Haystack connector

Overview

The Haystack connector reads asset data from a Haystack server. Haystack is an open-source AI framework for building AI agents and Retrieval-Augmented Generation (RAG) applications.

Note: For more information about Haystack refer to Haystack documentation.

How to use the Haystack connector

To use the Haystack connector, pass an object with the parameters you require, from the schema table listed below. Also refer to the various code samples listed below.

Haystack Connector schema

Property	Type	Description	Required
_orchcomp	String	Enter “J2_connector”.	Required
_name	String	Enter a descriptive name for what the component does in the orchestrator.	Optional
_sequenceno	Integer	Enter a number that defines the task’s order in the sequence of tasks to execute.	Required
_actualparams	Object	Pass an object that contains formal parameters or alternative parameters you want to add as key-value pairs, such as “userType”: “<your-user-type>”.	Optional
_actualparams.action	String	Enter one of the following SQL statements: readall, readbyid, readhistory, eval, config.	Required
_actualparams.project	String	Takes the project name as a string.	Required
_actualparams.url	String	Takes a datasource URL as a string.	Required
_actualparams.uname	String	Takes the database username as a string.	Required
_actualparams.pwd	String	Takes the database password as a string.	Required
_actualparams.cmd	Object	Pass a command.	Optional
_actualparams.to	String	To save the request response to File Service, enter "file" as the value.	Optional
_actualparams.fileName	String	Enter the file name including the file extension.	Optional

Generic code example

Refer to the sample code below, with a generic SQL action type.

…
  {
    "_orchcomp": "J2_connector", 
    "_name": "Haystack Connector", 
    "_sequenceno": 1, 
    "_actualparams": {
      "action": "readall" | "readybid" | "readhistory" | "eval" | "config" , 
      "project": "project_name",
      "url": "<URL-to-your-database>",
      "uname": "johnDoe", 
      "pwd": "secret",
      "cmd": "",
      "to": "file",
      "fileName": "haystack-connector-responses.json"
    }
  }, 
…

Code examples for each SQL action type

To use a Haystack connector with a specific SQL action type, refer to the code examples below.

Action type: readall

"_actualparams": {
    "params": {
        "action":"readall",
        "cmd":"point and equipRef == @mechericenter",
        "project": "project_name",
        "url": "<URL-to-your-database>",
        "uname": "<username>", 
        "pwd": "<password>"
    }
}

Action type: readbyid

"_actualparams": {
    "params": {
        "action":"readybid",
        "id": "p:mechericenter:r:2660ba41-851d6a87"
        "project": "project_name",
        "url": "<URL-to-your-database>",
        "uname": "<username>", 
        "pwd": "<password>"
    }
}

Action type: readhistory

"_actualparams": {
    "params": {
        "action":"readhistory",
        "id": "p:mechericenter:r:2660ba41-851d6a87",
        "range":"today",
        "project": "project_name",
        "url": "<URL-to-your-database>",
        "uname": "<username>", 
        "pwd": "<password>"
    }
}

Action type: config

"_actualparams": {
    "params": {
        "action": "config"
        "project": "project_name",
        "url": "<URL-to-your-database>",
        "uname": "<username>", 
        "pwd": "<password>"
    }
}

Action type: eval

"_actualparams": {
    "params": {
        "action":"eval",
        "cmd": "readAll(point).hisRead(date(2024,08,02)..now()).hisClip()",
        "project": "project_name",
        "url": "<URL-to-your-database>",
        "uname": "<username>", 
        "pwd": "<password>"
    }
}