JPH01124060A - Device for forming network graphic - Google Patents

Abstract

PURPOSE:To automatically lay out a network graphic while holding the easiness of ovservation by defining the relation of nodes defined like a network temporarily as the relation of tree structure, laying out the whole nodes and connecting these nodes through lines. CONSTITUTION:A line (h) out of lines D g E h D constituting a closed loop is cut out to form a directional graph having no closed loop. In the directional graph, the longest path A B D E G is defined as the trunk of the tree structure and nodes C, F, J, H to be applied to the trunk area applied in the order of the route. Consequently, the arrow direction of the network line is fixed and the operator's observation of the network graphic can be allowed to coincide with the flow direction of the layout. Consequently, rapid layout can be attained while keeping the easiness of observation.

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention relates to a network diagram creation device, and more particularly to a network diagram automatic layout creation device that greatly improves work efficiency while ensuring ease of viewing. It is.

[Conventional technology]

Conventionally, the following three methods have been used in network diagram creation work.

The first method is a method similar to a CAD system in which all the information necessary for the layerer 1 is provided to the operator (worker) and the layout is executed based on the input layout information.

The second method is a method in which the operator arranges the parts other than the lines, executes the layout based on the input layout information, and connects the parts by lines when the layerer 1~ is completed. The second method will be explained with reference to FIGS. 2 and 3. In FIG. 2, A-J are nodes to be laid out, and a-0 are lines connecting the nodes.

The numbers enclosed in squares represent the strength of the connections between nodes.

The higher this number, the stronger the connection. The specific operating procedure is shown in Fig. 3. At stage A, first, the system for Layer 1 is activated and the space to be laid out is displayed. In step B, nodes A to 1, etc. are set on the displayed space. At this time, the system stores each node and the location where it should be displayed. At stage C, the system automatically generates the necessary lines between each node.

The third method examines all combinations on the layout,
For example, an LSI that executes a layout that minimizes the area
It is a method similar to CAD.

[Problem that the invention seeks to solve]

In the first and second methods described above, it is necessary to store not only the relationship between the nodes of the network but also the positional information on the layout of the nodes.

Therefore, when nodes are added or deleted, it is necessary to redo the layout and storage of position information. In addition, if the layout space changes, or if you output the same information to devices with different display areas or display methods and need to execute the optimal layout for each device, you can simply redo the layout for each device. Instead, you have to memorize all the layouts for each device.

On the other hand, in the third method, trial and error is performed many times on the layout, and it takes a lot of time to determine the final layout. Therefore, even if you try to change data or perform layout operations interactively on the terminal screen, the response is slow and unusable. In addition, because the layout is done mechanically without considering the strength of the connection between each node, nodes that are originally intended to be laid out together semantically may be laid out apart, which impairs the legibility of the diagram. There was a drawback that it was damaged.

An object of the present invention is to provide a network diagram creation device that allows nodes with semantic connections to be laid out in close proximity using only semantic information of the network and without substantial trial and error. .

[Means for solving problems]

The above object is achieved by once considering the relationships between nodes defined in a network as relationships in a tree structure, laying out all the nodes, and connecting the nodes with lines.

That is, in order to achieve the above object, the present invention provides a device for inputting data having a network-like relationship, a layerer 1-device for laying out the relationship of the data as a network diagram, and a layerer 1-device for outputting the laid-out network diagram. In the screw 1 to wire 9 diagram creation device, which is configured with an output device for data
A means for converting a work-like relationship into a directed graph, a means for deleting the inverse directed graph of a closed loop when a closed loop exists in the directed graph, a means for examining the longest path of the directed graph from which the closed loop has been removed, and a method for determining the lower order of the longest path. A screw 1 comprising means for converting a directed graph into a tree structure by sequentially attaching related nodes to its longest path, and means for connecting each node of the tree structure with a layerer 1 and connecting each node with a line. - We propose a diagram creation device.

The means for converting into a directed graph includes a means for storing the strength of the semantic connection of each effective graph, and the means for checking the longest path of the directed graph preferentially employs the digraph with the strong semantic connection at the branch node. It is desirable to have a means to do so.

[Effect]

When converting from a network to a tree structure, the network-like specification is treated as a directed graph that does not include closed loops, and then the longest path of the directed graph is taken as the tree trunk. Nodes are sequentially added to this trunk starting from the node that is related to the lowest node of the nodes on this needle. With this method,
The direction of the arrow on the network line is a constant direction. When the arrows have a fixed direction, the way the operator views the network diagram can match the flow direction of the layout. - Also, when determining the longest path of a tree structure, for nodes that need to be laid out nearby, priorities are set for lines between nodes, and nodes with high priorities are preferentially adopted as the trunk.

In this way, since the network layout is reduced to a tree-structured layout, high-speed layout is possible while maintaining visibility. In other words, in the conventional method,
As the number of nodes increases, the amount of calculation for layout increases exponentially, but in the present invention, the amount of calculation is proportional to the number of nodes.

Furthermore, since the tree structure is created based on the longest path, the lines between nodes can be made into natural shapes.

Regarding the example of FIG. 2, an overview of conversion from a directed graph to a tree structure in the present invention will be explained with reference to FIGS. 4 to 6.

Here, it is assumed that nodes are input together with semantic connection information between each node, provisional arrowing is completed as a directed graph, and FIG. 2 is obtained.

First, l) −+ g→ which is a closed loop in Figure 2
E→h-) Cut the line D to create a directed graph without closed loops. The digraph after the cut is shown in FIG.

Next, examine the longest path in the directed graph of FIG. The longest path is A→B→[) −> E −> Q. However, if the priority (f is 1 and g is 0) is evaluated at D, then A→B−+D→G. This will be the trunk of the tree structure. Figure 5 shows the trunk of the tree structure.

Therefore, we examine the nodes attached to the trunk in order from the leaves to the root of the tree structure. In Figure 5, A −) B −+ D→E→
When G is used as the trunk, C, F, J, and H become corresponding nodes. This is shown in FIG. 6(a).

Also, in the case of A→B-+D-+G, C, E, J.

H is the corresponding node. This is shown in FIG. 6(b).

Furthermore, it continues to examine nodes attached to the trunk until all nodes are connected to the trunk.

〔Example〕

Next, the configuration of an embodiment of the network diagram creation apparatus according to the present invention will be described with reference to FIG.

This device includes an input device 10, a layout device 12, and an output device 28. The layout device 12 includes an input device driver 14, an arithmetic unit 16, an output device driver 18, and a working memory 26. The calculation unit 16 includes a specification input unit 20 and a layout unit 22, and the working memory 26 includes a specification storage table 24.

The input device driver 14 converts data from the input device 10 into a data format that the calculation unit 16 can understand. on the other hand,
The output device driver 18 outputs a layout diagram or the like in a signal format suitable for the operation of an output device (such as a CRT display device or an XY plotter) 28 .

Note that the device on the input device 10 side and the device on the output device 28 side may share a single CRT display device, keyboard, etc., and input data and output figures may be freely displayed in a multi-window format. be.

Data regarding the object to be laid out input from the input device 10 is sent to the input device driver 14 and the specification input unit 20.
It is stored in the specification storage table 24 via. The layout unit 22 uses the data to calculate a layout diagram, and outputs the result to the output device 28 via the output device driver 18.

An example of the specification storage table 24 is shown in FIG.

This specification storage table 24 is not a so-called image memory, but is in a table format. The specification storage table 24 stores the nodes of the network to be created on the vertical and horizontal axes, and indicates that there is a layout line between these nodes by setting a flag 1 at the intersection of the table. That is, the existence of a line in the direction from the starting point node 301 (for example, 5 points) to the ending point node (for example, point E) on the vertical axis is expressed by setting 1 at the intersection 303, and if there is no line, "-" is stored. Further, if a line exists, the priority level 304 of the line is stored as an integer such as 1°O. The larger this number, the higher the priority. For example, since D→G is 1 and D→E is O, D−+G has a high priority.

The layout section 22 operates based on the specification storage table 24 of FIG. 7, using the layout table of FIG. 8, and the algorithm of FIG. 9.

FIG. 8 is a layout table that represents the layouts of FIGS. 2, 4, etc. in tabular form. The layout table has the nodes that make up the network on the vertical and horizontal axes. At the intersection of the vertical axis and the horizontal axis, a node number area 401 that stores how many nodes exist at the end of each line, and a classification flag 402 that distinguishes the line used as the trunk of the tree structure are stored. For example, when looking in the direction of node B from node A, there are four nodes B, D, E, and F, so the number to be stored in the node number area at the intersection of starting point node A and ending point node B is 4. Become. Similarly, in the direction from A to J, J, E,
It becomes 3 of F. However, 0 is stored in the deleted portion of the reverse directed graph from E to D. Classification flag 402
By tracing from the starting point node A in 1, to B in 2, to the glue in 3, and to E in 4, you can find the node that became the trunk.

Next, the algorithm shown in FIG. 9 will be explained. It is assumed that the state shown in FIG. 7 is that information indicating the interrelationship of nodes constituting a network is input from the input device (a state in which a flag 1 is set at the intersection of related nodes and a priority value is also input).

Step 100 searches for networks.

Let O be the value in the node number area of the layout table for a line that is a closed loop. this is.

When a closed loop exists in a directed graph, this is the stage of deleting the inverse directed graph of the closed loop.

In step 110, a starting point node with no lines is found in the specification storage table shown in FIG. 7, and all lines that have that node as an end point node are stored. The starting nodes with no lines are C9F, G, and I. Therefore, lines ending at these nodes are stored.

In step 120, in the layout table of FIG. 8, the value of the node number area of the portion corresponding to the line stored in step 110 is set to 1. For example, 1 is stored at the intersection of 13-)C.

In step 130, it is determined whether all values in the number of nodes area 401 of the layout table in FIG. 8 have been determined.

If not yet, go to step 140 and store the starting point node for which all values in the line node number area have been determined in the layout table.

In step 150, the maximum value of the node number area among the lines coming from each starting point node stored in step 140 is stored.

In step 160, the value in the number of nodes area of the line whose end point is the starting point node stored in step 140 is set to step 1.
The value stored in 50 is +1. In other words, it memorizes that it is a higher level node.

If these steps are continued in sequence, a state will be reached in step 130 in which all values in the number of nodes area of the layout table have been determined.

Steps 10 to 160 are steps for converting the network-like relationship of input data into a directed graph.

In step 170, the node name of the end node that does not have a line is stored. In this case, it is A.

In step 180, the classification flag of the line having the highest priority and the largest value in the node number area among the lines whose starting point node is the node in step 170 is set to 1. A → B corresponds to this.

In step 190, the node name set in step 180 is memorized, and step 180 is repeated until the node for which all values in the number of nodes area are not defined (the classification flag continues to count up).

Steps 170 to 190 are stages for creating the trunk of the tree structure. Here, A→B-+D→E is created as the trunk.

In step 200, a search is made for end nodes for which no classification flag is set. That is, the tree structure is created starting from the lowest node.

In step 210, all starting point notes corresponding to the end point node line found in step 200 are examined. If only one classification flag is defined for the starting point node, that starting point node is stored. If there are two or more classification flags defined for the starting point node, the starting point node with the smaller numerical value of the classification flag is stored. This is because in a tree structure, multiple lines can exit from one node, but multiple lines cannot enter.

In step 220, the value of the division flag of the end point node found in step 200 of the line corresponding to the start node stored in step 210 is set to the value of the division flag of step 2]0 +
Set to 1. That is, the value of the division flag is increased as the hierarchy goes down, and the directed graph is converted into a tree structure.

Steps 200 to 220 are steps for converting a directed graph into a tree structure by sequentially attaching related nodes to the longest path from lower nodes to the longest path.

In step 230, nodes are laid out based on the layout table.

In step 240, lines are drawn based on the layout table.

In this way, the layout diagram is completed.

The data of the layout diagram is sent to the output device and displayed.

〔Effect of the invention〕

According to the present invention, it is possible to automatically lay out a network diagram while ensuring visibility, so even a diagram in which only semantic relationships are defined, such as a software specification, can be output as a network diagram. Compared to the conventional method of inputting, managing, and modifying image information,
Work efficiency is greatly improved.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the configuration of an embodiment of a network diagram creation device according to the present invention, FIG. 2 is a diagram showing an example of information to be laid out in a directed graph format, and FIG. 3 is an example of a conventional layout procedure. Figure 4 is a diagram showing an example of a directed graph, Figure 5 is a diagram showing the trunk of a tree structure, Figure 6 is a diagram showing how to check nodes attached to the trunk from leaves, and Figure 7 is a specification storage table. FIG. 8 is a diagram showing a layout table, and FIG. 9 is a diagram showing an algorithm of the layerer 1-device of the present invention. 10... Input device, 12... Layout device, 16
... Arithmetic section, 20... Specification input section, 22... Layout section, 24... Specification storage table, 26... Working, memory, 28... Output device, 301... Starting point node, 302... End point node, 303... Intersection, 304... Priority, 401... Node number area, 4
02... Classification flag.

Claims (1)

[Claims] 1. A network diagram comprising a device for inputting data having a network-like relationship, a layout device for laying out the relationship of the data as a network diagram, and an output device for outputting the laid out network diagram. In the creation device, the layout device includes: means for converting the network-like relationship of the data into a directed graph; means for deleting an inverse directed graph of the closed loop when a closed loop exists in the directed graph; and a method in which the closed loop has been removed. means for examining the longest path of a directed graph; means for converting the directed graph into a tree structure by sequentially adding related nodes from lower nodes of the longest path to the longest path; and means for laying out each node of the tree structure and connecting each node. 1. A network diagram creation device, comprising: means for connecting the two with lines. 2. In claim 1, the means for converting into a directed graph includes means for storing the strength of the semantic connection of each effective graph, and the means for examining the longest path of the directed graph includes converting the plurality of valid graphs into a directed graph at a branch node. A network diagram creation device characterized by including means for preferentially adopting a directed graph with strong semantic connections.