JP5658364B2 - Program visualization device - Google Patents

Description

  The present invention relates to a program visualization apparatus that extracts information from a program, performs processing necessary for program analysis using the extracted information, and displays an analysis result.

  In a wide variety of technical fields such as elevators, automobiles, construction machinery, home appliances, and the like, a technique for controlling a device (control target) by so-called embedded software incorporated in a control device is performed in order to realize a specific function. A control device incorporating such software can be said to be an embedded control device executed by a computer.

  Embedded software has advantages in that it can realize flexible and advanced control, and can develop many derivative products by partial modification of software, compared to conventional mechanical mechanisms and electric circuit systems.

  In such a built-in control device, such as an elevator control device, a task is started in response to a certain period or interrupt, and a control variable is updated based on an input of a sensor such as a destination floor designation button or a door safety sensor, thereby A so-called data-driven calculation model that controls actuators such as an opening / closing motor and a car driving motor is employed.

  There is a problem that the contents of control processing required for an embedded control device become more complicated year by year, and the dependency between control variables and subroutines becomes complicated, making it difficult to understand software. For example, even in control software that has been updated with thousands of variables, it is indispensable to verify the software by identifying the affected variables when the calculation method of a certain variable is changed. . Therefore, there is a demand for a visualization device that analyzes embedded software (program) and visualizes the relationship between variables that affect each other and displays the relationship to a user.

  The following has been proposed as a conventional program visualization apparatus.

  Patent Document 1 discloses definition information that defines a component for clarifying the nature of a component of software to be developed when program development is performed, and entity information that represents an entity that exists based on the definition. In addition, the display information for easily recognizing the interrelationship (analysis result) between each component is recorded independently with the ID, and the display information of any aspect is created by relating the information. A display software system analysis support apparatus has been proposed.

  Patent Document 2 discloses a program structure display device that analyzes a source program described in an object-oriented programming language and displays the structure of the source program. This program structure display device is configured to display not only the relationship between classes in the object-oriented program structure, but also the dependency of attributes and methods, using the shapes of nodes and lines, changing the colors and shapes. ing.

  Patent Document 3 shows an example of visualization of source code (program) described in an object-oriented language. That is, an object list and relationships between objects are collected from a log obtained by test execution, an object structure extraction unit extracts object relationships, creates an object diagram, and displays it through a display unit. In this display, the line type, color, thickness, and the like are shown separately depending on the calling direction and the ownership relationship, which are relationships between objects.

  Patent Document 4 shows an example of analyzing and visualizing dependency relationships between program components. In the present technology, the analysis result of the dependency relationship is illustrated in a matrix format, and the matrix element is highlighted according to the analysis result.

JP 7-175642 A Japanese Patent Laid-Open No. 11-95990 WO2008 / 096632 US 7,512,929 B2

  (I) Conventionally, as a program visualization device, a structure creation algorithm that analyzes a program to be analyzed to create a graph structure, a structure analysis algorithm that analyzes the structure, and a diagram that diagrams the structure and structure analysis results A series of optimization algorithms were developed in an integrated manner. As the programs to be analyzed become more diverse, it is necessary to improve the development efficiency of the program visualization device. In particular, with regard to the structure creation algorithm, structure analysis algorithm, and diagramming algorithm described above, if any of these algorithms is changed, the development of a new program visualization device each time reduces the program development efficiency. Therefore, the improvement was desired.

  (II) Various structural analysis algorithms have been proposed with the recent progress of large-scale data analysis technology. Among structural analysis algorithms, there are structural analysis algorithms that can be applied only to specific network characteristics, such as community structure analysis that can be applied only to graph structures having the characteristics of so-called complex networks. The selection of a structural analysis algorithm applicable to the characteristics of the analysis target program requires specialized knowledge, so it is desirable to clearly show the correspondence to the user.

  No prior art including the above-mentioned Patent Document 1 to Patent Document 4 has been proposed so far to satisfy the above-described requirements regarding (I) and (II). That is, in the conventional program visualization apparatus, software from program analysis to display is composed of a series of flows, and if a part of the software is changed, the entire software must be recreated. Further, in the conventional technique, consideration is not given to the user so that the applicable structural analysis algorithm can be understood according to the network characteristics of the analysis target program.

  The present invention has been made to solve the above-described problems. For example, when analyzing (analyzing) embedded software or the like for verification, at least one of the above (I) and (II) is required. It is an object of the present invention to provide a program visualization device that can satisfy one requirement and improve the efficiency of analysis work.

In order to realize the above object (I), the program visualization apparatus according to the present invention analyzes a program and outputs a graph structure composed of nodes and links, and performs an analysis process on the graph structure. A structure analysis means for outputting an analysis result; and an image drawing means for displaying the graph structure and the analysis result on an image display device in a predetermined drawing format.
These three means are software executed via a computer, and are configured as components that can be changed in units of each means independently .
The structure analysis means has a plurality of structure analysis algorithms for executing different analysis methods on the graph structure,
The program visualization device includes structural analysis algorithm suitability determination means for determining whether each of the structural analysis algorithms is applicable to the graph structure from the characteristics of the analysis target program,
The image display device has at least one screen.
(I) a first window for synthesizing and displaying the graph structure and an analysis result of a structure analysis algorithm applied to the graph structure through a drawing format of the image drawing means; (ii) the plurality of structures A function of displaying a name of an analysis algorithm, a function of designating a desired structure analysis algorithm through a screen from a plurality of displayed names of the structure analysis algorithm, and a function of displaying a determination result by the structure analysis algorithm suitability determination unit 2 windows are set .

In order to realize the purpose of (II), a program analysis unit that analyzes a program and outputs a graph structure including nodes and links, a structure analysis unit that performs an analysis process on the graph structure and outputs an analysis result; An image drawing means for displaying the graph structure and the analysis result on an image display device in a predetermined drawing format,
The structure analysis means has a plurality of structure analysis algorithms for executing different analysis methods on the graph structure,
The program visualization device includes structural analysis algorithm suitability determination means for determining whether each of the structural analysis algorithms is applicable to the graph structure from the characteristics of the analysis target program,
The image display device has at least one screen.
(I) a first window for synthesizing and displaying the graph structure and an analysis result of a structure analysis algorithm applied to the graph structure via a drawing format of the image drawing means; (ii) the plurality of structures A function of displaying a name of an analysis algorithm, a function of designating a desired structure analysis algorithm through a screen from a plurality of displayed names of the structure analysis algorithm, and a function of displaying a determination result by the structure analysis algorithm suitability determination unit 2 windows are set.

  According to the former program visualization apparatus according to the present invention, any one of a program analysis algorithm for analyzing a program and creating a graph structure, a structure analysis algorithm for analyzing the graph structure, and a drawing algorithm for displaying the analysis result Even if a change is made, other software can be reused without being affected by the change.

  In addition, according to the latter program visualization device, the user can easily understand the applicable structural analysis algorithm according to the characteristics of the analysis target program.

It is a figure which shows the display screen of the program visualization apparatus which concerns on one Example of this invention. It is a figure which shows the structure of the program visualization apparatus of the said Example. It is a figure which shows the flow of a process of the program visualization apparatus of the said Example. It is a figure which shows the structure of the program analysis layer 31 in the said Example. It is a figure which shows the structural example of analysis object program A211. It is a figure which shows an example of the output of the program A analysis part 311 in FIG. It is a figure which shows the structural example of analysis object program B221. It is a figure which shows the output of the program B analysis part 312 in FIG. It is a figure which shows the structure of the structural analysis layer 32 in the said Example. It is a figure which shows the structure of the complex network determination part 321 in the said Example. It is a figure which shows the flow of a process of the said complicated network determination part. It is a figure which shows the flow of a process of the scale free property determination part 3211 used for FIG. It is a figure which shows the structure of the program structure analysis part 322 in the said Example. It is a figure which shows the data structure 32271 of the structural analysis algorithm attribute memory | storage part 3227 in the said Example. It is a figure which shows the structure of the image drawing layer 33 in the said Example. It is a figure which shows the flow of a process of the analysis technique display preparation part 3321 used in FIG. It is a figure which shows the other example of the display screen of the program visualization apparatus which concerns on the said Example.

Embodiments of the present invention will be described based on the following examples.
(Example)
FIG. 2 is a configuration diagram of a program visualization device (: software visualization device) 100 according to an embodiment of the present invention. The program visualization device 100 is, for example, for analyzing and visualizing programs incorporated in various control devices, and displays a computer 1 that is a program analysis / visualization execution main body and a visualization result executed by the computer 1. A display 9 to be operated, and a mouse 7 and a keyboard 8 for receiving an operation from a user.

  The computer 1 includes a database 2 that stores a program to be analyzed, a memory 3 that stores a processing procedure (algorithm) of the program visualization device, a processor 4 that executes the processing procedure of the software visualization device, and a network interface 5. I have. The computer 1 is connected to other computers and databases via the network 10 using the network interface 5.

  FIG. 3 is a block diagram illustrating a processing flow of the software visualization apparatus 100. The database 2 includes a program A21 to be analyzed and a program B22. In this embodiment, as an example, two analysis target programs are provided in the database 2, but the number thereof is arbitrary.

  As an algorithm of the program visualization device, the memory 3 extracts a program to be analyzed (see FIGS. 5 and 7), analyzes the information, and outputs a graph structure including a set of nodes and links. (Program analysis means) 31, a structure analysis layer (structure analysis means) 32 that performs analysis processing based on the graph structure and outputs an analysis result, and an image display device that displays the graph structure and the analysis result in a predetermined drawing format And an image drawing layer (image drawing means) 33 to be displayed. The image drawing layer 33 accepts a selection operation of an analysis method and an image display method desired by the user with the mouse 7 and outputs a program analysis result based on the selection operation to the display 9.

  The program analysis layer 31, the structure analysis layer 32, and the image drawing layer 33 are configured by software (algorithm) that is independently made into parts, and can be changed in units of parts. Here, the change includes addition and deletion in addition to the software content change.

  FIG. 4 is a diagram showing a detailed configuration of the program analysis layer 31. The program analysis layer 31 includes a program A analysis unit 311 that is a program analysis unit for the program A21 and a program B analysis unit 312 that is a program analysis unit for the program B.

  Here, an example of program analysis performed by the program A analysis unit 311 will be described with reference to FIGS.

  FIG. 5 is a diagram showing a specific configuration according to an example of the program A21. The source code 211 of the program A21 includes a plurality of functions, and the program is configured by calling between functions.

  FIG. 6 is a diagram showing a program A graph structure 3111 that is an output result of the program A analysis unit 311.

  The algorithm of the program A analysis unit 311 performs program analysis using the function in the source code 211 (FIG. 5) as a node and the call relationship between functions as a link, and outputs the analysis result as a graph structure 3111 of the program A (FIG. 6). The content.

  As an example, the first to third lines of the source code 211 in FIG. 5 indicate that “function a calls function e”. In this case, the former (function a) is the start node and the latter (function e) is the end node. Similar analysis is performed for the subsequent lines, and a link ID (link identification number) is assigned to each inter-function call. There are various ways of attaching the link ID. Here, the order of youngest variable (alphabet) of the end node is given first priority, and then the order of youngest variable (alphabet) of the start node is given second priority (see FIG. 6). As a result, the above-mentioned “function a calls function e” (source code 211) indicates the calling relationship between functions as link ID 4 in the program A graph structure 3111 of FIG.

  Next, an example of program analysis performed by the program B analysis unit 312 will be described with reference to FIGS.

  FIG. 7 is a diagram showing a specific configuration according to an example of the program B22. The source code 221 of the program B22 has a plurality of functions. However, the program is configured not by calling relationships between functions but by variable substitution relationships.

  FIG. 8 is a diagram showing a program B graph structure 3121 that is an output result of the program B analysis unit 312.

  The algorithm of the program B analysis unit 312 performs program analysis using the variables in the source code 221 (FIG. 7) as nodes and the substitution relation between the variables as links, and outputs the analysis results as a program B graph structure 3121 (FIG. 8). Is the content.

  As an example, the second line of the source code 221 indicates that “the variable b is calculated based on the variable a”. In this case, the former (variable a) is the start point node and the latter (variable b) is the end point node. The same analysis is performed for the subsequent lines, and a link ID (link identification number) is assigned to the substitution relationship between the variables. The link ID is assigned in the same manner as the inter-function call described above. As a result, “the variable b is calculated based on the variable a” described above indicates the content as the link ID 1 in the program B graph structure 3121 of FIG.

As described above, the program visualization apparatus 100 is provided with a plurality of different program analysis units, for example, the program A analysis unit 311 and the program B analysis unit 312 as the plurality of program analysis layers 31, and the program analysis layer 31 and the structural analysis layer 32 are provided. By making the image drawing layer 33 into parts independently, it is possible to deal with a program having a different structure only by changing the program analysis unit. Therefore, when the program analysis unit is changed (including addition and deletion) according to the type of program, the program analysis layer 31 may be changed in units of each program analysis unit, and the change is another structural analysis. Since the layer 32 and the image drawing layer 33 are not affected, the reusability and development efficiency of the program visualization device are improved. Note that the structural analysis layer 32 and the image drawing layer 33 are also separately formed, and can be changed in units of each layer (this will be described later).

  FIG. 9 is a diagram showing a detailed configuration of the structural analysis layer 32. The structural analysis layer 32 includes a graph characteristic determination unit (graph characteristic determination unit) 320 and a program structure analysis unit 322. The graph characteristic determination unit includes a complex network determination unit 321.

  FIG. 10 is a diagram illustrating a detailed configuration of the complex network determination unit 321. A complex network is a scale-free property (power law of degree distribution) as a certain common property in various networks that exist in the real world (the network is represented by a graph structure consisting of a set of nodes and links), Since it has “small world” and “cluster”, and is a well-known study used to explain various phenomena in the real world in recent years, detailed description thereof will be omitted.

  The complex network determination unit 321 includes a scale-freeness determination unit 3211 that determines whether or not the degree distribution of the graph structure analyzed by the program analysis layer 31 is in accordance with a power law, and the arbitrarily selected two nodes. To determine whether there is a connection through intermediary nodes and links (distance between nodes), change this combination to all nodes and calculate the average shortest distance, and the average shortest distance is the size of the number of nodes A small world property determination unit 3212 that determines whether or not the value is smaller than that of the graph, and a cluster property determination unit 3213 that evaluates the value of the cluster coefficient of the graph structure.

  FIG. 11 is a diagram illustrating a processing flow of the complex network determination unit 321. Processing starts in step S32101.

  In step S32102, the scale freeness determination unit 3211 is executed to determine whether or not there is scale freeness. If it is determined that there is a scale-free property, the process proceeds to step S32103. If it is determined that there is no scale-free property, the process proceeds to step S32106, and the process ends.

  In step S32103, the small world property determination unit 3212 is executed to determine whether there is a small world property. If it is determined that there is a small world property, the process proceeds to step S32104. If it is determined that there is no small world, the process proceeds to step S32106 and the process ends.

  In step S32104, the clusterity determination unit 3213 is executed to determine the presence or absence of clusteriness. If it is determined that there is a cluster property, the process proceeds to step S32105. If it is determined that there is no cluster, the process proceeds to step S32106, and the process ends.

  If the process proceeds to step S32105, it is determined in this step that the graph structure of the analysis target program is a complex network, and the process proceeds to step S32106 and the process ends.

FIG. 12 is a diagram illustrating a processing flow of the scale-freeness determination unit 3211. The scale-free property is expressed when the probability distribution of the probability p (k) having a degree k of a node becomes a power law of p (k) ∝k− ^γ , and the following processing is performed for the determination. Done.

  Processing starts in step S32111.

  In step S32112, based on the degree that is the number of links connected to each node of the analysis target program, the degree distribution that is the correspondence between the degree and the number of nodes is acquired.

  In step S32113, the correlation coefficient of the degree distribution in both logarithms is calculated.

  In step S32114, it is determined whether the correlation coefficient exceeds a predetermined threshold. If so, the process proceeds to step S32115. If not, the process proceeds to step S32116 and the process is terminated.

  In step S32115, it is determined that the graph structure of the analysis target program has a scale-free property, and the process proceeds to step S32116 and the process ends.

  The determination process performed by the small world determination unit 3212 is not shown in the flowchart, but is performed roughly as follows. That is, a step of obtaining the number of nodes, a step of obtaining an average shortest distance between any two nodes, and determining whether or not the average shortest distance between the nodes is smaller than the number of nodes. However, in the case of taking a small value, there is a step of determining that it has a small world property.

  The determination process performed by the clusterity determination unit 3212 is also performed as follows, although not shown in the flowchart. In other words, the method includes a step of calculating a cluster coefficient C from a set of nodes and links in a graph structure, a step of determining whether or not the cluster coefficient is larger than a threshold value, and a step of evaluating the cluster coefficient if it is larger.

  FIG. 13 is a diagram illustrating a detailed configuration of the program structure analysis unit 322 in the structure analysis layer 32. The program configuration analysis unit 322 is composed of a plurality of types of analysis units that are independently made into parts. Each of these analysis units is configured by an analysis algorithm, and includes the following. For example, (i) a community structure analysis unit 3221 that analyzes a community (organization) made up of a specific common node and link in a graph structure made up of a network of nodes and links, and (ii) nodes that make up the network Among them, a node importance analysis unit 3222 for analyzing a node having a high importance level, and (iii) a node influence range analysis unit 3223 for analyzing an influence range of the node on other nodes by paying attention to a specific node; (Iv) an inter-node shortest path analysis unit 3224 that analyzes the shortest path when passing through several intermediary nodes and links that are linked from a specific node to another specific node; v) A structural similarity analysis unit 322 that analyzes the graph structure of a plurality of programs and analyzes the similarity of the graph structures. If, (vi) mediated centrality, i.e., mediated central analysis unit 3226 that analyzes the degree that can not reach other nodes when not passing through the node are prepared. Furthermore, the program structure analysis unit 322 includes a structure analysis algorithm attribute storage unit 3227 that stores attributes of the various analysis units (structure analysis algorithms) listed above.

  These various analysis units (structure analysis algorithms 3221 to 3226) can be selectively designated through the analysis technique display unit 913 of the display screen 91 shown in FIG. 1 (details of the display screen 91 will be described later).

  A plurality of structure analysis algorithms (3221 to 226) serving as the program structure analysis unit 322 are independent of each other, and are independent of the program analysis layer 31 and the image drawing layer 33. Therefore, even when the structure analysis algorithm is changed (including addition and deletion), the program analysis layer 31 and the image drawing layer 33 can be reused, and the development efficiency of the program visualization device can be improved.

  FIG. 14 is a diagram showing a detailed data configuration 32271 of the structure analysis algorithm attribute storage unit 3227. The data structure 32271 of the structure analysis algorithm attribute storage unit 3227 includes an analysis algorithm ID, the name of the algorithm, and an applicable graph structure. For example, the ID1 community structure analysis algorithm can be applied to the graph structure of a complex network dedicated program, and the ID4 internode shortest path analysis can be applied not only to the network dedicated program but also to a general-purpose graph structure.

  FIG. 15 is a diagram showing a detailed configuration of the image drawing layer 33.

  The image drawing layer 33 includes (i) a graph diagram drawing unit 331 for drawing the graph structure of the analysis target program with a specified drawing algorithm (in this example, any drawing algorithm of an effective graph, a dependency matrix, or a tree); (Ii) an analysis technique display creation unit 3321 for creating a list of names of the structure analysis algorithms (3221 to 3226 in FIG. 13: analysis technique) of the analysis target program, and (iii) the above drawing algorithm (3311 to 3313: A drawing format display creation unit 3322 for creating a list of names of (drawing formats), and (iv) a user interface unit 332 for accepting a user's selection (designation) regarding the analysis method and the drawing format through the screen of the display 9. Prepare.

  The drawing algorithm constituting the graph diagram drawing unit 331 of this embodiment includes a directed graph drawing algorithm (directed graph drawing unit) 3311 for displaying a graph structure composed of nodes and links by giving directions to the links, and a dependency relationship between the nodes. A dependency matrix drawing algorithm (dependency matrix drawing unit) 3312 that displays in a matrix manner and a tree drawing unit 3313 that draws node relationships in a tree shape via links are provided.

  These various drawing units (drawing algorithms 3311 to 3313) can be selectively designated through the drawing format display unit 924 of the display screen 91 shown in FIG. 1 (details of the display screen 91 will be described later).

  A plurality of drawing algorithms (3311-1313) constituting the graph diagram drawing unit 331 are independent of each other, and are independent of the program analysis layer 31 and the graph structure analysis layer 32. Therefore, even when the drawing algorithm is changed (including addition and deletion), the program analysis layer 31 and the graph structure analysis layer (structure analysis unit) 32 can be reused to improve the development efficiency of the program visualization device. Can do.

  The analysis technique display creation unit 3321 creates a list display of the structure analysis algorithms (analysis techniques) for analyzing the graph structure, and each structure analysis algorithm is analyzed with respect to the analyzed graph structure from the characteristics of the analysis target program. It has a structure analysis algorithm suitability judging function (structure analysis algorithm suitability judging means) for judging whether or not it is applicable. This determination is performed based on the data structure 32271 of the structure analysis algorithm attribute storage unit 3227 in FIG. 14, and the determination result is reflected in the creation of a list display of the structure analysis algorithm.

  FIG. 16 is a diagram showing the flow of processing of the structural analysis algorithm suitability determination function executed by the analysis technique display creation unit 3321.

  The analysis technique display creation unit 3321 starts structural analysis algorithm suitability determination processing from step S33211 after creating a structure analysis algorithm (analysis technique) list display as shown in the analysis technique display unit 913 of FIG.

  In step S33212, 1 is set to the ID of the structure analysis algorithm. ID1, ID2, ID3... IDn correspond to the analysis algorithm names in the list display shown in FIG.

  In step S33213, the network characteristic of the current analysis target program is acquired from the graph characteristic determination unit 320.

  In step S33214, based on the data in the structure analysis algorithm attribute storage unit 3227, it is determined whether or not the structure analysis algorithm of the current ID is applicable to the network characteristics of the analysis target program. If applicable, the process proceeds to step S33215, and if not applicable, the process proceeds to step S33216.

  In step S33215, the fact that the current ID structure analysis algorithm is applicable is displayed on the display 9 (analysis technique display unit 923 in FIG. 1).

  In step S33216, the fact that the structure analysis algorithm of the current ID is not applicable is displayed on the display 9 (analysis technique display unit 923 in FIG. 1).

  In step S33217, it is determined whether the structure analysis algorithm is the final ID. If it is the final ID, the process proceeds to step S33219, and the process is terminated. If it is not the final ID, the process proceeds to step S33218.

  In step S33218, 1 is added to ID, and the flow advances to step S33214. The determination result of the suitability of the above structural analysis algorithm is reflected on the screen display of the display of the program visualization device 100 described below.

  The drawing format display creation unit 3322 creates a list display of drawing format names and applies each drawing format (drawing algorithms 3311 to 313 in FIG. 15) to the graph structure analysis executed in the structural analysis layer 32. A drawing format suitability judging function (drawing format suitability judging means) for judging whether or not it is possible is provided. This discrimination result is reflected in creating a list display of a drawing algorithm (drawing format).

  FIG. 1 is a diagram showing a screen 91 in which the analysis result of the program A21 is displayed on the display 9 by the program visualization apparatus 100 according to the present configuration. As shown in the figure, a graph structure display unit (first window) 911, an analysis target program display unit 912, an analysis method display unit (second window) 913, and a drawing format (drawing method) are displayed on one screen 91. ) Display section (third window) 914 is set.

  The graph structure display unit (first window) 911 is created by the graph diagram drawing unit 331 shown in FIG. In FIG. 1, as an example, the node influence range is displayed in the directed graph format as the analysis result of the program A in the graph structure display unit 911. Since the mouse cursor points to the node d, the variable connected by the link that is affected by the change of the node d is highlighted in black. That is, in the graph structure display unit (first window) 911, the graph structure of the program analysis result and the analysis result of the structure analysis algorithm applied to the graph structure are designated by the image drawing layer 33. It is synthesized and displayed via a drawing format.

  As an example, the analysis target display unit 912 displays an analysis target program name “program A” and a graph structure characteristic “non-complex network”.

  An analysis technique display unit (second window) 913 displays the name of an analysis technique (structural analysis algorithm) used for structural analysis. Further, the analysis method display unit 913 includes a function for designating a desired analysis method through a screen from the names of a plurality of analysis methods (structure analysis algorithms) displayed in a list, and an analysis method display creation unit (structure analysis algorithm suitability determination unit) ) 3321 is displayed (see FIGS. 15 and 16).

  For example, as a function of designating a desired analysis technique in the analysis technique display unit (second window) 913, as shown in FIG. 1, a method of clicking a designation mark input part of a desired analysis technique with a mouse pointer (in FIG. 1) , Click the designation mark input part of the node influence range analysis). The designation method may be performed by double-clicking the name of a desired analysis technique instead of the designation mark input unit, or by clicking the name of each analysis technique indicated by a button. In the graph structure display unit (first window) 911, the analysis result of the analysis method specified by the analysis method display unit (second window) 913 is displayed together with the graph structure.

In this embodiment, the analysis algorithm suitability determination result executed by the analysis technique display creation unit 3321 (FIG. 15) is displayed superimposed on the name of the analysis algorithm (analysis technique). For example, in FIG. 1, as an example, the analysis target program A (graph characteristic: non-complex network) is not applicable to community structure analysis and node importance analysis, and a strikethrough is drawn on their names. Applicable analysis algorithms are displayed as they are without a strikethrough. Furthermore, in this embodiment, the analysis algorithms that can be applied to the analysis target program and the analysis algorithms that cannot be applied are displayed by changing the display order. In FIG. 1, the screen after the display order is changed is shown, and an analysis method that cannot be applied to the analysis method is moved to a lower position in the second window 913 . In addition, a present Example is an example of the display method, The display method is not limited to the thing of an Example. For example, the analysis method applicable to the analysis target program and the analysis method that cannot be applied may be identified and displayed using at least one of hue, lightness, and saturation.

  A drawing format display section (third window) 914 shows a list of drawing formats to be specified. Here, as an example, three drawing formats of an effective graph, a tree, and a dependency relationship matrix are displayed as a list. The drawing format display unit 914 has the following functions in addition to displaying a list of drawing formats. One has a function of designating a desired drawing format through a screen from names of a plurality of drawing formats displayed in a list. This designation method may be performed in the same manner as the method of designating a desired analysis technique from the above-described analysis technique list. The other has a function of reflecting the result of the drawing format suitability determination executed by the drawing format display creating unit 3322 (FIG. 15) in the drawing format list display. Also in this case, as in the case of the suitability of the analysis algorithm described above, identification display may be performed using any one of a cancellation line, shading, hue, brightness, and saturation. Furthermore, the display priority order may be changed according to the suitability determination result.

  As shown in this example, through the screen display, (ii) by switching the display method between the analysis method that can be applied to the analysis target program and the analysis method that cannot be applied, (ii) it is possible to draw according to the graph structure By switching the display method between a simple drawing format and an impossible drawing format, a user who does not have expertise in graph theory or statistical analysis can select an appropriate analysis method and drawing format.

  FIG. 17 is a diagram showing a screen 92 in which the analysis result of the program B22 is displayed on the display 9 by the program visualization device 100 according to the present configuration. The screen 92 includes a graph structure display unit 921, an analysis target program display unit 922, an analysis technique display unit 923, and a drawing format display unit 924.

  In the graph structure display unit 921, the community structure is displayed in the directed graph format as the analysis result of the program B.

  The analysis target display unit 922 displays the analysis target program name “program B” and the graph structure characteristic “complex network”.

  The analysis technique display unit 923 shows an analysis technique used for the structural analysis. Program B is a complex network, and all analysis methods can be selected in this embodiment.

  The drawing format display unit 924 shows a list of drawing formats.

  The display forms and functions of these display units 921 to 924 are the same as those of the display units 911 to 914 in FIG.

  In the example of FIG. 1, the graph structure of the node influence range analysis is displayed as a directed graph as an analysis method, and in the example of FIG. 17, the community structure analysis is displayed as a directed graph as an analysis method. The drawing display of graph structure analysis by other analysis methods and drawing formats is not shown. The reason is that other analysis methods and drawing formats are well known per se, and drawing creation itself can be easily carried out by those skilled in the art without showing specific examples in this specification. Since it is a thing, illustration is abbreviate | omitted. Incidentally, the drawing of the dependency matrix is disclosed in Patent Document 4 introduced in the prior art of this specification.

100: Program visualization device, 31: Program analysis layer (program analysis means), 32: Graph structure analysis layer (graph structure analysis means), 33: Image drawing layer (image drawing analysis means), 91: Display screen, 911: Graph Structure display unit (first window), 912: analysis target program display unit, 913: analysis technique display unit (second window), 914: drawing format display unit (third window).

Claims (17)

A program visualization device that extracts information from a program, performs processing necessary for program analysis using the extracted information, and displays an analysis result,
Program analysis means for analyzing the program and outputting a graph structure composed of nodes and links, structure analysis means for performing analysis processing on the graph structure and outputting an analysis result, the graph structure and its analysis result Image rendering means for displaying on the image display device in accordance with the rendering format of
These three means are software executed via a computer, and are configured as components that can be changed in units of each means independently .
The structure analysis means has a plurality of structure analysis algorithms for executing different analysis methods on the graph structure,
The program visualization device includes structural analysis algorithm suitability determination means for determining whether each of the structural analysis algorithms is applicable to the graph structure from the characteristics of the analysis target program,
The image display device has at least one screen.
(I) a first window for synthesizing and displaying the graph structure and an analysis result of a structure analysis algorithm applied to the graph structure through a drawing format of the image drawing means; (ii) the plurality of structures A function of displaying a name of an analysis algorithm, a function of designating a desired structure analysis algorithm through a screen from a plurality of displayed names of the structure analysis algorithm, and a function of displaying a determination result by the structure analysis algorithm suitability determination unit A program visualization device in which two windows are set. In claim 1,
The program analysis means has a plurality of program analysis algorithms for analyzing each of a plurality of different types of programs and creating a graph structure, and each of these program analysis algorithms is divided into parts and changed for each program analysis algorithm unit. Configured and possible
The structure analysis means has a plurality of structure analysis algorithms for executing different analysis methods on the graph structure, and each of these structure analysis algorithms is made into a component and can be changed for each structure analysis algorithm. Composed of
The program visualizing device, wherein the image drawing means has a plurality of drawing algorithms for executing different drawing formats, and each drawing algorithm is made into a component and can be changed for each drawing algorithm. In claim 1,
The image drawing means has a plurality of drawing algorithms for executing different drawing formats,
The program visualization device includes a drawing format suitability determining unit that determines whether or not a drawing format executed by each of the drawing algorithms is applicable from the characteristics of the analysis method of the structure analyzing unit,
In addition to the first and second windows, the image display device has one screen,
(Iii) A function for displaying a name of a drawing format executed by the drawing algorithm and designating a desired drawing format through a screen from a plurality of displayed drawing format names, and a determination result by the drawing format suitability determining means A program visualization apparatus in which a third window having a function of displaying is set. In any one of Claims 1 thru | or 3,
The said structure analysis means is a program visualization apparatus provided with the graph characteristic determination means which classify | categorizes the characteristic of the graph structure of an analysis object program. In claim 4,
The program visualization device, wherein the image drawing means changes a name display form of an analysis technique of the structure analysis means on a screen based on a determination result of the graph characteristic determination means. In claim 5,
The program analysis means analyzes the program using a variable as a node and a substitution relationship between variables as a link, or a function as a node and a call relationship between functions as a link, depending on the type of analysis target program. A program visualization device that analyzes programs. In claim 5,
The program visualization apparatus comprising a complex network determination unit that determines whether the analysis target program has a complex network characteristic based on the graph structure. In claim 1,
The structure analysis means comprises at least one structure analysis algorithm, and has a structure analysis algorithm attribute storage means for storing network characteristics to which the structure analysis algorithm can be applied. A program visualization device that extracts information from a program, performs processing necessary for program analysis using the extracted information, and displays an analysis result,
Program analysis means for analyzing the program and outputting a graph structure composed of nodes and links, structure analysis means for performing analysis processing on the graph structure and outputting an analysis result, the graph structure and its analysis result Image rendering means for displaying on the image display device in accordance with the rendering format of
The structure analysis means has a plurality of structure analysis algorithms for executing different analysis methods on the graph structure,
The program visualization device includes structural analysis algorithm suitability determination means for determining whether each of the structural analysis algorithms is applicable to the graph structure from the characteristics of the analysis target program,
The image display device has at least one screen.
(I) a first window for synthesizing and displaying the graph structure and an analysis result of a structure analysis algorithm applied to the graph structure through a drawing format of the image drawing means; (ii) the plurality of structures A function of displaying a name of an analysis algorithm, a function of designating a desired structure analysis algorithm through a screen from a plurality of displayed names of the structure analysis algorithm, and a function of displaying a determination result by the structure analysis algorithm suitability determination unit A program visualization device characterized in that two windows are set. In claim 9,
The image drawing means has a plurality of drawing algorithms for executing different drawing formats,
The program visualization device includes a drawing format suitability determining unit that determines whether or not a drawing format executed by each of the drawing algorithms is applicable from the characteristics of the analysis method of the structure analyzing unit,
In addition to the first and second windows, the image display device has one screen,
(Iii) A function for displaying a name of a drawing format executed by the drawing algorithm and designating a desired drawing format through a screen from a plurality of displayed drawing format names, and a determination result by the drawing format suitability determining means A program visualization apparatus in which a third window having a function of displaying is set. In claim 9 or 10,
In the second window, for the structural analysis algorithm that cannot be applied to the analysis target program, the image display device visualizes a program that displays information for identifying that the structural analysis algorithm cannot be applied to the name of the structural analysis algorithm. apparatus. In claim 9 or 10,
In the second window, the image display device displays a structural analysis algorithm that can be applied to an analysis target program and a structural analysis algorithm that cannot be applied to the analysis target program by changing at least one of hue, brightness, and saturation. apparatus. In any one of claims 9 to 12,
The image display device is a program visualization device that displays a structural analysis algorithm that can be applied to an analysis target program and a structural analysis algorithm that cannot be applied to the analysis target program by changing a display order in the second window. In any one of claims 9 to 13,
In the second window, the image display apparatus is a program visualization apparatus in which a function for making it impossible to select a structural analysis algorithm that cannot be applied to an analysis target program through a screen is set. In claim 10,
In the third window, the image display device superimposes and displays information for identifying that the drawing format that cannot be applied to the analysis target program is not applicable to the name of the drawing format in the third window. In claim 10,
The image display device is a program visualization device that displays, in the third window, a drawing format applicable to the analysis target program and a drawing format that cannot be applied by changing at least one of hue, brightness, and saturation. In any one of Claims 9 thru | or 16,
The image display device is a program visualization device that displays network characteristics of an analysis target program on one screen together with the first, second window or the first, second, and third windows.

Images