Objective: Take a sample feed, parse it, augment it, and output individual JSON objects for each product in it.
product.txt
- The file contains several fields of basic product information for 1000 products.
- Data for each product is on one line, with fields separated by a vertical pipe character.
This particular retailer classifies products into a particular store, department, category, and subcategory.
Example >
- A Blender might have the following "taxonomy" or category classification.
Home & Garden > Kitchen & Dining > Appliances > Blenders
Store
Department
Category
Subcategory
store_id
department_id
category_id
subcategory_id
These IDs are also represented in a sort of shorthand.
- They should all be prefixed by the first three letters of their respective field name.
A store_id of 1 becomes sto1
A department_id of 2 becomes dep2
A category_id of 3 becomes cat3
A subcategory_id of 4 becomes sub4
targets.txt
- A system to keep track of these various classifications
- Example
- A store_id of sto40 maps to a “target value” of 187298 in PromoteIQ's system
Output Example:
The desired output for the first product is one line of JSON that when pretty-printed looks like:
{
"sku": "22067130",
"name": "Brainboxes ES-571 Brainboxes Es-571 Industrial Isolated Ethernet to Serial + Switch - 1 x Network (R",
"description": "Brainboxes ES-571 Brand New Includes Lifetime Warranty Product # ES-571 Factory floors can be harsh environments.",
"priceList": 387.95,
"taxonomy": [
"sto2",
"dep7",
"cat270",
"sub18532"
],
"targetValues": [
187167,
187177,
187950,
196404
]
}
The taxonomy IDs have been mapped to their long form and placed in the taxonomy property.
- Each taxonomy ID has also been mapped to its corresponding target value ID from the targets.txt file
- and that result is stored in the targetValues
- The priceList property is the product price as a JSON float, not a string
