaluation step, 2011... Own coding part evaluation result output step, 2101...
・Comment line including pattern ID, 2102...Pattern ID, 2103 Pa block call statement, 2201.
...From the pattern, 2202.2204...Pattern/own coding identification information, 2203,2205.
...Identification ID, 2501...Pattern ID storage step, 2502...Identifier determination step, 2701...
- Identifier presence/absence determination step, 2702... Identifier determination step, 2704... Parts information input step, 27
o5... Pattern/component start information storage step, 27
06...Parts analysis step, 2901-2903...
- Other program call statement, 3011...Block range and structure analysis step, 3012...Measurement result evaluation step, 3o13...Structure diagram output step, 301
4...Various numerical value measurement step, 3201...Applicable program determination step, 32o3...Calling relationship analysis step, 32o4...Evaluation result accumulation step, 32
05...Evaluation result output step, 4001...Sentence correction step, 4002...Analysis result output step, 4
101-4106... line message, 4201-42
09...Text information, 5001...Evaluation history information input step, 5002...Evaluation result comparison step, 50o
3... Evaluation comparison result output step, 51o1... Created book name, 5102... Sample name, 5103... Evaluation point group, 6001... Analysis result output step. ￥ Fu diagram system 4 Yama daisu no life diagram envy diagram 8 Figure 1 λ (A) (8) Figure 1 λ Figure arrogance and rebuttal 3 Diagram S Diagram tb Cause 1? Figure Certain Noah Bacteria No. 9 Group Kuzu 20 Figure! S2 Figure 1 Figure 3 Figure 24 (Hi) (Shin (C) (d) 2b Button No. 5 I5 27th Cause (Hi) <-'o) Certain 28 Figure Certain 0 Listen (Q) Whisper q Figure 1 Figure 32 Figure 4 Figure 3 Figure 5 3b Dan 3rd r'I Figure 9

Claims (1)

[Claims] 1. Analyze at least one program, use the analysis information to detect descriptive information in the program that has a maintainability problem, and associate it with information indicating the cause of the maintainability problem. A software evaluation method characterized by outputting the following information. 2. Measure the presence or number of descriptive information that has maintainability problems or that is an index of maintainability in the detected program, and based on the measured information, according to established evaluation criteria, 2. The software evaluation method according to claim 1, further comprising calculating evaluation points and outputting the evaluation points in a chart format. 3. The user sets or changes the program description information type or evaluation criteria to be detected and measured, and based on this program description information type information, the description information with maintainability problems and its cause information are output. 3. The software evaluation method according to claim 1, further comprising: outputting the measurement information, or obtaining an evaluation result using the measurement information and the evaluation criteria and outputting it in a chart format. 4. GOTO statement or ALTER in the source program
Detects program description information that includes information about statements that have maintainability problems or that violate the standard, and based on this program description information, identifies the program description information that is determined to be problematic. Software evaluation according to claim 1 or 2, characterized in that the cause information is output, or the descriptive information is measured and output, or the evaluation result is obtained using this measurement information and the evaluation criteria and output in a chart format. Method. 5. Program description information that includes syntax that determines that there is a problem with maintainability of IF statements in the source program that do not have END-IF added at the end, or IF statements that do not include both THEN and ELSE. Detected and based on this program description information, output the description information determined to be problematic and its cause information, or measure and output the description information, or obtain evaluation results using this measurement information and the above evaluation criteria, 3. The software evaluation method according to claim 1, wherein the software evaluation method is output in a chart format. 6. Detects program description information in the source program that includes executable statements that are determined to have maintainability problems, such as GOTO statements or PERFORM statements whose destination is above the program, and based on this program description information, determines maintainability. At least one of the following: output of descriptive information that determines that there is a problem and its cause information, measurement and output of the descriptive information, or obtaining evaluation results using this measurement information and the above evaluation criteria and outputting them in chart format. The software evaluation method according to claim 1 or 2, characterized in that: 7. Detect program description information that includes lexical delimiter information that determines that "," and ";" have maintainability problems in the source program, and use the method described above to improve maintainability based on this program description information. Output of descriptive information that determines that there is a problem and its cause information, or measurement and output of the descriptive information,
3. The software evaluation method according to claim 1, further comprising the step of obtaining an evaluation result using this measurement information and the evaluation criteria and outputting it in a chart format. 8. Places in the source program that have multiple statements on the same line, sentences that are not indented among the target sentences,
Alternatively, detect program description information that includes a statement that determines that there is a maintainability problem regarding the program description position of a statement that has not been returned from the indentation range within the statement that has come out of the indentation target range, and Based on the above, output descriptive information that determines that there is a maintainability problem and its cause information, or measure and output the descriptive information, or obtain evaluation results using this measurement information and the above evaluation criteria and output in chart format. The software evaluation method according to claim 1 or 2, characterized by: 9. Measure the program description information including program internal procedure size information such as the number of lines or sentences of program internal procedures, or the average value or maximum value of at least one of them, in the source program, and add this program description information to the program description information. Based on the above method, output descriptive information that determines that there is a maintainability problem and its cause information, or measure and output the descriptive information, or obtain historical value results using this measured information and the above evaluation criteria, 3. The software evaluation method according to claim 1, wherein at least one of the following is output in a chart format. 10. Measure the program description information including information on the number or ratio of comment lines in the source program, and based on this program description information, use the method described above to determine the description information that determines that there is a maintainability problem and its cause. 3. The software evaluation method according to claim 1, further comprising outputting the information, measuring and outputting the descriptive information, or obtaining an evaluation result using the measured information and the evaluation criteria and outputting it in a chart format. 11. Measure the program description information including information about the overall size of the program, such as the number of program lines in the source program, the number of procedure division lines, the number of statements in the program, or the number of executable statements in the procedure division, and based on this program description information. , by using the above method, output the descriptive information that determines that there is a maintainability problem and its cause information, or measure and output the descriptive information, or obtain evaluation results using this measurement information and the above evaluation criteria, and output it in chart format. 3. The software evaluation method according to claim 1, wherein the software evaluation method is output as follows. 12. Analyze at least one program and comment on whether or not a program pattern/component, which is a semi-completed program created in advance, is used in the program, or identify a part based on a program pattern/part and a part with additional coding. This is done based on the identification information within, and outputs whether or not program patterns and parts are used, or additional coding parts, program patterns, and parts.
3. The software evaluation method according to claim 1, further comprising outputting size information such as the number of lines, the number of executable statements, or a ratio thereof. 13. Identification of program pattern/component parts and additional coding parts for program analysis information is as follows:
Input the identification information of the parts and additional coding parts according to the program pattern/parts that correspond to the corresponding pattern,
Using that information, the process outputs whether the program pattern/component is used or not, and size information such as the number of lines, the number of executable statements, or their ratio of the additional coding part/program pattern/part part. 3. The software evaluation method according to claim 1, further comprising the step of outputting, or evaluating and outputting the maintainability of the program in the detected additional coding portion. 14. Acquisition of information regarding the size of the program pattern/component part/additional coding part, or evaluation of the maintainability of the additional coding part is based on the corresponding pattern/part.
Claim 12 characterized in that measurement information serving as an index of maintainability regarding the part is input and the calculation is performed using that information.
Software evaluation method described in section. 15. When analyzing a program, by detecting the start information of each internal procedure and the call statements between internal procedures,
Create internal structure information within the program, detect program descriptions that are indicators of program maintainability using analysis information, measure that information, and calculate evaluation points based on the measured information for each internal procedure unit, The software according to claim 1 or 2, wherein the program internal structure diagram information is created based on the internal structure information, and the evaluation for each internal procedure is visually outputted by distinguishing it by color or shape. Evaluation method. 16. Perform program analysis, measure maintainability information, and calculate evaluation points for each program constituting the specified program group, and detect call relationships between programs based on other program call statements in each program;
3. The software evaluation method according to claim 1, wherein relation diagram information between programs is created based on the call relationship information, and evaluation points for each program are distinguished by color or shape and output visually. . 17. Analyze at least one program, use the analysis information to detect a program description in the program that has a maintainability problem, and use at least one of the analysis information of the program description to be determined in order to solve the problem. A software conversion method characterized in that the program is changed using a modified method, and a program is regenerated and output from the changed analysis information. 18. The detection program description information includes at least one of the following: an IF statement in which END-IF is not used, and an IF statement in which THEN and ELSE are not used together;
For F statements, unused END-IF, THEN
, ELSE is inserted into the analysis information to change the analysis information, and the software regenerates and outputs a program with the relevant portion changed from the analysis information. Conversion method. 19. The program description detection information includes at least one of the lexical delimiter information determined to have a maintainability problem,
18. The software conversion method according to claim 17, further comprising deleting the lexical delimiter information from the analysis information, and regenerating and outputting a program with the relevant portion changed from the analysis information. 20. Program description detection information is for locations with multiple statements on the same line, statements that are not indented among the indented statements included in an IF statement or PERFORM statement, or statements that are indented. Contains a program description that has been determined to have a maintainability problem regarding the program description position of a statement that has come out of the range but has not been returned from the indentation, and has multiple statements on the same line for the analysis information. Converting sentences after the second in a passage to separate lines, changing the first column position of a sentence that is not indented among the sentences to be indented to the indentation column position, and changing the position of a sentence that is out of the indentation target range. Make at least one of the changes, such as changing the first column position of the sentence that has not been returned from indentation to the column position for returning from indentation, and regenerate and output the program from this changed analysis information. 18. The software conversion method according to claim 17. 21. For automatically modified programs, detect information that is an index of maintainability, measure this information, calculate evaluation points based on the measured information and established evaluation criteria, and output the evaluation points in a chart format. 21. The software conversion method according to claim 17, wherein program description information having problems in maintainability in a program is detected by the method according to claim 1, and outputted in association with the program description information. 22. Perform program analysis, measure maintainability information, and calculate evaluation points for multiple programs to obtain maintainability evaluation information, and aggregate the evaluation information by person, group, or time series. 3. The software evaluation method according to claim 2, wherein a comparison output is performed. 23. Analyze at least one source program,
Based on the analysis information, descriptive information with maintainability problems in the program is detected, and lines containing the problematic descriptive information in the source program are matched with information indicating the cause of maintainability problems. A software evaluation method that is characterized by outputting the following information. 24. Analyzing at least one source program,
Based on the analysis information, descriptive information that has maintainability problems in the program is detected, and the cause of the maintainability problem is indicated for the lines in the source program that include the problematic descriptive information. A software evaluation method characterized by adding and outputting information.