Workflows Introduction

What is a Workflow?

A workflow is a sequence of tasks executed in a specific order on an Android device. It defines the logic, conditions, and flow of automation actions.

Internal Structure: Graph-Based

Internally, workflows are represented as a directed graph with: - Nodes: Individual tasks (Start, End, Normal tasks, Conditions) - Edges/Links: Connections between nodes defining the execution flow - Entry Point: Single START node where execution begins - Exit Points: One or more END nodes where execution terminates

The workflow engine traverses the graph starting from START, follows the links, and executes each node until reaching an END node.

Example JSON structure:

{
  "Start": [
    {
      "id": "0",
      "title": "Network Connectivity Test",
      "variables": [
        {
          "variableName": "$network_status",
          "variableValue": "unknown",
          "is_kpi": false
        }
      ]
    }
  ],

  "CmdStage": [
    {
      "id": "1",
      "title": "Ping Google DNS",
      "cmd_text": "ping -c 1 8.8.8.8",
      "cmd_result_output": "$PING_RESULT",
      "cmd_error_output": "$PING_ERROR"
    }
  ],

  "CompareText": [
    {
      "id": "2",
      "title": "Check if Network Unreachable",
      "text_x": "$PING_ERROR",
      "text_y": "unreachable",
      "compare_type": 2
    }
  ],

  "TextReport": [
    {
      "id": "3",
      "title": "Log Network Down",
      "texte": "Network is unreachable: $PING_ERROR"
    },
    {
      "id": "4",
      "title": "Log Network Up",
      "texte": "Network is OK: $PING_RESULT"
    }
  ],

  "End": [
    {
      "id": "100",
      "title": "Workflow Complete"
    }
  ],

  "Links": [
    {
      "from": "0",
      "to": "1"
    },
    {
      "from": "1",
      "to": "2"
    },
    {
      "from": "2",
      "true": "3",
      "false": "4"
    },
    {
      "from": "3",
      "to": "100"
    },
    {
      "from": "4",
      "to": "100"
    }
  ]
}

Graph visualization:

START (id:0)
    ↓
TextReport (id:1)
    ↓
CompareNumber (id:2)
    ├─→ TRUE → END (id:100)
    └─→ FALSE → TextReport (id:1) [loop back]

---

Understanding Links (Connections)

The Links section defines how nodes are connected and the execution flow:

Link Types

1. Sequential Link (Normal Task)

Connects one task directly to the next:

{
  "from": "0",
  "to": "1"
}

Meaning: After node 0 completes, immediately execute node 1.

---

2. Conditional Link (Decision Node)

Branches execution based on condition result:

{
  "from": "2",
  "true": "100",
  "false": "1"
}

Meaning: - If condition in node 2 is TRUE → go to node 100 - If condition in node 2 is FALSE → go to node 1

---

Graph Traversal Algorithm

The workflow engine:

1. Starts at the START node (id: 0 or similar)
2. Follows the "from" → "to" links
3. Evaluates conditions and branches (true/false)
4. Executes each task at every node visited
5. Terminates when reaching an END node

---

Complete Workflow Example with Graph Structure

Scenario: Check temperature and log result

{
  "Start": [
    {
      "id": "0",
      "title": "Initialize",
      "variables": [
        {
          "variableName": "$temperature",
          "variableValue": "25",
          "is_kpi": true
        }
      ]
    }
  ],

  "TextReport": [
    {
      "id": "1",
      "title": "Log Temperature",
      "texte": "Current temperature: $temperature°C"
    }
  ],

  "CompareNumber": [
    {
      "id": "2",
      "title": "Temperature Check",
      "num_x": "$temperature",
      "num_y": "30",
      "compare_type": 1
    }
  ],

  "TextReport": [
    {
      "id": "3",
      "title": "Too Hot Alert",
      "texte": "Temperature exceeds 30°C: $temperature°C"
    },
    {
      "id": "4",
      "title": "Normal Temperature",
      "texte": "Temperature is OK: $temperature°C"
    }
  ],

  "End": [
    {
      "id": "100"
    }
  ],

  "Links": [
    {
      "from": "0",
      "to": "1"
    },
    {
      "from": "1",
      "to": "2"
    },
    {
      "from": "2",
      "true": "3",
      "false": "4"
    },
    {
      "from": "3",
      "to": "100"
    },
    {
      "from": "4",
      "to": "100"
    }
  ]
}

Graph Flow:

flowchart TD
    Start([START id:0<br/>temperature=25]) --> TextReport1[TextReport id:1<br/>Log Temperature]
    TextReport1 --> CompareNum[CompareNumber id:2<br/>25 > 30?]
    CompareNum -->|TRUE| TextReport3[TextReport id:3<br/>Too Hot Alert]
    CompareNum -->|FALSE| TextReport4[TextReport id:4<br/>Normal Temperature]
    TextReport3 --> End[END id:100]
    TextReport4 --> End

    style Start fill:#e3f2fd
    style End fill:#c8e6c9
    style CompareNum fill:#fff9c4
    style TextReport1 fill:#f3e5f5
    style TextReport3 fill:#ffe0e0
    style TextReport4 fill:#e0f0e0

---

Workflow Components Summary

Component	Purpose	Quantity	JSON Key	Example ID	Description
START Task	Entry point, initializes workflow	Exactly 1	"Start"	"0"	Begins execution.
Normal Tasks	Execute sequentially without conditions	0 or more	Task type (CmdStage, Sleep, etc.)	"1", "2"	Execute one after another in sequence.
Conditional Tasks	Branch execution based on conditions	0 or more	CompareNumber, CompareText	"2"	Evaluate true/false and branch execution.
END Task	Exit point, terminates workflow	1 or more	"End"	"100"	Ends execution. Can have multiple endpoints.
Links	Define connections between nodes	Required	"Links"	N/A	Maps node IDs with "from", "to", "true", "false"

---

Workflow Execution Flow

Basic Flow (Linear)

Linear execution: START → Task 1 → Task 2 → END

JSON Example:

{
  "Start": [{"id": "0"}],
  "CmdStage": [{"id": "1", "cmd_text": "ping -c 1 8.8.8.8"}],
  "Sleep": [{"id": "2", "Time_sleep": 1000}],
  "End": [{"id": "100"}],
  "Links": [
    {"from": "0", "to": "1"},
    {"from": "1", "to": "2"},
    {"from": "2", "to": "100"}
  ]
}

Mermaid Diagram:

flowchart TD
    Start([START id:0]) --> Task1[CmdStage id:1<br/>ping]
    Task1 --> Task2[Sleep id:2<br/>1000ms]
    Task2 --> End[END id:100]

    style Start fill:#e3f2fd
    style End fill:#c8e6c9
    style Task1 fill:#f3e5f5
    style Task2 fill:#fff9c4

---

Branching Flow (Conditional)

Branching execution with true/false paths: START → Task 1 → Condition → (TRUE→Task2 | FALSE→Task3) → END

JSON Example:

{
  "Start": [{"id": "0"}],
  "CmdStage": [{"id": "1", "cmd_text": "ping -c 1 8.8.8.8", "cmd_error_output": "$ERROR"}],
  "CompareText": [{"id": "2", "text_x": "$ERROR", "text_y": "unreachable", "compare_type": 2}],
  "TextReport": [
    {"id": "3", "texte": "Network down"},
    {"id": "4", "texte": "Network up"}
  ],
  "End": [{"id": "100"}],
  "Links": [
    {"from": "0", "to": "1"},
    {"from": "1", "to": "2"},
    {"from": "2", "true": "3", "false": "4"},
    {"from": "3", "to": "100"},
    {"from": "4", "to": "100"}
  ]
}

Mermaid Diagram:

flowchart TD
    Start([START id:0]) --> Task1[CmdStage id:1<br/>ping]
    Task1 --> Condition[CompareText id:2<br/>ERROR contains unreachable?]
    Condition -->|TRUE| Task3[TextReport id:3<br/>Network Down]
    Condition -->|FALSE| Task4[TextReport id:4<br/>Network Up]
    Task3 --> End[END id:100]
    Task4 --> End

    style Start fill:#e3f2fd
    style End fill:#c8e6c9
    style Condition fill:#fff9c4
    style Task1 fill:#f3e5f5
    style Task3 fill:#ffe0e0
    style Task4 fill:#e0f0e0

---