{
"cells": [
{
"cell_type": "markdown",
"id": "640208dc",
"metadata": {
"tags": [
"papermill-error-cell-tag"
]
},
"source": [
"An Exception was encountered at 'In [2]'."
]
},
{
"cell_type": "markdown",
"id": "8c387081",
"metadata": {
"papermill": {
"duration": 0.012814,
"end_time": "2026-08-05T13:26:48.860275+00:00",
"exception": false,
"start_time": "2026-08-05T13:26:48.847461+00:00",
"status": "completed"
},
"tags": []
},
"source": [
"# Phase 2 — Semantic Normalization / Identity Extraction\n",
"\n",
"**Goal:** Normalize brands, extract identity attributes from original titles, parse variant attributes, build core titles and identity hashes.\n",
"\n",
"**Input:** `phase1_output.parquet` (4,440 × 14)\n",
"\n",
"**Output:** `phase2_output.parquet` (4,440 × 34) + `phase2_validation.json` + `phase2_statistics.json`\n",
"\n",
"**CRITICAL:** Identity information is extracted from the ORIGINAL title BEFORE any destructive normalization.\n",
"\n",
"**No LLM used. All deterministic/rule-based.**"
]
},
{
"cell_type": "code",
"execution_count": 1,
"id": "45fbf818",
"metadata": {
"execution": {
"iopub.execute_input": "2026-08-05T13:26:48.882368Z",
"iopub.status.busy": "2026-08-05T13:26:48.881790Z",
"iopub.status.idle": "2026-08-05T13:26:49.992622Z",
"shell.execute_reply": "2026-08-05T13:26:49.988873Z"
},
"papermill": {
"duration": 1.121741,
"end_time": "2026-08-05T13:26:49.994614+00:00",
"exception": false,
"start_time": "2026-08-05T13:26:48.872873+00:00",
"status": "completed"
},
"tags": []
},
"outputs": [],
"source": [
"import json\n",
"import re\n",
"from datetime import datetime, timezone\n",
"\n",
"import pandas as pd\n",
"import numpy as np\n",
"\n",
"VERSION = \"2.0.0\"\n",
"TIMESTAMP = datetime.now(timezone.utc).isoformat()\n",
"\n",
"def log(msg):\n",
" print(f\"[Phase2] {msg}\")"
]
},
{
"cell_type": "markdown",
"id": "49786aa1",
"metadata": {
"papermill": {
"duration": 0.004796,
"end_time": "2026-08-05T13:26:50.004290+00:00",
"exception": false,
"start_time": "2026-08-05T13:26:49.999494+00:00",
"status": "completed"
},
"tags": []
},
"source": [
"## A. Load Phase 1 Output"
]
},
{
"cell_type": "markdown",
"id": "71172033",
"metadata": {
"tags": [
"papermill-error-cell-tag"
]
},
"source": [
"Execution using papermill encountered an exception here and stopped:"
]
},
{
"cell_type": "code",
"execution_count": 2,
"id": "184d1d7d",
"metadata": {
"execution": {
"iopub.execute_input": "2026-08-05T13:26:50.019388Z",
"iopub.status.busy": "2026-08-05T13:26:50.017616Z",
"iopub.status.idle": "2026-08-05T13:26:50.607235Z",
"shell.execute_reply": "2026-08-05T13:26:50.603169Z"
},
"papermill": {
"duration": 0.599348,
"end_time": "2026-08-05T13:26:50.609679+00:00",
"exception": true,
"start_time": "2026-08-05T13:26:50.010331+00:00",
"status": "failed"
},
"tags": []
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"[Phase2] Loading Phase 1 output ...\n"
]
},
{
"ename": "FileNotFoundError",
"evalue": "[Errno 2] No such file or directory: 'phase1_output.parquet'",
"output_type": "error",
"traceback": [
"\u001b[31m---------------------------------------------------------------------------\u001b[39m",
"\u001b[31mFileNotFoundError\u001b[39m Traceback (most recent call last)",
"\u001b[36mCell\u001b[39m\u001b[36m \u001b[39m\u001b[32mIn[2]\u001b[39m\u001b[32m, line 2\u001b[39m\n\u001b[32m 1\u001b[39m log(\u001b[33m\"Loading Phase 1 output ...\"\u001b[39m)\n\u001b[32m----> \u001b[39m\u001b[32m2\u001b[39m df = pd.read_parquet(\u001b[33m\"phase1_output.parquet\"\u001b[39m)\n\u001b[32m 3\u001b[39m log(f\"Loaded: {df.shape[\u001b[32m0\u001b[39m]} rows, {df.shape[\u001b[32m1\u001b[39m]} columns\")\n\u001b[32m 4\u001b[39m df.head()\n",
"\u001b[36mFile \u001b[39m\u001b[32m~/gsoc/compliments-reference-db/.venv/lib/python3.13/site-packages/pandas/io/parquet.py:671\u001b[39m, in \u001b[36mread_parquet\u001b[39m\u001b[34m(path, engine, columns, storage_options, dtype_backend, filesystem, filters, to_pandas_kwargs, **kwargs)\u001b[39m\n\u001b[32m 668\u001b[39m impl = get_engine(engine)\n\u001b[32m 669\u001b[39m check_dtype_backend(dtype_backend)\n\u001b[32m--> \u001b[39m\u001b[32m671\u001b[39m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[30;43mimpl\u001b[39;49m\u001b[30;43m.\u001b[39;49m\u001b[30;43mread\u001b[39;49m\u001b[30;43m(\u001b[39;49m\n\u001b[32m 672\u001b[39m \u001b[30;43m \u001b[39;49m\u001b[30;43mpath\u001b[39;49m\u001b[30;43m,\u001b[39;49m\n\u001b[32m 673\u001b[39m \u001b[30;43m \u001b[39;49m\u001b[30;43mcolumns\u001b[39;49m\u001b[30;43m=\u001b[39;49m\u001b[30;43mcolumns\u001b[39;49m\u001b[30;43m,\u001b[39;49m\n\u001b[32m 674\u001b[39m \u001b[30;43m \u001b[39;49m\u001b[30;43mfilters\u001b[39;49m\u001b[30;43m=\u001b[39;49m\u001b[30;43mfilters\u001b[39;49m\u001b[30;43m,\u001b[39;49m\n\u001b[32m 675\u001b[39m \u001b[30;43m \u001b[39;49m\u001b[30;43mstorage_options\u001b[39;49m\u001b[30;43m=\u001b[39;49m\u001b[30;43mstorage_options\u001b[39;49m\u001b[30;43m,\u001b[39;49m\n\u001b[32m 676\u001b[39m \u001b[30;43m \u001b[39;49m\u001b[30;43mdtype_backend\u001b[39;49m\u001b[30;43m=\u001b[39;49m\u001b[30;43mdtype_backend\u001b[39;49m\u001b[30;43m,\u001b[39;49m\n\u001b[32m 677\u001b[39m \u001b[30;43m \u001b[39;49m\u001b[30;43mfilesystem\u001b[39;49m\u001b[30;43m=\u001b[39;49m\u001b[30;43mfilesystem\u001b[39;49m\u001b[30;43m,\u001b[39;49m\n\u001b[32m 678\u001b[39m \u001b[30;43m \u001b[39;49m\u001b[30;43mto_pandas_kwargs\u001b[39;49m\u001b[30;43m=\u001b[39;49m\u001b[30;43mto_pandas_kwargs\u001b[39;49m\u001b[30;43m,\u001b[39;49m\n\u001b[32m 679\u001b[39m \u001b[30;43m \u001b[39;49m\u001b[30;43m*\u001b[39;49m\u001b[30;43m*\u001b[39;49m\u001b[30;43mkwargs\u001b[39;49m\u001b[30;43m,\u001b[39;49m\n\u001b[32m 680\u001b[39m \u001b[30;43m\u001b[39;49m\u001b[30;43m)\u001b[39;49m\n",
"\u001b[36mFile \u001b[39m\u001b[32m~/gsoc/compliments-reference-db/.venv/lib/python3.13/site-packages/pandas/io/parquet.py:253\u001b[39m, in \u001b[36mPyArrowImpl.read\u001b[39m\u001b[34m(self, path, columns, filters, dtype_backend, storage_options, filesystem, to_pandas_kwargs, **kwargs)\u001b[39m\n\u001b[32m 240\u001b[39m \u001b[38;5;28;01mdef\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[34mread\u001b[39m(\n\u001b[32m 241\u001b[39m \u001b[38;5;28mself\u001b[39m,\n\u001b[32m 242\u001b[39m path,\n\u001b[32m (...)\u001b[39m\u001b[32m 249\u001b[39m **kwargs,\n\u001b[32m 250\u001b[39m ) -> DataFrame:\n\u001b[32m 251\u001b[39m kwargs[\u001b[33m\"\u001b[39m\u001b[33muse_pandas_metadata\u001b[39m\u001b[33m\"\u001b[39m] = \u001b[38;5;28;01mTrue\u001b[39;00m\n\u001b[32m--> \u001b[39m\u001b[32m253\u001b[39m path_or_handle, handles, filesystem = \u001b[30;43m_get_path_or_handle\u001b[39;49m\u001b[30;43m(\u001b[39;49m\n\u001b[32m 254\u001b[39m \u001b[30;43m \u001b[39;49m\u001b[30;43mpath\u001b[39;49m\u001b[30;43m,\u001b[39;49m\n\u001b[32m 255\u001b[39m \u001b[30;43m \u001b[39;49m\u001b[30;43mfilesystem\u001b[39;49m\u001b[30;43m,\u001b[39;49m\n\u001b[32m 256\u001b[39m \u001b[30;43m \u001b[39;49m\u001b[30;43mstorage_options\u001b[39;49m\u001b[30;43m=\u001b[39;49m\u001b[30;43mstorage_options\u001b[39;49m\u001b[30;43m,\u001b[39;49m\n\u001b[32m 257\u001b[39m \u001b[30;43m \u001b[39;49m\u001b[30;43mmode\u001b[39;49m\u001b[30;43m=\u001b[39;49m\u001b[30;43m\"\u001b[39;49m\u001b[30;43mrb\u001b[39;49m\u001b[30;43m\"\u001b[39;49m\u001b[30;43m,\u001b[39;49m\n\u001b[32m 258\u001b[39m \u001b[30;43m \u001b[39;49m\u001b[30;43m)\u001b[39;49m\n\u001b[32m 259\u001b[39m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[32m 260\u001b[39m pa_table = \u001b[38;5;28mself\u001b[39m.api.parquet.read_table(\n\u001b[32m 261\u001b[39m path_or_handle,\n\u001b[32m 262\u001b[39m columns=columns,\n\u001b[32m (...)\u001b[39m\u001b[32m 265\u001b[39m **kwargs,\n\u001b[32m 266\u001b[39m )\n",
"\u001b[36mFile \u001b[39m\u001b[32m~/gsoc/compliments-reference-db/.venv/lib/python3.13/site-packages/pandas/io/parquet.py:141\u001b[39m, in \u001b[36m_get_path_or_handle\u001b[39m\u001b[34m(path, fs, storage_options, mode, is_dir)\u001b[39m\n\u001b[32m 131\u001b[39m handles = \u001b[38;5;28;01mNone\u001b[39;00m\n\u001b[32m 132\u001b[39m \u001b[38;5;28;01mif\u001b[39;00m (\n\u001b[32m 133\u001b[39m \u001b[38;5;129;01mnot\u001b[39;00m fs\n\u001b[32m 134\u001b[39m \u001b[38;5;129;01mand\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m is_dir\n\u001b[32m (...)\u001b[39m\u001b[32m 139\u001b[39m \u001b[38;5;66;03m# fsspec resources can also point to directories\u001b[39;00m\n\u001b[32m 140\u001b[39m \u001b[38;5;66;03m# this branch is used for example when reading from non-fsspec URLs\u001b[39;00m\n\u001b[32m--> \u001b[39m\u001b[32m141\u001b[39m handles = \u001b[30;43mget_handle\u001b[39;49m\u001b[30;43m(\u001b[39;49m\n\u001b[32m 142\u001b[39m \u001b[30;43m \u001b[39;49m\u001b[30;43mpath_or_handle\u001b[39;49m\u001b[30;43m,\u001b[39;49m\u001b[30;43m \u001b[39;49m\u001b[30;43mmode\u001b[39;49m\u001b[30;43m,\u001b[39;49m\u001b[30;43m \u001b[39;49m\u001b[30;43mis_text\u001b[39;49m\u001b[30;43m=\u001b[39;49m\u001b[30;43;01mFalse\u001b[39;49;00m\u001b[30;43m,\u001b[39;49m\u001b[30;43m \u001b[39;49m\u001b[30;43mstorage_options\u001b[39;49m\u001b[30;43m=\u001b[39;49m\u001b[30;43mstorage_options\u001b[39;49m\n\u001b[32m 143\u001b[39m \u001b[30;43m \u001b[39;49m\u001b[30;43m)\u001b[39;49m\n\u001b[32m 144\u001b[39m fs = \u001b[38;5;28;01mNone\u001b[39;00m\n\u001b[32m 145\u001b[39m path_or_handle = handles.handle\n",
"\u001b[36mFile \u001b[39m\u001b[32m~/gsoc/compliments-reference-db/.venv/lib/python3.13/site-packages/pandas/io/common.py:939\u001b[39m, in \u001b[36mget_handle\u001b[39m\u001b[34m(path_or_buf, mode, encoding, compression, memory_map, is_text, errors, storage_options)\u001b[39m\n\u001b[32m 930\u001b[39m handle = \u001b[38;5;28mopen\u001b[39m(\n\u001b[32m 931\u001b[39m handle,\n\u001b[32m 932\u001b[39m ioargs.mode,\n\u001b[32m (...)\u001b[39m\u001b[32m 935\u001b[39m newline=\u001b[33m\"\u001b[39m\u001b[33m\"\u001b[39m,\n\u001b[32m 936\u001b[39m )\n\u001b[32m 937\u001b[39m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[32m 938\u001b[39m \u001b[38;5;66;03m# Binary mode\u001b[39;00m\n\u001b[32m--> \u001b[39m\u001b[32m939\u001b[39m handle = \u001b[30;43mopen\u001b[39;49m\u001b[30;43m(\u001b[39;49m\u001b[30;43mhandle\u001b[39;49m\u001b[30;43m,\u001b[39;49m\u001b[30;43m \u001b[39;49m\u001b[30;43mioargs\u001b[39;49m\u001b[30;43m.\u001b[39;49m\u001b[30;43mmode\u001b[39;49m\u001b[30;43m)\u001b[39;49m\n\u001b[32m 940\u001b[39m handles.append(handle)\n\u001b[32m 942\u001b[39m \u001b[38;5;66;03m# Convert BytesIO or file objects passed with an encoding\u001b[39;00m\n",
"\u001b[31mFileNotFoundError\u001b[39m: [Errno 2] No such file or directory: 'phase1_output.parquet'"
]
}
],
"source": [
"log(\"Loading Phase 1 output ...\")\n",
"df = pd.read_parquet(\"phase1_output.parquet\")\n",
"log(f\"Loaded: {df.shape[0]} rows, {df.shape[1]} columns\")\n",
"df.head()"
]
},
{
"cell_type": "markdown",
"id": "a571e057",
"metadata": {
"papermill": {
"duration": null,
"end_time": null,
"exception": null,
"start_time": null,
"status": "pending"
},
"tags": []
},
"source": [
"## B. Brand Normalization"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "119609b0",
"metadata": {
"papermill": {
"duration": null,
"end_time": null,
"exception": null,
"start_time": null,
"status": "pending"
},
"tags": []
},
"outputs": [],
"source": [
"BRAND_MAP = {\n",
" \"compliments organic\": (\"Compliments\", \"Organic\"),\n",
" \"compliments balance\": (\"Compliments\", \"Balance\"),\n",
" \"compliments naturally simple\": (\"Compliments\", \"Naturally Simple\"),\n",
" \"compliments green care\": (\"Compliments\", \"Green\"),\n",
" \"compliments green\": (\"Compliments\", \"Green\"),\n",
" \"compliments little ones\": (\"Compliments\", \"Little Ones\"),\n",
" \"sensations\": (\"Sensations\", \"Sensations\"),\n",
" \"compliments\": (\"Compliments\", \"Core\"),\n",
" \"compliments \": (\"Compliments\", \"Core\"),\n",
" \" compliments\": (\"Compliments\", \"Core\"),\n",
"}\n",
"\n",
"def normalize_brand(clean_brand):\n",
" if pd.isna(clean_brand):\n",
" return (\"Unknown\", \"Core\")\n",
" key = str(clean_brand).strip().lower()\n",
" if key in BRAND_MAP:\n",
" return BRAND_MAP[key]\n",
" if \"compliments\" in key:\n",
" return (\"Compliments\", \"Core\")\n",
" if \"sensations\" in key:\n",
" return (\"Sensations\", \"Sensations\")\n",
" return (str(clean_brand).strip(), \"Core\")\n",
"\n",
"brand_col = \"brand_clean\" if \"brand_clean\" in df.columns else \"brand\"\n",
"brand_results = df[brand_col].apply(normalize_brand)\n",
"df[\"brand_norm\"] = brand_results.apply(lambda x: x[0])\n",
"df[\"product_line\"] = brand_results.apply(lambda x: x[1])\n",
"\n",
"print(\"Brand normalization:\")\n",
"print(df[\"brand_norm\"].value_counts().to_string())\n",
"print(\"\\nProduct lines:\")\n",
"print(df[\"product_line\"].value_counts().to_string())"
]
},
{
"cell_type": "markdown",
"id": "47c0f9f3",
"metadata": {
"papermill": {
"duration": null,
"end_time": null,
"exception": null,
"start_time": null,
"status": "pending"
},
"tags": []
},
"source": [
"## C. Identity Attribute Extraction (from ORIGINAL title)"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "71e095e4",
"metadata": {
"papermill": {
"duration": null,
"end_time": null,
"exception": null,
"start_time": null,
"status": "pending"
},
"tags": []
},
"outputs": [],
"source": [
"def extract_identity_flags(title):\n",
" t = title.lower()\n",
" return {\n",
" \"is_organic\": bool(re.search(r\"\\borganic\\b\", t)),\n",
" \"is_gluten_free\": bool(re.search(r\"\\bgluten[\\s-]+free\\b\", t)),\n",
" \"is_naturally_simple\": bool(re.search(r\"\\bnaturally\\s+simple\\b\", t)),\n",
" \"is_sugar_free\": bool(re.search(\n",
" r\"\\bsugar[\\s-]+free\\b|\\b(?:no sugar added|unsweetened)\\b|\\bzero\\s+sugar\\b\", t)),\n",
" \"is_unsalted\": bool(re.search(r\"\\bunsalted\\b|\\bno salt\\b\", t)),\n",
" \"is_lactose_free\": bool(re.search(r\"\\blactose[\\s-]+free\\b\", t)),\n",
" \"is_peanut_free\": bool(re.search(r\"\\bpeanut[\\s-]+free\\b\", t)),\n",
" \"is_plant_based\": bool(re.search(r\"\\bplant[\\s-]*based\\b\", t)),\n",
" \"is_reduced_sodium\": bool(re.search(\n",
" r\"\\breduced\\s+sodium\\b|\\blow\\s+sodium\\b|\\bno\\s+salt\\s+added\\b\", t)),\n",
" }\n",
"\n",
"identity_flags = df[\"title\"].apply(extract_identity_flags)\n",
"id_df = pd.DataFrame(identity_flags.tolist())\n",
"for col in id_df.columns:\n",
" df[col] = id_df[col].values\n",
"\n",
"print(\"Identity flags:\")\n",
"for col in [\"is_organic\", \"is_gluten_free\", \"is_naturally_simple\", \"is_sugar_free\",\n",
" \"is_unsalted\", \"is_lactose_free\", \"is_peanut_free\", \"is_plant_based\",\n",
" \"is_reduced_sodium\"]:\n",
" print(f\" {col}: {int(df[col].sum())}\")"
]
},
{
"cell_type": "markdown",
"id": "cc1fd7e9",
"metadata": {
"papermill": {
"duration": null,
"end_time": null,
"exception": null,
"start_time": null,
"status": "pending"
},
"tags": []
},
"source": [
"## D. Fat Info + Flavour + Formulation Extraction"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "0ff388a0",
"metadata": {
"papermill": {
"duration": null,
"end_time": null,
"exception": null,
"start_time": null,
"status": "pending"
},
"tags": []
},
"outputs": [],
"source": [
"def extract_fat_info(title):\n",
" t = title.lower()\n",
" fat_level = \"regular\"\n",
" if re.search(r\"\\b(?:light|lite|reduced fat|low fat|lean)\\b\", t):\n",
" fat_level = \"reduced_fat\"\n",
" elif re.search(r\"\\bfat[\\s-]+free\\b\", t):\n",
" fat_level = \"fat_free\"\n",
" fat_percentage = None\n",
" pct_match = re.search(r\"(\\d+(?:\\.\\d+)?)\\s*%\", title)\n",
" if pct_match:\n",
" val = float(pct_match.group(1))\n",
" skip_context = any(w in t for w in [\"cocoa\", \"alcohol\", \"isopropyl\", \"peanuts\"])\n",
" if val < 100 and not skip_context:\n",
" fat_percentage = val\n",
" if 0 < val <= 0.7:\n",
" fat_level = \"fat_free\"\n",
" return {\"fat_level\": fat_level, \"fat_percentage\": fat_percentage}\n",
"\n",
"fat_info = df[\"title\"].apply(extract_fat_info)\n",
"fat_df = pd.DataFrame(fat_info.tolist())\n",
"df[\"fat_level\"] = fat_df[\"fat_level\"].values\n",
"df[\"fat_percentage\"] = fat_df[\"fat_percentage\"].values\n",
"\n",
"FLAVOUR_KEYWORDS = [\n",
" \"almond\", \"apple\", \"banana\", \"blueberry\", \"caramel\", \"cherry\",\n",
" \"chocolate\", \"cinnamon\", \"coconut\", \"cranberry\", \"honey\", \"lemon\",\n",
" \"lime\", \"mango\", \"maple\", \"mixed berry\", \"peach\", \"peanut\",\n",
" \"peppermint\", \"pineapple\", \"pomegranate\", \"raspberry\", \"strawberry\",\n",
" \"tropical\", \"vanilla\", \"watermelon\", \"white chocolate\", \"berry\",\n",
"]\n",
"\n",
"def extract_flavours(title):\n",
" t = title.lower()\n",
" return sorted(set(kw for kw in FLAVOUR_KEYWORDS if kw in t))\n",
"\n",
"FORMULATION_KEYWORDS = [\n",
" \"smooth\", \"crunchy\", \"creamy\", \"chunky\", \"whole\", \"halves\",\n",
" \"sliced\", \"ground\", \"chopped\", \"breaded\", \"fresh\", \"frozen\",\n",
" \"roasted\", \"smoked\",\n",
"]\n",
"\n",
"def extract_formulation(title):\n",
" t = title.lower()\n",
" return sorted(set(kw for kw in FORMULATION_KEYWORDS if kw in t))\n",
"\n",
"df[\"flavour\"] = df[\"title\"].apply(extract_flavours)\n",
"df[\"formulation\"] = df[\"title\"].apply(extract_formulation)\n",
"\n",
"print(f\"Products with flavour: {(df['flavour'].apply(len) > 0).sum()}\")\n",
"print(f\"Products with formulation: {(df['formulation'].apply(len) > 0).sum()}\")\n",
"print(f\"\\nFat levels:\")\n",
"print(df[\"fat_level\"].value_counts().to_string())"
]
},
{
"cell_type": "markdown",
"id": "1f69d434",
"metadata": {
"papermill": {
"duration": null,
"end_time": null,
"exception": null,
"start_time": null,
"status": "pending"
},
"tags": []
},
"source": [
"## E. Variant Attributes Parsing + Core Title Extraction"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "bc3a5247",
"metadata": {
"papermill": {
"duration": null,
"end_time": null,
"exception": null,
"start_time": null,
"status": "pending"
},
"tags": []
},
"outputs": [],
"source": [
"def build_variant_attributes(size_str):\n",
" if not size_str or pd.isna(size_str):\n",
" return {}\n",
" s = str(size_str).strip()\n",
" attrs = {}\n",
" m = re.match(r\"(\\d+(?:\\.\\d+)?)\\s*x\\s*(\\d+(?:\\.\\d+)?)\\s*(g|kg|ml|l|oz|lb)s?\\s*$\", s, re.IGNORECASE)\n",
" if m:\n",
" return {\"qty\": int(float(m.group(1))), \"amount\": float(m.group(2)), \"unit\": m.group(3).lower()}\n",
" m = re.match(r\"(\\d+(?:\\.\\d+)?)\\s*(g|kg|ml|l|oz|lb)s?\\s*(?:x\\s*(\\d+))?\\s*$\", s, re.IGNORECASE)\n",
" if m:\n",
" attrs = {\"amount\": float(m.group(1)), \"unit\": m.group(2).lower()}\n",
" if m.group(3):\n",
" attrs[\"multiplier\"] = int(m.group(3))\n",
" return attrs\n",
" m = re.match(r\"(\\d+)\\s*(?:per\\s+pack|count|ea|pack|piece|slice|cups?|pound)s?\\s*$\", s, re.IGNORECASE)\n",
" if m:\n",
" return {\"count\": int(m.group(1))}\n",
" return {\"raw\": s}\n",
"\n",
"df[\"variant_attributes\"] = df[\"size\"].apply(\n",
" lambda x: json.dumps(build_variant_attributes(x)) if pd.notna(x) else \"{}\"\n",
")\n",
"\n",
"BRAND_PREFIXES = [\n",
" (\"Compliments Naturally Simple \", \"Compliments \"),\n",
" (\"Compliments Balance \", \"Compliments Balance \"),\n",
" (\"Compliments Organic \", \"Compliments Organic \"),\n",
" (\"Compliments Green Care \", \"Compliments Green \"),\n",
" (\"Compliments Little Ones \", \"Compliments Little Ones \"),\n",
" (\"Compliments \", \"\"),\n",
" (\"Sensations \", \"\"),\n",
"]\n",
"SIZE_PATTERNS = [\n",
" r\"\\s+\\d[\\d,.]*\\s*(?:g|kg|ml|l|oz|lb|count|ea|pack|piece|slice|cups?|litre|liters?|pound)s?\\s*$\",\n",
" r\"\\s+\\d+\\s*x\\s+\\d+\\s*(?:g|kg|ml|l)\\s*$\",\n",
" r\"\\s+\\d+\\s*x\\s*$\",\n",
" r\"\\s+\\d+\\s+per\\s+pack\\s*$\",\n",
" r\"\\s+\\d+\\s+count\\s*$\",\n",
"]\n",
"BRACKET_PATTERN = r\"\\s*\\([^)]*\\)\\s*\"\n",
"\n",
"def extract_core_title(title, product_line):\n",
" t = title.strip()\n",
" for prefix, replacement in BRAND_PREFIXES:\n",
" if t.lower().startswith(prefix.lower()):\n",
" t = replacement + t[len(prefix):]\n",
" break\n",
" for pat in SIZE_PATTERNS:\n",
" t = re.sub(pat, \"\", t, flags=re.IGNORECASE)\n",
" t = re.sub(BRACKET_PATTERN, \" \", t)\n",
" return re.sub(r\"\\s+\", \" \", t).strip()\n",
"\n",
"df[\"core_title\"] = df.apply(lambda row: extract_core_title(row[\"title\"], row[\"product_line\"]), axis=1)\n",
"print(f\"Core titles: {df['core_title'].nunique()} unique\")"
]
},
{
"cell_type": "markdown",
"id": "8c9e0071",
"metadata": {
"papermill": {
"duration": null,
"end_time": null,
"exception": null,
"start_time": null,
"status": "pending"
},
"tags": []
},
"source": [
"## F. Identity Hash"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "4774b526",
"metadata": {
"papermill": {
"duration": null,
"end_time": null,
"exception": null,
"start_time": null,
"status": "pending"
},
"tags": []
},
"outputs": [],
"source": [
"def build_identity_hash(row):\n",
" parts = []\n",
" for col in [\"is_organic\", \"is_gluten_free\", \"is_naturally_simple\",\n",
" \"is_sugar_free\", \"is_unsalted\", \"is_lactose_free\",\n",
" \"is_peanut_free\", \"is_plant_based\", \"is_reduced_sodium\"]:\n",
" parts.append(\"1\" if row.get(col) else \"0\")\n",
" fp = row.get(\"fat_percentage\")\n",
" if pd.notna(fp) and fp is not None:\n",
" parts.append(f\"fat{fp:.2g}\" if fp else \"fat0\")\n",
" else:\n",
" parts.append(\"fat_none\")\n",
" parts.append(str(row.get(\"fat_level\", \"regular\")))\n",
" parts.append(str(row.get(\"product_line\", \"Core\")))\n",
" fl = row.get(\"flavour\", [])\n",
" parts.append(\",\".join(sorted(fl)) if fl else \"\")\n",
" fm = row.get(\"formulation\", [])\n",
" parts.append(\",\".join(sorted(fm)) if fm else \"\")\n",
" return \"|\".join(parts)\n",
"\n",
"df[\"identity_hash\"] = df.apply(lambda row: build_identity_hash(row.to_dict()), axis=1)\n",
"print(f\"Identity hashes: {df['identity_hash'].nunique()} unique\")"
]
},
{
"cell_type": "markdown",
"id": "45d3a284",
"metadata": {
"papermill": {
"duration": null,
"end_time": null,
"exception": null,
"start_time": null,
"status": "pending"
},
"tags": []
},
"source": [
"## G. Save Outputs"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "3582ddf5",
"metadata": {
"papermill": {
"duration": null,
"end_time": null,
"exception": null,
"start_time": null,
"status": "pending"
},
"tags": []
},
"outputs": [],
"source": [
"df.to_parquet(\"phase2_output.parquet\", index=False)\n",
"log(\"Saved phase2_output.parquet\")\n",
"print(f\"\\n=== PHASE 2 COMPLETE (v{VERSION}) ===\")\n",
"print(f\"Input: 4440 rows\")\n",
"print(f\"Output: {df.shape[0]} rows, {df.shape[1]} columns\")\n",
"print(f\"Columns added: {sorted(set(df.columns) - set(pd.read_parquet('phase1_output.parquet').columns))}\")"
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.13.7"
},
"papermill": {
"default_parameters": {},
"duration": 5.303654,
"end_time": "2026-08-05T13:26:51.442995+00:00",
"environment_variables": {},
"exception": true,
"input_path": "code/notebooks/phase2.ipynb",
"output_path": "code/notebooks/phase2.ipynb",
"parameters": {},
"start_time": "2026-08-05T13:26:46.139341+00:00",
"version": "2.7.0"
}
},
"nbformat": 4,
"nbformat_minor": 5
}