Smart shape insert

BACKGROUND

0001Computing devices executing the appropriate computer software are often used to create information graphics such as, but not limited to, diagrams, charts, graphs, and maps. For ease of discussion an information graphic is generically referred to herein as a “diagram.” A diagram is a visual representation of a data model, i.e., a software structure in which computer software programs store persistent information, e.g., a data graph. Computer software programs that can create diagrams visually representing data models are generically referred to herein as “diagramming applications.”

0002Diagramming applications are usually supported by a graphical user interface (GUI). A GUI is a computer software component, executed by a computing device, that presents visual elements in a display. A typical diagramming application provides a plurality of types of information graphics, i.e., diagram types, to enable the representation of different aspects of data models. Indeed, a diagramming application becomes more useful with each diagram type the diagramming application provides. Often, diagrams of differing types not only differ in appearance but also in how the diagrams are assembled. A command or control in a diagramming application used to insert visual elements into diagrams may not produce the same results in diagrams of differing types. Many diagram types often require a large number of actions to insert and position visual elements in a diagram.

SUMMARY

0003This summary is provided to introduce a selection of concepts in a simplified form that are further described below in the Detailed Description. This summary is not intended to identify key features of the claimed subject matter, nor is it intended to be used as an aid in determining the scope of the claimed subject matter.

0004A method and apparatus, including computer-readable medium, for determining the location of a shape inserted into a diagram representing a data graph is disclosed. The method comprises collecting information about a starting shape in the diagram; selecting a diagram definition in accordance with the diagram's type; extracting information from the diagram definition; and using the information extracted from the diagram definition and the information about the starting shape.

0005The information extracted from the diagram definition and the information about the starting shape is used to determine the type of node to associate with the inserted shape, the location of the associated node in the data graph, and the inserted shape's location in the diagram.

0006Information about the starting shape includes, but is not limited to, the starting shape's type, associated node, and parent node. Information about the starting shape also includes the number of child nodes of the starting shape's node and the number of child nodes of the parent node of the starting shape's node. The information extracted from the diagram definition comprises the preferred number of child nodes for the starting shape's node and the preferred number of child nodes for the parent node of the starting shape's node.

0007The preferred number of child nodes of the starting shape's node is compared to the number of child nodes of the starting shape's node. The preferred number of child nodes for the parent node of the starting shape's node is compared to the number of child nodes of the parent node of the starting shape's node. The results of the comparisons are used to determine the type of node to associate with the inserted shape, the location of the associated node in the data graph, and the inserted shape's location in the diagram.

0008The results of the comparisons are also used to determine whether or not to disable user interface controls for inserting shapes in a diagram. The insertion of a shape based on the results of the comparison are able to be overridden by a selection in a user interface control for inserting shapes in a diagram.

0009With no modification, the method is able to provide smart shape insertion for diagram types for which there are diagram descriptions.

0010A starting shape in a diagram may be a selected shape or predetermined default shape. An inserted shape may be a shape, a semantic bullet, or a format bullet.

DESCRIPTION OF THE DRAWINGS

0011The foregoing aspects and many of the attendant advantages of this invention will become more readily appreciated as the same become better understood by reference to the following detailed description, when taken in conjunction with the accompanying drawings, wherein:

0012<figref idref="DRAWINGS">FIG. 1</figref> is a pictorial illustration of an exemplary pane in an exemplary GUI containing an exemplary radial diagram;

0013<figref idref="DRAWINGS">FIG. 2</figref> is a pictorial illustration of an exemplary pane in an exemplary GUI containing an exemplary process diagram;

0014<figref idref="DRAWINGS">FIG. 3</figref> is a pictorial illustration of an exemplary pane in an exemplary GUI containing an exemplary radial diagram, including one peripheral circle, with the center circle selected;

0015<figref idref="DRAWINGS">FIG. 4</figref> is a pictorial illustration of an exemplary pane in an exemplary GUI containing an exemplary radial diagram, including two peripheral circles, with the center circle selected;

0016<figref idref="DRAWINGS">FIG. 5</figref> is a pictorial illustration of an exemplary pane in an exemplary GUI containing an exemplary radial diagram, including four peripheral circles, with a lower peripheral circle selected;

0017<figref idref="DRAWINGS">FIG. 6</figref> is a pictorial illustration of an exemplary pane in an exemplary GUI containing an exemplary process diagram, including two blocks, with the left block selected;

0018<figref idref="DRAWINGS">FIG. 7</figref> is a pictorial illustration of an exemplary pane in an exemplary GUI containing an exemplary process diagram, including three blocks, with a middle block selected;

0019<figref idref="DRAWINGS">FIG. 8</figref> is a pictorial illustration of an exemplary pane in an exemplary GUI containing an exemplary process diagram, including four blocks, with the second from right block selected;

0020<figref idref="DRAWINGS">FIG. 9</figref> is a pictorial illustration of an exemplary pane in an exemplary GUI containing an exemplary process diagram, including four blocks, with the second block selected;

0021<figref idref="DRAWINGS">FIG. 10</figref> is a pictorial illustration of an exemplary pane in an exemplary user interface containing an exemplary process diagram, including four blocks, with the middle block selected and containing three bullets;

0022<figref idref="DRAWINGS">FIG. 11</figref> is a pictorial illustration of an exemplary pane in an exemplary user interface containing an exemplary process diagram, including four blocks, with a bullet block in front of the selected middle block;

0023<figref idref="DRAWINGS">FIG. 12A</figref> is an exemplary functional flow diagram of an exemplary subroutine for determining if a child shape should be added to a diagram.

0024<figref idref="DRAWINGS">FIGS. 12B-12C</figref> comprise an exemplary functional flow diagram, which uses the subroutine shown in <figref idref="DRAWINGS">FIG. 12A</figref> and illustrates how shapes are inserted into diagrams.

DETAILED DESCRIPTION

0025Diagrams are information graphics that represent data models or aspects of data models. Data models are often graph structures i.e., data graphs. A type of data graph is a data tree, i.e., tree. A tree is a data structure containing a plurality of nodes that occupy a plurality of levels and are connected in a hierarchical fashion by links. While a node often contains, refers to, or is associated with data, a node may also be empty, i.e., a placeholder. A node can connect to one node in the level above the node, the node's parent node. A node can connect to many nodes in the level below the node, the node's child nodes. Nodes with the same parent node are sibling nodes. Nodes on the same node level are peer nodes. The root node, i.e., root, is the only node in the top level of a tree. The root can have zero or more child nodes, which occupy the next level of the tree. Each child in the level can have zero or more child nodes, which occupy the next level and so on.

0026Diagrams representing data models or aspects of data models may be created using applications, i.e., diagramming applications. Diagramming need not be the primary purpose of a diagramming application. An application that uses diagramming functions to support a purpose other than diagramming may also be considered a diagramming application. Diagramming applications often support a plurality of diagram types designed to present different aspects of data models. Diagramming applications, with support from a GUI, enable users to view diagrams and manipulate the visual elements comprising the diagrams.

0027To enable users to manipulate or otherwise interact with visual elements, a GUI displays a “cursor,” i.e., a small moveable graphic symbol, e.g., an arrow, used to point to visual elements. Cursor movement is usually controlled by a pointing device such as a mouse, trackball, etc. but may also be controlled by a touch pad, touch screen, keyboard, keypad, etc. A pointing device contains a multidirectional detection device and/or one or more buttons or keys. Actuation of the pointing device directs the movement of the cursor on the display. To point to a displayed visual element, a pointing device, e.g., a mouse, is used to move the cursor over the visual element. Often, the appearance of the visual element changes to indicate that the cursor is positioned over the visual element.

0028A commonly applied appearance change is “highlighting.” When a visual element is highlighted, the appearance of the visual element is altered in a way that calls attention to the visual element while keeping the visual element recognizable. Examples of highlighting include, but are not limited to, reversing the intensity of the colors of a visual element to create a “negative” image; changing the color or thickness of a border or background; overlaying a visual element with a transparent color rectangle or other shape; changing the hue, saturation, or value of the colors of a visual element; etc. Highlighting is often used to indicate that a particular visual element has “focus,” i.e., is the target of an action. After a cursor is positioned over a visual element, i.e., the visual element has focus and is highlighted, the visual element can be selected by pressing and releasing a mouse button without moving the cursor. Such “clicking” of a mouse button causes actions to be initiated in the computing device, the display, or both. For example, clicking on a visual element presented in a diagramming application may cause a change in a visual element's appearance.

0029The effect of changing the appearance of a clicked visual element is used to provide “soft” controls such as “soft” buttons. Soft buttons are visual elements that mimic physical buttons such as the buttons on a pointing device. In the following discussion, the term “button” refers to soft buttons and should not be confused with physical buttons, such as mouse buttons. The appearance of a button changes to reflect the state of the button. Button states include, but are not limited to, unpushed, pushed, and disabled. Diagramming applications often use buttons and other soft controls to enable user interaction.

0030A visual element may also be moved in the display by “dragging” the visual element. Dragging is performed by pressing a button or key and holding down the button or key while moving the mouse causing the cursor and the visual element the cursor is over to move. In both clicking and dragging, the button or key is usually a part of the pointing device, e.g., a mouse button, but may be a part of another device, e.g. a key on a keyboard. Clicking and dragging visual elements often occurs inside of windows and panes. A window is a bounded region of a display that is dedicated to presenting visual elements and/or providing a particular set of functions, i.e., actions. A pane is a bounded subregion within a window that is usually dedicated to working with a subset of the visual elements and/or functions provided by the containing window. Windows and panes are normally rectangular but may be any two dimensional shape.

0031While moving the focus, e.g., highlight, from one visual element to another, selecting visual elements, and invoking actions may be supported by a mouse having buttons, such activities may also be supported by keys on a keyboard. For example, pressing a tab key may cause the focus to move one visual element to another and pressing a return key may select a visual element that has focus. Thus, actions, described herein as invoked using a mouse with buttons, may also be invoked using a keyboard with keys.

0032Diagramming applications provide one or more windows and/or panes in which to view diagrams and controls, including but not limited to, buttons to manipulate the visual elements comprising the diagrams. In a window or pane of a diagramming application, simple hierarchical diagrams can be formed using only text to represent nodes and using indentation to represent levels and relationships between levels. Alternately, nodes may be represented by visual elements shaped as rectangles, circles, etc., i.e., shapes. Shapes may or may not contain lines or blocks of text. <figref idref="DRAWINGS">FIGS. 1 and 2</figref> illustrate examples of diagrams that use shapes containing text. <figref idref="DRAWINGS">FIG. 1</figref> illustrates an exemplary radial diagram <b>100</b> and <figref idref="DRAWINGS">FIG. 2</figref> illustrates an exemplary process diagram <b>150</b>. Note that the shapes used to represent nodes depend on the diagram type, the data contained in the node, and/or the relationships of the node to the surrounding nodes. For example, in the exemplary radial diagram <b>100</b> shown in <figref idref="DRAWINGS">FIG. 1</figref> nodes are represented by circles containing text and in the exemplary process diagram <b>150</b> shown in <figref idref="DRAWINGS">FIG. 2</figref> nodes are represented by round cornered rectangles, i.e., rounded rectangles.

0033A radial diagram, such as the radial diagram <b>100</b> shown in <figref idref="DRAWINGS">FIG. 1</figref>, is a diagram in which a center circle is surrounded by a plurality of peripheral circles. Radial diagrams are used to represent relationships between each of a plurality of items and one central item. Each item is stored in a node in the data graph represented by the diagram. Each node is represented in the radial diagram as a circle. Each circle in the plurality of peripheral circles is connected to the center circle by a line. The lines between the central circle the peripheral circles are “connectors.” A connector is a visual element representing a relationship between the nodes represented by the shapes, i.e., circles, connected by the connector. The exemplary radial diagram <b>100</b> in <figref idref="DRAWINGS">FIG. 1</figref> illustrates relationships between a city and a plurality of cities. The radial diagram <b>100</b> comprises one central circle <b>105</b> surrounded by four peripheral circles <b>110</b>, <b>115</b>, <b>120</b>, and <b>125</b>, each connected to the central circle by a line. Each circle is labeled with the name of the city associated with the node that the circle represents. Central circle <b>105</b> is labeled SEATTLE; peripheral circle <b>110</b> is labeled DENVER; peripheral circle <b>115</b> is labeled CHICAGO; peripheral circle <b>120</b> is labeled DALLAS; and peripheral circle <b>125</b> is labeled ATLANTA. The lines between the central circle <b>105</b> and each of the four peripheral circles <b>110</b>, <b>115</b>, <b>120</b>, and <b>125</b> are connectors that may, for example, indicate that there are direct airline flights between Seattle and the connected cities. Note that a radial diagram, such as radial diagram <b>100</b> shown in <figref idref="DRAWINGS">FIG. 1</figref> is an example a diagram that represents a hierarchical data model. Specifically, the node levels of the data tree are implied by the size of the circles but not otherwise explicitly represented.

0034Process diagrams, such as the process diagram <b>150</b> shown in <figref idref="DRAWINGS">FIG. 2</figref>, are used to represent linear processes. A linear process is a process in which the stages in the process occur one after the other with no branching. Each stage is stored in a data graph as a node. Each node is represented in the process diagram as a rounded rectangle. The exemplary process represented by the process diagram <b>150</b> shown in <figref idref="DRAWINGS">FIG. 2</figref> has four stages, each stored as a node in a data graph. The four nodes in the data graph are represented in the process diagram <b>150</b> as four rounded rectangles. The first node in the data graph, i.e., the first stage in the process, is represented by rounded rectangle <b>155</b> labeled ITEM IS SELECTED. To the right of rounded rectangle <b>155</b> is the second rounded rectangle <b>160</b> labeled PAYMENT TYPE IS SELECTED. To the right of rounded rectangle <b>165</b> is the third rounded rectangle <b>165</b> labeled PAYMENT IS SUBMITTED. Below and to the left of rounded rectangle <b>165</b> is the fourth and final rounded rectangle <b>170</b> labeled ITEM IS SHIPPED. Connectors representing the advancement from stage to stage are arrows.

0035In the past, diagrams have been assembled and modified by inserting and locating visual elements like shapes, bullets, and text. For example, a shape, e.g., peripheral circle <b>110</b> shown in <figref idref="DRAWINGS">FIG. 1</figref>, may be inserted into a diagram by clicking a button, e.g., an Insert Shape button. Text may be entered into the shape, e.g., DENVER. A corner of the shape may be dragged to resize the shape. The shape may be dragged to a location. While, preferably, connectors are automatically inserted as each shape is inserted, a diagramming program may require manual insertion of connectors. Instead of, or in addition to, dragging shapes to a location, selecting special purpose buttons may be used to insert a shape before, after, above, or below an existing shape. Special purpose buttons include, but are not limited to, Insert Before, Insert After, Insert Level Above, and Insert Level Below.

0036The results of selecting special purpose buttons are most easily understood in diagrams that explicitly show a hierarchical structure, e.g., an organization diagram of a company. The results of selecting special purpose buttons is less clear for a diagram with a flat, i.e., non-tree, structure such as a process diagram. For example, if a special purpose button such as Insert Level Above is selected, the effect of the special purpose button selection on a process diagram, e.g., the process diagram <b>150</b> shown in <figref idref="DRAWINGS">FIG. 2</figref>, is not clear. Even after extensive experimentation, users may not be sure what special controls cause what behaviors.

0037In order to avoid the foregoing, it is desirable for a diagramming application to enable creating diagrams without relying on special purpose buttons that require users to know what button to select, the diagram type, and the type of node to be inserted. It is also desirable for a diagramming application to minimize the number of actions, e.g., selections, a user must perform to insert and locate visual elements in a diagram and reduce the detailed knowledge required. Instead of presenting a user with many controls, the present diagramming applications may use “smart insertion” to aid the user in building a diagram that makes efficient use of the diagram's nodes, i.e., a well balance diagram. “Smart insertion” is used by a diagramming application to enable the diagramming application to predict what a user intends when inserting shapes in a diagram. A shape that is inserted through a process of smart insertion is said to be “smartly inserted.”

0038The effects of using smart insertion can be overridden using more direct and/or specific controls. For example, if a diagram is built using smart insertion and a node is restricted to having no more than two child nodes, the restriction may be overridden and additional child nodes may be inserted using a user control, e.g., a “split button.” A split button is a button comprising an icon representing an action, i.e., an action icon, and a downward pointing arrowhead icon. If the action icon is selected, the action represented by the action icon is invoked. If the arrowhead icon is selected, a menu providing alternate actions to the action of the action icon is displayed near the action icon. If a menu item is selected, the action for the menu item is invoked. For example, a split button may comprise an Insert Shape button that invokes a smart insertion action and an arrowhead icon that provides a menu comprising menu items to invoke Insert Before, Insert After, Insert Level Above, and Insert Level Below actions. If the Insert Shape icon is selected the shape is smartly inserted as described above. If a menu item is selected, e.g., Insert Level Above, the action represented by the icon is invoked, e.g., a shape is inserted at a level above the selected shape. Controls other than a split button, e.g., keys on a keyboard, may be used to override smart insertion. The use of a split button to override smart insertion should be construed as exemplary and not limiting.

0039Examples of the results of smart insertion progressively applied to radial diagrams are shown in <figref idref="DRAWINGS">FIGS. 3-5</figref>. A functional flow diagram showing how the smart insertion process operates is illustrated in <figref idref="DRAWINGS">FIGS. 12A-12C</figref> and described below. <figref idref="DRAWINGS">FIG. 3</figref> illustrates an exemplary radial diagram <b>200</b> in which the first few exemplary stages of diagram assembly have been executed. The exemplary radial diagram <b>200</b> comprises a central circle <b>205</b> labeled SEATTLE and a peripheral circle <b>210</b> labeled DENVER. Preferably, when a radial diagram such as radial diagram <b>200</b> is started, the central circle is visible and accessible. For example, central circle <b>205</b> is visible and accessible for entering text, e.g. SEATTLE. The central circle <b>205</b> is selected as indicated by the heavy line outlining. Selection may also be indicated by other graphic techniques such as, but not limited to, highlighting. The use of heavy outlining to indicate selection should be construed as exemplary and not limiting. With central circle <b>205</b> selected, an action, e.g., selecting Insert Shape, occurs and causes a peripheral circle <b>210</b> to be inserted into the radial diagram <b>200</b>. Because central circle <b>205</b> is selected, and because in this example only child nodes of central circle <b>205</b>'s node may be inserted, smart insertion causes the peripheral circle <b>210</b> to be inserted into the radial diagram <b>200</b> above central circle <b>205</b>. A straight line is automatically inserted between central circle <b>205</b> and the peripheral circle <b>210</b>. A text label is inserted into the peripheral circle <b>210</b>, i.e., DENVER, by the user.

0040<figref idref="DRAWINGS">FIG. 4</figref> illustrates an exemplary radial diagram <b>220</b> that is partially assembled using stages similar to stages illustrated in <figref idref="DRAWINGS">FIG. 3</figref> and in which additional exemplary stages of diagram assembly are executed. As in the radial diagram <b>200</b> shown in <figref idref="DRAWINGS">FIG. 3</figref>, the radial diagram <b>220</b> in <figref idref="DRAWINGS">FIG. 4</figref> contains a central circle <b>225</b>, labeled SEATTLE and peripheral circle <b>230</b>, labeled DENVER. The central circle <b>225</b> in the radial diagram <b>220</b> is selected. An action, e.g., selecting Insert Shape, occurs and causes an additional peripheral circle to be inserted into the radial diagram <b>220</b>. Because the central circle <b>225</b> is selected and only child nodes of central circle <b>225</b>'s node may be inserted, smart insertion causes the peripheral circle <b>235</b> to be inserted into the radial diagram <b>220</b> to the right of the central circle <b>225</b>. A straight line is automatically inserted between the central circle <b>225</b> and the peripheral circle <b>235</b>. A text label is inserted into the peripheral circle <b>235</b>, i.e., CHICAGO, by the user.

0041<figref idref="DRAWINGS">FIG. 5</figref> illustrates an exemplary radial diagram <b>250</b> that is partially assembled using stages similar to stages illustrated in <figref idref="DRAWINGS">FIGS. 3 and 4</figref> and in which additional exemplary stages of diagram assembly are executed. As in the radial diagram <b>220</b> shown in <figref idref="DRAWINGS">FIG. 4</figref>, the radial diagram <b>250</b> shown in <figref idref="DRAWINGS">FIG. 5</figref> contains a central circle <b>255</b>, labeled SEATTLE, and peripheral circles <b>260</b>, <b>265</b>, and <b>270</b>, respectively labeled DENVER, CHICAGO, and DALLAS. Peripheral circle <b>270</b> in radial diagram <b>250</b> is selected. An action, e.g., selecting Insert Shape, occurs and causes an additional peripheral circle to be inserted into the radial diagram <b>250</b>. Notice that even though peripheral circle <b>270</b> is selected, smart insertion causes a peripheral circle <b>275</b> to appear in the radial diagram <b>250</b> to the left of central circle <b>255</b>. The peripheral circle <b>275</b> is inserted because in this example, the “diagram definition” for the radial diagram <b>250</b> allows only the node for the central circle <b>255</b> to have child nodes. Diagram definitions specify the layout of diagrams and are explained in more detail below. A straight line is automatically inserted between the central circle <b>255</b> and the peripheral circle <b>275</b>. A text label is inserted into the peripheral circle <b>275</b>, i.e., ATLANTA, by the user.

0042As illustrated in <figref idref="DRAWINGS">FIGS. 3-5</figref> and described above, smart insertion enables diagram assembly by intelligently inserting appropriate shapes in appropriate locations in diagrams such as the diagrams illustrated in <figref idref="DRAWINGS">FIGS. 3-5</figref>. Smart insertion may also be applied to less hierarchical, i.e., more linear, diagrams illustrated in <figref idref="DRAWINGS">FIGS. 6-11</figref> as well as the more hierarchical diagrams illustrated in <figref idref="DRAWINGS">FIGS. 3-5</figref>. <figref idref="DRAWINGS">FIG. 6</figref> is an exemplary process diagram <b>300</b> in which the first part of a diagram assembly of the type illustrated in <figref idref="DRAWINGS">FIG. 2</figref> are shown. The exemplary process diagram <b>300</b> comprises first and second rounded rectangles <b>305</b> and <b>310</b>. Preferably, when a process diagram such as process diagram <b>300</b> is started, the first rounded rectangle <b>305</b> is visible and accessible. For example, the first rounded rectangle <b>305</b> is visible and accessible for entering text, e.g. ITEM IS SELECTED, by the user. When the first rounded rectangle <b>305</b> is selected and an action, e.g., selecting Insert Shape, occurs, the second rounded rectangle <b>310</b> is inserted into the process diagram <b>300</b>. It is important to note that although the process diagram <b>300</b> has a linear visual structure, the process diagram <b>300</b> represents a data graph with a root node and a plurality of child nodes of the root. The plurality of child nodes occupy one level, the level below the root. The root node is not represented in the process diagram <b>300</b>. Only the child nodes of the root are represented. For example, the first rounded rectangle <b>305</b> represents a child node of the unrepresented, i.e., hidden, root node. Because only child nodes of the hidden root may be represented and because the child nodes of the hidden root occupy the same level, when a rounded rectangle is inserted into the process diagram <b>300</b>, the rounded rectangle appears on the same level as the existing rounded rectangles. Given these rules, smart insertion causes the second rounded rectangle <b>310</b> to appear in the process diagram <b>300</b> to the right of the first rounded rectangle <b>305</b>. An arrow is automatically inserted pointing from the first rounded rectangle <b>305</b> to the second rounded rectangle <b>310</b>. Text is inserted into the second rounded rectangle <b>310</b>, i.e., PAYMENT TYPE IS SELECTED, by the user.

0043<figref idref="DRAWINGS">FIG. 7</figref> is an exemplary process diagram <b>320</b> that includes the partial assembly illustrated in <figref idref="DRAWINGS">FIG. 6</figref> and described above and an additional (third) rounded rectangle. The process diagram <b>320</b> shown in <figref idref="DRAWINGS">FIG. 7</figref> includes two rounded rectangles <b>325</b> and <b>330</b>, similar to the rounded rectangles <b>305</b> and <b>310</b> illustrated in <figref idref="DRAWINGS">FIG. 6</figref>, plus a third rounded rectangle <b>335</b>. As with <figref idref="DRAWINGS">FIG. 6</figref>, the first rounded rectangle <b>325</b> is labeled ITEM IS SELECTED, and the second rounded rectangle <b>330</b> is labeled PAYMENT TYPE IS SELECTED. The second rounded rectangle <b>330</b> in process diagram <b>320</b> is selected as indicated by a heavy outline. When an action, e.g., selecting Insert Shape, occurs an additional (third) rounded rectangle is inserted into the process diagram <b>320</b>. Because the second rounded rectangle <b>330</b> is selected and because only child nodes of the hidden root may be inserted, smart insertion causes the third rounded rectangle <b>335</b> to appear in the process diagram <b>320</b> to the right of the second rounded rectangle <b>330</b>. An arrow is automatically inserted pointing from the second rounded rectangle <b>330</b> to the third rounded rectangle <b>335</b>. Text, i.e., PAYMENT IS SUBMITTED, is inserted into rounded rectangle <b>335</b>, by the user.

0044<figref idref="DRAWINGS">FIG. 8</figref> is an exemplary process diagram <b>350</b> that includes the partial assemblies illustrated in <figref idref="DRAWINGS">FIGS. 6 and 7</figref> and an additional (fourth) rounded rectangle. Also, the sizes of the rounded rectangles have been reduced to allow the additional (fourth) rounded rectangle to be included in the process diagram <b>350</b>. Reducing the size of the rounded rectangles is an example of how a diagram definition may define how shapes are laid out in a diagram. As with <figref idref="DRAWINGS">FIGS. 6 and 7</figref>, the first rounded rectangle <b>355</b> is labeled ITEM IS SELECTED, the second rounded rectangle <b>360</b> is labeled PAYMENT TYPE IS SELECTED, and the third rounded rectangle <b>365</b> is labeled PAYMENT IS SUBMITTED. The third rounded rectangle <b>365</b> is selected as indicated by heavy outline. When an action, e.g., clicking Insert Shape, occurs the additional (fourth) rounded rectangle is inserted into the process diagram <b>350</b>. Because the third rounded rectangle <b>365</b> is selected and because only child nodes of the hidden root may be inserted, smart insertion causes the fourth rounded rectangle <b>370</b> to appear in the process diagram <b>350</b> to the left of the third rounded rectangle <b>365</b>. An arrow is automatically inserted pointing from the third rounded rectangle <b>365</b> to the fourth rounded rectangle <b>370</b>. Text, i.e., ITEM IS SHIPPED, is inserted into rounded rectangle <b>370</b>, by the user.

0045<figref idref="DRAWINGS">FIG. 9</figref> is an exemplary process diagram <b>375</b> that is similar to the process diagram <b>350</b> shown in <figref idref="DRAWINGS">FIG. 8</figref>. As with <figref idref="DRAWINGS">FIG. 8</figref>, the first rounded rectangle <b>380</b> is labeled ITEM IS SELECTED, the second rounded rectangle <b>385</b> is labeled PAYMENT TYPE IS SELECTED, the third rounded rectangle <b>390</b> is labeled PAYMENT IS SUBMITTED, and the fourth rounded rectangle <b>395</b> is labeled ITEM IS SHIPPED. Unlike <figref idref="DRAWINGS">FIG. 8</figref>, in <figref idref="DRAWINGS">FIG. 9</figref>, the second rounded rectangle <b>385</b> is selected as indicated by heavy outline. When an action, e.g., clicking Insert Shape, occurs, the third rounded rectangle <b>395</b> is inserted into the process diagram <b>350</b>. Because the second rounded rectangle <b>385</b> is selected and because only child nodes of the hidden root may be inserted, smart insertion causes the third rounded rectangle <b>390</b> to appear in the process diagram <b>395</b> after the second rounded rectangle <b>385</b>. An arrow is automatically inserted pointing from the second rounded rectangle <b>385</b> to the third rounded rectangle <b>390</b>. Text, i.e., PAYMENT IS SUBMITTED, is inserted into rounded rectangle <b>390</b>, by the user.

0046As illustrated in <figref idref="DRAWINGS">FIGS. 3-9</figref> and described above, smart insertion enables diagram assembly by intelligently inserting appropriate shapes in appropriate locations in hierarchical and non-hierarchical, i.e., linear, diagrams. Smart insertion may also be used to insert “bullets” into diagrams as illustrated in <figref idref="DRAWINGS">FIGS. 10-11</figref> and described below. The term “bullet” is often used to refer to small graphic symbols used to accent text but may also be used to refer to the “bulleted” text or the combination of bullet and text. In the following discussion, the term “bullet graphic” refers to a graphic symbol used to accent text; the term “bulleted text” refers to text accented by a bullet; and “bullet” refers to a combination of a bullet graphic and bulleted text. A shape may or may not support bullets, i.e., allow bullets to be added to the shape. Bullets may be placed in a shape, usually below a line of text or text block in the shape. Bullets may also be placed in an “adornment shape” which may overlap or is otherwise visually associated with a shape.

0047Bullets may be semantic or non-semantic. A semantic bullet is represented in a data model as a node whereas a non-semantic bullet is represented as an attribute of text within the node, i.e., as a formatting property on the text. A semantic bullet is recognized by smart insertion and is included in calculations and decisions involving the insertion and in the placement of shapes and bullets. A non-semantic bullet, i.e., a format bullet, is not recognized by smart insertion and is not counted in calculations or used in smart insertion and layout decisions. Both semantic and format bullets are affected by insertion and layout decisions normally made by the diagramming application. If a shape supports bullets, bullets may be smartly inserted into that shape. If an adornment shape supports bullets, bullets are may be smartly inserted into the adornment shape instead of the shape with which the adornment shape is associated.

0048<figref idref="DRAWINGS">FIGS. 10-11</figref> illustrate examples of the results of smart insertion of bullets in process diagrams. <figref idref="DRAWINGS">FIGS. 10-11</figref> illustrate exemplary process diagrams similar to the process diagrams shown in <figref idref="DRAWINGS">FIGS. 6-9</figref>; however, the process diagrams in FIGURES <b>10</b>-<b>11</b> also include bullets. The exemplary process diagram <b>400</b> shown in <figref idref="DRAWINGS">FIG. 10</figref> includes four rounded rectangles <b>405</b>, <b>410</b>, <b>415</b>, and <b>420</b> connected by arrows. The text in the first rounded rectangle <b>405</b> is ITEM IS SELECTED. An arrow points from the first rounded rectangle <b>405</b> to the second rounded rectangle <b>410</b>. The text in the second rounded rectangle <b>410</b> is PAYMENT TYPE IS SELECTED. The second rounded rectangle <b>410</b> includes three bullets below this text, namely CASH, CHECK, and CREDIT. An arrow points from the second rounded rectangle <b>410</b> to third rounded rectangle <b>415</b>. The text in the third rounded rectangle <b>415</b> is PAYMENT IS SUBMITTED. An arrow points from the third rounded rectangle <b>415</b> to the fourth rounded rectangle <b>420</b>. The text in the fourth rounded rectangle <b>420</b> is ITEM IS SHIPPED. Because the second rounded rectangle <b>410</b> is selected and because the node for the second rounded rectangle <b>410</b> allows bullet child nodes, when an action, e.g., selecting Insert Shape, occurs, bullets are smartly inserted into the second rounded rectangle <b>410</b>.

0049The exemplary process diagram <b>450</b> shown in <figref idref="DRAWINGS">FIG. 11</figref> includes four rounded rectangles <b>455</b>, <b>460</b>, <b>470</b>, and <b>475</b> connected by arrows. The text in the first rounded rectangle <b>455</b> is ITEM IS SELECTED. An arrow points from the first rounded rectangle <b>455</b> to the second rounded rectangle <b>460</b>. The text in the second rounded rectangle <b>460</b> is PAYMENT TYPE IS SELECTED. An adornment shape <b>465</b> overlaps the lower right corner of the second rounded rectangle <b>460</b> and contains three bullets: CASH, CHECK, and CREDIT. An arrow points from the second rounded rectangle <b>460</b> to the third rounded rectangle <b>470</b>. The text in the third rounded rectangle <b>470</b> is PAYMENT IS SUBMITTED. An arrow points from the third rounded rectangle <b>470</b> to the fourth rounded rectangle <b>475</b>. The text in the fourth rounded rectangle <b>475</b> is ITEM IS SHIPPED. Because the second rounded rectangle <b>460</b> is selected and because the node for the second rounded rectangle <b>460</b> allows bullet child nodes with bullets in an adornment shape, when an action, e.g., selecting Insert Shape, occurs, a bullet is smartly inserted into the adornment shape <b>465</b> for the second rounded rectangle <b>460</b>. If the inserted bullet is the first bullet to be inserted, the adornment shape <b>465</b> is laid over the second rounded rectangle <b>460</b> and the first bullet is inserted in the adornment shape <b>465</b>.

0050As illustrated in <figref idref="DRAWINGS">FIGS. 3-11</figref> and described above, smart insertion enables a diagramming application to intelligently determine an appropriate shape or bullet to insert into a diagram in response to a request for a shape or bullet. Smart insertion does not require users to know if a node for an inserted shape or bullet is a peer node or child node of the node of a selected shape in order to add a new shape or bullet to a diagram. Smart insertion enables users to build balanced, easy to read diagrams without prior knowledge of the underlying data graphs. One exemplary way showing how smart insertion is accomplished illustrated in <figref idref="DRAWINGS">FIGS. 12A-12C</figref> and described below.

0051Smart insertion may be used by diagramming applications to determine the type and location of shapes and bullets inserted into diagrams representing a data graph. Smart insertion uses “diagram definitions” to control the overall appearance of diagrams of a diagram type and to control the location of shapes and bullets in diagrams. The diagram definition describes how multiple shapes are laid out in a diagram. For example, in a radial diagram such the radial diagram <b>100</b> shown in <figref idref="DRAWINGS">FIG. 1</figref>, the diagram definition may indicate that shapes should be inserted in a clockwise fashion around the central circle. In another example, a diagram definition for a process diagram such the process diagram <b>350</b> shown in <figref idref="DRAWINGS">FIG. 8</figref>, may indicate that if a new shape needs to be inserted into the diagram, the shapes already existing in the diagram should be reduced in size to make room for the new shape.

0052Preferably, a diagram type is described in a computer file, i.e., a diagram definition file, written in a declarative language, e.g., XML. Diagram definition files are read by an engine that interprets the declarative language and exposes the data in the diagram definition to the smart insertion mechanism in a diagramming application. Because diagram definitions are written in an easily understandable format, diagram definitions may be easily written or modified without depending on computer programming knowledge. Using diagram definition files enables describing attractive diagrams without writing and compiling computer instructions thus saving computer programmer time.

0053In a diagram definition file, node types are described. Associated with each node type are a “child preference” attribute and a plurality of shapes and bullet adornment shapes used by nodes of the node type. The child preference attribute describes the total number of child nodes allowed for the node. For example, if a node description in a diagram definition file has a child preference of three, up to three child nodes can be attached to nodes of the node type. It is possible to have a child preference of zero indicating that no child nodes may be attached to a node. It is also possible to have a child preference of −1 to indicate that an infinite, i.e., unlimited, number of child nodes may be attached to a node. Preferably, a node description also specifies which shape or bullet to use to represent a node in a diagram depending on the node's location in the data graph, the type of the parent node, etc. Other data may be included in node and diagram definitions. The aforementioned data included in node descriptions and in diagram definitions should be construed as exemplary and not limiting.

0054The smart insertion process begins with collecting information about a “starting shape” in the diagram. A starting shape is the first shape considered in smart insertion. A starting shape may be a selected shape or a default shape set in a diagram definition or elsewhere in the data used by the diagramming application. Preferably, the information about the starting shape comprises the starting shape's type; the starting shape's node in the data graph; the starting shape's node's parent node in the data graph; the number of child nodes of the starting shape's node; and the number of child nodes of the parent node of the starting shape's node, i.e., the starting node. The smart insertion process continues by selecting a diagram definition in accordance with the diagram's type; extracting information from the diagram definition; and using the information extracted from the diagram definition and the information about the starting shape. The diagram definition information and starting shape information are used to determine the type of node to associate with the inserted shape, the location of the associated node in the data graph, and the inserted shape's location in the diagram. As explained above, the child preference indicates the number of child nodes allowed to be attached to a node. The parent node of the starting node is located in the data graph and the child preference of the parent node is read. If the parent node prefers to have more child nodes, i.e., the child preference is not exceeded, a new node is inserted and becomes a sibling of the starting node. If the parent node does not prefer to have more child nodes, the child preference of the starting node is used to determine if a child node may be attached to the starting node. If the starting node can accept another child node, a new child node is attached to the starting node. If the starting node cannot accept another child node, i.e., the child preference is exceeded, a peer of the starting node is checked.

0055How the smart insertion process intelligently assembles diagrams is illustrated in <figref idref="DRAWINGS">FIGS. 12A-12C</figref> and described below. <figref idref="DRAWINGS">FIG. 12A</figref> is a functional flow diagram illustrating a subroutine used in the functional flow diagram illustrated in <figref idref="DRAWINGS">FIGS. 12B-12C</figref>, described below. The exemplary title of the subroutine illustrated in <figref idref="DRAWINGS">FIG. 12A</figref>, SHOULD_ADD_CHILD, is shown in block <b>500</b>. The purpose of the SHOULD_ADD_CHILD subroutine is to determine if a child node should be added to a node of a shape submitted to the subroutine, i.e., an input shape. The SHOULD_ADD_CHILD subroutine is called from within the functional flow diagram shown in <figref idref="DRAWINGS">FIG. 12B</figref>. The SHOULD_ADD_CHILD subroutine shown in <figref idref="DRAWINGS">FIG. 12A</figref> starts with block <b>500</b> where an input shape is submitted to the subroutine. At block <b>505</b>, the node of the input shape is acquired from the data graph. At block <b>507</b>, the properties of the input shape's node are acquired. At block <b>510</b>, the diagram definition is queried, i.e., read, to get the node type's preferred number of child nodes, i.e., the node's child preference. At block <b>515</b>, the node is read to get the node's current number of child nodes. At block <b>520</b>, the node type's preferred number of child nodes is compared to the node's current number of child nodes. If the preferred number of child nodes is less than the current number of child nodes, the subroutine returns YES at block <b>530</b>. If the preferred number of child nodes is not less than the current number of child nodes, the subroutine returns NO at block <b>525</b>.

0056<figref idref="DRAWINGS">FIGS. 12B-12C</figref> comprise a functional flow diagram of an exemplary smart insertion process. The exemplary smart insertion process begins at block <b>540</b> of <figref idref="DRAWINGS">FIG. 12B</figref> by determining the starting shape. At block <b>545</b> a test is made to determine if the starting shape's node has a parent node. If the starting shape's node has no parent node, e.g., the starting shape represents a root node, the control flows to block <b>570</b>. If the starting shape has a parent node, the control flows to block <b>547</b>. For example, in radial diagram <b>250</b> shown in <figref idref="DRAWINGS">FIG. 5</figref>, the peripheral circle <b>270</b> is selected. Thus, the peripheral circle <b>270</b> is the starting shape. The node represented by the peripheral circle <b>270</b> has a parent node, i.e., the node of the central circle <b>255</b>, so the control flows to block <b>547</b>. If the central circle <b>255</b> had been selected, the central circle <b>255</b> would have been the starting shape. The node of central circle <b>255</b> has no parent node, i.e., represents a root node, so the control would flow to block <b>570</b>.

0057At block <b>547</b> of <figref idref="DRAWINGS">FIG. 12B</figref>, the parent node is acquired. At block <b>550</b>, the shape associated with the parent node, i.e., the parent shape, is acquired. For example, in the radial diagram <b>250</b> shown in <figref idref="DRAWINGS">FIG. 5</figref>, the peripheral circle <b>270</b> is the starting shape. The node represented by the peripheral circle <b>270</b> has a parent node, i.e., the node of central circle <b>255</b>. The central circle <b>255</b>'s node is represented by a shape, i.e., central circle <b>255</b>. Central circle <b>255</b>'s node is used to acquire the shape that represents the node, i.e., central circle <b>255</b>. Thus, the parent shape of the peripheral circle <b>270</b> is acquired, i.e., central circle <b>255</b>. Next, at block <b>555</b> (<figref idref="DRAWINGS">FIG. 12B</figref>), the SHOULD_ADD_CHILD subroutine (<figref idref="DRAWINGS">FIG. 12A</figref>) is called for the parent shape, i.e., central circle <b>255</b>. The SHOULD_ADD_CHILD subroutine begins at block <b>505</b> with the input shape, i.e., central circle <b>255</b>. The input shape's node, i.e., central circle <b>255</b>'s node, is acquired. At block <b>507</b>, the type of the input shape's node, i.e., the central circle type, is acquired. At block <b>510</b>, the diagram definition is queried to get the preferred number of child nodes for central circle type. At block <b>515</b>, the current number of child nodes of central circle <b>255</b>'s node are acquired. At block <b>520</b> the preferred number of child nodes of the central circle type are compared to the current number of child nodes of central circle <b>255</b>'s node. If the preferred number of child nodes of central circle <b>255</b>'s node is less than the current number of child nodes of central circle <b>255</b>'s node, YES is returned at block <b>530</b>. Otherwise, NO is returned at block <b>525</b>.

0058Returning to <figref idref="DRAWINGS">FIG. 12B</figref>, at block <b>555</b>, if the SHOULD_ADD_CHILD subroutine called for the parent shape, i.e., central circle <b>255</b>, returns YES, control flows to block <b>560</b>. At block <b>560</b>, a sibling shape of the starting shape, i.e., a child shape of the parent shape, is added to the diagram and the sibling shape's node is added to the data graph. For example, peripheral circle <b>275</b> is inserted in the radial diagram <b>250</b> shown in <figref idref="DRAWINGS">FIG. 5</figref> and peripheral circle <b>275</b>'s node is added to the data graph. Peripheral circle <b>275</b>'s node is a child node of central circle <b>255</b>'s node and hence is a sibling node of peripheral circle <b>270</b>'s node. The location of the sibling shape is determined by information associated with the node type of the sibling shape's node in the diagram definition file. After block <b>560</b>, the process ends. If, at block <b>555</b>, the SHOULD_ADD_CHILD subroutine called for the parent shape, i.e., central circle <b>255</b>, returns NO, control flows to block <b>570</b>. At block <b>570</b>, the SHOULD_ADD_CHILD subroutine is called for the starting shape, i.e., peripheral circle <b>270</b>. If, at block <b>570</b>, the SHOULD_ADD_CHILD subroutine called for the starting shape returns YES, control flows to block <b>565</b>. At block <b>565</b> a child shape of the starting shape is added, i.e., peripheral circle <b>270</b>, and the child shape's node is added to the data graph. After block <b>565</b>, the process ends. If, at block <b>570</b>, the SHOULD_ADD_CHILD subroutine called for the starting shape, i.e., peripheral circle <b>270</b>, returns NO, control flows to block <b>575</b>. At block <b>575</b>, the start shape, i.e., peripheral circle <b>270</b>, is set to the current shape.

0059The control flows from block <b>575</b> to block <b>580</b>. At block <b>580</b>, child shapes of the current shape, i.e., peripheral circle <b>270</b>, are added to the end of a queue. Peripheral circle <b>270</b> has no child shapes, hence, no items are added to the end of the queue. The control flows from block <b>580</b> to block <b>585</b>. If, at block <b>585</b>, the queue does not contain any items, the control flows to block <b>590</b>. At block <b>590</b>, the “Insert Shape” button is disabled thus restricting a user from inserting more shapes. After block <b>590</b>, the process ends.

0060If, at block <b>585</b>, the queue contains items, e.g., peripheral circle <b>270</b> has child shapes, the control flows to block <b>600</b> shown in <figref idref="DRAWINGS">FIG. 12C</figref>. At block <b>600</b>, the “Insert Shape” button, with controls providing a menu of items for alternate actions, e.g., a split button, is enabled. At block <b>605</b>, it is determined if a menu item has been selected from the menu. If a menu item has been selected, control flows to block <b>610</b>. At block <b>610</b>, a shape is inserted as directed by the menu selection, i.e., a shape is inserted by invoking the action represented by the menu selection. After block <b>610</b> the process ends.

0061If, at block <b>605</b> shown in <figref idref="DRAWINGS">FIG. 12C</figref>, it is determined that no menu item has been selected, i.e., a smart insertion shape action has been invoked, the control flows to block <b>615</b>. At block <b>615</b>, the first item in the queue is removed. At block <b>620</b>, the first item removed from the queue, e.g., a child shape of peripheral circle <b>270</b>, is set to the current shape. Control flows from block <b>620</b> to block <b>625</b>. At block <b>625</b>, the SHOULD_ADD_CHILD subroutine is called for the current shape. If, at block <b>625</b>, the call to the SHOULD_ADD_CHILD subroutine for the current shape returns YES, the control flows to block <b>625</b>. At block <b>630</b>, a child shape is added after all existing child shapes of the current shape and a child node is added in the data graph after all the child nodes of the current shape's node. After block <b>630</b>, the process ends.

0062If, at block <b>625</b>, the call to the SHOULD_ADD_CHILD subroutine for the current shape returns NO, the control flows back to block <b>580</b>, shown in <figref idref="DRAWINGS">FIG. 12B</figref>. If, at block <b>585</b>, it is determined that the queue does not contain any items, the flow proceeds to block <b>590</b> where the “Insert Shape” button is disabled. Then the process ends.

0063The smart insertion process illustrated in <figref idref="DRAWINGS">FIGS. 12A-12C</figref> and described above, may also be used to insert bullets. For example, a diagram definition may specify that the child nodes of a shape's node are bullets. When child nodes are added to the shape's node, the child nodes are represented by bullets. Smart insertion may be used to insert shapes and bullets into new diagrams containing no shapes or bullets. Smart insertion may also be applied to default diagrams containing default initial shapes and/or bullets. Preferably, by using smart insertion, balanced, attractive diagrams may be assembled by repeatedly selecting an Add Shape button or a similar control. <figref idref="DRAWINGS">FIGS. 12A-12C</figref> and the above description demonstrate that if a diagram definition is available for a diagram type, the smart insertion process may be applied to a diagram.

0064While illustrative embodiments have been illustrated and described, it will be appreciated that various changes can be made therein without departing from the spirit and scope of the invention. For example, smart insertion may be applied to other information graphics in addition to diagrams such as, but not limited to, charts, graphs, and maps.