#!/bin/bash
# First ingestion on the production node (run ON the node, from ~/apps/satelliteindex). Idempotent.
# CelesTrak refuses a second download of the same GP group within 2 h (HTTP 403 "has not updated") — the connector treats it as
# unchanged. If a local snapshot exists in ~/satelliteindex-data/seed/ (active.json, satcat.csv) it is used for the very first run.
set -euo pipefail
cd "$(dirname "$0")/.."
export PATH="/opt/homebrew/opt/postgresql@17/bin:/opt/homebrew/bin:$PATH"
export DATABASE_URL=${DATABASE_URL:-postgresql+asyncpg://satelliteindex:satelliteindex@127.0.0.1:5432/satelliteindex}
export REDIS_URL=${REDIS_URL:-redis://127.0.0.1:6379/4}
export SI_DATA_DIR=${SI_DATA_DIR:-$HOME/satelliteindex-data}
export SI_LOG_JSON=0
SI=.venv/bin/si
SEED=$SI_DATA_DIR/seed

$SI migrate | tail -1
$SI seed | tail -1
n=$(psql "postgresql://satelliteindex:satelliteindex@127.0.0.1:5432/satelliteindex" -Atc "select count(*) from satellites")
echo "satellites in db: $n"
if [ "$n" -lt 1000 ] && [ -s "$SEED/satcat.csv" ]; then
  echo "seeding SATCAT from local snapshot"; $SI run celestrak_satcat --file satcat.csv="$SEED/satcat.csv" || true
fi
$SI run celestrak_satcat || true
gp=$(psql "postgresql://satelliteindex:satelliteindex@127.0.0.1:5432/satelliteindex" -Atc "select count(*) from orbital_state")
if [ "$gp" -lt 1000 ] && [ -s "$SEED/active.json" ]; then
  echo "seeding GP from local snapshot"; $SI run celestrak_gp --file active="$SEED/active.json" || true
fi
$SI run celestrak_gp || true
$SI run celestrak_groups || true
$SI run derived_analytics
$SI stats