FInish the process

.gitignore

@@ -22,10 +22,3 @@
 # virtual machine crash logs, see http://www.java.com/en/download/help/error_hotspot.xml
 hs_err_pid*
 replay_pid*
-
-# irrelevant files
-*.pdf
-*.html
-
-# data
-/demo/neo4j2graphar/

.gitmodules

@@ -0,0 +1,3 @@
+[submodule "demo/Neo4j2GraphAr/GraphAr"]
+	path = demo/Neo4j2GraphAr/GraphAr
+	url = https://github.com/alibaba/GraphAr.git

README.md

@@ -1 +1,58 @@
 # data-migration
+
+![导入流程](process.png)
+
+## 利用spark将数据从neo4j导出到GraphAr
+
+### 环境搭建
+
+- 操作系统: `Linux`, 已在`centos7`和`ubuntu20`上进行测试
+- 安装`git`, `curl`, `java-11`, `maven`, 配置JAVA_HOME路径, maven配置文件加入mirror镜像
+
+### 软件安装
+
+此处将软件安装在`HOME`路径, 并添加临时环境变量, 具体可根据实际需要调整, 软件版本也可以按需调整
+
+#### spark3.2.4(已有可忽略)
+
+```bash
+curl https://mirrors.tuna.tsinghua.edu.cn/apache/spark/spark-3.2.4/spark-3.2.4-bin-hadoop3.2.tgz | tar -xz -C ${HOME}/
+export SPARK_HOME="${HOME}/spark-3.2.4-bin-hadoop3.2"  
+export PATH="${SPARK_HOME}/bin":"${PATH}"
+```
+
+#### neo4j-community-4.4.23(已有可忽略)
+
+```bash
+curl https://dist.neo4j.org/neo4j-community-4.4.23-unix.tar.gz | tar -xz -C ${HOME}/
+export NEO4J_HOME="${HOME}/neo4j-community-4.4.23"
+export PATH="${NEO4J_HOME}/bin":"${PATH}"
+```
+
+#### GraphAr spark
+
+```bash
+git clone https://github.com/alibaba/GraphAr.git
+cd GraphAr/spark
+mvn clean package -DskipTests
+```
+
+### 使用说明
+
+1. 参考`GraphAr/spark/import/neo4j.json`编写导出信息, 包含生成gar的配置, neo4j连接的配置, 导出schema的配置，按字段填写
+
+2. `./neo4j.sh neo4j.json`
+
+## GraphAr数据导入tugraph-db
+
+### 导入说明
+
+1. 使用lgraph_import工具, 设置配置文件为gar数据中的XXX.graph.yml, 注意为绝对路径, 设置gar=true即可
+2. 示例
+   `./lgraph_import -c /root/movie-gar/movie.graph.yml --gar true`
+
+### 注意
+
+1. GraphAr数据中的yml配置文件可以修改, 比如可以在graph.yml中删除指定的vertex或者edge, 也可以在对应的vertex.yml用prefix指定数据路径，数据存储路径可以与配置存储路径分离。
+2. GraphAr支持相同的edge_label可以具有不同的properties, 这类数据在tugraph-db中不支持, 导入过程中会报错指出对应edge_label, 可以根据报错信息在GraphAr数据中做修改
+

demo/Neo4j2GraphAr/README.md

@@ -0,0 +1,27 @@
+# 默认示例运行说明
+
+GraphAr提供了一些shell脚本, 可以安装neo4j, spark, 以及运行导入程序
+
+下面是一个运行示例
+
+```bash
+cd GraphAr/spark
+scripts/get-spark-to-home.sh
+export SPARK_HOME="${HOME}/spark-3.2.2-bin-hadoop3.2"
+export PATH="${SPARK_HOME}/bin":"${PATH}"
+
+scripts/get-neo4j-to-home.sh
+export NEO4J_HOME="${HOME}/neo4j-community-4.4.23"
+export PATH="${NEO4J_HOME}/bin":"${PATH}"
+neo4j-admin set-initial-password neo4j
+
+scripts/deploy-neo4j-movie-data.sh
+
+scripts/build.sh
+
+export NEO4J_USR="neo4j"
+export NEO4J_PWD="neo4j"
+
+cd import
+./neo4j.sh neo4j.json
+```

demo/ShowGraphAr/.gitignore

@@ -0,0 +1 @@
+/build/

demo/ShowGraphAr/CMakeLists.txt

@@ -1,14 +1,23 @@
 cmake_minimum_required(VERSION 3.15)
 
-set(GAR_MAJOR_VERSION 0)
-set(GAR_MINOR_VERSION 1)
-set(GAR_PATCH_VERSION 0)
-set(GAR_VERSION ${GAR_MAJOR_VERSION}.${GAR_MINOR_VERSION}.${GAR_PATCH_VERSION})
-
-project(graph-archive-demo LANGUAGES C CXX VERSION ${GAR_VERSION})
+project(graph-archive-demo)
 
+find_package(Threads REQUIRED)
 add_definitions(-DGAR_NAMESPACE=GraphArchive)
 set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -std=c++17 -Wall")
 
-add_executable(show show.cpp)
-target_link_libraries(show PRIVATE gar stdc++fs)
+set(TARGET "show")
+
+find_package(Arrow REQUIRED)
+find_package(ArrowDataset REQUIRED)
+find_package(ArrowAcero REQUIRED)
+find_package(Parquet REQUIRED)
+find_package(gar REQUIRED)
+find_package(Threads REQUIRED)
+
+add_executable(${TARGET} ${TARGET}.cpp)
+target_link_libraries(${TARGET} PRIVATE gar stdc++fs Threads::Threads 
+    Arrow::arrow_static
+    Parquet::parquet_static
+    ArrowDataset::arrow_dataset_static
+    ArrowAcero::arrow_acero_static)