Skip to content

index.md

Latest commit

 

History

History
193 lines (148 loc) · 7.33 KB

index.md

File metadata and controls

193 lines (148 loc) · 7.33 KB

PATSTAT Explorer Documentation Index

Generated: 2026-02-01 | Scan Level: Deep | Project Type: Data Analytics

Project Overview

Attribute Value
Type Monolith - Data Analytics Application (Modular Architecture)
Primary Language Python
UI Framework Streamlit >= 1.28.0
Database Google BigQuery (~450 GB)
AI Integration Anthropic Claude (Query Builder)
Data Source EPO PATSTAT 2025 Autumn Edition

Quick Reference

Item Value
Entry Point app.py (72 lines)
Modules modules/ (1,802 lines total)
Queries queries_bq.py (42 queries, 3,893 lines)
Tests tests/ (5 test files)
Live Demo patstatexplorer.depa.tech

TIP4PATLIB Alignment

This application serves as a bridge to TIP for 300 PATLIB centres, delivering:

  • 42 ready-to-use parameterized queries
  • No-code parameter adjustment
  • AI-powered query generation
  • Direct export to TIP Jupyter environment

Generated Documentation

Core Documentation

Document Description
Project Overview Architecture, tech stack, deployment guide
Query Design Patterns NEW - How queries are structured, SQL patterns
What Worked Well NEW - Lessons learned, key decisions
Query Catalog Complete reference for 42 queries
BigQuery Schema PATSTAT table definitions (27 tables)
Data Loading Guide Loading PATSTAT data to BigQuery

Architecture

┌─────────────────────────────────────────────────────────────┐
│                Coolify (patstatexplorer.depa.tech)           │
│  ┌───────────────────────────────────────────────────────┐  │
│  │                     app.py (72 lines)                 │  │
│  │                    Entry Point + Router               │  │
│  └───────────────────────┬───────────────────────────────┘  │
│                          │                                  │
│  ┌───────────────────────▼───────────────────────────────┐  │
│  │                   modules/ (1,802 lines)              │  │
│  │  config.py │ data.py │ logic.py │ ui.py │ utils.py   │  │
│  └───────────────────────┬───────────────────────────────┘  │
│                          │                                  │
│  ┌───────────────────────▼───────────────────────────────┐  │
│  │              queries_bq.py (42 queries)               │  │
│  └───────────────────────┬───────────────────────────────┘  │
└──────────────────────────┼──────────────────────────────────┘
                           │
                           ▼
                  ┌─────────────────┐
                  │  Google BigQuery │
                  │   27 tables      │
                  │   ~450 GB        │
                  └─────────────────┘

Key Features

Feature Description
Question-Based Navigation Queries presented as questions, filtered by category/stakeholder
Dynamic Parameters Year range, jurisdictions, tech fields via UI controls
AI Query Builder Natural language → SQL using Claude
Take to TIP Export ready-to-run code for TIP Jupyter
Visualization Auto-generated Altair charts
Contribution Users can submit new queries

Query Summary

Total: 42 queries organized by category and stakeholder

By Category

Category Count Description
Trends 12 Time-series analysis, growth patterns
Competitors 10 Applicant rankings, market share
Regional 8 Geographic analysis, NUTS regions
Technology 12 IPC/CPC/WIPO field analysis

By Stakeholder

Tag Count Description
PATLIB 25 Patent libraries, information centers
BUSINESS 20 Companies, industry users
UNIVERSITY 15 Researchers, academia

Module Structure

Module Lines Responsibility
config.py 123 Constants, color palette, AI prompt
data.py 156 BigQuery client, query execution
logic.py 238 Business logic, AI, filtering
ui.py 1,193 All Streamlit rendering
utils.py 84 Pure helper functions

Development Guide

Local Setup

git clone https://github.com/herrkrueger/patstat.git
cd patstat
python -m venv .venv && source .venv/bin/activate
pip install -r requirements.txt
cp .env.example .env  # Add credentials
streamlit run app.py

Adding a New Query

  1. Add entry to queries_bq.py with unique ID
  2. Include metadata: title, tags, category, description
  3. Define parameters for dynamic queries
  4. Set visualization config for charts
  5. Run pytest tests/test_query_metadata.py

Testing

pytest tests/ -v

Reference Materials

Document Location
TIP4PATLIB Specification context/BPM001977_Technical_Specifications__TIP4PATLIB.pdf
README README.md

Getting Started

For PATLIB Users

  1. Visit patstatexplorer.depa.tech
  2. Browse queries by category (Competitors, Trends, Regional, Technology)
  3. Filter by stakeholder tag (PATLIB, BUSINESS, UNIVERSITY)
  4. Run a query, adjust parameters, download results
  5. Use "Take to TIP" to export to Jupyter

For Developers

  1. Read Project Overview for architecture
  2. Study Query Design Patterns for SQL patterns
  3. Review What Worked Well for design decisions
  4. Check tests/ for validation patterns

For New Projects

  1. Copy Query Design Patterns for query structure
  2. Apply lessons from What Worked Well
  3. Use query-as-data pattern for rich metadata
  4. Build TIP export from day one

Contact