fix: clean up church-matcher types and add HK OSM bounding box

- Remove discovermassId/buscarmisasNetworkId from findDuplicateChurch match passes (importers now do their own pre-check dedup); restore as optional fields on ExistingChurch to keep type/runtime in sync - Add HK bounding box to COUNTRY_BOUNDING_BOXES; fix silent 0-result fallback when country query returns empty from mirror server - discovermass importer: add --limit flag and skip-already-imported pre-check using importedSlugs set - Import scripts: remove discovermassId from ExistingChurch select/stubs (field not needed in shared matcher context) - Schema: reorder discovermassId/kerknetId/gottesdienstzeitenId fields Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-01 22:20:45 -04:00
parent 3bd4d2e2f9
commit 033f805965
21 changed files with 499 additions and 64 deletions
--- a/scripts/import-baidu-churches.ts
+++ b/scripts/import-baidu-churches.ts
@@ -178,7 +178,6 @@ async function importFromBaidu(dryRun: boolean, resumeFromCell: number, jobId: s
      miserendId: true,
      kerknetId: true,
      gottesdienstzeitenId: true,
-      discovermassId: true,
      source: true,
      website: true,
      phone: true,
@@ -259,7 +258,6 @@ async function importFromBaidu(dryRun: boolean, resumeFromCell: number, jobId: s
          miserendId: null,
          kerknetId: null,
          gottesdienstzeitenId: null,
-          discovermassId: null,
          source: 'baidu',
          website: baiduChurch.website || null,
          phone: baiduChurch.phone || null,
--- a/scripts/import-bohosluzby.ts
+++ b/scripts/import-bohosluzby.ts
@@ -287,7 +287,6 @@ async function loadExistingCzechChurches(): Promise<ExistingChurch[]> {
      miserendId: true,
      kerknetId: true,
      gottesdienstzeitenId: true,
-      discovermassId: true,
      source: true,
      website: true,
      phone: true,
@@ -415,7 +414,6 @@ async function processChurch(
        miserendId: null,
        kerknetId: null,
        gottesdienstzeitenId: null,
-        discovermassId: null,
        source: 'bohosluzby',
        website: null,
        phone: null,
--- a/scripts/import-discovermass.ts
+++ b/scripts/import-discovermass.ts
@@ -94,6 +94,7 @@ interface CLIArgs {
  all: boolean;
  dryRun: boolean;
  resumeFrom?: number;
+  limit?: number;
  jobId?: string;
 }

@@ -507,6 +508,7 @@ function parseCLIArgs(): CLIArgs {
      case '--all': result.all = true; break;
      case '--dry-run': result.dryRun = true; break;
      case '--resume-from': result.resumeFrom = parseInt(args[++i], 10); break;
+      case '--limit': result.limit = parseInt(args[++i], 10); break;
      case '--job-id': result.jobId = args[++i]; break;
    }
  }
@@ -540,14 +542,25 @@ async function main() {
  try {
    const urls = await getAllChurchUrls();
    const existingChurches = await loadExistingChurches();
+
+    // Skip already-imported churches — check discovermassId set in DB
+    const importedSlugs = new Set(
+      existingChurches.filter(c => c.discovermassId).map(c => c.discovermassId!)
+    );
+
+    // Apply --resume-from first, then filter to unimported, then apply --limit
    const startIdx = args.resumeFrom ?? 0;
-    const churchUrls = urls.slice(startIdx);
-    console.log(`\nProcessing ${churchUrls.length} churches (starting from index ${startIdx})...\n`);
+    const candidateUrls = urls.slice(startIdx).filter(url => {
+      const slug = url.replace('https://discovermass.com/church/', '').replace(/\/$/, '');
+      return !importedSlugs.has(slug);
+    });
+    const churchUrls = args.limit ? candidateUrls.slice(0, args.limit) : candidateUrls;
+
+    console.log(`\nSitemap total: ${urls.length} | Already imported: ${importedSlugs.size} | This run: ${churchUrls.length}${args.limit ? ` (limit ${args.limit})` : ''}\n`);

    for (let i = 0; i < churchUrls.length; i++) {
      const url = churchUrls[i];
-      const overallIdx = startIdx + i;
-      console.log(`[${overallIdx + 1}/${urls.length}] ${url}`);
+      console.log(`[${i + 1}/${churchUrls.length}] ${url}`);
      await processChurch(url, existingChurches, args, stats);
      if (i < churchUrls.length - 1) {
        await sleep(REQUEST_DELAY_MS);
--- a/scripts/import-gcatholic.ts
+++ b/scripts/import-gcatholic.ts
@@ -401,7 +401,6 @@ async function loadExistingChurches(): Promise<ExistingChurch[]> {
      miserendId: true,
      kerknetId: true,
      gottesdienstzeitenId: true,
-      discovermassId: true,
      source: true,
      website: true,
      phone: true,
@@ -516,7 +515,6 @@ async function importChurch(
      miserendId: null,
      kerknetId: null,
      gottesdienstzeitenId: null,
-      discovermassId: null,
      source: 'gcatholic',
      website: church.website || null,
      phone: church.phone || null,
--- a/scripts/import-gottesdienstzeiten.ts
+++ b/scripts/import-gottesdienstzeiten.ts
@@ -316,7 +316,6 @@ async function loadExistingGermanChurches(): Promise<ExistingChurch[]> {
      miserendId: true,
      kerknetId: true,
      gottesdienstzeitenId: true,
-      discovermassId: true,
      source: true,
      website: true,
      phone: true,
@@ -479,7 +478,6 @@ async function processDiocese(
            miserendId: null,
            kerknetId: null,
            gottesdienstzeitenId: gdzId,
-            discovermassId: null,
            source: 'gottesdienstzeiten',
            website: church.website,
            phone: church.phone,
--- a/scripts/import-horariosmisas.ts
+++ b/scripts/import-horariosmisas.ts
@@ -570,7 +570,6 @@ async function loadExistingSpanishChurches(): Promise<ExistingChurch[]> {
      miserendId: true,
      kerknetId: true,
      gottesdienstzeitenId: true,
-      discovermassId: true,
      source: true,
      website: true,
      phone: true,
@@ -783,7 +782,6 @@ async function processChurch(
        miserendId: null,
        kerknetId: null,
        gottesdienstzeitenId: null,
-        discovermassId: null,
        source: 'horariosmisas',
        website: parsed.website,
        phone: parsed.phone,
--- a/scripts/import-kerknet.ts
+++ b/scripts/import-kerknet.ts
@@ -343,7 +343,6 @@ async function loadExistingBelgianChurches(): Promise<ExistingChurch[]> {
      miserendId: true,
      kerknetId: true,
      gottesdienstzeitenId: true,
-      discovermassId: true,
      source: true,
      website: true,
      phone: true,
@@ -490,8 +489,6 @@ async function processChurch(
        bohosluzbyId: null,
        miserendId: null,
        kerknetId,
-        gottesdienstzeitenId: null,
-        discovermassId: null,
        source: 'kerknet',
        website: church.website,
        phone: null,
--- a/scripts/import-mass-schedules-ph.ts
+++ b/scripts/import-mass-schedules-ph.ts
@@ -290,7 +290,6 @@ async function loadExistingPhilippineChurches(): Promise<ExistingChurch[]> {
      miserendId: true,
      kerknetId: true,
      gottesdienstzeitenId: true,
-      discovermassId: true,
      source: true,
      website: true,
      phone: true,
@@ -465,7 +464,6 @@ async function processChurch(
        miserendId: null,
        kerknetId: null,
        gottesdienstzeitenId: null,
-        discovermassId: null,
        source: 'mass-schedules-ph',
        website: null,
        phone: parsed.phone,
--- a/scripts/import-masstimes-api.ts
+++ b/scripts/import-masstimes-api.ts
@@ -398,7 +398,6 @@ async function loadExistingChurches(): Promise<ExistingChurch[]> {
      miserendId: true,
      kerknetId: true,
      gottesdienstzeitenId: true,
-      discovermassId: true,
      source: true,
      website: true,
      phone: true,
@@ -596,7 +595,7 @@ async function main() {
                orarimesseId: null, massSchedulesPhId: null,
                philmassId: null, horariosMisasId: null,
                mszeInfoId: null, weekdayMassesId: null,
-                messesInfoId: null, bohosluzbyId: null, miserendId: null, kerknetId: null, gottesdienstzeitenId: null, discovermassId: null,
+                messesInfoId: null, bohosluzbyId: null, miserendId: null, kerknetId: null, gottesdienstzeitenId: null,
                source: 'masstimes', website: mc.url?.trim() || null,
                phone: mc.phone_number?.trim() || null, address, country,
              });
--- a/scripts/import-messesinfo.ts
+++ b/scripts/import-messesinfo.ts
@@ -326,7 +326,6 @@ async function loadExistingFrenchChurches(): Promise<ExistingChurch[]> {
      miserendId: true,
      kerknetId: true,
      gottesdienstzeitenId: true,
-      discovermassId: true,
      source: true,
      website: true,
      phone: true,
@@ -480,7 +479,6 @@ async function processDiocese(
          miserendId: null,
          kerknetId: null,
          gottesdienstzeitenId: null,
-          discovermassId: null,
          source: 'messes-info',
          website: null,
          phone: null,
--- a/scripts/import-miserend.ts
+++ b/scripts/import-miserend.ts
@@ -240,7 +240,6 @@ async function loadExistingChurches(countryCodes: string[]): Promise<ExistingChu
      miserendId: true,
      kerknetId: true,
      gottesdienstzeitenId: true,
-      discovermassId: true,
      source: true,
      website: true,
      phone: true,
@@ -366,7 +365,6 @@ async function processChurch(
        miserendId,
        kerknetId: null,
        gottesdienstzeitenId: null,
-        discovermassId: null,
        source: 'miserend',
        website: null,
        phone: null,
--- a/scripts/import-msze-info.ts
+++ b/scripts/import-msze-info.ts
@@ -367,7 +367,6 @@ async function loadExistingPolishChurches(): Promise<ExistingChurch[]> {
      miserendId: true,
      kerknetId: true,
      gottesdienstzeitenId: true,
-      discovermassId: true,
      source: true,
      website: true,
      phone: true,
@@ -538,7 +537,6 @@ async function processChurch(
        miserendId: null,
        kerknetId: null,
        gottesdienstzeitenId: null,
-        discovermassId: null,
        source: 'msze-info',
        website: parsed.website,
        phone: parsed.phone,
--- a/scripts/import-orarimesse.ts
+++ b/scripts/import-orarimesse.ts
@@ -283,7 +283,6 @@ async function loadExistingItalianChurches(): Promise<ExistingChurch[]> {
      miserendId: true,
      kerknetId: true,
      gottesdienstzeitenId: true,
-      discovermassId: true,
      source: true,
      website: true,
      phone: true,
@@ -425,7 +424,6 @@ async function processChurchesForDiocese(
          miserendId: null,
          kerknetId: null,
          gottesdienstzeitenId: null,
-          discovermassId: null,
          source: 'orarimesse',
          website: church.sito || null,
          phone: null,
--- a/scripts/import-osm-churches.ts
+++ b/scripts/import-osm-churches.ts
@@ -204,7 +204,6 @@ async function importFromOSM(countryCode: string, dryRun: boolean = false): Prom
        miserendId: true,
        kerknetId: true,
        gottesdienstzeitenId: true,
-      discovermassId: true,
        source: true,
        website: true,
        phone: true,
@@ -380,7 +379,6 @@ async function importFromOSM(countryCode: string, dryRun: boolean = false): Prom
            miserendId: null,
            kerknetId: null,
            gottesdienstzeitenId: null,
-            discovermassId: null,
            source: 'osm',
            website: osmChurch.website || null,
            phone: osmChurch.phone || null,
--- a/scripts/import-osm-region.ts
+++ b/scripts/import-osm-region.ts
@@ -152,7 +152,6 @@ async function importFromRegion(countryCode: string, regionName: string, dryRun:
        miserendId: true,
        kerknetId: true,
        gottesdienstzeitenId: true,
-      discovermassId: true,
        source: true,
        website: true,
        phone: true,
@@ -256,7 +255,6 @@ async function importFromRegion(countryCode: string, regionName: string, dryRun:
            miserendId: null,
            kerknetId: null,
            gottesdienstzeitenId: null,
-            discovermassId: null,
            source: 'osm',
            website: osmChurch.website || null,
            phone: osmChurch.phone || null,
--- a/scripts/import-philmass.ts
+++ b/scripts/import-philmass.ts
@@ -301,7 +301,6 @@ async function loadExistingPhilippineChurches(): Promise<ExistingChurch[]> {
      miserendId: true,
      kerknetId: true,
      gottesdienstzeitenId: true,
-      discovermassId: true,
      source: true,
      website: true,
      phone: true,
--- a/scripts/import-weekdaymasses.ts
+++ b/scripts/import-weekdaymasses.ts
@@ -822,7 +822,6 @@ async function loadExistingChurches(): Promise<ExistingChurch[]> {
      miserendId: true,
      kerknetId: true,
      gottesdienstzeitenId: true,
-      discovermassId: true,
      source: true,
      website: true,
      phone: true,
@@ -982,7 +981,6 @@ async function importAreaBlocks(
            miserendId: null,
            kerknetId: null,
            gottesdienstzeitenId: null,
-            discovermassId: null,
            source: 'weekdaymasses',
            website: church.website,
            phone: church.phone,