48c3b13c37
- Re-ran identity_bind.py to restore identity_id on face_detections - Dedup cleanup had removed rows with identity_id, kept NULL rows - 70691 face_detections now have identity_id, 428 identities - Full package rebuild: 169MB sqlite, 1358MB tar.gz - identities.json: 428 identities + 5483 bindings + 5483 trace maps - TMDB matching complete: Audrey Hepburn 843 traces, Cary Grant 482
101 lines
3.6 KiB
Bash
101 lines
3.6 KiB
Bash
#!/bin/bash
|
|
# ===========================================================
|
|
# Package Release — Standard Deliverable for M4
|
|
# Usage: bash scripts/package_release.sh <version> [file_uuid]
|
|
# version: e.g. v1.0.2
|
|
# file_uuid: default aeed71342a899fe4b4c57b7d41bcb692
|
|
# ===========================================================
|
|
set -euo pipefail
|
|
|
|
VERSION="${1:?Usage: $0 <version> [file_uuid]}"
|
|
UUID="${2:-aeed71342a899fe4b4c57b7d41bcb692}"
|
|
PG_BIN="/Users/accusys/pgsql/18.3/bin"
|
|
PROJECT="/Users/accusys/momentry_core_0.1"
|
|
OUTPUT="$PROJECT/release/phase1/${VERSION}"
|
|
M4_DIR="$PROJECT/docs_v1.0/M4_HANDOVER"
|
|
T0=$(date +%s)
|
|
|
|
echo "=========================================="
|
|
echo " Package Release ${VERSION}"
|
|
echo " UUID: ${UUID}"
|
|
echo "=========================================="
|
|
echo ""
|
|
|
|
mkdir -p "$OUTPUT"
|
|
|
|
# ── Step 1: DB backup (post-correction) ──
|
|
echo "[1/5] DB backup (post-correction)..."
|
|
$PG_BIN/pg_dump -U accusys -d momentry \
|
|
--schema=dev \
|
|
--table=dev.chunk \
|
|
--table=dev.chunk_vectors \
|
|
--data-only --column-inserts \
|
|
> "$OUTPUT/dev_backup_post_correction.sql" 2>/dev/null
|
|
echo " $(ls -lh "$OUTPUT/dev_backup_post_correction.sql" | awk '{print $5}')"
|
|
|
|
# ── Step 2: Verify counts ──
|
|
echo "[2/5] Verify counts..."
|
|
SENTENCE=$($PG_BIN/psql -U accusys -d momentry -t -A -c "SELECT count(*) FROM dev.chunk WHERE file_uuid='$UUID' AND chunk_type='sentence';" 2>/dev/null)
|
|
VECTORS=$($PG_BIN/psql -U accusys -d momentry -t -A -c "SELECT count(*) FROM dev.chunk_vectors WHERE uuid='$UUID';" 2>/dev/null)
|
|
MATCHED=$($PG_BIN/psql -U accusys -d momentry -t -A -c "
|
|
SELECT count(*) FROM dev.chunk_vectors cv
|
|
JOIN dev.chunk c ON c.file_uuid=cv.uuid AND c.chunk_id=cv.chunk_id
|
|
WHERE cv.uuid='$UUID';" 2>/dev/null)
|
|
echo " sentence chunks: $SENTENCE"
|
|
echo " vectors: $VECTORS"
|
|
echo " matched: $MATCHED"
|
|
|
|
if [ "$SENTENCE" != "$VECTORS" ] || [ "$VECTORS" != "$MATCHED" ]; then
|
|
echo " ❌ MISMATCH — aborting"
|
|
exit 1
|
|
fi
|
|
echo " ✅ All counts consistent"
|
|
|
|
# ── Step 3: Source code archive ──
|
|
echo "[3/5] Source code (git archive)..."
|
|
cd "$PROJECT"
|
|
git archive --format=tar.gz -o "$OUTPUT/momentry_core_${VERSION}_source.tar.gz" HEAD 2>/dev/null
|
|
echo " $(ls -lh "$OUTPUT/momentry_core_${VERSION}_source.tar.gz" | awk '{print $5}')"
|
|
|
|
# ── Step 4: Correction record ──
|
|
echo "[4/5] Correction record (asr-1.json)..."
|
|
ASR1="/Users/accusys/momentry/output_dev/${UUID}.asr-1.json"
|
|
if [ -f "$ASR1" ]; then
|
|
cp "$ASR1" "$OUTPUT/"
|
|
echo " $(ls -lh "$OUTPUT/${UUID}.asr-1.json" | awk '{print $5}')"
|
|
else
|
|
echo " ⚠️ No asr-1.json found for $UUID"
|
|
fi
|
|
|
|
# ── Step 5: Copy to M4_HANDOVER ──
|
|
echo "[5/5] Deploy to M4_HANDOVER..."
|
|
rm -rf "${M4_DIR:?}"/*
|
|
cp "$OUTPUT/dev_backup_post_correction.sql" "$M4_DIR/"
|
|
cp "$OUTPUT/momentry_core_${VERSION}_source.tar.gz" "$M4_DIR/"
|
|
if [ -f "$OUTPUT/${UUID}.asr-1.json" ]; then
|
|
cp "$OUTPUT/${UUID}.asr-1.json" "$M4_DIR/"
|
|
fi
|
|
# Scripts
|
|
for s in generate_asr1.py apply_asr_corrections.py clean_sentence_text.py pipeline_status.py; do
|
|
cp "$PROJECT/scripts/$s" "$M4_DIR/"
|
|
done
|
|
# Handover doc + test script
|
|
cp "$PROJECT/docs_v1.0/API_V1.0.0/RELEASE/PHASE1_HANDOVER_V1.0.0.md" "$M4_DIR/HANDOVER_${VERSION}.md"
|
|
cp "/tmp/test_api.sh" "$M4_DIR/api_test.sh"
|
|
|
|
# Create RELEASE_INFO
|
|
cat > "$M4_DIR/RELEASE_INFO.txt" << EOF
|
|
Release: ${VERSION}
|
|
Date: $(date +%Y-%m-%d)
|
|
UUID: ${UUID}
|
|
Pipeline: $(python3 $PROJECT/scripts/pipeline_status.py 2>/dev/null | grep "TOTAL" | tr -d ' ')
|
|
EOF
|
|
|
|
ELAPSED=$(($(date +%s) - T0))
|
|
echo ""
|
|
echo "=========================================="
|
|
echo " Package ${VERSION} complete (${ELAPSED}s)"
|
|
echo " Output: $OUTPUT"
|
|
echo " M4: $M4_DIR"
|
|
echo "=========================================="
|