FedMCP - Federal Parliamentary Information

/** * News fetcher for retrieving articles mentioning MPs from Canadian news sources * Uses Google News RSS feeds which are free and don't require an API key */ import { XMLParser } from 'fast-xml-parser'; export interface NewsArticle { title: string; url: string; source: string; published_date: string | null; description: string; image_url: string | null; last_updated: string; } // Canadian news sources for relevance filtering const CANADIAN_SOURCES = [ 'cbc.ca', 'globalnews.ca', 'ctvnews.ca', 'theglobeandmail.com', 'nationalpost.com', 'thestar.com', 'hilltimes.com', 'macleans.ca', 'winnipegfreepress.com', 'calgaryherald.com', 'montrealgazette.com', 'ottawacitizen.com', ]; const SOURCE_NAMES: Record<string, string> = { 'cbc.ca': 'CBC News', 'globalnews.ca': 'Global News', 'ctvnews.ca': 'CTV News', 'theglobeandmail.com': 'The Globe and Mail', 'nationalpost.com': 'National Post', 'thestar.com': 'Toronto Star', 'hilltimes.com': 'The Hill Times', 'macleans.ca': "Maclean's", 'winnipegfreepress.com': 'Winnipeg Free Press', 'calgaryherald.com': 'Calgary Herald', 'montrealgazette.com': 'Montreal Gazette', 'ottawacitizen.com': 'Ottawa Citizen', }; /** * Fetch news articles mentioning an MP */ export async function fetchMPNews( mpName: string, limit: number = 10 ): Promise<NewsArticle[]> { try { // Capture timestamp when fetch occurs const lastUpdated = new Date().toISOString(); // Build search query - add "Canada" and "MP" for relevance const searchQuery = `"${mpName}" MP Canada`; // Google News RSS feed URL const params = new URLSearchParams({ q: searchQuery, hl: 'en-CA', // Canadian English gl: 'CA', // Canada region ceid: 'CA:en', }); const url = `https://news.google.com/rss/search?${params.toString()}`; // Fetch RSS feed const response = await fetch(url, { headers: { 'User-Agent': 'Mozilla/5.0 (compatible; FedMCP/1.0; +https://github.com/yourusername/fedmcp)', }, }); if (!response.ok) { console.error(`Failed to fetch news for ${mpName}: ${response.statusText}`); return []; } const xmlText = await response.text(); // Parse RSS XML const parser = new XMLParser({ ignoreAttributes: false, attributeNamePrefix: '@_', }); const parsed = parser.parse(xmlText); // Extract items from RSS feed const items = parsed?.rss?.channel?.item || []; const itemsArray = Array.isArray(items) ? items : [items]; const articles: NewsArticle[] = itemsArray .slice(0, limit) .map((item: any) => { const title = item.title || 'No title'; const url = item.link || ''; const description = item.description || ''; const pubDate = item.pubDate || null; // Extract source - handle both string and object formats let source: string; if (typeof item.source === 'string') { source = item.source; } else if (item.source && typeof item.source === 'object') { // RSS source element can have text content and url attribute source = item.source['#text'] || extractSourceFromUrl(item.source['@_url'] || url); } else { source = extractSourceFromUrl(url); } // Parse and format date let publishedDate: string | null = null; if (pubDate) { try { const date = new Date(pubDate); publishedDate = date.toISOString(); } catch (e) { publishedDate = pubDate; } } // Extract image URL from various possible locations in RSS feed let imageUrl: string | null = null; // Try enclosure tag (common in RSS feeds) if (item.enclosure && item.enclosure['@_url']) { imageUrl = item.enclosure['@_url']; } // Try media:content tag else if (item['media:content'] && item['media:content']['@_url']) { imageUrl = item['media:content']['@_url']; } // Try media:thumbnail tag else if (item['media:thumbnail'] && item['media:thumbnail']['@_url']) { imageUrl = item['media:thumbnail']['@_url']; } // Try to extract from description HTML if present else if (description) { const imgMatch = description.match(/<img[^>]+src="([^">]+)"/); if (imgMatch && imgMatch[1]) { imageUrl = imgMatch[1]; } } return { title, url, source, published_date: publishedDate, description, image_url: imageUrl, last_updated: lastUpdated, }; }); // Filter to Canadian sources if we have enough results const filtered = articles.filter((article) => CANADIAN_SOURCES.some((source) => article.url.toLowerCase().includes(source)) ); // Use filtered results if we have at least half of what we wanted if (filtered.length >= Math.min(5, limit / 2)) { return filtered.slice(0, limit); } return articles; } catch (error) { console.error(`Error fetching news for ${mpName}:`, error); return []; } } /** * Extract and format news source name from URL */ function extractSourceFromUrl(url: string): string { if (!url) return 'Unknown'; try { const urlObj = new URL(url); const domain = urlObj.hostname.toLowerCase().replace('www.', ''); // Return mapped name or capitalize domain return ( SOURCE_NAMES[domain] || domain.split('.')[0].charAt(0).toUpperCase() + domain.split('.')[0].slice(1) ); } catch { return 'Unknown'; } }