# Paper knowledge base: Provenance + Citation graph + Semantic search
# A single Postgres instance (pgvector) covers all three layers.
# GROBID parses references / structure from PDFs without an arXiv source.
#
# Quick start:
#   docker compose up -d
#   psql "${DATABASE_URL}"          # see .env.example for default value

services:
  db:
    image: pgvector/pgvector:pg17       # official pgvector image
    container_name: papers-db
    environment:
      POSTGRES_DB: papers
      POSTGRES_USER: researcher
      POSTGRES_PASSWORD: change_me      # change before production use
    ports:
      - "5432:5432"
    volumes:
      - pgdata:/var/lib/postgresql/data
      # schema.sql is executed automatically on first start:
      - ./schema.sql:/docker-entrypoint-initdb.d/01-schema.sql:ro
    healthcheck:
      test: ["CMD-SHELL", "pg_isready -U researcher -d papers"]
      interval: 10s
      timeout: 5s
      retries: 5
    restart: unless-stopped

  grobid:
    # CRF-only image: sufficient for reference extraction (/api/processReferences)
    # and ~500MB vs ~10GB for the full deep-learning variant.
    #
    # The `-crf` tag is also the ONLY GROBID variant published as a multi-arch
    # manifest that includes linux/arm64 — the full image (grobid/grobid:0.9.0)
    # and every 0.8.x tag are amd64-only. So this is what lets GROBID run
    # *natively* on the aarch64 Jetson (no slow/RAM-heavy qemu emulation). arm64
    # builds start at 0.9.0-crf; there is no arm64 0.8.x.
    # Deploy + docker-permission fix: docs/infra/grobid-jetson.md
    image: grobid/grobid:0.9.0-crf
    container_name: papers-grobid
    ports:
      - "8070:8070"
    init: true            # reap zombie children (GROBID docs recommend --init)
    ulimits:
      core: 0             # disable core dumps (GROBID docs: --ulimit core=0)
    # On-demand only: GROBID is needed solely during reference extraction
    # (scripts/ra_grobid_backfill.py), not for normal corpus operations — start it
    # for an R-A run, stop it after to reclaim ~3.5GB. "no" = no auto-start on boot.
    # See docs/infra/grobid-jetson.md ("Run on-demand").
    restart: "no"
    # GROBID is RAM-hungry; cap usage on the constrained Jetson (Orin Nano, 8GB
    # RAM shared with Postgres). 3GB suffices for references; 4g leaves headroom.
    # Honored by `docker compose up` under Compose v2 (no swarm needed).
    deploy:
      resources:
        limits:
          memory: 4g

volumes:
  pgdata: