Advanced: nested columns example

This example shows using nested sections to extract repeating vertical section groups.

Overview

To give a brief overview of using vertical sections for columns, the following image shows extracting numbered headings and their columns with these steps:

define a section group
define a nested vertical section group

Click to enlarge

The following config uses abbreviated YML notation to give an overview of the more complex SenseML JSON:

fields:
  - id: parentSections
    type: section
    range:
      anchor: heading
    fields:
      - id: sectionTitle
        method:
          id: passthrough
        anchor: heading
    sections:
      - id: nestedColumns
        range:
          direction: vertical
          anchor: column
        fields:
          - id: columnTitle
            method:
              id: passthrough
            anchor: column

With this approach, you can output something like the following, using abbreviated YML notation to give an overview of the more complex JSON extraction response:

---
parentSections:
- sectionTitle:
    type: string
    value: Heading 1
  nestedColumns:
  - columnTitle:
      type: string
      value: Column A
  - columnTitle:
      type: string
      value: Column B
  - columnTitle:
      type: string
      value: Column C
- sectionTitle:
    type: string
    value: Heading 2
  nestedColumns:
  - columnTitle:
      type: string
      value: Column D
  - columnTitle:
      type: string
      value: Column E
  - columnTitle:
      type: string
      value: Column F

Details

The following elaborates on the preceding overview using JSON instead of YML.

Config

{
  "fields": [
    {
      "id": "parentSections",
      "type": "sections",
      "range": {
        "anchor": {
          "match": {
            "type": "startsWith",
            "text": "heading"
          }
        }
      },
      "fields": [
        {
          "id": "sectionTitle",
          "method": {
            "id": "passthrough"
          },
          "anchor": {
            "match": {
              "type": "startsWith",
              "text": "heading"
            }
          }
        }
      ],
      "sections": [
        {
          "id": "nestedColumns",
          "range": {
            "direction": "vertical",
            "anchor": {
              "match": {
                "type": "startsWith",
                "text": "column"
              }
            }
          },
          "fields": [
            {
              "id": "columnTitle",
              "method": {
                "id": "passthrough"
              },
              "anchor": {
                "match": {
                  "type": "startsWith",
                  "text": "column"
                }
              }
            }
          ]
        }
      ]
    }
  ]
}

Example document
The following image shows the example document used with this example config:

Click to enlarge

Example document	Download link

Output

{
  "parentSections": [
    {
      "sectionTitle": {
        "type": "string",
        "value": "Heading 1"
      },
      "nestedColumns": [
        {
          "columnTitle": {
            "type": "string",
            "value": "Column A"
          }
        },
        {
          "columnTitle": {
            "type": "string",
            "value": "Column B"
          }
        },
        {
          "columnTitle": {
            "type": "string",
            "value": "Column C"
          }
        }
      ]
    },
    {
      "sectionTitle": {
        "type": "string",
        "value": "Heading 2"
      },
      "nestedColumns": [
        {
          "columnTitle": {
            "type": "string",
            "value": "Column D"
          }
        },
        {
          "columnTitle": {
            "type": "string",
            "value": "Column E"
          }
        },
        {
          "columnTitle": {
            "type": "string",
            "value": "Column F"
          }
        }
      ]
    }
  ]
}