Workflow Explorer

The workflow explorer shows how the different nodes form the current workflow. Nodes can receive data via one or more inputs on their left side and output data via one ore more outputs on their right side (figure 2). Each node represents one step in a workflow. By chaining different analyses and visualizations it becomes possible to build a workflow graph that is fitting for a specific task or question.

The state of a node can be seen by the colored circle that is displayed with each node:

Processing (Light Green)

The node is currently computing, e.g. because its input or the value of one of its parameters has been changed. Once the computation has finished successfully, the nodes that succeed it in the graph will start their computation. Note that while a node is computing its output can not be considered valid. Especially visualizations might still display the data as it was before the new computation was started. This allows a better comparison between the old and the new state, once the computation is finished.

Valid (Dark Green)

The node was able to compute its last task successfully and its outputs are valid.

Failed (Red)

There was an error during the last computation by this node. Such an error can occur e.g. because the input data is not fitting for the node or because some parameters in its settings pane do not allow a successful computation. Hover over the red circle to receive more information about the cause of the error.

Invalid (Gray)

The node's output data is outdated or otherwise invalid. The main reason for this state is that no valid input data exists for this node and therefore no computation can be done. This might be because not all inputs of this node are connected, or because an failed state (rede) exists in a preceding node.

All nodes can be deleted by clicking the symbol on the node.

The main Area of the Analysis Explorer displays the following Columns:

The workflow explorer displays the data flow and how the different nodes are connected. When a node is hovered over with the mouse, it is highlit in a light blue. The connections to other nodes are shown in the same color, and all nodes that either influence it (via input) or are influenced by it (via output) are shown with a box in the same color.

In a similar manner the part of the graph influenced by or influencing other nodes can be shown. This can be set via the workflow explorer options accessible in its top right corner.

Highlight graph of nodes in active window:

Show the graph of all nodes displayed in the active window (or of the active node, if the active node does not display a window). This graph is displayed in the same color as the header of the active window.

Highlight graph of nodes in visible windows:

Show the graphs of all nodes displayed in currently visible windows. This graph is displayed in the same color as the header of non-active visible windows.

Show all connections:

Show all connections in the graph, regardless of whether the nodes are displayed in a visible window or not.

Hide nodes that are not part of highlighted graphs:

Hides all nodes completely that are not part of currently displayed graphs. Which nodes exactly will be hidden depends on the first three options.

It can be very useful to change the flow of data in the workflow explorer, e.g. if you have created a workflow that answers a specific question and want to apply that workflow to another source of data. There are three ways of influencing the workflow: Creating new connections, changing existing connections, and deleting connections..

To create a new connection, click on the little arrow that denotes an output, hold the mouse button and start dragging. You can then drop the connection onto any node that has an input. The connection will be created between the output you started dragging on and the first input of the node where you released the mouse button. If you need to connect to an input other than the first, you need to release the mouse button while hovering the arrow symbol of the input you want to target. If the input you are targeting is already connected to another node, this existing connection will be deleted. While you hover over a node or an input, a preview and a description of the new connections is shown. Sometimes, creating such a new connection will result in several connections, or a new connection is not possible, because other nodes are displayed in the same window. If you want to avoid that, make sure that the target node is displayed in a window of its own.

To change a connection, click on the little arrow that denotes an input, hold the mouse button and start dragging. When you drop the connection, the result will be the same as for a new connection, with the exception that the connection that already existed between an output and the input where you started the operation will be deleted.

To delete a connection, right-click on the input at the end of the connection you want to remove and select "Remove Connection".

The nodes displayed in the workflow explorer all have a context menu. Even though it might differ a little depending on the column and the type of the node, it generally looks somewhat as shown in figure 3

Analyses:

Add a new analysis to the first output of the node, if the node is not a visualization. If you want to add an analysis to another output than the first, activate the context-menu directly on the little arrow icon of the output. If the node is a visualization, add a new analysis to the same output of the previous node, that this node is currently attached to, and add the new analysis to the window that displays the current node.

Visualizations:

Add a new Visualization to the first output of the node, if the node is not a visualization. If you want to add a visualization to another output than the first, activate the context-menu directly on the little arrow icon of the output. If the node is a visualization, add a new visualization to the same output of the previous node, that this node is currently attached to, and add the new visualization to the window that displays the current node.

Duplicate Marking:

Create a copy of the marking created by this node, and store the copy in the "Markings" column. This option is only available for nodes that output a marking.

Set as Current Marking:

Set the current marking to be the same as the marking output by this node. This option is only available for nodes that output a marking and are not the current marking node.

Add to Window:

Add settings and visualization of this node to another window. This option is only available for analyses and visualizations, and only windows of other nodes displaying the same data can be selected. By using this option it is possible to couple different visualizations, e.g. a line graph and a parallel scan path.

Display in Separate Window:

Display the settings and visualization of this node in a separate window. This option is only available if the node is displayed in a window containing more than one node.

Rename:

Rename the node.

Input:

A submenu for the input of the node. Note that in cases where the node has several inputs, several such submenus will exist, one for each input, and with the same name as the respective input.

Export Selected Items to File(s):

Export all the data that is currently selected in the Input Selection to one or several files. This allows the precise selection of the data you want to export. If you want to export all data, it might be advisable to use the export command on the output of the predecessor node.

Copy Selected Items to New Data Node:

Copy all the data that is currently selected in the Input Selection to a new data node.

Output:

A submenu for the output of the node. In cases where the node has several outputs, several such submenus will exist, one for each output.

Export to File(s):

Export all data on this node's output to one or several files.

Copy to New Data Node:

Copy all the data on this node's output to a new data node.

The Input and Output submenus are also available by activating the context menu directly on the little arrow icons for inputs and outputs.