7 |
58 | );
59 | }
60 |
--------------------------------------------------------------------------------
/cloudbuild.yaml:
--------------------------------------------------------------------------------
1 | steps:
2 | # 1. Build Backend Image
3 | - name: 'gcr.io/cloud-builders/docker'
4 | args: ['build', '-t', 'us-central1-docker.pkg.dev/$PROJECT_ID/polarbear-repo/backend:$COMMIT_SHA', './backend']
5 | id: 'Build Backend'
6 | waitFor: ['-'] # Start immediately
7 |
8 | # 2. Build Frontend Image
9 | - name: 'gcr.io/cloud-builders/docker'
10 | args: ['build', '-t', 'us-central1-docker.pkg.dev/$PROJECT_ID/polarbear-repo/frontend:$COMMIT_SHA', './frontend']
11 | id: 'Build Frontend'
12 | waitFor: ['-'] # Start immediately (Parallel with Backend)
13 |
14 | # 3. Push Images to Artifact Registry
15 | - name: 'gcr.io/cloud-builders/docker'
16 | args: ['push', 'us-central1-docker.pkg.dev/$PROJECT_ID/polarbear-repo/backend:$COMMIT_SHA']
17 | id: 'Push Backend'
18 | waitFor: ['Build Backend']
19 |
20 | - name: 'gcr.io/cloud-builders/docker'
21 | args: ['push', 'us-central1-docker.pkg.dev/$PROJECT_ID/polarbear-repo/frontend:$COMMIT_SHA']
22 | id: 'Push Frontend'
23 | waitFor: ['Build Frontend']
24 |
25 | # 4. Deploy Backend to Cloud Run
26 | - name: 'gcr.io/google.com/cloudsdktool/cloud-sdk'
27 | entrypoint: gcloud
28 | args:
29 | - 'run'
30 | - 'deploy'
31 | - 'polarbear-backend'
32 | - '--image'
33 | - 'us-central1-docker.pkg.dev/$PROJECT_ID/polarbear-repo/backend:$COMMIT_SHA'
34 | - '--region'
35 | - 'us-central1'
36 | - '--platform'
37 | - 'managed'
38 | - '--memory'
39 | - '1Gi'
40 | - '--allow-unauthenticated'
41 | # Set environment variables for Meilisearch connection (to be replaced by user manually or via secret manager in prod)
42 | # For now, we assume the user will set these in Cloud Run console or we can pass them if we had them.
43 | # We'll leave them as placeholders or rely on the VM script output.
44 | id: 'Deploy Backend'
45 | waitFor: ['Push Backend']
46 |
47 | # 5. Deploy Frontend to Cloud Run
48 | - name: 'gcr.io/google.com/cloudsdktool/cloud-sdk'
49 | entrypoint: gcloud
50 | args:
51 | - 'run'
52 | - 'deploy'
53 | - 'polarbear-frontend'
54 | - '--image'
55 | - 'us-central1-docker.pkg.dev/$PROJECT_ID/polarbear-repo/frontend:$COMMIT_SHA'
56 | - '--region'
57 | - 'us-central1'
58 | - '--platform'
59 | - 'managed'
60 | - '--allow-unauthenticated'
61 | # Pass the Backend URL to the Frontend
62 | # Note: We might need to know the backend URL beforehand or update it later.
63 | # For simplicity in this MVP CI/CD, we might need to hardcode or use a fixed service name URL if internal.
64 | # But since they are separate services, we'll need the public URL.
65 | # A common pattern is to deploy backend, get URL, then deploy frontend with that arg.
66 | # However, Cloud Run URLs are deterministic based on service name + project.
67 | # So we can predict it: https://polarbear-backend-
8 |
57 |
9 |
21 | 10 | PolarBear 11 |
12 |
22 | The Open-Source Hybrid Search Engine for SMEs.
23 |
24 | Powerful, AI-enhanced, and easy to use.
25 |
28 |
32 | Start Searching
33 |
34 |
38 | Upload Data
39 |
40 |
41 |
42 |
43 |
56 |
44 |
47 | Hybrid Search
45 |Combines keyword and vector search for best results.
46 |
48 |
51 | Analytics
49 |Track clicks, orders, and missed searches.
50 |
52 |
55 | Product Mgmt
53 |Easily manage your catalog and images.
54 |
48 |
114 | );
115 | }
116 |
--------------------------------------------------------------------------------
/README.md:
--------------------------------------------------------------------------------
1 |
49 |
113 |
50 | Data Ingestion
51 |
52 | 53 | Upload Product Catalog 54 |
55 | 56 |
57 |
79 |
80 |
58 |
76 |
77 |
78 | Instructions
59 |
60 |
75 | Required CSV Columns:
61 |-
62 |
id(Unique ID)
63 | title(Product Name)
64 | description(Product Description)
65 | price(Number)
66 | category(e.g., Apparel)
67 | brand(e.g., Nike)
68 | tags(Comma-separated, e.g., "summer, cotton")
69 |
71 | Note: Uploading a new file will merge with existing products. 72 | Existing IDs will be updated, and new IDs will be added. 73 |
74 |
81 |
84 |
96 |
97 |
98 |
106 |
107 | {status && (
108 |
109 | {status}
110 |
111 | )}
112 |
2 | 
3 |
18 |
19 | ---
20 |
21 | ## 🌟 Introduction
22 |
23 | **PolarBear** is a powerful, no-code, AI-enhanced search engine designed specifically for Small and Medium-sized Enterprises (SMEs). It democratizes access to advanced search technology, allowing business owners to create a professional search experience for their products, services, or inventory in minutes—completely free and open source.
24 |
25 | Unlike complex enterprise solutions, PolarBear focuses on simplicity without compromising on power. It combines **Keyword Search** (Meilisearch) and **Semantic Vector Search** (FAISS) to deliver results that are both accurate and contextually relevant.
26 |
27 | ## 🚀 Features
28 |
29 | - **🔍 Hybrid Search**: seamlessly blends keyword matching (BM25) with AI-powered semantic search (Embeddings) for superior result relevance.
30 | - **⚡ No-Code Ingestion**: Upload your data via CSV, Excel, or Google Sheets. No coding required.
31 | - **🧠 AI-Ready**: Built-in vectorization pipeline using state-of-the-art embedding models.
32 | - **📊 Insights Dashboard**: Track user behavior, top queries, zero-result searches, and conversion metrics.
33 | - **🛍️ Product Management**: Built-in catalog management to edit products and upload images directly.
34 | - **☁️ Cloud-Native**: Dockerized for easy deployment on Google Cloud Run, AWS, or your own server.
35 | - **🔓 Open Source**: 100% free to use, modify, and distribute.
36 |
37 | ## 🛠️ Tech Stack
38 |
39 | | Component | Technology | Description |
40 | |-----------|------------|-------------|
41 | | **Frontend** | Next.js (React) | Modern, responsive admin and search UI. |
42 | | **Backend** | FastAPI (Python) | High-performance API for ingestion and search. |
43 | | **Search** | Meilisearch | Lightning-fast keyword search engine. |
44 | | **Vector DB** | FAISS | Efficient similarity search for embeddings. |
45 | | **Infrastructure** | Docker | Containerized for consistent deployment. |
46 |
47 | ## 🚀 Getting Started
48 |
49 | ### Prerequisites
50 | - **Node.js** 18+
51 | - **Python** 3.11+
52 | - **Docker** & **Docker Compose**
53 |
54 | ### Quick Start
55 |
56 | 1. **Clone the Repository**
57 | ```bash
58 | git clone https://github.com/dukesky/PolarBear.git
59 | cd PolarBear
60 | ```
61 |
62 | 2. **Start Infrastructure**
63 | ```bash
64 | cd infrastructure
65 | docker-compose up -d
66 | ```
67 |
68 | 3. **Start Backend**
69 | ```bash
70 | cd backend
71 | poetry install
72 | poetry run uvicorn app.main:app --reload --port 8000
73 | ```
74 |
75 | 4. **Start Frontend**
76 | ```bash
77 | cd frontend
78 | npm install
79 | npm run dev
80 | ```
81 |
82 | 5. **Experience PolarBear**
83 | - **Upload Data**: Go to `http://localhost:3000/upload` and upload a CSV (e.g., `sample_products.csv`).
84 | - **Search**: Visit `http://localhost:3000/search` to try the hybrid search.
85 | - **Insights**: Check `http://localhost:3000/insights` for analytics and product management.
86 |
87 | ## 📚 Documentation
88 |
89 | Detailed walkthroughs for each development phase:
90 |
91 | - [**Phase 1: Setup & MVP Core**](docs/phase1/walkthrough_phase1.md) - Infrastructure and basic search.
92 | - [**Phase 2: Ingestion & Indexing**](docs/phase2/walkthrough_phase2.md) - CSV parsing and hybrid indexing pipeline.
93 | - [**Phase 3: Search Interface**](docs/phase3/walkthrough_phase3.md) - Frontend UI and search logic.
94 | - [**Phase 4: Analytics**](docs/phase4/walkthrough_phase4.md) - Tracking user queries and dashboard.
95 | - [**Phase 5: UX & Product Mgmt**](docs/phase5/walkthrough_phase5_extended.md) - Image support, catalog editing, and advanced UX.
96 |
97 | ## 🤝 Contributing
98 |
99 | We welcome contributions from the community! Whether it's fixing bugs, improving documentation, or suggesting new features, your help is appreciated.
100 |
101 | 1. Fork the Project
102 | 2. Create your Feature Branch (`git checkout -b feature/AmazingFeature`)
103 | 3. Commit your Changes (`git commit -m 'Add some AmazingFeature'`)
104 | 4. Push to the Branch (`git push origin feature/AmazingFeature`)
105 | 5. Open a Pull Request
106 |
107 | ## 📄 License
108 |
109 | Distributed under the MIT License. See `LICENSE` for more information.
110 |
111 | ## 📬 Contact
112 |
113 | Project Link: [https://github.com/dukesky/PolarBear](https://github.com/dukesky/PolarBear)
114 |
115 | ---
116 |
3 | PolarBear 🐻❄️
4 |The Open-Source Hybrid Search Engine for SMEs
5 | 6 | [](https://opensource.org/licenses/MIT) 7 | [](https://www.python.org/) 8 | [](https://nextjs.org/) 9 | [](https://www.docker.com/) 10 | 11 |12 | Features • 13 | Getting Started • 14 | Documentation • 15 | Contributing 16 |
17 |
117 | Made with ❤️ for the Open Source Community
118 |
119 |
--------------------------------------------------------------------------------
/backend/app/services/search/indexer.py:
--------------------------------------------------------------------------------
1 | import meilisearch
2 | import faiss
3 | import numpy as np
4 | from sentence_transformers import SentenceTransformer
5 | from app.core.config import settings
6 | import os
7 | import pickle
8 |
9 | class HybridIndexer:
10 | def __init__(self):
11 | # Meilisearch Client
12 | self.meili_client = meilisearch.Client(settings.MEILI_HOST, settings.MEILI_MASTER_KEY)
13 | self.index_name = "products"
14 |
15 | # Embedding Model
16 | # Using a lightweight model for CPU efficiency
17 | self.model = SentenceTransformer('all-MiniLM-L6-v2')
18 |
19 | # FAISS Index Path
20 | self.faiss_index_path = "faiss_index.bin"
21 | self.doc_map_path = "doc_map.pkl" # Maps FAISS ID to Product ID
22 |
23 | def index_data(self, new_documents: list[dict]):
24 | """
25 | Performs hybrid indexing with Merge & Rebuild strategy:
26 | 1. Fetch ALL existing documents from Meilisearch.
27 | 2. Merge with new_documents (deduplicate by ID).
28 | 3. Re-index ALL documents to Meilisearch.
29 | 4. Re-build FAISS index from scratch with ALL documents.
30 | """
31 | print(f"Received {len(new_documents)} new documents.")
32 |
33 | # 1. Fetch Existing Documents
34 | existing_docs = []
35 | try:
36 | # For MVP, we assume < 10k items. In prod, use pagination.
37 | results = self.meili_client.index(self.index_name).get_documents({'limit': 10000})
38 | # Meilisearch python client v0.20+ returns an object with .results
39 | # Older versions might return list. Let's handle object.
40 | if hasattr(results, 'results'):
41 | existing_docs = [dict(d) for d in results.results]
42 | else:
43 | existing_docs = results # Fallback if it returns list
44 | except Exception as e:
45 | print(f"Could not fetch existing docs (might be empty index): {e}")
46 | existing_docs = []
47 |
48 | print(f"Found {len(existing_docs)} existing documents.")
49 |
50 | # 2. Merge Documents
51 | # Create a dict keyed by ID for easy merging
52 | doc_map = {str(d['id']): d for d in existing_docs}
53 |
54 | # Update/Add new documents
55 | for doc in new_documents:
56 | doc_id = str(doc['id'])
57 | doc_map[doc_id] = doc # Overwrite if exists, add if new
58 |
59 | all_documents = list(doc_map.values())
60 | print(f"Total documents after merge: {len(all_documents)}")
61 |
62 | # 3. Meilisearch Indexing (Re-index ALL)
63 | print("Indexing all documents to Meilisearch...")
64 | index = self.meili_client.index(self.index_name)
65 | index.update_settings({
66 | 'searchableAttributes': ['title', 'description', 'brand', 'category'],
67 | 'filterableAttributes': ['brand', 'category', 'price', 'tags'],
68 | 'displayedAttributes': ['*'] # Ensure all attributes are returned
69 | })
70 | # deleteAll is optional but cleaner to avoid ghosts if we were removing items.
71 | # But here we are just adding/updating. add_documents upserts.
72 | index.add_documents(all_documents)
73 |
74 | # 4. Vector Indexing (FAISS) - Rebuild from scratch
75 | print("Generating embeddings for all documents...")
76 | texts = [f"{doc.get('title', '')} {doc.get('description', '')}" for doc in all_documents]
77 | embeddings = self.model.encode(texts)
78 |
79 | # Convert to float32 for FAISS
80 | embeddings = np.array(embeddings).astype('float32')
81 | dimension = embeddings.shape[1]
82 |
83 | print(f"Rebuilding FAISS index with dimension {dimension}...")
84 | faiss_index = faiss.IndexFlatL2(dimension)
85 | faiss_index.add(embeddings)
86 |
87 | # Save Index
88 | faiss.write_index(faiss_index, self.faiss_index_path)
89 |
90 | # Save ID Mapping (FAISS internal ID -> Document ID)
91 | # Order matches 'all_documents' list order
92 | doc_ids = [doc['id'] for doc in all_documents]
93 | with open(self.doc_map_path, 'wb') as f:
94 | pickle.dump(doc_ids, f)
95 |
96 | print("Hybrid indexing (Merge & Rebuild) complete.")
97 |
98 | def search_vectors(self, query: str, k: int = 10):
99 | """
100 | Search FAISS index for query
101 | """
102 | if not os.path.exists(self.faiss_index_path):
103 | return []
104 |
105 | index = faiss.read_index(self.faiss_index_path)
106 | query_vector = self.model.encode([query]).astype('float32')
107 |
108 | distances, indices = index.search(query_vector, k)
109 |
110 | # Load ID mapping
111 | with open(self.doc_map_path, 'rb') as f:
112 | doc_ids = pickle.load(f)
113 |
114 | results = []
115 | for i, idx in enumerate(indices[0]):
116 | if idx != -1 and idx < len(doc_ids):
117 | results.append({
118 | "id": doc_ids[idx],
119 | "score": float(distances[0][i])
120 | })
121 |
122 | return results
123 |
--------------------------------------------------------------------------------
/backend/app/services/search/searcher.py:
--------------------------------------------------------------------------------
1 | import meilisearch
2 | import faiss
3 | import numpy as np
4 | from sentence_transformers import SentenceTransformer
5 | from app.core.config import settings
6 | import pickle
7 | import os
8 |
9 | class HybridSearcher:
10 | def __init__(self):
11 | # Meilisearch Client
12 | self.meili_client = meilisearch.Client(settings.MEILI_HOST, settings.MEILI_MASTER_KEY)
13 | self.index_name = "products"
14 |
15 | # Embedding Model
16 | self.model = SentenceTransformer('all-MiniLM-L6-v2')
17 |
18 | # FAISS Index & Map Paths
19 | self.faiss_index_path = "faiss_index.bin"
20 | self.doc_map_path = "doc_map.pkl"
21 |
22 | def search(self, query: str, limit: int = 20) -> list[dict]:
23 | """
24 | Performs hybrid search:
25 | 1. Get Keyword results from Meilisearch.
26 | 2. Get Vector results from FAISS.
27 | 3. Merge and rank results.
28 | """
29 | # 1. Keyword Search (Meilisearch)
30 | try:
31 | meili_results = self.meili_client.index(self.index_name).search(query, {'limit': limit})
32 | keyword_hits = meili_results.get('hits', [])
33 | except Exception as e:
34 | print(f"Meilisearch error: {e}")
35 | keyword_hits = []
36 |
37 | # 2. Vector Search (FAISS)
38 | vector_hits = []
39 | if os.path.exists(self.faiss_index_path) and os.path.exists(self.doc_map_path):
40 | try:
41 | index = faiss.read_index(self.faiss_index_path)
42 | with open(self.doc_map_path, 'rb') as f:
43 | doc_ids = pickle.load(f)
44 |
45 | query_vector = self.model.encode([query]).astype('float32')
46 | distances, indices = index.search(query_vector, limit)
47 |
48 | for i, idx in enumerate(indices[0]):
49 | if idx != -1 and idx < len(doc_ids):
50 | vector_hits.append({
51 | "id": doc_ids[idx],
52 | "vector_score": float(distances[0][i]) # Lower is better for L2
53 | })
54 | except Exception as e:
55 | print(f"FAISS error: {e}")
56 |
57 | # 3. Merge Results (Simple Linear Combination)
58 | # We need to retrieve full documents for vector hits from Meilisearch if they aren't in keyword hits
59 |
60 | # Create a map of all unique IDs found
61 | all_ids = set([h['id'] for h in keyword_hits] + [h['id'] for h in vector_hits])
62 |
63 | # Retrieve full documents for all IDs from Meilisearch to ensure we have data
64 | # (Optimization: In production, we might store data in a DB, but here Meilisearch acts as DB)
65 | final_results = []
66 | if all_ids:
67 | try:
68 | # Meilisearch 'get_documents' can fetch by ID
69 | docs = self.meili_client.index(self.index_name).get_documents({'filter': f"id IN [{','.join(all_ids)}]", 'limit': len(all_ids)})
70 | doc_map = {d.id: d for d in docs.results} # Meilisearch python client returns objects or dicts depending on version
71 | # Let's assume it returns objects with attributes or dicts. The python client usually returns objects that can be accessed as dicts or attributes.
72 | # Actually, standard client returns object with .results which is a list of dicts usually?
73 | # Let's check standard behavior or use a safer retrieval.
74 | # Safer: use search with filter id IN [...] to get full docs
75 |
76 | # Alternative: Just use the data we have.
77 | # Keyword hits have data. Vector hits only have ID.
78 | # We MUST fetch data for vector-only hits.
79 | pass
80 | except Exception:
81 | pass
82 |
83 | # RERANKING LOGIC (Simplified)
84 | # We will score items.
85 | # Keyword Score: 1.0 / (rank + 1) (Reciprocal Rank) or just use Meilisearch score if available? Meilisearch doesn't expose score easily in standard search response without showRankingScore=True.
86 | # Vector Score: 1.0 / (1.0 + distance)
87 |
88 | scores = {}
89 |
90 | # Process Keyword Hits
91 | for i, hit in enumerate(keyword_hits):
92 | pid = hit['id']
93 | # Score: High for top results.
94 | # Simple approach: 1.0 for top 1, 0.9 for top 2... or just 1.0 * weight
95 | # Let's use Reciprocal Rank: 1 / (i + 1)
96 | k_score = 1.0 / (i + 1)
97 | scores[pid] = {'score': k_score * 0.3, 'doc': hit} # Weight 0.3
98 |
99 | # Process Vector Hits
100 | for hit in vector_hits:
101 | pid = hit['id']
102 | # L2 Distance: Lower is better. Convert to similarity score.
103 | # Simple inversion: 1 / (1 + distance)
104 | v_score = 1.0 / (1.0 + hit['vector_score'])
105 |
106 | if pid in scores:
107 | scores[pid]['score'] += v_score * 0.7 # Weight 0.7
108 | else:
109 | # We need to fetch the doc content if it wasn't in keyword hits
110 | # For MVP, we will do a quick fetch from Meilisearch for this ID
111 | try:
112 | doc = self.meili_client.index(self.index_name).get_document(pid)
113 | # get_document returns a dict usually
114 | scores[pid] = {'score': v_score * 0.7, 'doc': doc}
115 | except:
116 | # If doc not found (sync issue?), skip
117 | continue
118 |
119 | # Sort by final score
120 | sorted_pids = sorted(scores.keys(), key=lambda x: scores[x]['score'], reverse=True)
121 |
122 | # Format Output
123 | output = []
124 | for pid in sorted_pids:
125 | doc = scores[pid]['doc']
126 | # Add debug score info if needed
127 | # doc['_score'] = scores[pid]['score']
128 | output.append(doc)
129 |
130 | return output[:limit]
131 |
--------------------------------------------------------------------------------
/frontend/src/app/search/page.tsx:
--------------------------------------------------------------------------------
1 | 'use client';
2 |
3 | import { useState } from 'react';
4 |
5 | interface Product {
6 | id: string;
7 | title: string;
8 | description: string;
9 | price: number;
10 | category: string;
11 | brand: string;
12 | tags: string;
13 | image_url?: string;
14 | }
15 |
16 | interface SearchResponse {
17 | query: string;
18 | total: number;
19 | results: Product[];
20 | }
21 |
22 | export default function SearchPage() {
23 | const [query, setQuery] = useState('');
24 | const [results, setResults] = useState
73 |
162 | );
163 | }
164 |
--------------------------------------------------------------------------------
/frontend/src/app/insights/page.tsx:
--------------------------------------------------------------------------------
1 | 'use client';
2 |
3 | import { useState, useEffect } from 'react';
4 |
5 | interface AnalyticsData {
6 | total_searches: number;
7 | top_queries: { query: string; count: number }[];
8 | zero_results: { query: string; count: number }[];
9 | product_stats: { product_id: string; title: string; clicks: number; orders: number }[];
10 | }
11 |
12 | export default function InsightsPage() {
13 | const [data, setData] = useState
74 |
161 |
75 |
83 |
84 | {/* Search Bar */}
85 |
103 |
104 | {/* Results */}
105 | PolarBear Search
76 |Hybrid Search for your Product Catalog
77 | 82 |
106 | {hasSearched && results.length === 0 && !isSearching && (
107 |
160 |
108 | No results found for "{query}".
109 |
110 | )}
111 |
112 | {results.map((product) => (
113 | trackEvent('click', product)}
116 | className="bg-white p-6 rounded-xl shadow-sm hover:shadow-md transition-shadow border border-gray-100 cursor-pointer flex gap-6"
117 | >
118 | {/* Image */}
119 |
158 | ))}
159 |
120 | {product.image_url ? (
121 | 
122 | ) : (
123 |
128 |
129 |
124 | No Image
125 |
126 | )}
127 |
130 |
157 |
131 |
153 | {product.title}
132 |{product.description}
133 |
134 |
135 | {product.brand}
136 |
137 |
138 | {product.category}
139 |
140 | {product.tags.split(',').map(tag => (
141 |
142 | #{tag.trim()}
143 |
144 | ))}
145 |
146 |
152 |
154 | ${product.price}
155 |
156 | Loading...
;
36 | }
37 |
38 | if (!data) {
39 | return Failed to load data.
;
40 | }
41 |
42 | return (
43 |
44 |
165 | );
166 | }
167 |
168 | function ProductCatalog() {
169 | const [products, setProducts] = useState
45 |
161 |
162 | {/* Product Catalog Section */}
163 |
46 |
54 |
55 | {/* Stats Grid */}
56 |
47 |
50 |
51 | Go to Search →
52 |
53 | Insights Dashboard
48 |Search Analytics & Product Performance
49 |
57 |
63 |
64 |
58 |
61 | {/* Add more cards here later (e.g., Total Products) */}
62 | Total Searches
59 |{data.total_searches}
60 |
65 | {/* Top Queries */}
66 |
124 |
125 | {/* Product Performance */}
126 |
67 |
71 |
72 |
92 |
93 |
94 | {/* Zero Results */}
95 |
68 |
70 | Top Search Queries
69 || Query | 74 |Count | 75 |
|---|---|
| No data yet. | 81 ||
| {item.query} | 86 |{item.count} | 87 |
96 |
101 |
102 |
122 |
123 |
97 |
100 | Missed Opportunities (0 Results)
98 |Users searched for these but found nothing.
99 || Query | 104 |Count | 105 |
|---|---|
| No missed searches yet. | 111 ||
| {item.query} | 116 |{item.count} | 117 |
127 |
132 |
133 |
159 |
160 |
128 |
131 | Product Performance
129 |Clicks and Orders tracking.
130 || Product | 135 |Clicks | 136 |Orders | 137 |Conversion Rate | 138 |
|---|---|---|---|
| No product activity yet. | 144 ||||
| {item.title} | 149 |{item.clicks} | 150 |{item.orders} | 151 |152 | {item.clicks > 0 ? ((item.orders / item.clicks) * 100).toFixed(1) : '0.0'}% 153 | | 154 |
234 |
235 |
241 |
242 |
236 |
239 |
240 | Product Catalog
237 |Manage your inventory.
238 |
243 |
244 |
245 |
274 |
275 |
276 | {/* Edit Modal */}
277 | {editingProduct && (
278 | | Image | 247 |Title | 248 |Price | 249 |Actions | 250 |
|---|---|---|---|
|
256 |
257 | {p.image_url &&
259 | |
260 | {p.title} | 261 |${p.price} | 262 |263 | 269 | | 270 |
279 |
280 |