fix: robust pdfjs-dist worker resolution + Docker deployment

.dockerignore

@@ -0,0 +1,22 @@
+node_modules
+.pnpm-store
+.next
+.turbo
+dist
+coverage
+.git
+.env
+.env.local
+.env.*.local
+*.log
+*.tsbuildinfo
+local-blobs
+test-parse.*
+e2e/test-results
+e2e/report
+e2e/.auth
+.DS_Store
+.vscode
+.idea
+.claude
+docs

apps/web/Dockerfile

@@ -0,0 +1,58 @@
+FROM node:20-alpine AS base
+
+RUN corepack enable && corepack prepare pnpm@9.15.4 --activate
+
+WORKDIR /app
+
+# Copy workspace root files
+COPY package.json pnpm-workspace.yaml pnpm-lock.yaml* turbo.json tsconfig.json ./
+
+# Copy all package.json files for dependency resolution
+COPY apps/web/package.json apps/web/
+COPY packages/common/package.json packages/common/
+COPY packages/database/package.json packages/database/
+COPY packages/blob-storage/package.json packages/blob-storage/
+COPY packages/ai/package.json packages/ai/
+COPY packages/ingestion/package.json packages/ingestion/
+COPY packages/events/package.json packages/events/
+COPY packages/sharing/package.json packages/sharing/
+
+# Install dependencies (shamefully-hoist for Next.js bundler compatibility)
+RUN echo "shamefully-hoist=true" > .npmrc && pnpm install --frozen-lockfile
+
+# Copy all source
+COPY packages/ packages/
+COPY apps/web/ apps/web/
+
+# Copy root .env.example as build-time .env (real values come at runtime)
+COPY .env.example .env
+
+# Build the Next.js app via turbo (resolves workspace deps correctly)
+RUN pnpm turbo build --filter=@openvitals/web
+
+# Production stage
+FROM node:20-alpine AS runner
+WORKDIR /app
+
+ENV NODE_ENV=production
+ENV NEXT_TELEMETRY_DISABLED=1
+
+RUN addgroup --system --gid 1001 nodejs && \
+    adduser --system --uid 1001 nextjs
+
+# Create blob directory with correct ownership
+RUN mkdir -p /data/blobs && chown nextjs:nodejs /data/blobs
+
+# Copy standalone output
+COPY --from=base /app/apps/web/.next/standalone ./
+COPY --from=base /app/apps/web/.next/static ./apps/web/.next/static
+COPY --from=base /app/apps/web/public ./apps/web/public
+
+USER nextjs
+
+EXPOSE 3000
+
+ENV PORT=3000
+ENV HOSTNAME="0.0.0.0"
+
+CMD ["node", "apps/web/server.js"]

apps/web/next.config.ts

@@ -1,13 +1,14 @@
-import type { NextConfig } from 'next';
+import type { NextConfig } from "next";
 
 const nextConfig: NextConfig = {
+  output: "standalone",
   transpilePackages: [
-    '@openvitals/common',
-    '@openvitals/database',
-    '@openvitals/blob-storage',
-    '@openvitals/ai',
-    '@openvitals/events',
-    '@openvitals/sharing',
+    "@openvitals/common",
+    "@openvitals/database",
+    "@openvitals/blob-storage",
+    "@openvitals/ai",
+    "@openvitals/events",
+    "@openvitals/sharing",
   ],
 };
 

docker-compose.prod.yml

@@ -0,0 +1,84 @@
+services:
+  postgres:
+    image: postgres:16-alpine
+    restart: unless-stopped
+    environment:
+      POSTGRES_USER: ${POSTGRES_USER:-postgres}
+      POSTGRES_PASSWORD: ${POSTGRES_PASSWORD}
+      POSTGRES_DB: ${POSTGRES_DB:-openvitals}
+    volumes:
+      - pgdata:/var/lib/postgresql/data
+    healthcheck:
+      test: ["CMD-SHELL", "pg_isready -U ${POSTGRES_USER:-postgres}"]
+      interval: 10s
+      timeout: 5s
+      retries: 5
+
+  web:
+    build:
+      context: .
+      dockerfile: apps/web/Dockerfile
+    restart: unless-stopped
+    ports:
+      - "${WEB_PORT:-3000}:3000"
+    environment:
+      DATABASE_URL: postgresql://${POSTGRES_USER:-postgres}:${POSTGRES_PASSWORD}@postgres:5432/${POSTGRES_DB:-openvitals}
+      BETTER_AUTH_SECRET: ${BETTER_AUTH_SECRET}
+      BETTER_AUTH_URL: ${BETTER_AUTH_URL:-http://localhost:3000}
+      NEXT_PUBLIC_APP_URL: ${NEXT_PUBLIC_APP_URL:-http://localhost:3000}
+      OPENROUTER_API_KEY: ${OPENROUTER_API_KEY}
+      AI_DEFAULT_MODEL: ${AI_DEFAULT_MODEL:-google/gemini-2.5-flash}
+      BLOB_STORAGE_PROVIDER: local
+      LOCAL_BLOB_DIR: /data/blobs
+      RENDER_WORKER_URL: http://worker:4000
+      RENDER_WEBHOOK_SECRET: ${RENDER_WEBHOOK_SECRET}
+      ENCRYPTION_KEY: ${ENCRYPTION_KEY}
+      NODE_ENV: production
+    volumes:
+      - blobdata:/data/blobs
+    depends_on:
+      postgres:
+        condition: service_healthy
+      migrate:
+        condition: service_completed_successfully
+
+  worker:
+    build:
+      context: .
+      dockerfile: services/ingestion-worker/Dockerfile
+    restart: unless-stopped
+    environment:
+      DATABASE_URL: postgresql://${POSTGRES_USER:-postgres}:${POSTGRES_PASSWORD}@postgres:5432/${POSTGRES_DB:-openvitals}
+      OPENROUTER_API_KEY: ${OPENROUTER_API_KEY}
+      AI_DEFAULT_MODEL: ${AI_DEFAULT_MODEL:-google/gemini-2.5-flash}
+      BLOB_STORAGE_PROVIDER: local
+      LOCAL_BLOB_DIR: /data/blobs
+      RENDER_WEBHOOK_SECRET: ${RENDER_WEBHOOK_SECRET}
+      NODE_ENV: production
+    volumes:
+      - blobdata:/data/blobs
+    depends_on:
+      postgres:
+        condition: service_healthy
+      migrate:
+        condition: service_completed_successfully
+
+  # Run migrations and seed on deploy (one-shot, exits after completion)
+  migrate:
+    build:
+      context: .
+      dockerfile: apps/web/Dockerfile
+      target: base
+    restart: "no"
+    environment:
+      DATABASE_URL: postgresql://${POSTGRES_USER:-postgres}:${POSTGRES_PASSWORD}@postgres:5432/${POSTGRES_DB:-openvitals}
+    working_dir: /app
+    command: >
+      sh -c "cd packages/database && pnpm db:migrate"
+    depends_on:
+      postgres:
+        condition: service_healthy
+
+volumes:
+  pgdata:
+  blobdata:

services/ingestion-worker/Dockerfile

@@ -35,6 +35,7 @@ COPY --from=base /app/services/ingestion-worker/dist ./dist
 COPY --from=base /app/services/ingestion-worker/package.json ./package.json
 COPY --from=base /app/node_modules/.pnpm/pdfjs-dist@*/node_modules/pdfjs-dist ./node_modules/pdfjs-dist
 
+
 EXPOSE 4000
 
 CMD ["node", "dist/server.js"]

services/ingestion-worker/src/lib/pdf.ts

@@ -14,17 +14,27 @@ interface TextItem {
  * with their column headers in table-based documents like lab trend reports.
  */
 export async function extractTextFromPdf(buffer: Buffer): Promise<string> {
-  const { getDocument } = await import('pdfjs-dist/legacy/build/pdf.mjs');
-  const doc = await getDocument({ data: new Uint8Array(buffer) }).promise;
+  const pdfjs = await import("pdfjs-dist/legacy/build/pdf.mjs");
+  // Resolve the worker file from node_modules (not relative to this source file)
+  const { createRequire } = await import("module");
+  const require = createRequire(import.meta.url);
+  pdfjs.GlobalWorkerOptions.workerSrc =
+    require.resolve("pdfjs-dist/legacy/build/pdf.worker.mjs");
+  const doc = await pdfjs.getDocument({
+    data: new Uint8Array(buffer),
+    useWorkerFetch: false,
+    isEvalSupported: false,
+    useSystemFonts: true,
+  }).promise;
 
-  let text = '';
+  let text = "";
   for (let i = 1; i <= doc.numPages; i++) {
     const page = await doc.getPage(i);
     const content = await page.getTextContent();
     const items = content.items as TextItem[];
 
     if (items.length === 0) {
-      text += '\n';
+      text += "\n";
       continue;
     }
 
@@ -34,7 +44,7 @@ export async function extractTextFromPdf(buffer: Buffer): Promise<string> {
     const rows: { y: number; items: { x: number; str: string }[] }[] = [];
 
     for (const item of items) {
-      if (!item.str.trim() && !item.str.includes(' ')) continue;
+      if (!item.str.trim() && !item.str.includes(" ")) continue;
       const x = item.transform[4]!;
       const y = item.transform[5]!;
 
@@ -54,23 +64,23 @@ export async function extractTextFromPdf(buffer: Buffer): Promise<string> {
       row.items.sort((a, b) => a.x - b.x);
 
       // Insert tab separators when there's a significant horizontal gap
-      let line = '';
+      let line = "";
       let prevEnd = -Infinity;
       for (const item of row.items) {
         const gap = item.x - prevEnd;
         if (prevEnd > -Infinity && gap > 15) {
-          line += '\t';
+          line += "\t";
         } else if (prevEnd > -Infinity && gap > 2) {
-          line += ' ';
+          line += " ";
         }
         line += item.str;
         // Estimate end position: x + approximate character width
-        prevEnd = item.x + (item.str.length * 5);
+        prevEnd = item.x + item.str.length * 5;
       }
-      text += line + '\n';
+      text += line + "\n";
     }
 
-    text += '\n';
+    text += "\n";
   }
 
   doc.destroy();