My solution for the homework

.vscode/launch.json

@@ -0,0 +1,15 @@
+{
+    // Use IntelliSense to learn about possible attributes.
+    // Hover to view descriptions of existing attributes.
+    // For more information, visit: https://go.microsoft.com/fwlink/?linkid=830387
+    "version": "0.2.0",
+    "configurations": [
+        {
+            "name": "Python Debugger: Current File",
+            "type": "debugpy",
+            "request": "launch",
+            "program": "main.py",
+            "console": "integratedTerminal"
+        }
+    ]
+}

main.py

@@ -0,0 +1,43 @@
+import pandas as pd
+import json
+from utils import Utils
+
+# Load header mapping from JSON file
+with open('mapping.json', 'r') as json_file:
+    header_mapping = json.load(json_file)
+
+# Define input and output filenames
+input_file = 'homework.csv'
+output_file = 'formatted.csv'
+
+# Read input CSV file using pandas
+df = pd.read_csv(input_file)
+
+# Process each row using the header mapping and Utils
+formatted_rows = []
+for index, row in df.iterrows():
+    formatted_row = Utils.process_row(row, header_mapping)
+
+    # If using separate columns for width and depth
+    width = formatted_row.get('furniture seat width (inches)', None)
+    depth = formatted_row.get('furniture seat depth (inches)', None)
+
+    formatted_row['attrib__seat_width'] = width
+    formatted_row['attrib__seat_depth'] = depth
+
+    formatted_rows.append(formatted_row)
+
+# Create a DataFrame from the list of formatted rows
+formatted_df = pd.DataFrame(formatted_rows)
+
+# Drop columns starting with '_'
+columns_to_drop = [col for col in formatted_df.columns if col.startswith('_')]
+df_filtered = formatted_df.drop(columns=columns_to_drop)
+
+df_filtered['product__country_of_origin__alpha_3'] = df_filtered['product__country_of_origin__alpha_3'].apply(Utils.convert_country_name_to_iso2)
+df_filtered['attrib__ul_certified'] = df_filtered['attrib__ul_certified'].apply(Utils.convert_ul_certified)
+df_filtered['ean13'] = df_filtered['ean13'].astype(str)
+
+df_filtered.to_csv(output_file, index=False)
+
+print(f"Transformation completed. Transformed data written to '{output_file}'.")

mapping.json

@@ -0,0 +1,82 @@
+{
+    "manufacturer_sku": "item number",
+    "ean13": "upc",
+    "weight": "item weight (pounds)",
+    "length": "_function_",
+    "width": "item width (inches)",
+    "height": "item height (inches)",
+
+    "_depth": "item depth (inches)",
+    "_diameter": "item diameter (inches)",
+
+    "prop_65": "certified damp/wet",
+    "cost_price": "msrp ($)",
+    "min_price": "map ($)",
+    "made_to_order": "",
+    "product__product_class__name": "item category",
+    "product__brand__name": "brand",
+    "product__title": "description",
+    "product__description": "long description",
+    "product__bullets__0": "selling point 1",
+    "product__bullets__1": "selling point 2",
+    "product__bullets__2": "selling point 3",
+    "product__bullets__3": "selling point 4",
+    "product__bullets__4": "selling point 5",
+    "product__bullets__5": "selling point 6",
+    "product__bullets__6": "selling point 7",
+    "product__configuration__codes": "",
+    "product__multipack_quantity": "min order qty",
+    "product__country_of_origin__alpha_3": "country of origin",
+    "product__parent_sku": "related items",
+    "attrib__arm_height": "furniture arm height (inches)",
+    "attrib__assembly_required": "url instruction sheet",
+    "attrib__back_material": "",
+    "attrib__blade_finish": "",
+    "attrib__bulb_included": "bulb 1 included",
+    "attrib__bulb_type": "bulb 1 type",
+    "attrib__color": "primary color family",
+    "attrib__cord_length": "cord length (inches)",
+    "attrib__design_id": "",
+    "attrib__designer": "licensed by",
+    "attrib__distressed_finish": "",
+    "attrib__fill": "",
+    "attrib__finish": "item finish",
+    "attrib__frame_color": "",
+    "attrib__hardwire": "",
+    "attrib__kit": "conversion kit option",
+    "attrib__leg_color": "",
+    "attrib__leg_finish": "",
+    "attrib__material": "item materials",
+    "attrib__number_bulbs": "bulb 1 count",
+    "attrib__orientation": "",
+    "attrib__outdoor_safe": "outdoor",
+    "attrib__pile_height": "",
+    "attrib__seat_depth": "",
+    "attrib__seat_width": "",
+    "attrib__seat_height": "furniture seat height (inches)",
+    "attrib__shade": "shade shape",
+    "attrib__size": "",
+    "attrib__switch_type": "switch type",
+    "attrib__ul_certified": "safety rating",
+    "attrib__warranty_years": "",
+    "attrib__wattage": "voltage",
+    "attrib__weave": "",
+    "attrib__weight_capacity": "furniture weight capacity (pounds)",
+    "boxes__0__weight": "carton 1 weight (pounds)",
+    "boxes__0__length": "carton 1 length (inches)",
+    "boxes__0__height": "carton 1 height (inches)",
+    "boxes__0__width": "carton 1 width (inches)",
+    "boxes__1__weight": "carton 2 weight (pounds)",
+    "boxes__1__length": "carton 2 length (inches)",
+    "boxes__1__height": "carton 2 height (inches)",
+    "boxes__1__width": "carton 2 width (inches)",
+    "boxes__2__weight": "carton 3 weight (pounds)",
+    "boxes__2__length": "carton 3 length (inches)",
+    "boxes__2__height": "carton 3 height (inches)",
+    "boxes__2__width": "carton 3 width (inches)",
+    "boxes__3__weight": "",
+    "boxes__3__length": "",
+    "boxes__3__height": "",
+    "boxes__3__width": "",
+    "product__styles": "item style"
+}

requirements.txt

@@ -0,0 +1,37 @@
+asttokens==2.4.1
+comm==0.2.2
+country-converter==1.2
+debugpy==1.8.1
+decorator==5.1.1
+et-xmlfile==1.1.0
+exceptiongroup==1.2.1
+executing==2.0.1
+ipykernel==6.29.4
+ipython==8.24.0
+jedi==0.19.1
+jupyter_client==8.6.1
+jupyter_core==5.7.2
+matplotlib-inline==0.1.7
+nest-asyncio==1.6.0
+numpy==1.26.4
+openpyxl==3.1.2
+packaging==24.0
+pandas==2.2.2
+parso==0.8.4
+pexpect==4.9.0
+platformdirs==4.2.1
+prompt-toolkit==3.0.43
+psutil==5.9.8
+ptyprocess==0.7.0
+pure-eval==0.2.2
+Pygments==2.18.0
+python-dateutil==2.9.0.post0
+pytz==2024.1
+pyzmq==26.0.3
+six==1.16.0
+stack-data==0.6.3
+tornado==6.4
+traitlets==5.14.3
+typing_extensions==4.11.0
+tzdata==2024.1
+wcwidth==0.2.13

utils.py

@@ -0,0 +1,113 @@
+from datetime import datetime
+from decimal import Decimal
+
+
+class Utils:
+
+    CONVERTER_CM = 0.393701
+    CONVERTER_MM = 0.0393701
+    CONVERTER_KG_TO_POUNDS = 2.20462
+
+    @staticmethod
+    def process_row(row, header_mapping):
+        formatted_row = {}
+
+        for output_field, input_field in header_mapping.items():
+            try:
+                value = row[input_field]
+            except:
+                value = ''
+
+            if isinstance(value, str):
+                value = value.strip()  # Strip whitespace if it's a string
+
+            if any(field in output_field for field in ['width', 'height', 'depth']) and 'inches' not in output_field:
+                value = Utils.convert_to_inches(value, 'inches')
+            elif output_field == 'weight':
+                value = Utils.convert_to_pounds(value, 'pounds')
+            elif output_field == 'upc':
+                value = str(value)
+            elif output_field == 'system creation date':
+                value = Utils.format_date(value)
+            elif output_field in ['wholesale ($)', 'map ($)', 'msrp ($)']:
+                value = Utils.format_currency(value)
+            elif output_field == 'length':
+                # Calculate length using furniture dimensions
+                length = Utils.calculate_furniture_length(row)
+                formatted_row[output_field] = length
+                continue
+            formatted_row[output_field] = value
+
+        return formatted_row
+
+    @staticmethod
+    def convert_to_inches(value, unit):
+        try:
+            if unit.lower() == 'inches':
+                return float(value)
+            elif unit.lower() == 'cm':
+                return float(value) * Utils.CONVERTER_CM
+            elif unit.lower() == 'mm':
+                return float(value) * Utils.CONVERTER_MM
+            else:
+                return float(value)  # Assume already in inches if unknown unit
+        except:
+            return None
+
+    @staticmethod
+    def convert_to_pounds(value, unit):
+        if unit.lower() == 'pounds':
+            return float(value)
+        elif unit.lower() == 'kg':
+            return float(value) * Utils.CONVERTER_KG_TO_POUNDS
+        else:
+            return float(value)
+
+    @staticmethod
+    def format_currency(value):
+        try:
+            return Decimal(value).quantize(Decimal('0.01'))
+        except:
+            return None
+
+    @staticmethod
+    def format_date(date_str):
+        try:
+            return datetime.strptime(date_str, '%m/%d/%y').strftime('%Y-%m-%d')
+        except:
+            return None
+
+    @staticmethod
+    def calculate_furniture_length(row):
+        width = float(row['item width (inches)'])
+        depth = float(row['item depth (inches)'])
+        height = float(row['item height (inches)'])
+        diameter = float(row['item diameter (inches)'])
+        dimensions = [width, depth, height, diameter]
+        return max(dimensions)
+
+    @staticmethod
+    def convert_country_name_to_iso2(country_name):
+        ref = {
+            "China": "CHN",
+            "India": "IND",
+            "Indonesia": "IDN",
+            "Phillipines": "PHL",
+            "Thailand": "THA",
+            "Vietnam": "VNM",
+        }
+        try:
+            return ref[country_name]
+        except:
+            return None
+
+    @staticmethod
+    def convert_ul_certified(ul):
+        ref = {
+            "UL": 'True',
+            "UL/CUL": 'True',
+        }
+        try:
+            return ref[ul]
+        except:
+            return False