With the conditions available right now, if you use unique tags, you can use the On tagged node change trigger and perform your actions or functions when a specific node is reached.

You can also use the On any node change trigger coupled with the GetCurrentNodeTag expression to achieve the same effect.

Even though the current feature set is pretty bare bones, most of the things suggested up until now, can already be achieved.

Whatever method you choose to determine what node has been reached on a trigger, it is reasonable to have a unique event sheet that just checks which node has been reached and depending on that execute additional conditions and actions.

On a different note, there is a feature which the current example doesn't explore. The Flowchart Controller plugin is not single global, that means it can be added to a container, that means you can have unique flowcharts instances for a bunch of instances and manage them relatively easily. I was thinking of creating a different example showing a possible way to use that.

I think the fundamental problem with adding the functionality to execute logic to nodes is that the more logic capabilities flowcharts had, the more they would overlap with event sheets.

That poses two problems.

The first is having two different ways to handle logic, which in on itself is not great because it would be confusing. Specially for new people. The immediate question someone new would have is "What are better, Event Sheets or Flow charts?" and the answer would inevitably be "It depends on what you are doing, both have pitfalls... bla bla bla" :)

But the second problem is that Event Sheets are just better for programming and are Construct's way of doing things. Flowcharts could be used for small bits of logic, but they don't scale. If you have boxes connected with lines to handle logic, things become spaghetti rather quickly. I pray for the soul that has to debug that kind of visual systems.

So, if the best practice is to use the feature for relatively small systems, I think it's ok to have to use event sheets to glue things together.

XHXIAIEIN

6) Allow top-down organizational structure (Flowchart properties)

I don't think that will happen.

When I first started developing the feature, the diagrams I drew on paper had the outputs at the bottom of the node and the input on top, instead of the left and right respectively. That is how most people would draw tree diagrams.

The problem with that was that text fields are horizontal so having each output to the side of each name and value pair makes much more sense visually and because being on the side makes sense for the outputs, it becomes natural for the input to be on the opposite side rather than the top.

XHXIAIEIN

4) I definitely did not think about that last one. I don't think it is necessary though.

In the example the output values are used as raw text, but they don't need to be used like that all the time. They could be ids to a dictionary. With a few events it would be relatively easy to pick up the real values from the dictionary using the ids from the node outputs.

I think that's a reasonable and scalable solution that can be implemented in any project right now, no need to wait for us.

5) That sounds like something that might be pushed for later updates because there are already more important things to work on, but maybe if we don't get flooded with requests I might get around to do it.

XHXIAIEIN

1) I though about that myself, but because I spent more time than I would have liked just getting it to do what it is doing now. I though I would leave that as a future improvement.

2) You should be able to undo the deletion of a node if it happens on accident. I think that is much better than a confirmation dialog.

3) I actually didn't think about that the first time around. Will consider it for the future polish updates the feature will get.

Introduction

Starting with r370 we are introducing a new feature, Flow Charts. In the editor, a flow chart allows you to:

1. Create nodes
2. Add arbitrary data to those nodes
3. Stablish logical connections between the nodes

Later, at runtime, it is possible to create multiple instances of a flow chart and use the new Flowchart Controller plugin to navigate through the nodes. Each instance will keep it's state independently, namely the current node.

Since event sheets are the main method of managing any kind of logic in C3, flow charts themselves don't have any functionality to perform logic, instead they only serve as data holders and to model the connections between the nodes. Any logic that can be associated with the current state of a flow chart is handled by events.

Possible applications

This kind of data structure can be used to model any kind of system that might look like a tree structure, like dialog trees, skill trees or quest trees in an RPG. It could also be used as a state machine to manage the logic behind objects in a game that have moderately complex behaviours, like a boss or mini boss would have in a shoot 'em up game.

Those are just some examples I could come up off the top of my head and certainly not the only applications.

Creating a flow chart

To create a flow chart, just do so through the Project bar by bringing up the context menu in the new Flow charts sub folder and selecting the Add Flow chart option.

Editing a flow chart

Double clicking on a flow chart item in the Project bar will open a new tab with a Flow chart view where it is possible to edit it.

Some of the basic editing options are the following:

Adding nodes

The context menu that comes up by clicking on any empty space in the flow chart will have the "Insert node" option, which will create a new node in that position.

Deleting nodes

The context menu that comes up by clicking on any empty space in a node will have the "Delete node" option, which will delete the corresponding node.

Adding node outputs

A newly created node has one output, which has a name and a value. By clicking on the "Add" link at the bottom of a node, it is possible to add as many outputs as needed to a node.

Deleting node outputs

A node output can be deleted by bring up the context menu of the output and choosing the "Delete output" option. Nodes must have at least one output.

Edit output name

The name of an output can be edited by double clicking it or by bringing up the context menu and choosing the "Edit name" option.

Edit output value

The value of an output can be edited by double clicking it or by bringing up the context menu and choosing the "Edit value" option.

Connecting nodes

Each output of a node can be connected to the input of another node. To do so:

1. tap and hold on the starting input or output icon


2. drag to the other input or output


3. release the pointer device

Disconnecting nodes

Connected nodes can be disconnected by bringing up the context menu option in the corresponding input or output and choosing the "Unlink" option

Edit a node's tag

Every node has a tag property which can be used at runtime to identify it.

Sorting outputs

By tapping and holding an output and then dragging it it is possible to give it a different position in the node.

Asides from those options, nodes can also be moved by tapping and holding the caption of the node and resized by tapping and holding the borders of the node.

Moving and resizing is limited by other nodes.

Properties Bar

The Properties bar shows another view of the properties of a flow chart and more importantly the individual nodes. At the moment of writing there is only one option which can only be edited through the properties bar and not the nodes themselves.

That is the "Start Node" property which indicates which should be the default starting node when a new flow chart instance is created at runtime.

Using any of the "Reset" actions will also set the current node of a flow chart as the one set as the starting one.

Flow chart controller plugin

The plugin is used to instantiate flow charts at runtime, navigate through them, extract data and eventually release them when no longer needed.

Actions

Start flow chart & Start flow chart by name

Used to instantiate a new flow chart from the ones defined in the editor. The newly created flow chart starts at the node defined as the starting one. A tag needs to be passed when starting a flow chart to be able to control it later.

End flow chart & End flow chart by tag

Used to release all resources associated with a flow chart. The first action releases the currently active flow chart, while the "by tag" variant allows you to release any other flow chart.

Reset flow chart & Reset flow chart by tag

Reset a flow chart to it's initial state. The first action resets the currently active flow chart while the "by tag" variant targets a specific one. The current node is set as the one marked as the starting one in the editor.

Set flow chart

Sets a flow chart instance as the currently active one. Other actions, conditions and expressions will work according to the active flow chart. Some expressions can receive a flow chart tag as argument to work according to that specific flow chart rather than the active one.

Go to next node

Uses an index or a name and will follow the corresponding output into the next node.

Go to any node

Uses a tag to just jump to that node ignoring all connections.

Go to previous node

Goes to the previous node that was navigated to with one of the previous actions.

Go to parent node

uses an index or a tag to go to the parent of the current node. In this action the index or tag is only meaningful if a node has more than one parent.

Conditions

On any flow chart change

Triggered when the current flow chart changes.

On any node change

Triggered when the current node changes.

On tagged node change

Triggered when the current node changes and it's tag matches the one specified.

On tagged node change by flow chart

Triggered when the current node changes and it's tag matches the one specified, in the flow chart specified.

On any node change by flow chart

Triggered when the current node changes in the flow chart specified.

Is at start node

Check if the current node is the starting one.

Is at start node by flow chart

Check if the current node is the starting one, in the specified flow chart.

Has flow chart

Check if the plugin instance has a flow chart for the specified tag.

Expressions

GetCurrentFlowchartTag

Get the tag of the current flow chart.

GetCurrentNodeTag

Get the tag of the current node in the current flow chart. An optional flow chart argument can be passed to get the same value of another flow chart instance.

GetCurrentNodeOutputCount

Get the number of outputs in the current node of the current flow chart. An optional flow chart argument can be passed to get the same value of another flow chart instance.

GetCurrentNodeOutputNameAt

Get the name of an output in the current node of the current flow chart. Pass in an index or a name as argument. An optional flow chart argument can be passed to get the same value of another flow chart instance.

GetCurrentNodeOutputValueAt

Get the value of an output in the current node of the current flow chart. Pass in an index or a name as argument. An optional flow chart argument can be passed to get the same value of another flow chart instance.

GetCurrentNodeParentCount

Get the amount of parent nodes the current node in the current flow chart has. An optional flow chart argument can be passed to get the same value of another flow chart instance.

GetCurrentNodeParentTag

Get the tag of the parent of the current node in the current flow chart. In the case of having more than one parent an index can be specified to choose which tag to get. An optional flow chart argument can be passed to get the same value of another flow chart instance.

GetCurrentNodeParentIndex

Get the index of the parent of the current node in the current flow chart. In the case of having more than one parent a tag can be specified to choose which index to get. An optional flow chart argument can be passed to get the same value of another flow chart instance.

Example project

Lastly, make sure to check out the example browser for the new example, it is a moderately complex project, showing a possible use for this new feature. It is exactly 50 events, so it will work on the free edition.