Field Name Mapping

What is Field Name Mapping?

Field name mapping is the process of linking or aligning the names of fields from one data source or system to another, especially when they have different naming conventions or structures.

If you are using a custom data feed to index your products, the field names in your feed may differ from (or may not exist in) Doofinder’s default field names.

To successfully merge or integrate fields, you will need to use Doofinder’s field name mapping:

How to Add Field Name Mappings

You can access the Field name mapping section through the Admin Panel. Here’s how:

1. In your Admin Panel, go to Configuration > Indices.
2. On the Product tab, click the three dots next to the Product source. A dropdown menu will appear.
3. Click on Field name mapping.

4. The Field Name Mapping pop-up will appear with two sections:
• Normalized Field Name: indicates how the field is indexed in Doofinder.
• Field Aliases: indicates how the field appears in your feed.

  In the Field Aliases section, enter the field names used in your data feed that should correspond to Doofinder's normalized field names (if they are not already listed).

  For example:

  • Your product data feed uses g:availability, marque, product type.
  • While Doofinder uses and expects availability, brand, categories.

5. Save and index.

   During the next indexing process, Doofinder will apply the new aliases to retrieve and display the relevant data from your feed.

How to Add New Fields

You can also add new fields to index fields from your feed not covered in Doofinder’s default settings.

1. Click the "+Add new field" button.
2. Under Normalized Field Names, add the new field you would like to index in Doofinder.
3. Under Field Aliases, add the names of the fields from your data feed that should be mapped to the normalized field name.

Field Name Mapping Types

Doofinder supports two types of field mapping:

Basic Field Mapping

Used for mapping simple fields:

• id from product_id
• title from name
• price from base_price

xml in-feed field example:

<color> black </color> 

Deep Field Mapping

Used for mapping nested attributes within an object or list:

• color from attributes.color
• type from custom_fields.specifications.type
• attributes from custom_fields.attributes

xml in-feed field example:

<attributes>
<color> black </color>
<size>L</size>
<category>clothing</category>
</attributes> 

How Does Deep Mapping Work?

There are three ways to map nested attributes:

• Nested Fields

To access any nested attribute and use it as a first-level field, reference the full attribute path (there is no attribute level limitation): custom_fields.specifications.type

• Indexing Arrays

To retrieve the first item from a list or array, use square brackets with a number (starting at 0): images.[0]

• Filtering Arrays

To target specific items based on attribute values, use the following format: attributes.[name=color].value

This fetches the value of the item where name=color.

Your field name mapping will look like this:

Learn by Example - How to Map Nested Values with Deep Field Mapping

Imagine you have a product data feed that looks like this:

{
  "id": "12345",
  "name": "Premium Wireless Headphones",
  "price": 299.99,
  "custom_fields": {
    "attributes": [
      {
        "name": "color",
        "value": "black"
      },
      {
        "name": "warranty",
        "value": "2 years"
      },
      {
        "name": "brand",
        "value": "AudioTech"
      }
    ],
    "specifications": [
      {
        "type": "battery_life",
        "value": "30 hours"
      },
      {
        "type": "connectivity",
        "value": "Bluetooth 5.0"
      }
    ],
    "images": [
      "https://example.com/headphones-1.jpg",
      "https://example.com/headphones-2.jpg",
      "https://example.com/headphones-3.jpg"
    ],
    "variants": [
      {
        "sku": "WH-001-BLK",
        "price": 299.99,
        "stock": 50
      },
      {
        "sku": "WH-001-WHT", 
        "price": 299.99,
        "stock": 25
      }
    ]
  }
}

You want to map:

• The second image, using indexing: custom_fields.images.[1]

• The Bluetooth 5.0 value using filter: custom_fields.specifications.[type=connectivity].value

Your field name mapping will look like this:

This is the feed output:

{
  "id": "product_id",
  "name": "title",
  "price": "base_price",
  "custom_fields.attributes.[name=color].value": "product_color",
  "custom_fields.attributes.[name=brand].value": "brand_name", 
  "custom_fields.images.[1]": "main_image",
  "custom_fields.variants.[0].sku": "primary_sku",
  "custom_fields.specifications.[type=connectivity].value": "connectivity"
}

This is the output in Results Preview in the product information box:

{
  "product_id": "12345",
  "title": "Premium Wireless Headphones",
  "base_price": 299.99,
  "product_color": "black",
  "brand_name": "AudioTech",
  "main_image": "https://example.com/headphones-1.jpg",
  "primary_sku": "WH-001-BLK",
  "connectivity": "Bluetooth 5.0"
}

Troubleshooting

An Error Appears During First Indexation

To get your data feed properly indexed, Doofinder requires the ID and Title. If one or both of them have a field name in your feed that is not in the expected Field Aliases, Doofinder will not recognize it, and the index will deliver an error message (Unusual Issues 5.1 All other logs not belonging to any other category), such as this:

"Too much warnings. Last warning: id attribute(s) missing" See more details

You can use the Normalized Field Names section to fix this by setting the correspondence between the field name in the data feed and the normalized one. Once completed, save the changes.

The Layer Doesn’t Show Some Information in the Results

Doofinder should show the following fields in the results:

• image_link
• title
• price
• sale_price (if exists)
• link

If some of this data is missing, the problem could be a non-recognized field name in your feed. To fix it, detect the non-recognized field name and configure the alias.

Say, for instance, we need a title field sent to Doofinder. If the data you sent in the field is called product_name, you should customize the field title with the field product_name.

Depending on how the feed export is configured, you should know the name of this field from your feed.

Otherwise, you can open the feed and check it.

How to Recover an Object After Deep Field Mapping

If you have already mapped some nested values and want to add more but can’t recall which values are available (and you don’t have access to the full feed):

1. Temporarily remove the mappings from that object in Field name mapping.
2. Save and Index. The full object will reappear in your feed. Remember, they only disappear from the index.
3. Review its structure and add the desired fields.
4. Re-apply and apply new mappings and save.