#!/usr/bin/env bash
# Post-deploy smoke test.
#
#   ./scripts/check_deploy.sh                          # https://mcp.nearle.ai.in
#   ./scripts/check_deploy.sh http://localhost:3000    # a local container
#
#   ADMIN_PASSWORD=... ./scripts/check_deploy.sh       # also prove sign-in WORKS
#
# ADMIN_PASSWORD is read from the environment and never stored here: a committed
# smoke test must not carry a live credential. Without it the positive sign-in
# check is skipped with a WARN rather than failing.
#
# Separates the three failures that all look alike from a browser:
#   502 everywhere      - the container is not running (it exited at startup)
#   200 + database:false - the API is fine, Postgres is not
#   200 + database:true  - working
set -uo pipefail

BASE="${1:-https://mcp.nearle.ai.in}"
pass=0; fail=0

hit() { curl -sS -m 20 -o /tmp/_body -w '%{http_code}' "$BASE$1" 2>/dev/null || echo 000; }

check() { # path  expected  label
  local code; code=$(hit "$1")
  if [ "$code" = "$2" ]; then printf '  \033[32mPASS\033[0m  %-16s %s\n' "$1" "$3"; pass=$((pass+1))
  else printf '  \033[31mFAIL\033[0m  %-16s expected %s, got %s\n' "$1" "$2" "$code"; fail=$((fail+1)); fi
}

echo "Checking $BASE"
echo
echo "Is the container running at all?"
code=$(hit /)
if [ "$code" = "502" ] || [ "$code" = "000" ]; then
  echo "  FAIL  / -> $code"
  echo
  echo "  The container is not serving. It almost certainly exited at startup."
  echo "  Read the Dokploy logs; settings.py names the missing value explicitly."
  echo "  Confirm the image actually contains /app/.env - the build log must show"
  echo "  a 'COPY .env .' step, and .dockerignore must not list .env."
  exit 1
fi
printf '  \033[32mPASS\033[0m  %-16s container is up\n' "/"
pass=$((pass+1))

echo
echo "Routes that must not depend on the database:"
check /docs         200 "interactive API docs"
check /openapi.json 200 "OpenAPI schema"
check /api/health   200 "health endpoint answers"

echo
echo "Dependencies (reported in the body; 'false' does not mean the API is broken):"
health=$(curl -sS -m 20 "$BASE/api/health" 2>/dev/null)
db=$(printf '%s' "$health" | sed -n 's/.*"database":\([a-z]*\).*/\1/p')
ol=$(printf '%s' "$health" | sed -n 's/.*"ollama":\([a-z]*\).*/\1/p')
if [ "$db" = "true" ]; then printf '  \033[32mPASS\033[0m  database        connected\n'; pass=$((pass+1))
else
  printf '  \033[33mWARN\033[0m  database        NOT connected\n'
  echo "        The API works; catalog pages will be empty. Check DB_HOST/DB_USER/"
  echo "        DB_PASSWORD/DB_NAME. A wrong password logs 'password authentication"
  echo "        failed' rather than a timeout."
fi
[ "$ol" = "true" ] \
  && printf '  \033[32mPASS\033[0m  ollama          connected\n' \
  || printf '  \033[33mWARN\033[0m  ollama          not connected (/api/chat unavailable; expected if USE_OLLAMA=false)\n'

echo
echo "Auth:"
code=$(curl -sS -m 20 -o /dev/null -w '%{http_code}' -X POST "$BASE/api/auth/login" \
        -H 'Content-Type: application/json' -d '{"username":"admin","password":"definitely-not-the-password"}' 2>/dev/null)
if [ "$code" = "401" ]; then printf '  \033[32mPASS\033[0m  wrong password rejected (401)\n'; pass=$((pass+1))
elif [ "$code" = "200" ]; then
  printf '  \033[31mFAIL\033[0m  a WRONG PASSWORD WAS ACCEPTED - AUTH_ALLOW_ANY_LOGIN is true.\n'
  echo "        Anyone who finds this host can sign in as admin. Set it to false."
  fail=$((fail+1))
else printf '  \033[31mFAIL\033[0m  login endpoint returned %s\n' "$code"; fail=$((fail+1)); fi

# The complementary half, and the one that actually catches a bad deploy. The
# check above passes just as happily when NOBODY can sign in - which is exactly
# the outage this script failed to notice: prod rejected the correct password
# while every check here stayed green. Note the negative check runs FIRST on
# purpose, since a successful login calls _clear_failures() and would otherwise
# hand the wrong-password probe a fresh throttle bucket.
if [ -n "${ADMIN_PASSWORD:-}" ]; then
  code=$(curl -sS -m 20 -o /tmp/_login -w '%{http_code}' -X POST "$BASE/api/auth/login" \
          -H 'Content-Type: application/json' \
          -d "{\"username\":\"${ADMIN_USERNAME:-admin}\",\"password\":\"$ADMIN_PASSWORD\"}" 2>/dev/null)
  if [ "$code" = "200" ] && grep -q access_token /tmp/_login 2>/dev/null; then
    printf '  \033[32mPASS\033[0m  correct password accepted (200 + token)\n'; pass=$((pass+1))
  elif [ "$code" = "429" ]; then
    printf '  \033[33mWARN\033[0m  throttled (429) by the earlier failed attempt, not a\n'
    echo "        credential problem. Wait AUTH_LOCKOUT_SECONDS and rerun."
  elif [ "$code" = "401" ]; then
    printf '  \033[31mFAIL\033[0m  the CORRECT password was rejected (401).\n'
    echo "        This deployment is not carrying the credential you think it is."
    echo "        The fingerprint check below says which way; the server log names"
    echo "        the reason - grep it for 'reason=' and for 'Auth config:'."
    fail=$((fail+1))
  else
    printf '  \033[31mFAIL\033[0m  login with the correct password returned %s\n' "$code"; fail=$((fail+1))
  fi
  rm -f /tmp/_login
else
  printf '  \033[33mWARN\033[0m  sign-in not proven to WORK - set ADMIN_PASSWORD to check that.\n'
  echo "        Rejecting a wrong password is only half the test; an image with a"
  echo "        stale hash passes that half while nobody can sign in."
fi

# Which credential is this deployment actually running? The fingerprint is a
# digest of the configured hash, so comparing it against the local file settles
# "is my config live?" without either side revealing a secret.
fp=$(printf '%s' "$health" | sed -n 's/.*"password_hash_fingerprint":"\([a-z0-9]*\)".*/\1/p')
hash_ok=$(printf '%s' "$health" | sed -n 's/.*"password_hash_valid":\([a-z]*\).*/\1/p')
hsrc=$(printf '%s' "$health" | sed -n 's/.*"password_hash_source":"\([a-z-]*\)".*/\1/p')
if [ -z "$fp" ]; then
  printf '  \033[33mWARN\033[0m  no auth diagnostics in /api/health - this deployment predates\n'
  echo "        them and needs a rebuild before it can be compared."
else
  printf '  \033[32mPASS\033[0m  credential fingerprint %s (from %s)\n' "$fp" "$hsrc"; pass=$((pass+1))
  if [ "$hash_ok" = "false" ]; then
    printf '  \033[31mFAIL\033[0m  the configured password hash does not parse. NO password can\n'
    echo "        match it. Regenerate: python scripts/make_auth_secrets.py"
    fail=$((fail+1))
  fi
  local_env="$(dirname "$0")/../.env.production"
  # Probe interpreters by RUNNING one, not with `command -v`: on Windows,
  # /c/.../WindowsApps/python3 is an App Store stub that resolves fine, prints a
  # "Python was not found" notice and exits 0. Trusting the lookup made this
  # comparison skip silently - which reads as "checked, matches", the exact kind
  # of quiet pass this script exists to eliminate.
  py=""
  for candidate in python3 python py; do
    if [ "$("$candidate" -c 'print(1)' 2>/dev/null)" = "1" ]; then py="$candidate"; break; fi
  done
  if [ ! -f "$local_env" ]; then
    :
  elif [ -z "$py" ]; then
    printf '  \033[33mWARN\033[0m  no working python on PATH - cannot compare the deployment\n'
    echo "        against local .env.production."
  else
    local_fp=$("$py" "$(dirname "$0")/make_auth_secrets.py" --fingerprint "$local_env" 2>/dev/null | sed -n 's/^fingerprint *: *//p')
    if [ -z "$local_fp" ]; then
      printf '  \033[33mWARN\033[0m  could not read a fingerprint out of %s\n' "$local_env"
    elif [ "$local_fp" = "$fp" ]; then
      printf '  \033[32mPASS\033[0m  matches local .env.production\n'; pass=$((pass+1))
    else
      printf '  \033[31mFAIL\033[0m  local .env.production is %s, deployment is %s\n' "$local_fp" "$fp"
      if [ "$hsrc" = "process-env" ]; then
        echo "        It came from the process environment, which OVERRIDES the .env"
        echo "        baked into the image. Clear AUTH_ADMIN_PASSWORD_HASH from the"
        echo "        deployment platform's Environment tab."
      else
        echo "        It came from the image's own .env, so that image was built from a"
        echo "        different .env.production. This needs a REBUILD - the Dockerfile"
        echo "        copies the file at BUILD time, so a restart will not pick it up."
      fi
      fail=$((fail+1))
    fi
  fi
fi

echo
echo "MCP:"
code=$(curl -sS -m 20 -o /dev/null -w '%{http_code}' "$BASE/api/mcp/info" 2>/dev/null)
[ "$code" = "401" ] \
  && { printf '  \033[32mPASS\033[0m  /api/mcp/info guarded (401 without a token)\n'; pass=$((pass+1)); } \
  || printf '  \033[33mWARN\033[0m  /api/mcp/info returned %s (expected 401)\n' "$code"

echo
echo "-------- $pass passed, $fail failed --------"
[ "$fail" -eq 0 ]