Converting data from HTTP call into table

Help me Build my Workflow
1 
Hello, I am new to N8N and am using an HTTP call to access data from Boston’s MBTA and local weather data, but when the data comes in it isn’t in useful table format like it is in the original source. I have been trying to Set fields, split out, etc but have been unsuccessful. Does anyone know how to fix this? Code attached here.

{
  "name": "My workflow",
  "nodes": [
    {
      "parameters": {},
      "type": "n8n-nodes-base.manualTrigger",
      "typeVersion": 1,
      "position": [
        0,
        0
      ],
      "id": "da03b5c3-b1af-4a26-a0c1-9fa4ee3fbc37",
      "name": "When clicking ‘Execute workflow’"
    },
    {
      "parameters": {
        "url": "https://services1.arcgis.com/ceiitspzDAHrdGO1/arcgis/rest/services/GSE/FeatureServer/0/query?where=1%3D1&outFields=*&outSR=4326&f=json",
        "options": {}
      },
      "type": "n8n-nodes-base.httpRequest",
      "typeVersion": 4.2,
      "position": [
        208,
        -96
      ],
      "id": "ae6c4a74-5c0f-4e06-b05d-6f8a997fc430",
      "name": "MBTA Ridership"
    },
    {
      "parameters": {
        "url": "http://api.weatherapi.com/v1/history.json?key=90c12bad6a7a44d9858214722250710&q=Boston&dt=2025-07-31",
        "options": {}
      },
      "type": "n8n-nodes-base.httpRequest",
      "typeVersion": 4.2,
      "position": [
        208,
        96
      ],
      "id": "47a47f4f-8f15-48b3-952d-0966ae267e8b",
      "name": "Weather Data"
    }
  ],
  "pinData": {},
  "connections": {
    "When clicking ‘Execute workflow’": {
      "main": [
        [
          {
            "node": "MBTA Ridership",
            "type": "main",
            "index": 0
          },
          {
            "node": "Weather Data",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "MBTA Ridership": {
      "main": [
        []
      ]
    },
    "Weather Data": {
      "main": [
        []
      ]
    }
  },
  "active": false,
  "settings": {
    "executionOrder": "v1"
  },
  "versionId": "63dfb495-dbe6-4fb1-a289-d4ec5e04135b",
  "meta": {
    "templateCredsSetupCompleted": true,
    "instanceId": "e4bf23e9ff769f2de6563787ec1b52c6486bd7b604a7121ea8a23f8adff6f0df"
  },
  "id": "BYNzC4aglPWKelgx",
  "tags": []
2

Hey @Scott_Myers hope all is good. Welcome to the community.

When you say table, what do you mean exactly? How do you see the end result?
An html table? A CSV file? A Google Sheet?

3

Hi, in the original data source, the data is mapped well to a table like this. However, when I pull it into N8N, the table has the wrong headers and is not usable. I am viewing the end result by looking at the end result by looking at my HTTP call output in “table view”. Attached some photos here to show what I am seeing and happy to share the workflow with you as well. Thanks!

table
table2034×1280 266 KB

n8n
n8n948×1094 30.8 KB

4

Yet, again, where do you want the result to go. In n8n the data is in structured JSON format, which is very easy to work with when processing. What is the goal though? What do you need to do with the data? I doubt the end goal is to look at the table in n8n, right? :slight_smile:

5

Apologies, I need to compare how weather conditions affect daily ridership on the T. The end goal is looking to aggregate total ridership per day across all stations, and seeing different rates if it’s raining, based on temperature, etc. I suppose I would do that in excel but it looks like I can also use the Summarize node to do calculations as well.

6

Ok, fair enough. The format you get back from the http request node is the result of the f=json parameter in the URL. Other options are described here if json isn’t something you like.

As for the data being wrong, I assume you could be referring to the service date field represented by a number. This is a common practice to store dates as numbers. You can format it as date with some javascript, for example:

{{ $json.attributes.service_date.toDateTime('ms') }} will turn 1756612800000 into 2025-08-31T01:00:00.000-03:00 (that is for me, could will be different for you, given it is likely you are in a different timezone).

7

I tried those different formats and none of them seem to map the data to the original table format shown in the source. From watching videos it might have something to do with setting fields, but I have tried and haven’t gotten it to recognize the right fields. Notice how in N8N the field names are “objectIDFieldName” etc instead of “service_date” so I think something is confusing it there.

8

All the data is still there, if you look at the resulting data (ignore the fields array), each row is represented by a structure inside the “features” field

image
image1148×1279 132 KB

if you wish to split the features up and have many items where each item is a row from the table, and the service_date converted to a readable date, you could do something like this

9

The Split node was it, thanks so much for your help!

1 Like