Washington Law MCP Server

import axios from 'axios'; import * as cheerio from 'cheerio'; import Database from 'better-sqlite3'; import { join, dirname } from 'path'; import { fileURLToPath } from 'url'; import { getDocument } from 'pdfjs-dist/legacy/build/pdf.mjs'; const __filename = fileURLToPath(import.meta.url); const __dirname = dirname(__filename); const DB_PATH = join(__dirname, '../../data/washington-laws.db'); const BASE_URL = 'https://www.courts.wa.gov'; interface RPCRule { ruleSet: string; ruleNumber: string; ruleName: string; pdfUrl: string; } class RPCScraper { private db: Database.Database; private insertStmt: Database.Statement; constructor() { this.db = new Database(DB_PATH); this.insertStmt = this.db.prepare(` INSERT OR REPLACE INTO court_rules ( rule_set, rule_number, rule_name, full_text, updated_at ) VALUES (?, ?, ?, ?, CURRENT_TIMESTAMP) `); } async delay(ms: number): Promise<void> { return new Promise(resolve => setTimeout(resolve, ms)); } async scrapeAll(): Promise<void> { console.log('Starting RPC (Rules of Professional Conduct) scraper...\n'); await this.scrapeRPC(); this.printStats(); this.db.close(); } async scrapeRPC(): Promise<void> { console.log('Processing RPC - Rules of Professional Conduct'); try { // Get list of RPC rules const listUrl = `${BASE_URL}/court_rules/?fa=court_rules.list&group=ga&set=RPC`; console.log(` Fetching rule list from: ${listUrl}`); const response = await axios.get(listUrl); const $ = cheerio.load(response.data); const rules: RPCRule[] = []; // Find all PDF links for RPC rules $('a').each((_, element) => { const href = $(element).attr('href'); const text = $(element).text().trim(); // Match PDF links like "/court_rules/pdf/RPC/GA_RPC_01_07_00.pdf" if (href && href.includes('.pdf') && href.includes('RPC')) { // Extract rule number from filename (e.g., GA_RPC_01_07_00.pdf -> 1.7) const fileMatch = href.match(/GA_RPC_(\d+)_(\d+)_(\d+)\.pdf/i); if (fileMatch) { // Build rule number (e.g., 1.7 or 1.7a if third part is not 00) let ruleNumber = `${parseInt(fileMatch[1])}.${parseInt(fileMatch[2])}`; if (fileMatch[3] !== '00') { // Handle sub-rules like 1.7a const subRule = String.fromCharCode(96 + parseInt(fileMatch[3])); // Convert to letter ruleNumber += subRule; } // Build full URL let pdfUrl = href; if (href.startsWith('../')) { pdfUrl = `${BASE_URL}/${href.substring(3)}`; } else if (!href.startsWith('http')) { pdfUrl = `${BASE_URL}${href.startsWith('/') ? '' : '/'}${href}`; } // Extract rule name from text let ruleName = text; // Remove "RPC X.X" prefix if present ruleName = ruleName.replace(/^RPC\s+[\d.]+[a-z]?\s*[-–]\s*/i, '').trim(); if (!ruleName || ruleName === text) { // Try to extract from the full text const nameMatch = text.match(/RPC\s+[\d.]+[a-z]?\s*[-–]\s*(.+)/i); if (nameMatch) { ruleName = nameMatch[1].trim(); } else { ruleName = `Rule ${ruleNumber}`; } } // Avoid duplicates if (!rules.find(r => r.ruleNumber === ruleNumber)) { rules.push({ ruleSet: 'RPC', ruleNumber, ruleName, pdfUrl }); } } } }); console.log(` Found ${rules.length} RPC rules to download and parse`); // Sort rules by rule number for better progress tracking rules.sort((a, b) => { const aParts = a.ruleNumber.match(/(\d+)\.(\d+)([a-z]?)/); const bParts = b.ruleNumber.match(/(\d+)\.(\d+)([a-z]?)/); if (aParts && bParts) { const aMain = parseInt(aParts[1]) * 1000 + parseInt(aParts[2]); const bMain = parseInt(bParts[1]) * 1000 + parseInt(bParts[2]); if (aMain !== bMain) return aMain - bMain; return (aParts[3] || '').localeCompare(bParts[3] || ''); } return a.ruleNumber.localeCompare(b.ruleNumber); }); // Download and parse each PDF for (let i = 0; i < rules.length; i++) { const rule = rules[i]; console.log(` Processing RPC ${rule.ruleNumber}: ${rule.ruleName}...`); try { await this.downloadAndParseRule(rule); } catch (error) { console.error(` Error: ${(error as Error).message}`); } if ((i + 1) % 10 === 0) { console.log(` Processed ${i + 1}/${rules.length} rules`); } await this.delay(300); // Be respectful to the server } console.log(` ✓ Completed RPC: ${rules.length} rules`); } catch (error) { console.error(`Error scraping RPC:`, error); } } async downloadAndParseRule(rule: RPCRule): Promise<void> { try { // Download PDF const response = await axios.get(rule.pdfUrl, { responseType: 'arraybuffer', timeout: 30000 }); // Convert to Uint8Array for pdfjs-dist const pdfData = new Uint8Array(response.data); // Load PDF document const loadingTask = getDocument({ data: pdfData }); const pdf = await loadingTask.promise; // Extract text from all pages let fullText = ''; for (let pageNum = 1; pageNum <= pdf.numPages; pageNum++) { const page = await pdf.getPage(pageNum); const textContent = await page.getTextContent(); const pageText = textContent.items .map((item: any) => item.str) .join(' '); fullText += pageText + '\n'; } // Clean up the text fullText = fullText .replace(/\r\n/g, '\n') .replace(/\n{3,}/g, '\n\n') .replace(/\s+/g, ' ') .trim(); // Try to extract just the rule content (remove headers/footers) const ruleStart = fullText.search(new RegExp(`RPC\\s+${rule.ruleNumber.replace('.', '\\.')}`, 'i')); if (ruleStart > 0 && ruleStart < 100) { fullText = fullText.substring(ruleStart); } // Remove common footer/header text fullText = fullText .replace(/Page \d+ of \d+/gi, '') .replace(/Effective \d+\/\d+\/\d+/gi, '') .replace(/\[.*?Reserved\]/gi, '') .trim(); // Try to extract a better rule name from the PDF content if needed if (rule.ruleName === `Rule ${rule.ruleNumber}`) { const nameMatch = fullText.match(new RegExp(`RPC\\s+${rule.ruleNumber.replace('.', '\\.')}\\s*[-–]?\\s*([A-Z][^.\\n]+)`, 'i')); if (nameMatch) { rule.ruleName = nameMatch[1].trim(); } } // Save to database this.insertStmt.run( rule.ruleSet, rule.ruleNumber, rule.ruleName, fullText ); } catch (error) { throw new Error(`Failed to process PDF for RPC ${rule.ruleNumber}: ${(error as Error).message}`); } } private printStats(): void { const rpcCount = this.db.prepare(` SELECT COUNT(*) as count FROM court_rules WHERE rule_set = 'RPC' `).get() as any; console.log('\n=== RPC Scraping Complete ==='); console.log(` RPC: ${rpcCount.count} rules`); const total = this.db.prepare('SELECT COUNT(*) as count FROM court_rules').get() as any; console.log(` Total court rules in database: ${total.count}`); } } // Run the scraper if (import.meta.url === `file://${process.argv[1]}`) { const scraper = new RPCScraper(); scraper.scrapeAll().catch(console.error); }