Files
ScraperControl/scripts/import-philmass.ts

742 lines
25 KiB
TypeScript
Raw Normal View History

#!/usr/bin/env tsx
/**
* Import Catholic churches and mass schedules from Philmass.com
*
* Philmass.com provides rich Schema.org-annotated mass schedule data for
* Philippine churches. It has no coordinates, so we match against existing
* churches (OSM + mass-schedules.com) and only update matched records.
* Unmatched churches are logged for manual review.
*
* Discovery strategy:
* 1. Fetch Philippines page extract province URLs
* 2. For each province extract city listing URLs
* 3. For each city listing extract church mass-schedule URLs
* 4. Deduplicate all church URLs globally
* 5. For each church: parse JSON-LD + Schema.org Events, match, upsert
*
* Usage:
* npx tsx scripts/import-philmass.ts --all
* npx tsx scripts/import-philmass.ts --all --dry-run
* npx tsx scripts/import-philmass.ts --province Metro-Manila
* npx tsx scripts/import-philmass.ts --all --resume-from Cebu
* npx tsx scripts/import-philmass.ts --all --job-id {uuid}
*/
import dotenv from 'dotenv';
import path from 'path';
dotenv.config({ path: path.resolve(process.cwd(), '.env.local') });
dotenv.config({ path: path.resolve(process.cwd(), '.env') });
import { Pool } from 'pg';
import { PrismaPg } from '@prisma/adapter-pg';
import { PrismaClient } from '@prisma/client';
const dbUrl = process.env.DATABASE_URL || 'postgresql://postgres:postgres@localhost:5432/nearestmass';
console.log(`Connecting to database: ${dbUrl.replace(/:[^:@]+@/, ':***@')}`);
const pool = new Pool({
connectionString: dbUrl,
ssl: dbUrl.includes('neon') ? { rejectUnauthorized: false } : undefined,
});
const adapter = new PrismaPg(pool);
const prisma = new PrismaClient({ adapter });
import { findDuplicateChurch } from '../src/lib/church-matcher';
import type { ExistingChurch } from '../src/lib/church-matcher';
// ─── Constants ───────────────────────────────────────────────────────────────
const SITE_BASE = 'https://www.philmass.com';
const PHILIPPINES_URL = `${SITE_BASE}/Asia/Philippines.html`;
const USER_AGENT = 'NearestMass-Importer/1.0 (parish data aggregator; contact: privacy@nearestmass.com)';
const REQUEST_DELAY_MS = 2000;
// ─── Types ───────────────────────────────────────────────────────────────────
interface ProvinceInfo {
name: string;
url: string;
}
interface ChurchUrl {
url: string;
slug: string; // URL slug used as philmassId
province: string;
city: string;
}
interface ParsedPhilmassChurch {
name: string;
streetAddress: string | null;
city: string | null;
region: string | null;
}
interface ParsedSchedule {
dayOfWeek: number;
time: string;
}
interface ImportStats {
provincesProcessed: number;
citiesProcessed: number;
churchUrlsDiscovered: number;
churchesProcessed: number;
churchesMatched: number;
churchesUnmatched: number;
churchesSkipped: number;
schedulesUpdated: number;
massSchedulesCreated: number;
errors: number;
}
interface CLIArgs {
all: boolean;
province?: string;
dryRun: boolean;
resumeFrom?: string;
jobId?: string;
}
// ─── HTTP Client ─────────────────────────────────────────────────────────────
let requestCount = 0;
function delay(ms: number): Promise<void> {
return new Promise((resolve) => setTimeout(resolve, ms));
}
async function fetchPage(url: string): Promise<string | null> {
if (requestCount > 0) {
await delay(REQUEST_DELAY_MS);
}
requestCount++;
try {
const response = await fetch(url, {
headers: {
'User-Agent': USER_AGENT,
'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
},
});
if (!response.ok) {
console.error(` HTTP ${response.status} for ${url}`);
return null;
}
return await response.text();
} catch (error) {
console.error(` Fetch error for ${url}: ${error instanceof Error ? error.message : error}`);
return null;
}
}
// ─── Discovery: Province → City → Church URLs ───────────────────────────────
async function fetchProvinceUrls(): Promise<ProvinceInfo[]> {
console.log(`Fetching Philippines page: ${PHILIPPINES_URL}`);
const html = await fetchPage(PHILIPPINES_URL);
if (!html) throw new Error('Failed to fetch Philippines page');
const provinces: ProvinceInfo[] = [];
const seen = new Set<string>();
// Pattern: href="https://www.philmass.com/Asia/Philippines/{Province}.html"
const regex = /href="(https:\/\/www\.philmass\.com\/Asia\/Philippines\/([^/"]+)\.html)"/g;
let match;
while ((match = regex.exec(html)) !== null) {
const url = match[1];
const name = match[2];
// Skip non-province pages (weekly-sunday, etc.)
if (name.includes('weekly') || name.includes('Roman-Catholic') || seen.has(name)) continue;
seen.add(name);
provinces.push({ name, url });
}
return provinces;
}
function decodeHtmlEntities(str: string): string {
return str
.replace(/&#(\d+);/g, (_, code: string) => String.fromCharCode(parseInt(code)))
.replace(/&amp;/g, '&')
.replace(/&lt;/g, '<')
.replace(/&gt;/g, '>')
.replace(/&quot;/g, '"');
}
async function fetchCityListingUrls(provinceUrl: string, provinceName: string): Promise<string[]> {
const html = await fetchPage(provinceUrl);
if (!html) return [];
const urls: string[] = [];
const seen = new Set<string>();
// Pattern: href=".../{Province}/{City}/Roman-Catholic-Churches-in-{City}...html"
const regex = /href="(https:\/\/www\.philmass\.com\/Asia\/Philippines\/[^"]*\/Roman-Catholic-Churches-in-[^"]*\.html)"/g;
let match;
while ((match = regex.exec(html)) !== null) {
const url = decodeHtmlEntities(match[1]);
if (seen.has(url)) continue;
seen.add(url);
urls.push(url);
}
return urls;
}
async function fetchChurchUrlsFromCityPage(cityUrl: string, provinceName: string): Promise<ChurchUrl[]> {
const html = await fetchPage(cityUrl);
if (!html) return [];
const churches: ChurchUrl[] = [];
const seen = new Set<string>();
// Pattern: href=".../Roman-Catholic-Churches/{Church-Name}/mass-schedule.html"
const regex = /href="(https:\/\/www\.philmass\.com\/Asia\/Philippines\/([^/]+)\/([^/]+)\/Roman-Catholic-Churches\/([^/]+)\/mass-schedule\.html)"/g;
let match;
while ((match = regex.exec(html)) !== null) {
const url = decodeHtmlEntities(match[1]);
const province = decodeURIComponent(decodeHtmlEntities(match[2]));
const city = decodeURIComponent(decodeHtmlEntities(match[3]));
const slug = decodeURIComponent(decodeHtmlEntities(match[4]));
if (seen.has(url)) continue;
seen.add(url);
churches.push({ url, slug, province, city });
}
return churches;
}
// ─── HTML Parsers ────────────────────────────────────────────────────────────
function parseChurchJsonLd(html: string): ParsedPhilmassChurch | null {
// Extract JSON-LD: <script type="application/ld+json">{...}</script>
const jsonLdMatch = html.match(/<script\s+type="application\/ld\+json"\s*>([\s\S]*?)<\/script>/i);
if (!jsonLdMatch) return null;
try {
const data = JSON.parse(jsonLdMatch[1]);
const church = data.mainEntityOfPage;
if (!church || church['@type'] !== 'PlaceOfWorship') return null;
const address = church.address || {};
return {
name: church.name || null,
streetAddress: address.streetAddress?.replace(/,\s*$/, '').trim() || null,
city: address.addressLocality || null,
region: address.addressRegion || null,
};
} catch {
return null;
}
}
function parseChurchNameFromH1(html: string): string | null {
// Fallback: <h1>Quiapo Church mass schedule 2026 - Minor Basilica of the Black Nazarene</h1>
const h1Match = html.match(/<h1>([^<]+)<\/h1>/i);
if (!h1Match) return null;
let name = h1Match[1].trim();
// Remove "mass schedule YYYY" and trailing " - "
name = name.replace(/\s*mass\s+schedule\s+\d{4}\s*/i, '');
name = name.replace(/^\s*-\s*/, '').replace(/\s*-\s*$/, '');
return name.trim() || null;
}
function parseScheduleFromStartDates(html: string): ParsedSchedule[] {
// Extract all startDate ISO timestamps from Schema.org Event microdata
// Pattern: itemprop="startDate" content="2026-02-22T05:00:00+08:00"
const schedules: ParsedSchedule[] = [];
const seen = new Set<string>();
const regex = /itemprop="startDate"\s+content="(\d{4}-\d{2}-\d{2})T(\d{2}):(\d{2}):\d{2}[^"]*"/g;
let match;
while ((match = regex.exec(html)) !== null) {
const dateStr = match[1];
const hours = match[2];
const minutes = match[3];
// Derive dayOfWeek from the date
const date = new Date(`${dateStr}T12:00:00`); // noon to avoid TZ issues
const dayOfWeek = date.getDay(); // 0=Sun, 1=Mon, ..., 6=Sat
const time = `${hours}:${minutes}`;
const key = `${dayOfWeek}:${time}`;
if (seen.has(key)) continue;
seen.add(key);
schedules.push({ dayOfWeek, time });
}
return schedules;
}
// ─── Database Operations ─────────────────────────────────────────────────────
async function loadExistingPhilippineChurches(): Promise<ExistingChurch[]> {
console.log('Loading existing Philippine churches for deduplication...');
const churches = await prisma.church.findMany({
where: { country: 'PH' },
select: {
id: true,
name: true,
latitude: true,
longitude: true,
osmId: true,
baiduId: true,
masstimesId: true,
orarimesseId: true,
massSchedulesPhId: true,
philmassId: true,
horariosMisasId: true,
mszeInfoId: true,
weekdayMassesId: true,
messesInfoId: true,
bohosluzbyId: true,
miserendId: true,
kerknetId: true,
gottesdienstzeitenId: true,
source: true,
website: true,
phone: true,
address: true,
},
});
console.log(`Loaded ${churches.length} existing Philippine churches`);
return churches;
}
// ─── Import Logic ────────────────────────────────────────────────────────────
async function processChurch(
churchUrl: ChurchUrl,
existingChurches: ExistingChurch[],
unmatchedLog: string[],
dryRun: boolean,
stats: ImportStats,
): Promise<void> {
stats.churchesProcessed++;
const html = await fetchPage(churchUrl.url);
if (!html) {
stats.errors++;
return;
}
// Parse church info from JSON-LD
const jsonLd = parseChurchJsonLd(html);
const churchName = jsonLd?.name || parseChurchNameFromH1(html);
if (!churchName) {
console.log(` Skipping ${churchUrl.slug}: no name found`);
stats.churchesSkipped++;
return;
}
// Parse schedules from Schema.org startDate attributes
const schedules = parseScheduleFromStartDates(html);
// Try to find a match by philmassId first
const existingByPhilmass = existingChurches.find((c) => c.philmassId === churchUrl.slug);
let matched = existingByPhilmass || null;
// If no philmassId match, try name-based matching against churches with coordinates
if (!matched) {
// Try matching by name similarity against all PH churches
// We can't use findDuplicateChurch() without coordinates, so do name-only matching
const normalizedName = churchName.toLowerCase()
.replace(/\bst\.\s/g, 'saint ')
.replace(/\bst\s/g, 'saint ')
.replace(/\bcatholic church\b/g, '')
.replace(/\bparish\b/g, '')
.replace(/\bchurch\b/g, '')
.replace(/[^\w\s]/g, '')
.replace(/\s+/g, ' ')
.trim();
// Filter to churches in the same city if possible
const cityName = jsonLd?.city || churchUrl.city.replace(/-/g, ' ');
const candidatesInCity = existingChurches.filter((c) => {
if (!c.address) return false;
return c.address.toLowerCase().includes(cityName.toLowerCase());
});
// Search in-city candidates first, then all PH churches
const searchPools = candidatesInCity.length > 0
? [candidatesInCity, existingChurches]
: [existingChurches];
for (const searchPool of searchPools) {
if (matched) break;
for (const existing of searchPool) {
const existingNorm = existing.name.toLowerCase()
.replace(/\bst\.\s/g, 'saint ')
.replace(/\bst\s/g, 'saint ')
.replace(/\bcatholic church\b/g, '')
.replace(/\bparish\b/g, '')
.replace(/\bchurch\b/g, '')
.replace(/[^\w\s]/g, '')
.replace(/\s+/g, ' ')
.trim();
// Require strong name match: one name contains the other, or very similar
// Guard against overly generic names ("chapel", "holy", etc.) by requiring
// that the shorter name is at least 8 chars after normalization
const shorter = normalizedName.length <= existingNorm.length ? normalizedName : existingNorm;
if (shorter.length >= 8) {
if (normalizedName.includes(existingNorm) || existingNorm.includes(normalizedName)) {
matched = existing;
break;
}
}
}
}
}
if (dryRun) {
if (matched) {
stats.churchesMatched++;
console.log(` [MATCH] "${churchName}" → existing "${matched.name}" (${matched.id})`);
} else {
stats.churchesUnmatched++;
unmatchedLog.push(`${churchName} | ${jsonLd?.city || churchUrl.city} | ${churchUrl.url}`);
console.log(` [UNMATCHED] "${churchName}" in ${jsonLd?.city || churchUrl.city}`);
}
if (schedules.length > 0) {
stats.massSchedulesCreated += schedules.length;
}
return;
}
if (!matched) {
stats.churchesUnmatched++;
unmatchedLog.push(`${churchName} | ${jsonLd?.city || churchUrl.city} | ${churchUrl.url}`);
return;
}
stats.churchesMatched++;
// Update existing church: set philmassId, fill missing fields
const updateData: Record<string, unknown> = {
philmassId: churchUrl.slug,
};
if (!matched.address && jsonLd?.streetAddress) {
const fullAddress = [jsonLd.streetAddress, jsonLd.city, jsonLd.region]
.filter(Boolean).join(', ');
updateData.address = fullAddress;
}
// Fill city/state from JSON-LD or URL
const dbRecord = await prisma.church.findUnique({
where: { id: matched.id },
select: { city: true, state: true },
});
if (dbRecord && !dbRecord.city && (jsonLd?.city || churchUrl.city)) {
updateData.city = jsonLd?.city || churchUrl.city.replace(/-/g, ' ');
}
if (dbRecord && !dbRecord.state && (jsonLd?.region || churchUrl.province)) {
updateData.state = jsonLd?.region || churchUrl.province.replace(/-/g, ' ');
}
try {
await prisma.church.update({
where: { id: matched.id },
data: updateData,
});
} catch (error) {
if (error instanceof Error && error.message.includes('Unique constraint')) {
stats.churchesSkipped++;
return;
}
throw error;
}
// Replace mass schedules if we have any
if (schedules.length > 0) {
try {
await prisma.$transaction(async (tx) => {
await tx.massSchedule.deleteMany({ where: { churchId: matched!.id } });
await tx.massSchedule.createMany({
data: schedules.map((s) => ({
churchId: matched!.id,
dayOfWeek: s.dayOfWeek,
time: s.time,
language: 'English',
})),
});
await tx.church.update({
where: { id: matched!.id },
data: { lastScrapedAt: new Date() },
});
});
stats.schedulesUpdated++;
stats.massSchedulesCreated += schedules.length;
} catch (error) {
stats.errors++;
console.error(` Error saving schedules for ${churchUrl.slug}: ${error instanceof Error ? error.message : error}`);
}
}
}
// ─── CLI ─────────────────────────────────────────────────────────────────────
function parseArgs(): CLIArgs {
const args = process.argv.slice(2);
const result: CLIArgs = {
all: false,
dryRun: false,
};
for (let i = 0; i < args.length; i++) {
switch (args[i]) {
case '--all':
result.all = true;
break;
case '--province':
result.province = args[++i];
break;
case '--dry-run':
result.dryRun = true;
break;
case '--resume-from':
result.resumeFrom = args[++i];
break;
case '--job-id':
result.jobId = args[++i];
break;
case '--help':
case '-h':
console.log(`
Usage: npx tsx scripts/import-philmass.ts [options]
Options:
--all Import from all provinces
--province <name> Import from a single province (e.g. "Metro-Manila")
--dry-run No database writes, just report what would happen
--resume-from <province> Skip provinces until reaching this one
--job-id <uuid> Background job tracking ID
--help, -h Show this help message
Examples:
npx tsx scripts/import-philmass.ts --province Metro-Manila --dry-run
npx tsx scripts/import-philmass.ts --all
npx tsx scripts/import-philmass.ts --all --resume-from Cebu
`);
process.exit(0);
}
}
if (!result.all && !result.province) {
console.error('Error: specify --all or --province <name>');
process.exit(1);
}
return result;
}
// ─── Helpers ─────────────────────────────────────────────────────────────────
function formatDuration(ms: number): string {
const seconds = Math.floor(ms / 1000);
const minutes = Math.floor(seconds / 60);
const hours = Math.floor(minutes / 60);
if (hours > 0) return `${hours}h ${minutes % 60}m ${seconds % 60}s`;
if (minutes > 0) return `${minutes}m ${seconds % 60}s`;
return `${seconds}s`;
}
// ─── Main ────────────────────────────────────────────────────────────────────
async function main() {
const args = parseArgs();
const startTime = Date.now();
console.log('\n' + '='.repeat(70));
console.log('PHILMASS.COM IMPORTER');
console.log('='.repeat(70));
console.log(`Mode: ${args.all ? 'All provinces' : `Single province: ${args.province}`}`);
console.log(`Dry run: ${args.dryRun ? 'YES (no DB writes)' : 'NO'}`);
if (args.resumeFrom) console.log(`Resume from: ${args.resumeFrom}`);
console.log(`Time: ${new Date().toISOString()}`);
console.log('='.repeat(70) + '\n');
// Update background job status if provided
if (args.jobId) {
try {
await prisma.backgroundJob.update({
where: { id: args.jobId },
data: { status: 'running', startedAt: new Date() },
});
} catch {
// Job might not exist yet
}
}
// Load existing Philippine churches for dedup
const existingChurches = await loadExistingPhilippineChurches();
// ─── Phase 1: Discover all church URLs ───────────────────────────────────
console.log('=== Phase 1: Discovering church URLs ===\n');
const allProvinces = await fetchProvinceUrls();
console.log(`Found ${allProvinces.length} provinces\n`);
// Filter to requested provinces
let provincesToProcess: ProvinceInfo[];
if (args.province) {
const found = allProvinces.find((p) => p.name === args.province);
if (!found) {
console.error(`Province "${args.province}" not found. Available: ${allProvinces.map((p) => p.name).join(', ')}`);
process.exit(1);
}
provincesToProcess = [found];
} else {
provincesToProcess = allProvinces;
}
// Handle --resume-from
if (args.resumeFrom) {
const idx = provincesToProcess.findIndex((p) => p.name === args.resumeFrom);
if (idx === -1) {
console.error(`Resume province "${args.resumeFrom}" not found.`);
process.exit(1);
}
console.log(`Resuming from province "${args.resumeFrom}" (skipping ${idx} provinces)\n`);
provincesToProcess = provincesToProcess.slice(idx);
}
// Collect all unique church URLs across all provinces/cities
const allChurchUrls = new Map<string, ChurchUrl>(); // keyed by URL to deduplicate
const stats: ImportStats = {
provincesProcessed: 0,
citiesProcessed: 0,
churchUrlsDiscovered: 0,
churchesProcessed: 0,
churchesMatched: 0,
churchesUnmatched: 0,
churchesSkipped: 0,
schedulesUpdated: 0,
massSchedulesCreated: 0,
errors: 0,
};
for (let pi = 0; pi < provincesToProcess.length; pi++) {
const province = provincesToProcess[pi];
const elapsed = formatDuration(Date.now() - startTime);
console.log(`[${pi + 1}/${provincesToProcess.length}] Province: ${province.name} [${elapsed} elapsed]`);
try {
// Get city listing URLs from province page
const cityUrls = await fetchCityListingUrls(province.url, province.name);
console.log(` Found ${cityUrls.length} city listing pages`);
for (const cityUrl of cityUrls) {
const churchUrls = await fetchChurchUrlsFromCityPage(cityUrl, province.name);
stats.citiesProcessed++;
for (const church of churchUrls) {
if (!allChurchUrls.has(church.url)) {
allChurchUrls.set(church.url, church);
}
}
}
stats.provincesProcessed++;
console.log(` Total unique churches so far: ${allChurchUrls.size}`);
} catch (error) {
stats.errors++;
console.error(` ERROR discovering ${province.name}: ${error instanceof Error ? error.message : error}`);
}
}
stats.churchUrlsDiscovered = allChurchUrls.size;
console.log(`\nDiscovery complete: ${allChurchUrls.size} unique church URLs across ${stats.citiesProcessed} city pages\n`);
// ─── Phase 2: Process each church ─────────────────────────────────────────
console.log('=== Phase 2: Processing churches ===\n');
const churchList = [...allChurchUrls.values()];
const unmatchedLog: string[] = [];
for (let i = 0; i < churchList.length; i++) {
const church = churchList[i];
const elapsed = formatDuration(Date.now() - startTime);
if ((i + 1) % 50 === 0 || i === 0) {
console.log(`[${i + 1}/${churchList.length}] Processing churches... [${elapsed} elapsed]`);
}
try {
await processChurch(church, existingChurches, unmatchedLog, args.dryRun, stats);
} catch (error) {
stats.errors++;
console.error(` ERROR processing ${church.slug}: ${error instanceof Error ? error.message : error}`);
}
}
// Print summary
const totalTime = Date.now() - startTime;
console.log('\n' + '='.repeat(70));
console.log(`IMPORT SUMMARY ${args.dryRun ? '(DRY RUN)' : ''}`);
console.log('='.repeat(70));
console.log(`Provinces processed: ${stats.provincesProcessed}`);
console.log(`Cities processed: ${stats.citiesProcessed}`);
console.log(`Church URLs discovered: ${stats.churchUrlsDiscovered}`);
console.log(`Churches processed: ${stats.churchesProcessed}`);
console.log(` Matched (updated): ${stats.churchesMatched}`);
console.log(` Unmatched (skipped): ${stats.churchesUnmatched}`);
console.log(` Skipped (other): ${stats.churchesSkipped}`);
console.log(`Schedules updated: ${stats.schedulesUpdated}`);
console.log(`Mass schedules created: ${stats.massSchedulesCreated}`);
console.log(`Errors: ${stats.errors}`);
console.log(`Total time: ${formatDuration(totalTime)}`);
console.log(`HTTP requests: ${requestCount}`);
console.log('='.repeat(70));
// Log unmatched churches for manual review
if (unmatchedLog.length > 0) {
console.log(`\nUnmatched churches (${unmatchedLog.length}):`);
console.log('-'.repeat(70));
for (const line of unmatchedLog) {
console.log(` ${line}`);
}
console.log('-'.repeat(70));
}
console.log('');
// Update background job
if (args.jobId) {
try {
await prisma.backgroundJob.update({
where: { id: args.jobId },
data: {
status: stats.errors > 0 ? 'completed_with_errors' : 'completed',
completedAt: new Date(),
result: JSON.stringify(stats),
},
});
} catch {
// Ignore
}
}
}
main()
.catch((error) => {
console.error('Fatal error:', error);
process.exit(1);
})
.finally(async () => {
await prisma.$disconnect();
await pool.end();
});