Connection Creator is a visual builder. You place nodes on a canvas and connect them with arrows. Together they form a scenario = a flow that runs whenever something happens in PalDock.
Nodes and connections
- Node = one action. Send an HTTP request, set a variable, modify a field, wait, finish.
- Connection = an arrow between two nodes. It decides what runs next. Arrows can carry conditions.
Scenario and run
- Scenario is the flow you build and save.
- Run is one execution of it. Every lead, postback or form submit starts its own run. Runs are logged separately, so you can always see what happened to one specific lead.
Types of scenarios
A scenario always belongs to one context. The context decides what triggers it and what data it receives.
- Integration: sends a lead to an advertiser and processes the answer.
- Structure:
- validation: checks a value while the form is being filled in, and can block the submit.
- modification: fills a field with a value from an external service (Autofill).
- autocomplete: suggests values to the visitor while typing.
- Tracking:
- S2S postback (incoming): processes a conversion or status update sent to PalDock.
- Tracking API (outgoing): sends conversion and status data from PalDock to affiliates or other systems, and can also poll an advertiser for a result.
- Feed: pulls product or offer data from an external source.
How data moves
The first node receives the incoming data – such as form fields, tracking parameters, system tags and others. Each node then reads that data, does its job, and writes its own values back. Everything travels forward. A node can use any value produced by any node before it. Nothing is passed backwards.
Order
Nodes run one after another, following the arrows.
- If a node has several outgoing arrows, every arrow whose condition is met is followed. Arrows are evaluated in the order they are listed on the node, so if two of them can both apply, the one higher in the list runs first.
- If a condition is not met, that arrow is skipped.
- A branch can end on its own without stopping the other branches.
Immediate or background
Some scenarios must return an answer while the visitor is waiting such as form validation, autocomplete or a lead sent over the API. These run immediately and the caller waits for the result. Others run in the background such as outgoing Tracking API. They are queued and processed without blocking anything.
Pausing
A Wait node pauses the run for a set time. A Webhook node pauses it until an external system calls back. The run stays open, keeps its data, and continues from the same place.
Errors and limits
- A node runs once, with no automatic retry. Every part of the flow also has a time limit, and a scenario that runs while a visitor waits has the tightest one of all. See Limits and timeouts.
- Failures are counted and written to the run log. Use response conditions to decide what should happen next instead of relying on the failure itself.
- A Breaker node limits how many times a loop may repeat, so a flow can never run forever.
- A scenario can start another scenario, but a scenario cannot call itself in a circle.
Where to look when something breaks
Every run stores the input and output of each node. Open the run and follow it node by node and you will see the exact data each node received and returned.
