kgrag/suzhu-kgrag_migration/01_copy_files.sh
2026-06-30 13:35:52 +08:00

81 lines
2.6 KiB
Bash
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

#!/bin/bash
# =============================================================================
# 步骤 1从原数据机拉取所有文件到迁移目标机全量版
# 在【迁移目标机】上执行本脚本
# 配置项请修改 migration_config.sh
# =============================================================================
SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"
source "${SCRIPT_DIR}/migration_config.sh"
set -e
echo "=========================================="
echo " kgrag 文件迁移脚本 - 步骤 1全量"
echo " 原数据机: ${SRC_SERVER}:${SRC_APP_DIR}"
echo " 目标代码目录: ${DST_CODE_DIR}"
echo " 目标文件目录: ${DST_FILES_DIR}"
echo "=========================================="
mkdir -p "${DST_CODE_DIR}"
mkdir -p "${DST_FILES_DIR}/uploads_v1"
mkdir -p "${DST_CODE_DIR}/kg_output"
mkdir -p "${DST_CODE_DIR}/fault_wenjian"
echo ""
echo "[1/6] 迁移原始文档目录 → ${DST_FILES_DIR}/ ..."
rsync -avz --progress ${RSYNC_EXTRA} \
-e "ssh -p ${SRC_SSH_PORT}" \
--exclude="uploads_v1/" \
"${SRC_SERVER}:${SRC_APP_DIR}/files/" \
"${DST_FILES_DIR}/"
echo ""
echo "[2/6] 迁移高亮注释 PDF → ${DST_FILES_DIR}/uploads_v1/ ..."
rsync -avz --progress ${RSYNC_EXTRA} \
-e "ssh -p ${SRC_SSH_PORT}" \
"${SRC_SERVER}:${SRC_APP_DIR}/files/uploads_v1/" \
"${DST_FILES_DIR}/uploads_v1/"
echo ""
echo "[3/6] 迁移抽取结果 JSON → ${DST_CODE_DIR}/kg_output/ ..."
rsync -avz --progress ${RSYNC_EXTRA} \
-e "ssh -p ${SRC_SSH_PORT}" \
"${SRC_SERVER}:${SRC_APP_DIR}/kg_output/" \
"${DST_CODE_DIR}/kg_output/"
echo ""
echo "[4/6] 迁移 kg_snapshot.json → ${DST_CODE_DIR}/ ..."
rsync -avz --progress ${RSYNC_EXTRA} \
-e "ssh -p ${SRC_SSH_PORT}" \
"${SRC_SERVER}:${SRC_APP_DIR}/kg_snapshot.json" \
"${DST_CODE_DIR}/kg_snapshot.json"
echo ""
echo "[5/6] 迁移 fault_wenjian/ → ${DST_CODE_DIR}/fault_wenjian/ ..."
rsync -avz --progress ${RSYNC_EXTRA} \
-e "ssh -p ${SRC_SSH_PORT}" \
"${SRC_SERVER}:${SRC_APP_DIR}/fault_wenjian/" \
"${DST_CODE_DIR}/fault_wenjian/"
echo ""
echo "[6/6] 迁移代码文件 → ${DST_CODE_DIR}/ ..."
rsync -avz --progress ${RSYNC_EXTRA} \
-e "ssh -p ${SRC_SSH_PORT}" \
--exclude="__pycache__/" \
--exclude="*.pyc" \
--exclude="*.log" \
--exclude="files/" \
--exclude="kg_output/" \
--exclude="kg_snapshot.json" \
--exclude="fault_wenjian/" \
--exclude="mineru_output/" \
"${SRC_SERVER}:${SRC_APP_DIR}/" \
"${DST_CODE_DIR}/"
echo ""
echo "=========================================="
echo " 文件迁移完成!"
echo " 下一步:执行 bash 02_load_neo4j.sh"
echo "=========================================="