mirror of
https://github.com/alibaba/DataX.git
synced 2025-05-03 01:30:50 +08:00
gfbfabcd123
GitHub
commit
ec7bfc0aa4
31
README.md
31
README.md
@ -26,7 +26,7 @@ DataX本身作为数据同步框架，将不同数据源的同步抽象为从源
|
|||||||
|
|
||||||
# Quick Start
|
# Quick Start
|
||||||
|
|
||||||
##### Download [DataX下载地址](https://datax-opensource.oss-cn-hangzhou.aliyuncs.com/202303/datax.tar.gz)
|
##### Download [DataX下载地址](https://datax-opensource.oss-cn-hangzhou.aliyuncs.com/202308/datax.tar.gz)
|
||||||
|
|
||||||
|
|
||||||
##### 请点击：[Quick Start](https://github.com/alibaba/DataX/blob/master/userGuid.md)
|
##### 请点击：[Quick Start](https://github.com/alibaba/DataX/blob/master/userGuid.md)
|
||||||
@ -56,7 +56,8 @@ DataX目前已经有了比较全面的插件体系，主流的RDBMS数据库、N
|
|||||||
| | AnalyticDB For PostgreSQL | | √ | 写 |
|
| | AnalyticDB For PostgreSQL | | √ | 写 |
|
||||||
| 阿里云中间件 | datahub | √ | √ | 读 、写 |
|
| 阿里云中间件 | datahub | √ | √ | 读 、写 |
|
||||||
| | SLS | √ | √ | 读 、写 |
|
| | SLS | √ | √ | 读 、写 |
|
||||||
| 阿里云图数据库 | GDB | √ | √ | [读](https://github.com/alibaba/DataX/blob/master/gdbreader/doc/gdbreader.md) 、[写](https://github.com/alibaba/DataX/blob/master/gdbwriter/doc/gdbwriter.md) |
|
| 图数据库 | 阿里云 GDB | √ | √ | [读](https://github.com/alibaba/DataX/blob/master/gdbreader/doc/gdbreader.md) 、[写](https://github.com/alibaba/DataX/blob/master/gdbwriter/doc/gdbwriter.md) |
|
||||||
|
| | Neo4j | | √ | [写](https://github.com/alibaba/DataX/blob/master/neo4jwriter/doc/neo4jwriter.md) |
|
||||||
| NoSQL数据存储 | OTS | √ | √ | [读](https://github.com/alibaba/DataX/blob/master/otsreader/doc/otsreader.md) 、[写](https://github.com/alibaba/DataX/blob/master/otswriter/doc/otswriter.md) |
|
| NoSQL数据存储 | OTS | √ | √ | [读](https://github.com/alibaba/DataX/blob/master/otsreader/doc/otsreader.md) 、[写](https://github.com/alibaba/DataX/blob/master/otswriter/doc/otswriter.md) |
|
||||||
| | Hbase0.94 | √ | √ | [读](https://github.com/alibaba/DataX/blob/master/hbase094xreader/doc/hbase094xreader.md) 、[写](https://github.com/alibaba/DataX/blob/master/hbase094xwriter/doc/hbase094xwriter.md) |
|
| | Hbase0.94 | √ | √ | [读](https://github.com/alibaba/DataX/blob/master/hbase094xreader/doc/hbase094xreader.md) 、[写](https://github.com/alibaba/DataX/blob/master/hbase094xwriter/doc/hbase094xwriter.md) |
|
||||||
| | Hbase1.1 | √ | √ | [读](https://github.com/alibaba/DataX/blob/master/hbase11xreader/doc/hbase11xreader.md) 、[写](https://github.com/alibaba/DataX/blob/master/hbase11xwriter/doc/hbase11xwriter.md) |
|
| | Hbase1.1 | √ | √ | [读](https://github.com/alibaba/DataX/blob/master/hbase11xreader/doc/hbase11xreader.md) 、[写](https://github.com/alibaba/DataX/blob/master/hbase11xwriter/doc/hbase11xwriter.md) |
|
||||||
@ -66,7 +67,7 @@ DataX目前已经有了比较全面的插件体系，主流的RDBMS数据库、N
|
|||||||
| | Cassandra | √ | √ | [读](https://github.com/alibaba/DataX/blob/master/cassandrareader/doc/cassandrareader.md) 、[写](https://github.com/alibaba/DataX/blob/master/cassandrawriter/doc/cassandrawriter.md) |
|
| | Cassandra | √ | √ | [读](https://github.com/alibaba/DataX/blob/master/cassandrareader/doc/cassandrareader.md) 、[写](https://github.com/alibaba/DataX/blob/master/cassandrawriter/doc/cassandrawriter.md) |
|
||||||
| 数仓数据存储 | StarRocks | √ | √ | 读 、[写](https://github.com/alibaba/DataX/blob/master/starrockswriter/doc/starrockswriter.md) |
|
| 数仓数据存储 | StarRocks | √ | √ | 读 、[写](https://github.com/alibaba/DataX/blob/master/starrockswriter/doc/starrockswriter.md) |
|
||||||
| | ApacheDoris | | √ | [写](https://github.com/alibaba/DataX/blob/master/doriswriter/doc/doriswriter.md) |
|
| | ApacheDoris | | √ | [写](https://github.com/alibaba/DataX/blob/master/doriswriter/doc/doriswriter.md) |
|
||||||
| | ClickHouse | | √ | 写 |
|
| | ClickHouse | √ | √ | [读](https://github.com/alibaba/DataX/blob/master/clickhousereader/doc/clickhousereader.md) 、[写](https://github.com/alibaba/DataX/blob/master/clickhousewriter/doc/clickhousewriter.md) |
|
||||||
| | Databend | | √ | [写](https://github.com/alibaba/DataX/blob/master/databendwriter/doc/databendwriter.md) |
|
| | Databend | | √ | [写](https://github.com/alibaba/DataX/blob/master/databendwriter/doc/databendwriter.md) |
|
||||||
| | Hive | √ | √ | [读](https://github.com/alibaba/DataX/blob/master/hdfsreader/doc/hdfsreader.md) 、[写](https://github.com/alibaba/DataX/blob/master/hdfswriter/doc/hdfswriter.md) |
|
| | Hive | √ | √ | [读](https://github.com/alibaba/DataX/blob/master/hdfsreader/doc/hdfsreader.md) 、[写](https://github.com/alibaba/DataX/blob/master/hdfswriter/doc/hdfswriter.md) |
|
||||||
| | kudu | | √ | [写](https://github.com/alibaba/DataX/blob/master/hdfswriter/doc/hdfswriter.md) |
|
| | kudu | | √ | [写](https://github.com/alibaba/DataX/blob/master/hdfswriter/doc/hdfswriter.md) |
|
||||||
@ -99,7 +100,7 @@ DataX目前已经有了比较全面的插件体系，主流的RDBMS数据库、N
|
|||||||
- 整库迁移：https://help.aliyun.com/document_detail/137809.html
|
- 整库迁移：https://help.aliyun.com/document_detail/137809.html
|
||||||
- 批量上云：https://help.aliyun.com/document_detail/146671.html
|
- 批量上云：https://help.aliyun.com/document_detail/146671.html
|
||||||
- 更新更多能力请访问：https://help.aliyun.com/document_detail/137663.html
|
- 更新更多能力请访问：https://help.aliyun.com/document_detail/137663.html
|
||||||
|
-
|
||||||
|
|
||||||
# 我要开发新的插件
|
# 我要开发新的插件
|
||||||
|
|
||||||
@ -109,6 +110,28 @@ DataX目前已经有了比较全面的插件体系，主流的RDBMS数据库、N
|
|||||||
|
|
||||||
DataX 后续计划月度迭代更新，也欢迎感兴趣的同学提交 Pull requests，月度更新内容会介绍介绍如下。
|
DataX 后续计划月度迭代更新，也欢迎感兴趣的同学提交 Pull requests，月度更新内容会介绍介绍如下。
|
||||||
|
|
||||||
|
- [datax_v202309]（https://github.com/alibaba/DataX/releases/tag/datax_v202309)
|
||||||
|
- 支持Phoenix 同步数据添加 where条件
|
||||||
|
- 支持华为 GuassDB读写插件
|
||||||
|
- 修复ClickReader 插件运行报错 Can't find bundle for base name
|
||||||
|
- 增加 DataX调试模块
|
||||||
|
- 修复 orc空文件报错问题
|
||||||
|
- 优化obwriter性能
|
||||||
|
- txtfilewriter 增加导出为insert语句功能支持
|
||||||
|
- HdfsReader/HdfsWriter 支持parquet读写能力
|
||||||
|
|
||||||
|
- [datax_v202308]（https://github.com/alibaba/DataX/releases/tag/datax_v202308)
|
||||||
|
- OTS 插件更新
|
||||||
|
- databend 插件更新
|
||||||
|
- Oceanbase驱动修复
|
||||||
|
|
||||||
|
|
||||||
|
- [datax_v202306]（https://github.com/alibaba/DataX/releases/tag/datax_v202306)
|
||||||
|
- 精简代码
|
||||||
|
- 新增插件（neo4jwriter、clickhousewriter）
|
||||||
|
- 优化插件、修复问题（oceanbase、hdfs、databend、txtfile）
|
||||||
|
|
||||||
|
|
||||||
- [datax_v202303]（https://github.com/alibaba/DataX/releases/tag/datax_v202303)
|
- [datax_v202303]（https://github.com/alibaba/DataX/releases/tag/datax_v202303)
|
||||||
- 精简代码
|
- 精简代码
|
||||||
- 新增插件（adbmysqlwriter、databendwriter、selectdbwriter）
|
- 新增插件（adbmysqlwriter、databendwriter、selectdbwriter）
|
||||||
|
|||||||
344
clickhousereader/doc/clickhousereader.md
Normal file
344
clickhousereader/doc/clickhousereader.md
Normal file
@ -0,0 +1,344 @@
|
|||||||
|
|
||||||
|
# ClickhouseReader 插件文档
|
||||||
|
|
||||||
|
|
||||||
|
___
|
||||||
|
|
||||||
|
|
||||||
|
## 1 快速介绍
|
||||||
|
|
||||||
|
ClickhouseReader插件实现了从Clickhouse读取数据。在底层实现上，ClickhouseReader通过JDBC连接远程Clickhouse数据库，并执行相应的sql语句将数据从Clickhouse库中SELECT出来。
|
||||||
|
|
||||||
|
## 2 实现原理
|
||||||
|
|
||||||
|
简而言之，ClickhouseReader通过JDBC连接器连接到远程的Clickhouse数据库，并根据用户配置的信息生成查询SELECT SQL语句并发送到远程Clickhouse数据库，并将该SQL执行返回结果使用DataX自定义的数据类型拼装为抽象的数据集，并传递给下游Writer处理。
|
||||||
|
|
||||||
|
对于用户配置Table、Column、Where的信息，ClickhouseReader将其拼接为SQL语句发送到Clickhouse数据库；对于用户配置querySql信息，Clickhouse直接将其发送到Clickhouse数据库。
|
||||||
|
|
||||||
|
|
||||||
|
## 3 功能说明
|
||||||
|
|
||||||
|
### 3.1 配置样例
|
||||||
|
|
||||||
|
* 配置一个从Clickhouse数据库同步抽取数据到本地的作业:
|
||||||
|
|
||||||
|
```
|
||||||
|
{
|
||||||
|
"job": {
|
||||||
|
"setting": {
|
||||||
|
"speed": {
|
||||||
|
//设置传输速度 byte/s 尽量逼近这个速度但是不高于它.
|
||||||
|
// channel 表示通道数量，byte表示通道速度，如果单通道速度1MB，配置byte为1048576表示一个channel
|
||||||
|
"byte": 1048576
|
||||||
|
},
|
||||||
|
//出错限制
|
||||||
|
"errorLimit": {
|
||||||
|
//先选择record
|
||||||
|
"record": 0,
|
||||||
|
//百分比 1表示100%
|
||||||
|
"percentage": 0.02
|
||||||
|
}
|
||||||
|
},
|
||||||
|
"content": [
|
||||||
|
{
|
||||||
|
"reader": {
|
||||||
|
"name": "clickhousereader",
|
||||||
|
"parameter": {
|
||||||
|
// 数据库连接用户名
|
||||||
|
"username": "root",
|
||||||
|
// 数据库连接密码
|
||||||
|
"password": "root",
|
||||||
|
"column": [
|
||||||
|
"id","name"
|
||||||
|
],
|
||||||
|
"connection": [
|
||||||
|
{
|
||||||
|
"table": [
|
||||||
|
"table"
|
||||||
|
],
|
||||||
|
"jdbcUrl": [
|
||||||
|
"jdbc:clickhouse://[HOST_NAME]:PORT/[DATABASE_NAME]"
|
||||||
|
]
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
|
},
|
||||||
|
"writer": {
|
||||||
|
//writer类型
|
||||||
|
"name": "streamwriter",
|
||||||
|
// 是否打印内容
|
||||||
|
"parameter": {
|
||||||
|
"print": true
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
```
|
||||||
|
|
||||||
|
* 配置一个自定义SQL的数据库同步任务到本地内容的作业：
|
||||||
|
|
||||||
|
```
|
||||||
|
{
|
||||||
|
"job": {
|
||||||
|
"setting": {
|
||||||
|
"speed": {
|
||||||
|
"channel": 5
|
||||||
|
}
|
||||||
|
},
|
||||||
|
"content": [
|
||||||
|
{
|
||||||
|
"reader": {
|
||||||
|
"name": "clickhousereader",
|
||||||
|
"parameter": {
|
||||||
|
"username": "root",
|
||||||
|
"password": "root",
|
||||||
|
"where": "",
|
||||||
|
"connection": [
|
||||||
|
{
|
||||||
|
"querySql": [
|
||||||
|
"select db_id,on_line_flag from db_info where db_id < 10"
|
||||||
|
],
|
||||||
|
"jdbcUrl": [
|
||||||
|
"jdbc:clickhouse://1.1.1.1:8123/default"
|
||||||
|
]
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
|
},
|
||||||
|
"writer": {
|
||||||
|
"name": "streamwriter",
|
||||||
|
"parameter": {
|
||||||
|
"visible": false,
|
||||||
|
"encoding": "UTF-8"
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
|
}
|
||||||
|
```
|
||||||
|
|
||||||
|
|
||||||
|
### 3.2 参数说明
|
||||||
|
|
||||||
|
* **jdbcUrl**
|
||||||
|
|
||||||
|
* 描述：描述的是到对端数据库的JDBC连接信息，使用JSON的数组描述，并支持一个库填写多个连接地址。之所以使用JSON数组描述连接信息，是因为阿里集团内部支持多个IP探测，如果配置了多个，ClickhouseReader可以依次探测ip的可连接性，直到选择一个合法的IP。如果全部连接失败，ClickhouseReader报错。 注意，jdbcUrl必须包含在connection配置单元中。对于阿里集团外部使用情况，JSON数组填写一个JDBC连接即可。
|
||||||
|
|
||||||
|
jdbcUrl按照Clickhouse官方规范，并可以填写连接附件控制信息。具体请参看[Clickhouse官方文档](https://clickhouse.com/docs/en/engines/table-engines/integrations/jdbc)。
|
||||||
|
|
||||||
|
* 必选：是 <br />
|
||||||
|
|
||||||
|
* 默认值：无 <br />
|
||||||
|
|
||||||
|
* **username**
|
||||||
|
|
||||||
|
* 描述：数据源的用户名 <br />
|
||||||
|
|
||||||
|
* 必选：是 <br />
|
||||||
|
|
||||||
|
* 默认值：无 <br />
|
||||||
|
|
||||||
|
* **password**
|
||||||
|
|
||||||
|
* 描述：数据源指定用户名的密码 <br />
|
||||||
|
|
||||||
|
* 必选：是 <br />
|
||||||
|
|
||||||
|
* 默认值：无 <br />
|
||||||
|
|
||||||
|
* **table**
|
||||||
|
|
||||||
|
* 描述：所选取的需要同步的表。使用JSON的数组描述，因此支持多张表同时抽取。当配置为多张表时，用户自己需保证多张表是同一schema结构，ClickhouseReader不予检查表是否同一逻辑表。注意，table必须包含在connection配置单元中。<br />
|
||||||
|
|
||||||
|
* 必选：是 <br />
|
||||||
|
|
||||||
|
* 默认值：无 <br />
|
||||||
|
|
||||||
|
* **column**
|
||||||
|
|
||||||
|
* 描述：所配置的表中需要同步的列名集合，使用JSON的数组描述字段信息。用户使用\*代表默认使用所有列配置，例如['\*']。
|
||||||
|
|
||||||
|
支持列裁剪，即列可以挑选部分列进行导出。
|
||||||
|
|
||||||
|
支持列换序，即列可以不按照表schema信息进行导出。
|
||||||
|
|
||||||
|
支持常量配置，用户需要按照JSON格式:
|
||||||
|
["id", "`table`", "1", "'bazhen.csy'", "null", "to_char(a + 1)", "2.3" , "true"]
|
||||||
|
id为普通列名，\`table\`为包含保留在的列名，1为整形数字常量，'bazhen.csy'为字符串常量，null为空指针，to_char(a + 1)为表达式，2.3为浮点数，true为布尔值。
|
||||||
|
|
||||||
|
Column必须显示填写，不允许为空！
|
||||||
|
|
||||||
|
* 必选：是 <br />
|
||||||
|
|
||||||
|
* 默认值：无 <br />
|
||||||
|
|
||||||
|
* **splitPk**
|
||||||
|
|
||||||
|
* 描述：ClickhouseReader进行数据抽取时，如果指定splitPk，表示用户希望使用splitPk代表的字段进行数据分片，DataX因此会启动并发任务进行数据同步，这样可以大大提供数据同步的效能。
|
||||||
|
|
||||||
|
推荐splitPk用户使用表主键，因为表主键通常情况下比较均匀，因此切分出来的分片也不容易出现数据热点。
|
||||||
|
|
||||||
|
目前splitPk仅支持整形数据切分，`不支持浮点、日期等其他类型`。如果用户指定其他非支持类型，ClickhouseReader将报错！
|
||||||
|
|
||||||
|
splitPk如果不填写，将视作用户不对单表进行切分，ClickhouseReader使用单通道同步全量数据。
|
||||||
|
|
||||||
|
* 必选：否 <br />
|
||||||
|
|
||||||
|
* 默认值：无 <br />
|
||||||
|
|
||||||
|
* **where**
|
||||||
|
|
||||||
|
* 描述：筛选条件，MysqlReader根据指定的column、table、where条件拼接SQL，并根据这个SQL进行数据抽取。在实际业务场景中，往往会选择当天的数据进行同步，可以将where条件指定为gmt_create > $bizdate 。注意：不可以将where条件指定为limit 10，limit不是SQL的合法where子句。<br />
|
||||||
|
|
||||||
|
where条件可以有效地进行业务增量同步。
|
||||||
|
|
||||||
|
* 必选：否 <br />
|
||||||
|
|
||||||
|
* 默认值：无 <br />
|
||||||
|
|
||||||
|
* **querySql**
|
||||||
|
|
||||||
|
* 描述：在有些业务场景下，where这一配置项不足以描述所筛选的条件，用户可以通过该配置型来自定义筛选SQL。当用户配置了这一项之后，DataX系统就会忽略table，column这些配置型，直接使用这个配置项的内容对数据进行筛选，例如需要进行多表join后同步数据，使用select a,b from table_a join table_b on table_a.id = table_b.id <br />
|
||||||
|
|
||||||
|
`当用户配置querySql时，ClickhouseReader直接忽略table、column、where条件的配置`。
|
||||||
|
|
||||||
|
* 必选：否 <br />
|
||||||
|
|
||||||
|
* 默认值：无 <br />
|
||||||
|
|
||||||
|
* **fetchSize**
|
||||||
|
|
||||||
|
* 描述：该配置项定义了插件和数据库服务器端每次批量数据获取条数，该值决定了DataX和服务器端的网络交互次数，能够较大的提升数据抽取性能。<br />
|
||||||
|
|
||||||
|
`注意，该值过大(>2048)可能造成DataX进程OOM。`。
|
||||||
|
|
||||||
|
* 必选：否 <br />
|
||||||
|
|
||||||
|
* 默认值：1024 <br />
|
||||||
|
|
||||||
|
* **session**
|
||||||
|
|
||||||
|
* 描述：控制写入数据的时间格式，时区等的配置，如果表中有时间字段，配置该值以明确告知写入 clickhouse 的时间格式。通常配置的参数为：NLS_DATE_FORMAT,NLS_TIME_FORMAT。其配置的值为 json 格式，例如：
|
||||||
|
```
|
||||||
|
"session": [
|
||||||
|
"alter session set NLS_DATE_FORMAT='yyyy-mm-dd hh24:mi:ss'",
|
||||||
|
"alter session set NLS_TIMESTAMP_FORMAT='yyyy-mm-dd hh24:mi:ss'",
|
||||||
|
"alter session set NLS_TIMESTAMP_TZ_FORMAT='yyyy-mm-dd hh24:mi:ss'",
|
||||||
|
"alter session set TIME_ZONE='US/Pacific'"
|
||||||
|
]
|
||||||
|
```
|
||||||
|
`(注意"是 " 的转义字符串)`。
|
||||||
|
|
||||||
|
* 必选：否 <br />
|
||||||
|
|
||||||
|
* 默认值：无 <br />
|
||||||
|
|
||||||
|
|
||||||
|
### 3.3 类型转换
|
||||||
|
|
||||||
|
目前ClickhouseReader支持大部分Clickhouse类型，但也存在部分个别类型没有支持的情况，请注意检查你的类型。
|
||||||
|
|
||||||
|
下面列出ClickhouseReader针对Clickhouse类型转换列表:
|
||||||
|
|
||||||
|
|
||||||
|
| DataX 内部类型| Clickhouse 数据类型 |
|
||||||
|
| -------- |--------------------------------------------------------------------------------------------|
|
||||||
|
| Long | UInt8, UInt16, UInt32, UInt64, UInt128, UInt256, Int8, Int16, Int32, Int64, Int128, Int256 |
|
||||||
|
| Double | Float32, Float64, Decimal |
|
||||||
|
| String | String, FixedString |
|
||||||
|
| Date | DATE, Date32, DateTime, DateTime64 |
|
||||||
|
| Boolean | Boolean |
|
||||||
|
| Bytes | BLOB,BFILE,RAW,LONG RAW |
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
请注意:
|
||||||
|
|
||||||
|
* `除上述罗列字段类型外，其他类型均不支持`。
|
||||||
|
|
||||||
|
|
||||||
|
## 4 性能报告
|
||||||
|
|
||||||
|
### 4.1 环境准备
|
||||||
|
|
||||||
|
#### 4.1.1 数据特征
|
||||||
|
|
||||||
|
为了模拟线上真实数据，我们设计两个Clickhouse数据表，分别为:
|
||||||
|
|
||||||
|
#### 4.1.2 机器参数
|
||||||
|
|
||||||
|
* 执行DataX的机器参数为:
|
||||||
|
|
||||||
|
* Clickhouse数据库机器参数为:
|
||||||
|
|
||||||
|
### 4.2 测试报告
|
||||||
|
|
||||||
|
#### 4.2.1 表1测试报告
|
||||||
|
|
||||||
|
|
||||||
|
| 并发任务数| DataX速度(Rec/s)|DataX流量|网卡流量|DataX运行负载|DB运行负载|
|
||||||
|
|--------| --------|--------|--------|--------|--------|
|
||||||
|
|1| DataX 统计速度(Rec/s)|DataX统计流量|网卡流量|DataX运行负载|DB运行负载|
|
||||||
|
|
||||||
|
## 5 约束限制
|
||||||
|
|
||||||
|
### 5.1 主备同步数据恢复问题
|
||||||
|
|
||||||
|
主备同步问题指Clickhouse使用主从灾备，备库从主库不间断通过binlog恢复数据。由于主备数据同步存在一定的时间差，特别在于某些特定情况，例如网络延迟等问题，导致备库同步恢复的数据与主库有较大差别，导致从备库同步的数据不是一份当前时间的完整镜像。
|
||||||
|
|
||||||
|
针对这个问题，我们提供了preSql功能，该功能待补充。
|
||||||
|
|
||||||
|
### 5.2 一致性约束
|
||||||
|
|
||||||
|
Clickhouse在数据存储划分中属于RDBMS系统，对外可以提供强一致性数据查询接口。例如当一次同步任务启动运行过程中，当该库存在其他数据写入方写入数据时，ClickhouseReader完全不会获取到写入更新数据，这是由于数据库本身的快照特性决定的。关于数据库快照特性，请参看[MVCC Wikipedia](https://en.wikipedia.org/wiki/Multiversion_concurrency_control)
|
||||||
|
|
||||||
|
上述是在ClickhouseReader单线程模型下数据同步一致性的特性，由于ClickhouseReader可以根据用户配置信息使用了并发数据抽取，因此不能严格保证数据一致性：当ClickhouseReader根据splitPk进行数据切分后，会先后启动多个并发任务完成数据同步。由于多个并发任务相互之间不属于同一个读事务，同时多个并发任务存在时间间隔。因此这份数据并不是`完整的`、`一致的`数据快照信息。
|
||||||
|
|
||||||
|
针对多线程的一致性快照需求，在技术上目前无法实现，只能从工程角度解决，工程化的方式存在取舍，我们提供几个解决思路给用户，用户可以自行选择：
|
||||||
|
|
||||||
|
1. 使用单线程同步，即不再进行数据切片。缺点是速度比较慢，但是能够很好保证一致性。
|
||||||
|
|
||||||
|
2. 关闭其他数据写入方，保证当前数据为静态数据，例如，锁表、关闭备库同步等等。缺点是可能影响在线业务。
|
||||||
|
|
||||||
|
### 5.3 数据库编码问题
|
||||||
|
|
||||||
|
|
||||||
|
ClickhouseReader底层使用JDBC进行数据抽取，JDBC天然适配各类编码，并在底层进行了编码转换。因此ClickhouseReader不需用户指定编码，可以自动获取编码并转码。
|
||||||
|
|
||||||
|
对于Clickhouse底层写入编码和其设定的编码不一致的混乱情况，ClickhouseReader对此无法识别，对此也无法提供解决方案，对于这类情况，`导出有可能为乱码`。
|
||||||
|
|
||||||
|
### 5.4 增量数据同步
|
||||||
|
|
||||||
|
ClickhouseReader使用JDBC SELECT语句完成数据抽取工作，因此可以使用SELECT...WHERE...进行增量数据抽取，方式有多种：
|
||||||
|
|
||||||
|
* 数据库在线应用写入数据库时，填充modify字段为更改时间戳，包括新增、更新、删除(逻辑删)。对于这类应用，ClickhouseReader只需要WHERE条件跟上一同步阶段时间戳即可。
|
||||||
|
* 对于新增流水型数据，ClickhouseReader可以WHERE条件后跟上一阶段最大自增ID即可。
|
||||||
|
|
||||||
|
对于业务上无字段区分新增、修改数据情况，ClickhouseReader也无法进行增量数据同步，只能同步全量数据。
|
||||||
|
|
||||||
|
### 5.5 Sql安全性
|
||||||
|
|
||||||
|
ClickhouseReader提供querySql语句交给用户自己实现SELECT抽取语句，ClickhouseReader本身对querySql不做任何安全性校验。这块交由DataX用户方自己保证。
|
||||||
|
|
||||||
|
## 6 FAQ
|
||||||
|
|
||||||
|
***
|
||||||
|
|
||||||
|
**Q: ClickhouseReader同步报错，报错信息为XXX**
|
||||||
|
|
||||||
|
A: 网络或者权限问题，请使用Clickhouse命令行测试
|
||||||
|
|
||||||
|
|
||||||
|
如果上述命令也报错，那可以证实是环境问题，请联系你的DBA。
|
||||||
|
|
||||||
|
|
||||||
|
**Q: ClickhouseReader抽取速度很慢怎么办？**
|
||||||
|
|
||||||
|
A: 影响抽取时间的原因大概有如下几个：(来自专业 DBA 卫绾)
|
||||||
|
1. 由于SQL的plan异常，导致的抽取时间长； 在抽取时，尽可能使用全表扫描代替索引扫描;
|
||||||
|
2. 合理sql的并发度，减少抽取时间；
|
||||||
|
3. 抽取sql要简单，尽量不用replace等函数，这个非常消耗cpu，会严重影响抽取速度;
|
||||||
91
clickhousereader/pom.xml
Normal file
91
clickhousereader/pom.xml
Normal file
@ -0,0 +1,91 @@
|
|||||||
|
<?xml version="1.0" encoding="UTF-8"?>
|
||||||
|
<project xmlns="http://maven.apache.org/POM/4.0.0"
|
||||||
|
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
|
||||||
|
xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
|
||||||
|
<parent>
|
||||||
|
<artifactId>datax-all</artifactId>
|
||||||
|
<groupId>com.alibaba.datax</groupId>
|
||||||
|
<version>0.0.1-SNAPSHOT</version>
|
||||||
|
</parent>
|
||||||
|
|
||||||
|
<modelVersion>4.0.0</modelVersion>
|
||||||
|
<artifactId>clickhousereader</artifactId>
|
||||||
|
<name>clickhousereader</name>
|
||||||
|
<packaging>jar</packaging>
|
||||||
|
|
||||||
|
<dependencies>
|
||||||
|
<dependency>
|
||||||
|
<groupId>ru.yandex.clickhouse</groupId>
|
||||||
|
<artifactId>clickhouse-jdbc</artifactId>
|
||||||
|
<version>0.2.4</version>
|
||||||
|
</dependency>
|
||||||
|
<dependency>
|
||||||
|
<groupId>com.alibaba.datax</groupId>
|
||||||
|
<artifactId>datax-core</artifactId>
|
||||||
|
<version>${datax-project-version}</version>
|
||||||
|
</dependency>
|
||||||
|
<dependency>
|
||||||
|
<groupId>com.alibaba.datax</groupId>
|
||||||
|
<artifactId>datax-common</artifactId>
|
||||||
|
<version>${datax-project-version}</version>
|
||||||
|
</dependency>
|
||||||
|
<dependency>
|
||||||
|
<groupId>org.slf4j</groupId>
|
||||||
|
<artifactId>slf4j-api</artifactId>
|
||||||
|
</dependency>
|
||||||
|
|
||||||
|
<dependency>
|
||||||
|
<groupId>ch.qos.logback</groupId>
|
||||||
|
<artifactId>logback-classic</artifactId>
|
||||||
|
</dependency>
|
||||||
|
|
||||||
|
<dependency>
|
||||||
|
<groupId>com.alibaba.datax</groupId>
|
||||||
|
<artifactId>plugin-rdbms-util</artifactId>
|
||||||
|
<version>${datax-project-version}</version>
|
||||||
|
</dependency>
|
||||||
|
</dependencies>
|
||||||
|
|
||||||
|
<build>
|
||||||
|
<resources>
|
||||||
|
<resource>
|
||||||
|
<directory>src/main/java</directory>
|
||||||
|
<includes>
|
||||||
|
<include>**/*.properties</include>
|
||||||
|
</includes>
|
||||||
|
</resource>
|
||||||
|
</resources>
|
||||||
|
<plugins>
|
||||||
|
<!-- compiler plugin -->
|
||||||
|
<plugin>
|
||||||
|
<artifactId>maven-compiler-plugin</artifactId>
|
||||||
|
<configuration>
|
||||||
|
<source>${jdk-version}</source>
|
||||||
|
<target>${jdk-version}</target>
|
||||||
|
<encoding>${project-sourceEncoding}</encoding>
|
||||||
|
</configuration>
|
||||||
|
</plugin>
|
||||||
|
<!-- assembly plugin -->
|
||||||
|
<plugin>
|
||||||
|
<artifactId>maven-assembly-plugin</artifactId>
|
||||||
|
<configuration>
|
||||||
|
<descriptors>
|
||||||
|
<descriptor>src/main/assembly/package.xml</descriptor>
|
||||||
|
</descriptors>
|
||||||
|
<finalName>datax</finalName>
|
||||||
|
</configuration>
|
||||||
|
<executions>
|
||||||
|
<execution>
|
||||||
|
<id>dwzip</id>
|
||||||
|
<phase>package</phase>
|
||||||
|
<goals>
|
||||||
|
<goal>single</goal>
|
||||||
|
</goals>
|
||||||
|
</execution>
|
||||||
|
</executions>
|
||||||
|
</plugin>
|
||||||
|
</plugins>
|
||||||
|
</build>
|
||||||
|
|
||||||
|
|
||||||
|
</project>
|
||||||
35
clickhousereader/src/main/assembly/package.xml
Normal file
35
clickhousereader/src/main/assembly/package.xml
Normal file
@ -0,0 +1,35 @@
|
|||||||
|
<assembly
|
||||||
|
xmlns="http://maven.apache.org/plugins/maven-assembly-plugin/assembly/1.1.0"
|
||||||
|
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
|
||||||
|
xsi:schemaLocation="http://maven.apache.org/plugins/maven-assembly-plugin/assembly/1.1.0 http://maven.apache.org/xsd/assembly-1.1.0.xsd">
|
||||||
|
<id></id>
|
||||||
|
<formats>
|
||||||
|
<format>dir</format>
|
||||||
|
</formats>
|
||||||
|
<includeBaseDirectory>false</includeBaseDirectory>
|
||||||
|
<fileSets>
|
||||||
|
<fileSet>
|
||||||
|
<directory>src/main/resources</directory>
|
||||||
|
<includes>
|
||||||
|
<include>plugin.json</include>
|
||||||
|
<include>plugin_job_template.json</include>
|
||||||
|
</includes>
|
||||||
|
<outputDirectory>plugin/reader/clickhousereader</outputDirectory>
|
||||||
|
</fileSet>
|
||||||
|
<fileSet>
|
||||||
|
<directory>target/</directory>
|
||||||
|
<includes>
|
||||||
|
<include>clickhousereader-0.0.1-SNAPSHOT.jar</include>
|
||||||
|
</includes>
|
||||||
|
<outputDirectory>plugin/reader/clickhousereader</outputDirectory>
|
||||||
|
</fileSet>
|
||||||
|
</fileSets>
|
||||||
|
|
||||||
|
<dependencySets>
|
||||||
|
<dependencySet>
|
||||||
|
<useProjectArtifact>false</useProjectArtifact>
|
||||||
|
<outputDirectory>plugin/reader/clickhousereader/libs</outputDirectory>
|
||||||
|
<scope>runtime</scope>
|
||||||
|
</dependencySet>
|
||||||
|
</dependencySets>
|
||||||
|
</assembly>
|
||||||
@ -0,0 +1,85 @@
|
|||||||
|
package com.alibaba.datax.plugin.reader.clickhousereader;
|
||||||
|
|
||||||
|
import java.sql.Array;
|
||||||
|
import java.sql.ResultSet;
|
||||||
|
import java.sql.ResultSetMetaData;
|
||||||
|
import java.sql.SQLException;
|
||||||
|
import java.sql.Types;
|
||||||
|
import java.util.List;
|
||||||
|
|
||||||
|
import com.alibaba.datax.common.element.Record;
|
||||||
|
import com.alibaba.datax.common.element.StringColumn;
|
||||||
|
import com.alibaba.datax.common.plugin.RecordSender;
|
||||||
|
import com.alibaba.datax.common.plugin.TaskPluginCollector;
|
||||||
|
import com.alibaba.datax.common.spi.Reader;
|
||||||
|
import com.alibaba.datax.common.util.Configuration;
|
||||||
|
import com.alibaba.datax.common.util.MessageSource;
|
||||||
|
import com.alibaba.datax.plugin.rdbms.reader.CommonRdbmsReader;
|
||||||
|
import com.alibaba.datax.plugin.rdbms.util.DataBaseType;
|
||||||
|
import com.alibaba.fastjson2.JSON;
|
||||||
|
|
||||||
|
import org.slf4j.Logger;
|
||||||
|
import org.slf4j.LoggerFactory;
|
||||||
|
|
||||||
|
public class ClickhouseReader extends Reader {
|
||||||
|
|
||||||
|
private static final DataBaseType DATABASE_TYPE = DataBaseType.ClickHouse;
|
||||||
|
private static final Logger LOG = LoggerFactory.getLogger(ClickhouseReader.class);
|
||||||
|
|
||||||
|
public static class Job extends Reader.Job {
|
||||||
|
private Configuration jobConfig = null;
|
||||||
|
private CommonRdbmsReader.Job commonRdbmsReaderMaster;
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public void init() {
|
||||||
|
this.jobConfig = super.getPluginJobConf();
|
||||||
|
this.commonRdbmsReaderMaster = new CommonRdbmsReader.Job(DATABASE_TYPE);
|
||||||
|
this.commonRdbmsReaderMaster.init(this.jobConfig);
|
||||||
|
}
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public List<Configuration> split(int mandatoryNumber) {
|
||||||
|
return this.commonRdbmsReaderMaster.split(this.jobConfig, mandatoryNumber);
|
||||||
|
}
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public void post() {
|
||||||
|
this.commonRdbmsReaderMaster.post(this.jobConfig);
|
||||||
|
}
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public void destroy() {
|
||||||
|
this.commonRdbmsReaderMaster.destroy(this.jobConfig);
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
public static class Task extends Reader.Task {
|
||||||
|
|
||||||
|
private Configuration jobConfig;
|
||||||
|
private CommonRdbmsReader.Task commonRdbmsReaderSlave;
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public void init() {
|
||||||
|
this.jobConfig = super.getPluginJobConf();
|
||||||
|
this.commonRdbmsReaderSlave = new CommonRdbmsReader.Task(DATABASE_TYPE, super.getTaskGroupId(), super.getTaskId());
|
||||||
|
this.commonRdbmsReaderSlave.init(this.jobConfig);
|
||||||
|
}
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public void startRead(RecordSender recordSender) {
|
||||||
|
int fetchSize = this.jobConfig.getInt(com.alibaba.datax.plugin.rdbms.reader.Constant.FETCH_SIZE, 1000);
|
||||||
|
|
||||||
|
this.commonRdbmsReaderSlave.startRead(this.jobConfig, recordSender, super.getTaskPluginCollector(), fetchSize);
|
||||||
|
}
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public void post() {
|
||||||
|
this.commonRdbmsReaderSlave.post(this.jobConfig);
|
||||||
|
}
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public void destroy() {
|
||||||
|
this.commonRdbmsReaderSlave.destroy(this.jobConfig);
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
6
clickhousereader/src/main/resources/plugin.json
Normal file
6
clickhousereader/src/main/resources/plugin.json
Normal file
@ -0,0 +1,6 @@
|
|||||||
|
{
|
||||||
|
"name": "clickhousereader",
|
||||||
|
"class": "com.alibaba.datax.plugin.reader.clickhousereader.ClickhouseReader",
|
||||||
|
"description": "useScene: prod. mechanism: Jdbc connection using the database, execute select sql.",
|
||||||
|
"developer": "alibaba"
|
||||||
|
}
|
||||||
16
clickhousereader/src/main/resources/plugin_job_template.json
Normal file
16
clickhousereader/src/main/resources/plugin_job_template.json
Normal file
@ -0,0 +1,16 @@
|
|||||||
|
{
|
||||||
|
"name": "clickhousereader",
|
||||||
|
"parameter": {
|
||||||
|
"username": "username",
|
||||||
|
"password": "password",
|
||||||
|
"column": ["col1", "col2", "col3"],
|
||||||
|
"connection": [
|
||||||
|
{
|
||||||
|
"jdbcUrl": "jdbc:clickhouse://<host>:<port>[/<database>]",
|
||||||
|
"table": ["table1", "table2"]
|
||||||
|
}
|
||||||
|
],
|
||||||
|
"preSql": [],
|
||||||
|
"postSql": []
|
||||||
|
}
|
||||||
|
}
|
||||||
57
clickhousereader/src/test/resources/basic1.json
Executable file
57
clickhousereader/src/test/resources/basic1.json
Executable file
@ -0,0 +1,57 @@
|
|||||||
|
{
|
||||||
|
"job": {
|
||||||
|
"setting": {
|
||||||
|
"speed": {
|
||||||
|
"channel": 5
|
||||||
|
}
|
||||||
|
},
|
||||||
|
"content": [
|
||||||
|
{
|
||||||
|
"reader": {
|
||||||
|
"name": "clickhousereader",
|
||||||
|
"parameter": {
|
||||||
|
"username": "XXXX",
|
||||||
|
"password": "XXXX",
|
||||||
|
"column": [
|
||||||
|
"uint8_col",
|
||||||
|
"uint16_col",
|
||||||
|
"uint32_col",
|
||||||
|
"uint64_col",
|
||||||
|
"int8_col",
|
||||||
|
"int16_col",
|
||||||
|
"int32_col",
|
||||||
|
"int64_col",
|
||||||
|
"float32_col",
|
||||||
|
"float64_col",
|
||||||
|
"bool_col",
|
||||||
|
"str_col",
|
||||||
|
"fixedstr_col",
|
||||||
|
"uuid_col",
|
||||||
|
"date_col",
|
||||||
|
"datetime_col",
|
||||||
|
"enum_col",
|
||||||
|
"ary_uint8_col",
|
||||||
|
"ary_str_col",
|
||||||
|
"tuple_col",
|
||||||
|
"nullable_col",
|
||||||
|
"nested_col.nested_id",
|
||||||
|
"nested_col.nested_str",
|
||||||
|
"ipv4_col",
|
||||||
|
"ipv6_col",
|
||||||
|
"decimal_col"
|
||||||
|
],
|
||||||
|
"connection": [
|
||||||
|
{
|
||||||
|
"table": [
|
||||||
|
"all_type_tbl"
|
||||||
|
],
|
||||||
|
"jdbcUrl":["jdbc:clickhouse://XXXX:8123/default"]
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
|
},
|
||||||
|
"writer": {}
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
|
}
|
||||||
34
clickhousereader/src/test/resources/basic1.sql
Normal file
34
clickhousereader/src/test/resources/basic1.sql
Normal file
@ -0,0 +1,34 @@
|
|||||||
|
CREATE TABLE IF NOT EXISTS default.all_type_tbl
|
||||||
|
(
|
||||||
|
`uint8_col` UInt8,
|
||||||
|
`uint16_col` UInt16,
|
||||||
|
uint32_col UInt32,
|
||||||
|
uint64_col UInt64,
|
||||||
|
int8_col Int8,
|
||||||
|
int16_col Int16,
|
||||||
|
int32_col Int32,
|
||||||
|
int64_col Int64,
|
||||||
|
float32_col Float32,
|
||||||
|
float64_col Float64,
|
||||||
|
bool_col UInt8,
|
||||||
|
str_col String,
|
||||||
|
fixedstr_col FixedString(3),
|
||||||
|
uuid_col UUID,
|
||||||
|
date_col Date,
|
||||||
|
datetime_col DateTime,
|
||||||
|
enum_col Enum('hello' = 1, 'world' = 2),
|
||||||
|
ary_uint8_col Array(UInt8),
|
||||||
|
ary_str_col Array(String),
|
||||||
|
tuple_col Tuple(UInt8, String),
|
||||||
|
nullable_col Nullable(UInt8),
|
||||||
|
nested_col Nested
|
||||||
|
(
|
||||||
|
nested_id UInt32,
|
||||||
|
nested_str String
|
||||||
|
),
|
||||||
|
ipv4_col IPv4,
|
||||||
|
ipv6_col IPv6,
|
||||||
|
decimal_col Decimal(5,3)
|
||||||
|
)
|
||||||
|
ENGINE = MergeTree()
|
||||||
|
ORDER BY (uint8_col);
|
||||||
@ -5,6 +5,7 @@ import com.alibaba.datax.common.exception.DataXException;
|
|||||||
|
|
||||||
import java.math.BigDecimal;
|
import java.math.BigDecimal;
|
||||||
import java.math.BigInteger;
|
import java.math.BigInteger;
|
||||||
|
import java.sql.Time;
|
||||||
import java.util.Date;
|
import java.util.Date;
|
||||||
|
|
||||||
/**
|
/**
|
||||||
@ -14,15 +15,51 @@ public class DateColumn extends Column {
|
|||||||
|
|
||||||
private DateType subType = DateType.DATETIME;
|
private DateType subType = DateType.DATETIME;
|
||||||
|
|
||||||
|
private int nanos = 0;
|
||||||
|
|
||||||
|
private int precision = -1;
|
||||||
|
|
||||||
public static enum DateType {
|
public static enum DateType {
|
||||||
DATE, TIME, DATETIME
|
DATE, TIME, DATETIME
|
||||||
}
|
}
|
||||||
|
|
||||||
|
/**
|
||||||
|
* 构建值为time(java.sql.Time)的DateColumn，使用Date子类型为TIME，只有时间，没有日期
|
||||||
|
*/
|
||||||
|
public DateColumn(Time time, int nanos, int jdbcPrecision) {
|
||||||
|
this(time);
|
||||||
|
if (time != null) {
|
||||||
|
setNanos(nanos);
|
||||||
|
}
|
||||||
|
if (jdbcPrecision == 10) {
|
||||||
|
setPrecision(0);
|
||||||
|
}
|
||||||
|
if (jdbcPrecision >= 12 && jdbcPrecision <= 17) {
|
||||||
|
setPrecision(jdbcPrecision - 11);
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
public long getNanos() {
|
||||||
|
return nanos;
|
||||||
|
}
|
||||||
|
|
||||||
|
public void setNanos(int nanos) {
|
||||||
|
this.nanos = nanos;
|
||||||
|
}
|
||||||
|
|
||||||
|
public int getPrecision() {
|
||||||
|
return precision;
|
||||||
|
}
|
||||||
|
|
||||||
|
public void setPrecision(int precision) {
|
||||||
|
this.precision = precision;
|
||||||
|
}
|
||||||
|
|
||||||
/**
|
/**
|
||||||
* 构建值为null的DateColumn，使用Date子类型为DATETIME
|
* 构建值为null的DateColumn，使用Date子类型为DATETIME
|
||||||
* */
|
*/
|
||||||
public DateColumn() {
|
public DateColumn() {
|
||||||
this((Long)null);
|
this((Long) null);
|
||||||
}
|
}
|
||||||
|
|
||||||
/**
|
/**
|
||||||
|
|||||||
@ -77,8 +77,8 @@ public class VMInfo {
|
|||||||
garbageCollectorMXBeanList = java.lang.management.ManagementFactory.getGarbageCollectorMXBeans();
|
garbageCollectorMXBeanList = java.lang.management.ManagementFactory.getGarbageCollectorMXBeans();
|
||||||
memoryPoolMXBeanList = java.lang.management.ManagementFactory.getMemoryPoolMXBeans();
|
memoryPoolMXBeanList = java.lang.management.ManagementFactory.getMemoryPoolMXBeans();
|
||||||
|
|
||||||
osInfo = runtimeMXBean.getVmVendor() + " " + runtimeMXBean.getSpecVersion() + " " + runtimeMXBean.getVmVersion();
|
jvmInfo = runtimeMXBean.getVmVendor() + " " + runtimeMXBean.getSpecVersion() + " " + runtimeMXBean.getVmVersion();
|
||||||
jvmInfo = osMXBean.getName() + " " + osMXBean.getArch() + " " + osMXBean.getVersion();
|
osInfo = osMXBean.getName() + " " + osMXBean.getArch() + " " + osMXBean.getVersion();
|
||||||
totalProcessorCount = osMXBean.getAvailableProcessors();
|
totalProcessorCount = osMXBean.getAvailableProcessors();
|
||||||
|
|
||||||
//构建startPhyOSStatus
|
//构建startPhyOSStatus
|
||||||
|
|||||||
@ -0,0 +1,34 @@
|
|||||||
|
package com.alibaba.datax.common.util;
|
||||||
|
|
||||||
|
import org.apache.commons.lang3.StringUtils;
|
||||||
|
|
||||||
|
import java.util.HashMap;
|
||||||
|
import java.util.Map;
|
||||||
|
|
||||||
|
/**
|
||||||
|
* @author jitongchen
|
||||||
|
* @date 2023/9/7 9:47 AM
|
||||||
|
*/
|
||||||
|
public class LimitLogger {
|
||||||
|
|
||||||
|
private static Map<String, Long> lastPrintTime = new HashMap<>();
|
||||||
|
|
||||||
|
public static void limit(String name, long limit, LoggerFunction function) {
|
||||||
|
if (StringUtils.isBlank(name)) {
|
||||||
|
name = "__all__";
|
||||||
|
}
|
||||||
|
if (limit <= 0) {
|
||||||
|
function.apply();
|
||||||
|
} else {
|
||||||
|
if (!lastPrintTime.containsKey(name)) {
|
||||||
|
lastPrintTime.put(name, System.currentTimeMillis());
|
||||||
|
function.apply();
|
||||||
|
} else {
|
||||||
|
if (System.currentTimeMillis() > lastPrintTime.get(name) + limit) {
|
||||||
|
lastPrintTime.put(name, System.currentTimeMillis());
|
||||||
|
function.apply();
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
@ -0,0 +1,10 @@
|
|||||||
|
package com.alibaba.datax.common.util;
|
||||||
|
|
||||||
|
/**
|
||||||
|
* @author molin.lxd
|
||||||
|
* @date 2021-05-09
|
||||||
|
*/
|
||||||
|
public interface LoggerFunction {
|
||||||
|
|
||||||
|
void apply();
|
||||||
|
}
|
||||||
@ -29,7 +29,7 @@ public class MemoryChannel extends Channel {
|
|||||||
|
|
||||||
private ReentrantLock lock;
|
private ReentrantLock lock;
|
||||||
|
|
||||||
private Condition notInsufficient, notEmpty;
|
private Condition notSufficient, notEmpty;
|
||||||
|
|
||||||
public MemoryChannel(final Configuration configuration) {
|
public MemoryChannel(final Configuration configuration) {
|
||||||
super(configuration);
|
super(configuration);
|
||||||
@ -37,7 +37,7 @@ public class MemoryChannel extends Channel {
|
|||||||
this.bufferSize = configuration.getInt(CoreConstant.DATAX_CORE_TRANSPORT_EXCHANGER_BUFFERSIZE);
|
this.bufferSize = configuration.getInt(CoreConstant.DATAX_CORE_TRANSPORT_EXCHANGER_BUFFERSIZE);
|
||||||
|
|
||||||
lock = new ReentrantLock();
|
lock = new ReentrantLock();
|
||||||
notInsufficient = lock.newCondition();
|
notSufficient = lock.newCondition();
|
||||||
notEmpty = lock.newCondition();
|
notEmpty = lock.newCondition();
|
||||||
}
|
}
|
||||||
|
|
||||||
@ -75,7 +75,7 @@ public class MemoryChannel extends Channel {
|
|||||||
lock.lockInterruptibly();
|
lock.lockInterruptibly();
|
||||||
int bytes = getRecordBytes(rs);
|
int bytes = getRecordBytes(rs);
|
||||||
while (memoryBytes.get() + bytes > this.byteCapacity || rs.size() > this.queue.remainingCapacity()) {
|
while (memoryBytes.get() + bytes > this.byteCapacity || rs.size() > this.queue.remainingCapacity()) {
|
||||||
notInsufficient.await(200L, TimeUnit.MILLISECONDS);
|
notSufficient.await(200L, TimeUnit.MILLISECONDS);
|
||||||
}
|
}
|
||||||
this.queue.addAll(rs);
|
this.queue.addAll(rs);
|
||||||
waitWriterTime += System.nanoTime() - startTime;
|
waitWriterTime += System.nanoTime() - startTime;
|
||||||
@ -116,7 +116,7 @@ public class MemoryChannel extends Channel {
|
|||||||
waitReaderTime += System.nanoTime() - startTime;
|
waitReaderTime += System.nanoTime() - startTime;
|
||||||
int bytes = getRecordBytes(rs);
|
int bytes = getRecordBytes(rs);
|
||||||
memoryBytes.addAndGet(-bytes);
|
memoryBytes.addAndGet(-bytes);
|
||||||
notInsufficient.signalAll();
|
notSufficient.signalAll();
|
||||||
} catch (InterruptedException e) {
|
} catch (InterruptedException e) {
|
||||||
throw DataXException.asDataXException(
|
throw DataXException.asDataXException(
|
||||||
FrameworkErrorCode.RUNTIME_ERROR, e);
|
FrameworkErrorCode.RUNTIME_ERROR, e);
|
||||||
|
|||||||
@ -168,6 +168,7 @@ public final class ConfigParser {
|
|||||||
boolean isDefaultPath = StringUtils.isBlank(pluginPath);
|
boolean isDefaultPath = StringUtils.isBlank(pluginPath);
|
||||||
if (isDefaultPath) {
|
if (isDefaultPath) {
|
||||||
configuration.set("path", path);
|
configuration.set("path", path);
|
||||||
|
configuration.set("loadType","jarLoader");
|
||||||
}
|
}
|
||||||
|
|
||||||
Configuration result = Configuration.newDefault();
|
Configuration result = Configuration.newDefault();
|
||||||
|
|||||||
@ -15,7 +15,7 @@ import java.util.List;
|
|||||||
/**
|
/**
|
||||||
* 提供Jar隔离的加载机制，会把传入的路径、及其子路径、以及路径中的jar文件加入到class path。
|
* 提供Jar隔离的加载机制，会把传入的路径、及其子路径、以及路径中的jar文件加入到class path。
|
||||||
*/
|
*/
|
||||||
public class JarLoader extends URLClassLoader {
|
public class JarLoader extends URLClassLoader{
|
||||||
public JarLoader(String[] paths) {
|
public JarLoader(String[] paths) {
|
||||||
this(paths, JarLoader.class.getClassLoader());
|
this(paths, JarLoader.class.getClassLoader());
|
||||||
}
|
}
|
||||||
|
|||||||
@ -49,7 +49,7 @@ public class LoadUtil {
|
|||||||
/**
|
/**
|
||||||
* jarLoader的缓冲
|
* jarLoader的缓冲
|
||||||
*/
|
*/
|
||||||
private static Map<String, JarLoader> jarLoaderCenter = new HashMap<String, JarLoader>();
|
private static Map<String, JarLoader> jarLoaderCenter = new HashMap();
|
||||||
|
|
||||||
/**
|
/**
|
||||||
* 设置pluginConfigs，方便后面插件来获取
|
* 设置pluginConfigs，方便后面插件来获取
|
||||||
|
|||||||
@ -79,6 +79,8 @@ create table if not exsits datax.sample1(a string, b int64, c date, d timestamp,
|
|||||||
"writer": {
|
"writer": {
|
||||||
"name": "databendwriter",
|
"name": "databendwriter",
|
||||||
"parameter": {
|
"parameter": {
|
||||||
|
"writeMode": "replace",
|
||||||
|
"onConflictColumn": ["id"],
|
||||||
"username": "databend",
|
"username": "databend",
|
||||||
"password": "databend",
|
"password": "databend",
|
||||||
"column": ["a", "b", "c", "d", "e", "f", "g"],
|
"column": ["a", "b", "c", "d", "e", "f", "g"],
|
||||||
@ -149,6 +151,16 @@ create table if not exsits datax.sample1(a string, b int64, c date, d timestamp,
|
|||||||
* 必选: 否
|
* 必选: 否
|
||||||
* 默认值: 无
|
* 默认值: 无
|
||||||
* 示例: ["select count(*) from datax.sample1"]
|
* 示例: ["select count(*) from datax.sample1"]
|
||||||
|
* writeMode
|
||||||
|
* 描述：写入模式，支持 insert 和 replace 两种模式，默认为 insert。若为 replace，务必填写 onConflictColumn 参数
|
||||||
|
* 必选：否
|
||||||
|
* 默认值：insert
|
||||||
|
* 示例："replace"
|
||||||
|
* onConflictColumn
|
||||||
|
* 描述：on conflict 字段，指定 writeMode 为 replace 后，需要此参数
|
||||||
|
* 必选：否
|
||||||
|
* 默认值：无
|
||||||
|
* 示例：["id","user"]
|
||||||
|
|
||||||
### 3.3 类型转化
|
### 3.3 类型转化
|
||||||
DataX中的数据类型可以转换为databend中的相应数据类型。下表显示了两种类型之间的对应关系。
|
DataX中的数据类型可以转换为databend中的相应数据类型。下表显示了两种类型之间的对应关系。
|
||||||
|
|||||||
@ -142,6 +142,16 @@ create table if not exsits datax.sample1(a string, b int64, c date, d timestamp,
|
|||||||
* Description: A list of SQL statements that will be executed after the write operation.
|
* Description: A list of SQL statements that will be executed after the write operation.
|
||||||
* Required: no
|
* Required: no
|
||||||
* Default: none
|
* Default: none
|
||||||
|
* writeMode
|
||||||
|
* Description：The write mode, support `insert` and `replace` two mode.
|
||||||
|
* Required：no
|
||||||
|
* Default：insert
|
||||||
|
* Example："replace"
|
||||||
|
* onConflictColumn
|
||||||
|
* Description：On conflict fields list.
|
||||||
|
* Required：no
|
||||||
|
* Default：none
|
||||||
|
* Example：["id","user"]
|
||||||
|
|
||||||
### 3.3 Type Convert
|
### 3.3 Type Convert
|
||||||
Data types in datax can be converted to the corresponding data types in databend. The following table shows the correspondence between the two types.
|
Data types in datax can be converted to the corresponding data types in databend. The following table shows the correspondence between the two types.
|
||||||
|
|||||||
@ -17,7 +17,7 @@
|
|||||||
<dependency>
|
<dependency>
|
||||||
<groupId>com.databend</groupId>
|
<groupId>com.databend</groupId>
|
||||||
<artifactId>databend-jdbc</artifactId>
|
<artifactId>databend-jdbc</artifactId>
|
||||||
<version>0.0.7</version>
|
<version>0.1.0</version>
|
||||||
</dependency>
|
</dependency>
|
||||||
<dependency>
|
<dependency>
|
||||||
<groupId>com.alibaba.datax</groupId>
|
<groupId>com.alibaba.datax</groupId>
|
||||||
|
|||||||
@ -17,20 +17,17 @@ import java.sql.*;
|
|||||||
import java.util.List;
|
import java.util.List;
|
||||||
import java.util.regex.Pattern;
|
import java.util.regex.Pattern;
|
||||||
|
|
||||||
public class DatabendWriter extends Writer
|
public class DatabendWriter extends Writer {
|
||||||
{
|
|
||||||
private static final DataBaseType DATABASE_TYPE = DataBaseType.Databend;
|
private static final DataBaseType DATABASE_TYPE = DataBaseType.Databend;
|
||||||
|
|
||||||
public static class Job
|
public static class Job
|
||||||
extends Writer.Job
|
extends Writer.Job {
|
||||||
{
|
|
||||||
private static final Logger LOG = LoggerFactory.getLogger(Job.class);
|
private static final Logger LOG = LoggerFactory.getLogger(Job.class);
|
||||||
private Configuration originalConfig;
|
private Configuration originalConfig;
|
||||||
private CommonRdbmsWriter.Job commonRdbmsWriterMaster;
|
private CommonRdbmsWriter.Job commonRdbmsWriterMaster;
|
||||||
|
|
||||||
@Override
|
@Override
|
||||||
public void init()
|
public void init() throws DataXException {
|
||||||
{
|
|
||||||
this.originalConfig = super.getPluginJobConf();
|
this.originalConfig = super.getPluginJobConf();
|
||||||
this.commonRdbmsWriterMaster = new CommonRdbmsWriter.Job(DATABASE_TYPE);
|
this.commonRdbmsWriterMaster = new CommonRdbmsWriter.Job(DATABASE_TYPE);
|
||||||
this.commonRdbmsWriterMaster.init(this.originalConfig);
|
this.commonRdbmsWriterMaster.init(this.originalConfig);
|
||||||
@ -39,8 +36,7 @@ public class DatabendWriter extends Writer
|
|||||||
}
|
}
|
||||||
|
|
||||||
@Override
|
@Override
|
||||||
public void preCheck()
|
public void preCheck() {
|
||||||
{
|
|
||||||
this.init();
|
this.init();
|
||||||
this.commonRdbmsWriterMaster.writerPreCheck(this.originalConfig, DATABASE_TYPE);
|
this.commonRdbmsWriterMaster.writerPreCheck(this.originalConfig, DATABASE_TYPE);
|
||||||
}
|
}
|
||||||
@ -67,8 +63,7 @@ public class DatabendWriter extends Writer
|
|||||||
}
|
}
|
||||||
|
|
||||||
|
|
||||||
public static class Task extends Writer.Task
|
public static class Task extends Writer.Task {
|
||||||
{
|
|
||||||
private static final Logger LOG = LoggerFactory.getLogger(Task.class);
|
private static final Logger LOG = LoggerFactory.getLogger(Task.class);
|
||||||
|
|
||||||
private Configuration writerSliceConfig;
|
private Configuration writerSliceConfig;
|
||||||
@ -76,11 +71,10 @@ public class DatabendWriter extends Writer
|
|||||||
private CommonRdbmsWriter.Task commonRdbmsWriterSlave;
|
private CommonRdbmsWriter.Task commonRdbmsWriterSlave;
|
||||||
|
|
||||||
@Override
|
@Override
|
||||||
public void init()
|
public void init() {
|
||||||
{
|
|
||||||
this.writerSliceConfig = super.getPluginJobConf();
|
this.writerSliceConfig = super.getPluginJobConf();
|
||||||
|
|
||||||
this.commonRdbmsWriterSlave = new CommonRdbmsWriter.Task(DataBaseType.Databend){
|
this.commonRdbmsWriterSlave = new CommonRdbmsWriter.Task(DataBaseType.Databend) {
|
||||||
@Override
|
@Override
|
||||||
protected PreparedStatement fillPreparedStatementColumnType(PreparedStatement preparedStatement, int columnIndex, int columnSqltype, String typeName, Column column) throws SQLException {
|
protected PreparedStatement fillPreparedStatementColumnType(PreparedStatement preparedStatement, int columnIndex, int columnSqltype, String typeName, Column column) throws SQLException {
|
||||||
try {
|
try {
|
||||||
@ -224,8 +218,7 @@ public class DatabendWriter extends Writer
|
|||||||
}
|
}
|
||||||
|
|
||||||
@Override
|
@Override
|
||||||
public void destroy()
|
public void destroy() {
|
||||||
{
|
|
||||||
this.commonRdbmsWriterSlave.destroy(this.writerSliceConfig);
|
this.commonRdbmsWriterSlave.destroy(this.writerSliceConfig);
|
||||||
}
|
}
|
||||||
|
|
||||||
@ -238,9 +231,9 @@ public class DatabendWriter extends Writer
|
|||||||
public void post() {
|
public void post() {
|
||||||
this.commonRdbmsWriterSlave.post(this.writerSliceConfig);
|
this.commonRdbmsWriterSlave.post(this.writerSliceConfig);
|
||||||
}
|
}
|
||||||
|
|
||||||
@Override
|
@Override
|
||||||
public void startWrite(RecordReceiver lineReceiver)
|
public void startWrite(RecordReceiver lineReceiver) {
|
||||||
{
|
|
||||||
this.commonRdbmsWriterSlave.startWrite(lineReceiver, this.writerSliceConfig, this.getTaskPluginCollector());
|
this.commonRdbmsWriterSlave.startWrite(lineReceiver, this.writerSliceConfig, this.getTaskPluginCollector());
|
||||||
}
|
}
|
||||||
|
|
||||||
|
|||||||
@ -0,0 +1,33 @@
|
|||||||
|
package com.alibaba.datax.plugin.writer.databendwriter;
|
||||||
|
|
||||||
|
import com.alibaba.datax.common.spi.ErrorCode;
|
||||||
|
|
||||||
|
|
||||||
|
public enum DatabendWriterErrorCode implements ErrorCode {
|
||||||
|
CONF_ERROR("DatabendWriter-00", "配置错误."),
|
||||||
|
WRITE_DATA_ERROR("DatabendWriter-01", "写入数据时失败."),
|
||||||
|
;
|
||||||
|
|
||||||
|
private final String code;
|
||||||
|
private final String description;
|
||||||
|
|
||||||
|
private DatabendWriterErrorCode(String code, String description) {
|
||||||
|
this.code = code;
|
||||||
|
this.description = description;
|
||||||
|
}
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public String getCode() {
|
||||||
|
return this.code;
|
||||||
|
}
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public String getDescription() {
|
||||||
|
return this.description;
|
||||||
|
}
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public String toString() {
|
||||||
|
return String.format("Code:[%s], Description:[%s].", this.code, this.description);
|
||||||
|
}
|
||||||
|
}
|
||||||
@ -1,29 +1,53 @@
|
|||||||
package com.alibaba.datax.plugin.writer.databendwriter.util;
|
package com.alibaba.datax.plugin.writer.databendwriter.util;
|
||||||
|
|
||||||
|
import com.alibaba.datax.common.exception.DataXException;
|
||||||
import com.alibaba.datax.common.util.Configuration;
|
import com.alibaba.datax.common.util.Configuration;
|
||||||
import com.alibaba.datax.plugin.rdbms.writer.Constant;
|
import com.alibaba.datax.plugin.rdbms.writer.Constant;
|
||||||
import com.alibaba.datax.plugin.rdbms.writer.Key;
|
import com.alibaba.datax.plugin.rdbms.writer.Key;
|
||||||
|
|
||||||
|
import com.alibaba.datax.plugin.writer.databendwriter.DatabendWriterErrorCode;
|
||||||
|
import org.apache.commons.lang3.StringUtils;
|
||||||
import org.slf4j.Logger;
|
import org.slf4j.Logger;
|
||||||
import org.slf4j.LoggerFactory;
|
import org.slf4j.LoggerFactory;
|
||||||
|
|
||||||
|
import javax.xml.crypto.Data;
|
||||||
import java.util.List;
|
import java.util.List;
|
||||||
import java.util.StringJoiner;
|
import java.util.StringJoiner;
|
||||||
|
|
||||||
public final class DatabendWriterUtil
|
public final class DatabendWriterUtil {
|
||||||
{
|
|
||||||
private static final Logger LOG = LoggerFactory.getLogger(DatabendWriterUtil.class);
|
private static final Logger LOG = LoggerFactory.getLogger(DatabendWriterUtil.class);
|
||||||
|
|
||||||
private DatabendWriterUtil() {}
|
private DatabendWriterUtil() {
|
||||||
public static void dealWriteMode(Configuration originalConfig)
|
}
|
||||||
{
|
|
||||||
|
public static void dealWriteMode(Configuration originalConfig) throws DataXException {
|
||||||
List<String> columns = originalConfig.getList(Key.COLUMN, String.class);
|
List<String> columns = originalConfig.getList(Key.COLUMN, String.class);
|
||||||
|
List<String> onConflictColumns = originalConfig.getList(Key.ONCONFLICT_COLUMN, String.class);
|
||||||
|
StringBuilder writeDataSqlTemplate = new StringBuilder();
|
||||||
|
|
||||||
String jdbcUrl = originalConfig.getString(String.format("%s[0].%s",
|
String jdbcUrl = originalConfig.getString(String.format("%s[0].%s",
|
||||||
Constant.CONN_MARK, Key.JDBC_URL, String.class));
|
Constant.CONN_MARK, Key.JDBC_URL, String.class));
|
||||||
|
|
||||||
String writeMode = originalConfig.getString(Key.WRITE_MODE, "INSERT");
|
String writeMode = originalConfig.getString(Key.WRITE_MODE, "INSERT");
|
||||||
|
LOG.info("write mode is {}", writeMode);
|
||||||
|
if (writeMode.toLowerCase().contains("replace")) {
|
||||||
|
if (onConflictColumns == null || onConflictColumns.size() == 0) {
|
||||||
|
throw DataXException
|
||||||
|
.asDataXException(
|
||||||
|
DatabendWriterErrorCode.CONF_ERROR,
|
||||||
|
String.format(
|
||||||
|
"Replace mode must has onConflictColumn config."
|
||||||
|
));
|
||||||
|
}
|
||||||
|
|
||||||
StringBuilder writeDataSqlTemplate = new StringBuilder();
|
// for databend if you want to use replace mode, the writeMode should be: "writeMode": "replace"
|
||||||
|
writeDataSqlTemplate.append("REPLACE INTO %s (")
|
||||||
|
.append(StringUtils.join(columns, ",")).append(") ").append(onConFlictDoString(onConflictColumns))
|
||||||
|
.append(" VALUES");
|
||||||
|
|
||||||
|
LOG.info("Replace data [\n{}\n], which jdbcUrl like:[{}]", writeDataSqlTemplate, jdbcUrl);
|
||||||
|
originalConfig.set(Constant.INSERT_OR_REPLACE_TEMPLATE_MARK, writeDataSqlTemplate);
|
||||||
|
} else {
|
||||||
writeDataSqlTemplate.append("INSERT INTO %s");
|
writeDataSqlTemplate.append("INSERT INTO %s");
|
||||||
StringJoiner columnString = new StringJoiner(",");
|
StringJoiner columnString = new StringJoiner(",");
|
||||||
|
|
||||||
@ -33,8 +57,16 @@ public final class DatabendWriterUtil
|
|||||||
writeDataSqlTemplate.append(String.format("(%s)", columnString));
|
writeDataSqlTemplate.append(String.format("(%s)", columnString));
|
||||||
writeDataSqlTemplate.append(" VALUES");
|
writeDataSqlTemplate.append(" VALUES");
|
||||||
|
|
||||||
LOG.info("Write data [\n{}\n], which jdbcUrl like:[{}]", writeDataSqlTemplate, jdbcUrl);
|
LOG.info("Insert data [\n{}\n], which jdbcUrl like:[{}]", writeDataSqlTemplate, jdbcUrl);
|
||||||
|
|
||||||
originalConfig.set(Constant.INSERT_OR_REPLACE_TEMPLATE_MARK, writeDataSqlTemplate);
|
originalConfig.set(Constant.INSERT_OR_REPLACE_TEMPLATE_MARK, writeDataSqlTemplate);
|
||||||
}
|
}
|
||||||
|
|
||||||
|
}
|
||||||
|
|
||||||
|
public static String onConFlictDoString(List<String> conflictColumns) {
|
||||||
|
return " ON " +
|
||||||
|
"(" +
|
||||||
|
StringUtils.join(conflictColumns, ",") + ") ";
|
||||||
|
}
|
||||||
}
|
}
|
||||||
20
datax-example/datax-example-core/pom.xml
Normal file
20
datax-example/datax-example-core/pom.xml
Normal file
@ -0,0 +1,20 @@
|
|||||||
|
<?xml version="1.0" encoding="UTF-8"?>
|
||||||
|
<project xmlns="http://maven.apache.org/POM/4.0.0"
|
||||||
|
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
|
||||||
|
xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
|
||||||
|
<modelVersion>4.0.0</modelVersion>
|
||||||
|
<parent>
|
||||||
|
<groupId>com.alibaba.datax</groupId>
|
||||||
|
<artifactId>datax-example</artifactId>
|
||||||
|
<version>0.0.1-SNAPSHOT</version>
|
||||||
|
</parent>
|
||||||
|
|
||||||
|
<artifactId>datax-example-core</artifactId>
|
||||||
|
|
||||||
|
<properties>
|
||||||
|
<maven.compiler.source>8</maven.compiler.source>
|
||||||
|
<maven.compiler.target>8</maven.compiler.target>
|
||||||
|
<project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
|
||||||
|
</properties>
|
||||||
|
|
||||||
|
</project>
|
||||||
@ -0,0 +1,26 @@
|
|||||||
|
package com.alibaba.datax.example;
|
||||||
|
|
||||||
|
import com.alibaba.datax.common.util.Configuration;
|
||||||
|
import com.alibaba.datax.core.Engine;
|
||||||
|
import com.alibaba.datax.example.util.ExampleConfigParser;
|
||||||
|
|
||||||
|
/**
|
||||||
|
* {@code Date} 2023/8/6 11:22
|
||||||
|
*
|
||||||
|
* @author fuyouj
|
||||||
|
*/
|
||||||
|
|
||||||
|
public class ExampleContainer {
|
||||||
|
/**
|
||||||
|
* example对外暴露的启动入口
|
||||||
|
* 使用前最好看下 datax-example/doc/README.MD
|
||||||
|
* @param jobPath 任务json绝对路径
|
||||||
|
*/
|
||||||
|
public static void start(String jobPath) {
|
||||||
|
|
||||||
|
Configuration configuration = ExampleConfigParser.parse(jobPath);
|
||||||
|
|
||||||
|
Engine engine = new Engine();
|
||||||
|
engine.start(configuration);
|
||||||
|
}
|
||||||
|
}
|
||||||
@ -0,0 +1,23 @@
|
|||||||
|
package com.alibaba.datax.example;
|
||||||
|
|
||||||
|
|
||||||
|
import com.alibaba.datax.example.util.PathUtil;
|
||||||
|
|
||||||
|
/**
|
||||||
|
* @author fuyouj
|
||||||
|
*/
|
||||||
|
public class Main {
|
||||||
|
|
||||||
|
/**
|
||||||
|
* 1.在example模块pom文件添加你依赖的的调试插件，
|
||||||
|
* 你可以直接打开本模块的pom文件,参考是如何引入streamreader，streamwriter
|
||||||
|
* 2. 在此处指定你的job文件
|
||||||
|
*/
|
||||||
|
public static void main(String[] args) {
|
||||||
|
|
||||||
|
String classPathJobPath = "/job/stream2stream.json";
|
||||||
|
String absJobPath = PathUtil.getAbsolutePathFromClassPath(classPathJobPath);
|
||||||
|
ExampleContainer.start(absJobPath);
|
||||||
|
}
|
||||||
|
|
||||||
|
}
|
||||||
@ -0,0 +1,154 @@
|
|||||||
|
package com.alibaba.datax.example.util;
|
||||||
|
|
||||||
|
import com.alibaba.datax.common.exception.DataXException;
|
||||||
|
import com.alibaba.datax.common.util.Configuration;
|
||||||
|
import com.alibaba.datax.core.util.ConfigParser;
|
||||||
|
import com.alibaba.datax.core.util.FrameworkErrorCode;
|
||||||
|
import com.alibaba.datax.core.util.container.CoreConstant;
|
||||||
|
|
||||||
|
import java.io.File;
|
||||||
|
import java.io.IOException;
|
||||||
|
import java.net.URL;
|
||||||
|
import java.nio.file.Paths;
|
||||||
|
import java.util.*;
|
||||||
|
|
||||||
|
/**
|
||||||
|
* @author fuyouj
|
||||||
|
*/
|
||||||
|
public class ExampleConfigParser {
|
||||||
|
private static final String CORE_CONF = "/example/conf/core.json";
|
||||||
|
|
||||||
|
private static final String PLUGIN_DESC_FILE = "plugin.json";
|
||||||
|
|
||||||
|
/**
|
||||||
|
* 指定Job配置路径，ConfigParser会解析Job、Plugin、Core全部信息，并以Configuration返回
|
||||||
|
* 不同于Core的ConfigParser,这里的core,plugin 不依赖于编译后的datax.home,而是扫描程序编译后的target目录
|
||||||
|
*/
|
||||||
|
public static Configuration parse(final String jobPath) {
|
||||||
|
|
||||||
|
Configuration configuration = ConfigParser.parseJobConfig(jobPath);
|
||||||
|
configuration.merge(coreConfig(),
|
||||||
|
false);
|
||||||
|
|
||||||
|
Map<String, String> pluginTypeMap = new HashMap<>();
|
||||||
|
String readerName = configuration.getString(CoreConstant.DATAX_JOB_CONTENT_READER_NAME);
|
||||||
|
String writerName = configuration.getString(CoreConstant.DATAX_JOB_CONTENT_WRITER_NAME);
|
||||||
|
pluginTypeMap.put(readerName, "reader");
|
||||||
|
pluginTypeMap.put(writerName, "writer");
|
||||||
|
Configuration pluginsDescConfig = parsePluginsConfig(pluginTypeMap);
|
||||||
|
configuration.merge(pluginsDescConfig, false);
|
||||||
|
return configuration;
|
||||||
|
}
|
||||||
|
|
||||||
|
private static Configuration parsePluginsConfig(Map<String, String> pluginTypeMap) {
|
||||||
|
|
||||||
|
Configuration configuration = Configuration.newDefault();
|
||||||
|
|
||||||
|
//最初打算通过user.dir获取工作目录来扫描插件，
|
||||||
|
//但是user.dir在不同有一些不确定性，所以废弃了这个选择
|
||||||
|
|
||||||
|
for (File basePackage : runtimeBasePackages()) {
|
||||||
|
if (pluginTypeMap.isEmpty()) {
|
||||||
|
break;
|
||||||
|
}
|
||||||
|
scanPluginByPackage(basePackage, configuration, basePackage.listFiles(), pluginTypeMap);
|
||||||
|
}
|
||||||
|
if (!pluginTypeMap.isEmpty()) {
|
||||||
|
String failedPlugin = pluginTypeMap.keySet().toString();
|
||||||
|
String message = "\nplugin %s load failed ：ry to analyze the reasons from the following aspects.。\n" +
|
||||||
|
"1: Check if the name of the plugin is spelled correctly, and verify whether DataX supports this plugin\n" +
|
||||||
|
"2：Verify if the <resource></resource> tag has been added under <build></build> section in the pom file of the relevant plugin.\n<resource>" +
|
||||||
|
" <directory>src/main/resources</directory>\n" +
|
||||||
|
" <includes>\n" +
|
||||||
|
" <include>**/*.*</include>\n" +
|
||||||
|
" </includes>\n" +
|
||||||
|
" <filtering>true</filtering>\n" +
|
||||||
|
" </resource>\n [Refer to the streamreader pom file] \n" +
|
||||||
|
"3: Check that the datax-yourPlugin-example module imported your test plugin";
|
||||||
|
message = String.format(message, failedPlugin);
|
||||||
|
throw DataXException.asDataXException(FrameworkErrorCode.PLUGIN_INIT_ERROR, message);
|
||||||
|
}
|
||||||
|
return configuration;
|
||||||
|
}
|
||||||
|
|
||||||
|
/**
|
||||||
|
* 通过classLoader获取程序编译的输出目录
|
||||||
|
*
|
||||||
|
* @return File[/datax-example/target/classes,xxReader/target/classes,xxWriter/target/classes]
|
||||||
|
*/
|
||||||
|
private static File[] runtimeBasePackages() {
|
||||||
|
List<File> basePackages = new ArrayList<>();
|
||||||
|
ClassLoader classLoader = Thread.currentThread().getContextClassLoader();
|
||||||
|
Enumeration<URL> resources = null;
|
||||||
|
try {
|
||||||
|
resources = classLoader.getResources("");
|
||||||
|
} catch (IOException e) {
|
||||||
|
throw DataXException.asDataXException(e.getMessage());
|
||||||
|
}
|
||||||
|
|
||||||
|
while (resources.hasMoreElements()) {
|
||||||
|
URL resource = resources.nextElement();
|
||||||
|
File file = new File(resource.getFile());
|
||||||
|
if (file.isDirectory()) {
|
||||||
|
basePackages.add(file);
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
return basePackages.toArray(new File[0]);
|
||||||
|
}
|
||||||
|
|
||||||
|
/**
|
||||||
|
* @param packageFile 编译出来的target/classes根目录 便于找到插件时设置插件的URL目录，设置根目录是最保险的方式
|
||||||
|
* @param configuration pluginConfig
|
||||||
|
* @param files 待扫描文件
|
||||||
|
* @param needPluginTypeMap 需要的插件
|
||||||
|
*/
|
||||||
|
private static void scanPluginByPackage(File packageFile,
|
||||||
|
Configuration configuration,
|
||||||
|
File[] files,
|
||||||
|
Map<String, String> needPluginTypeMap) {
|
||||||
|
if (files == null) {
|
||||||
|
return;
|
||||||
|
}
|
||||||
|
for (File file : files) {
|
||||||
|
if (file.isFile() && PLUGIN_DESC_FILE.equals(file.getName())) {
|
||||||
|
Configuration pluginDesc = Configuration.from(file);
|
||||||
|
String descPluginName = pluginDesc.getString("name", "");
|
||||||
|
|
||||||
|
if (needPluginTypeMap.containsKey(descPluginName)) {
|
||||||
|
|
||||||
|
String type = needPluginTypeMap.get(descPluginName);
|
||||||
|
configuration.merge(parseOnePlugin(packageFile.getAbsolutePath(), type, descPluginName, pluginDesc), false);
|
||||||
|
needPluginTypeMap.remove(descPluginName);
|
||||||
|
|
||||||
|
}
|
||||||
|
} else {
|
||||||
|
scanPluginByPackage(packageFile, configuration, file.listFiles(), needPluginTypeMap);
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
|
||||||
|
private static Configuration parseOnePlugin(String packagePath,
|
||||||
|
String pluginType,
|
||||||
|
String pluginName,
|
||||||
|
Configuration pluginDesc) {
|
||||||
|
//设置path 兼容jarLoader的加载方式URLClassLoader
|
||||||
|
pluginDesc.set("path", packagePath);
|
||||||
|
Configuration pluginConfInJob = Configuration.newDefault();
|
||||||
|
pluginConfInJob.set(
|
||||||
|
String.format("plugin.%s.%s", pluginType, pluginName),
|
||||||
|
pluginDesc.getInternal());
|
||||||
|
return pluginConfInJob;
|
||||||
|
}
|
||||||
|
|
||||||
|
private static Configuration coreConfig() {
|
||||||
|
try {
|
||||||
|
URL resource = ExampleConfigParser.class.getResource(CORE_CONF);
|
||||||
|
return Configuration.from(Paths.get(resource.toURI()).toFile());
|
||||||
|
} catch (Exception ignore) {
|
||||||
|
throw DataXException.asDataXException("Failed to load the configuration file core.json. " +
|
||||||
|
"Please check whether /example/conf/core.json exists!");
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
@ -0,0 +1,26 @@
|
|||||||
|
package com.alibaba.datax.example.util;
|
||||||
|
|
||||||
|
|
||||||
|
import com.alibaba.datax.common.exception.DataXException;
|
||||||
|
|
||||||
|
import java.net.URI;
|
||||||
|
import java.net.URISyntaxException;
|
||||||
|
import java.net.URL;
|
||||||
|
import java.nio.file.Paths;
|
||||||
|
|
||||||
|
/**
|
||||||
|
* @author fuyouj
|
||||||
|
*/
|
||||||
|
public class PathUtil {
|
||||||
|
public static String getAbsolutePathFromClassPath(String path) {
|
||||||
|
URL resource = PathUtil.class.getResource(path);
|
||||||
|
try {
|
||||||
|
assert resource != null;
|
||||||
|
URI uri = resource.toURI();
|
||||||
|
return Paths.get(uri).toString();
|
||||||
|
} catch (NullPointerException | URISyntaxException e) {
|
||||||
|
throw DataXException.asDataXException("path error,please check whether the path is correct");
|
||||||
|
}
|
||||||
|
|
||||||
|
}
|
||||||
|
}
|
||||||
60
datax-example/datax-example-core/src/main/resources/example/conf/core.json
Executable file
60
datax-example/datax-example-core/src/main/resources/example/conf/core.json
Executable file
@ -0,0 +1,60 @@
|
|||||||
|
{
|
||||||
|
"entry": {
|
||||||
|
"jvm": "-Xms1G -Xmx1G",
|
||||||
|
"environment": {}
|
||||||
|
},
|
||||||
|
"common": {
|
||||||
|
"column": {
|
||||||
|
"datetimeFormat": "yyyy-MM-dd HH:mm:ss",
|
||||||
|
"timeFormat": "HH:mm:ss",
|
||||||
|
"dateFormat": "yyyy-MM-dd",
|
||||||
|
"extraFormats":["yyyyMMdd"],
|
||||||
|
"timeZone": "GMT+8",
|
||||||
|
"encoding": "utf-8"
|
||||||
|
}
|
||||||
|
},
|
||||||
|
"core": {
|
||||||
|
"dataXServer": {
|
||||||
|
"address": "http://localhost:7001/api",
|
||||||
|
"timeout": 10000,
|
||||||
|
"reportDataxLog": false,
|
||||||
|
"reportPerfLog": false
|
||||||
|
},
|
||||||
|
"transport": {
|
||||||
|
"channel": {
|
||||||
|
"class": "com.alibaba.datax.core.transport.channel.memory.MemoryChannel",
|
||||||
|
"speed": {
|
||||||
|
"byte": -1,
|
||||||
|
"record": -1
|
||||||
|
},
|
||||||
|
"flowControlInterval": 20,
|
||||||
|
"capacity": 512,
|
||||||
|
"byteCapacity": 67108864
|
||||||
|
},
|
||||||
|
"exchanger": {
|
||||||
|
"class": "com.alibaba.datax.core.plugin.BufferedRecordExchanger",
|
||||||
|
"bufferSize": 32
|
||||||
|
}
|
||||||
|
},
|
||||||
|
"container": {
|
||||||
|
"job": {
|
||||||
|
"reportInterval": 10000
|
||||||
|
},
|
||||||
|
"taskGroup": {
|
||||||
|
"channel": 5
|
||||||
|
},
|
||||||
|
"trace": {
|
||||||
|
"enable": "false"
|
||||||
|
}
|
||||||
|
|
||||||
|
},
|
||||||
|
"statistics": {
|
||||||
|
"collector": {
|
||||||
|
"plugin": {
|
||||||
|
"taskClass": "com.alibaba.datax.core.statistics.plugin.task.StdoutPluginCollector",
|
||||||
|
"maxDirtyNumber": 10
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
@ -0,0 +1,19 @@
|
|||||||
|
package com.alibaba.datax.example.util;
|
||||||
|
|
||||||
|
import org.junit.Assert;
|
||||||
|
import org.junit.Test;
|
||||||
|
|
||||||
|
/**
|
||||||
|
* {@code Author} FuYouJ
|
||||||
|
* {@code Date} 2023/8/19 21:38
|
||||||
|
*/
|
||||||
|
|
||||||
|
public class PathUtilTest {
|
||||||
|
|
||||||
|
@Test
|
||||||
|
public void testParseClassPathFile() {
|
||||||
|
String path = "/pathTest.json";
|
||||||
|
String absolutePathFromClassPath = PathUtil.getAbsolutePathFromClassPath(path);
|
||||||
|
Assert.assertNotNull(absolutePathFromClassPath);
|
||||||
|
}
|
||||||
|
}
|
||||||
@ -0,0 +1 @@
|
|||||||
|
{}
|
||||||
43
datax-example/datax-example-neo4j/pom.xml
Normal file
43
datax-example/datax-example-neo4j/pom.xml
Normal file
@ -0,0 +1,43 @@
|
|||||||
|
<?xml version="1.0" encoding="UTF-8"?>
|
||||||
|
<project xmlns="http://maven.apache.org/POM/4.0.0"
|
||||||
|
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
|
||||||
|
xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
|
||||||
|
<modelVersion>4.0.0</modelVersion>
|
||||||
|
<parent>
|
||||||
|
<groupId>com.alibaba.datax</groupId>
|
||||||
|
<artifactId>datax-example</artifactId>
|
||||||
|
<version>0.0.1-SNAPSHOT</version>
|
||||||
|
</parent>
|
||||||
|
|
||||||
|
<artifactId>datax-example-neo4j</artifactId>
|
||||||
|
|
||||||
|
<properties>
|
||||||
|
<maven.compiler.source>8</maven.compiler.source>
|
||||||
|
<maven.compiler.target>8</maven.compiler.target>
|
||||||
|
<project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
|
||||||
|
<test.container.version>1.17.6</test.container.version>
|
||||||
|
<neo4j-java-driver.version>4.4.9</neo4j-java-driver.version>
|
||||||
|
</properties>
|
||||||
|
<dependencies>
|
||||||
|
<dependency>
|
||||||
|
<groupId>com.alibaba.datax</groupId>
|
||||||
|
<artifactId>datax-example-core</artifactId>
|
||||||
|
<version>0.0.1-SNAPSHOT</version>
|
||||||
|
</dependency>
|
||||||
|
<dependency>
|
||||||
|
<groupId>org.testcontainers</groupId>
|
||||||
|
<artifactId>testcontainers</artifactId>
|
||||||
|
<version>${test.container.version}</version>
|
||||||
|
</dependency>
|
||||||
|
<dependency>
|
||||||
|
<groupId>com.alibaba.datax</groupId>
|
||||||
|
<artifactId>neo4jwriter</artifactId>
|
||||||
|
<version>0.0.1-SNAPSHOT</version>
|
||||||
|
</dependency>
|
||||||
|
<dependency>
|
||||||
|
<groupId>com.alibaba.datax</groupId>
|
||||||
|
<artifactId>datax-example-streamreader</artifactId>
|
||||||
|
<version>0.0.1-SNAPSHOT</version>
|
||||||
|
</dependency>
|
||||||
|
</dependencies>
|
||||||
|
</project>
|
||||||
@ -0,0 +1,138 @@
|
|||||||
|
package com.alibaba.datax.example.neo4j;
|
||||||
|
|
||||||
|
import com.alibaba.datax.example.ExampleContainer;
|
||||||
|
import com.alibaba.datax.example.util.PathUtil;
|
||||||
|
import org.junit.After;
|
||||||
|
import org.junit.Assert;
|
||||||
|
import org.junit.Before;
|
||||||
|
import org.junit.Test;
|
||||||
|
import org.neo4j.driver.*;
|
||||||
|
import org.neo4j.driver.types.Node;
|
||||||
|
import org.slf4j.Logger;
|
||||||
|
import org.slf4j.LoggerFactory;
|
||||||
|
import org.testcontainers.containers.GenericContainer;
|
||||||
|
import org.testcontainers.containers.Network;
|
||||||
|
import org.testcontainers.containers.output.Slf4jLogConsumer;
|
||||||
|
import org.testcontainers.lifecycle.Startables;
|
||||||
|
import org.testcontainers.shaded.org.awaitility.Awaitility;
|
||||||
|
import org.testcontainers.utility.DockerImageName;
|
||||||
|
import org.testcontainers.utility.DockerLoggerFactory;
|
||||||
|
|
||||||
|
import java.net.URI;
|
||||||
|
import java.util.Arrays;
|
||||||
|
import java.util.concurrent.TimeUnit;
|
||||||
|
import java.util.stream.Stream;
|
||||||
|
|
||||||
|
/**
|
||||||
|
* {@code Author} FuYouJ
|
||||||
|
* {@code Date} 2023/8/19 21:48
|
||||||
|
*/
|
||||||
|
|
||||||
|
public class StreamReader2Neo4jWriterTest {
|
||||||
|
private static final Logger LOGGER = LoggerFactory.getLogger(StreamReader2Neo4jWriterTest.class);
|
||||||
|
private static final String CONTAINER_IMAGE = "neo4j:5.9.0";
|
||||||
|
|
||||||
|
private static final String CONTAINER_HOST = "neo4j-host";
|
||||||
|
private static final int HTTP_PORT = 7474;
|
||||||
|
private static final int BOLT_PORT = 7687;
|
||||||
|
private static final String CONTAINER_NEO4J_USERNAME = "neo4j";
|
||||||
|
private static final String CONTAINER_NEO4J_PASSWORD = "Test@12343";
|
||||||
|
private static final URI CONTAINER_URI = URI.create("neo4j://localhost:" + BOLT_PORT);
|
||||||
|
|
||||||
|
protected static final Network NETWORK = Network.newNetwork();
|
||||||
|
|
||||||
|
private GenericContainer<?> container;
|
||||||
|
protected Driver neo4jDriver;
|
||||||
|
protected Session neo4jSession;
|
||||||
|
private static final int CHANNEL = 5;
|
||||||
|
private static final int READER_NUM = 10;
|
||||||
|
|
||||||
|
@Before
|
||||||
|
public void init() {
|
||||||
|
DockerImageName imageName = DockerImageName.parse(CONTAINER_IMAGE);
|
||||||
|
container =
|
||||||
|
new GenericContainer<>(imageName)
|
||||||
|
.withNetwork(NETWORK)
|
||||||
|
.withNetworkAliases(CONTAINER_HOST)
|
||||||
|
.withExposedPorts(HTTP_PORT, BOLT_PORT)
|
||||||
|
.withEnv(
|
||||||
|
"NEO4J_AUTH",
|
||||||
|
CONTAINER_NEO4J_USERNAME + "/" + CONTAINER_NEO4J_PASSWORD)
|
||||||
|
.withEnv("apoc.export.file.enabled", "true")
|
||||||
|
.withEnv("apoc.import.file.enabled", "true")
|
||||||
|
.withEnv("apoc.import.file.use_neo4j_config", "true")
|
||||||
|
.withEnv("NEO4J_PLUGINS", "[\"apoc\"]")
|
||||||
|
.withLogConsumer(
|
||||||
|
new Slf4jLogConsumer(
|
||||||
|
DockerLoggerFactory.getLogger(CONTAINER_IMAGE)));
|
||||||
|
container.setPortBindings(
|
||||||
|
Arrays.asList(
|
||||||
|
String.format("%s:%s", HTTP_PORT, HTTP_PORT),
|
||||||
|
String.format("%s:%s", BOLT_PORT, BOLT_PORT)));
|
||||||
|
Startables.deepStart(Stream.of(container)).join();
|
||||||
|
LOGGER.info("container started");
|
||||||
|
Awaitility.given()
|
||||||
|
.ignoreExceptions()
|
||||||
|
.await()
|
||||||
|
.atMost(30, TimeUnit.SECONDS)
|
||||||
|
.untilAsserted(this::initConnection);
|
||||||
|
}
|
||||||
|
|
||||||
|
//在neo4jWriter模块使用Example测试整个job,方便发现整个流程的代码问题
|
||||||
|
@Test
|
||||||
|
public void streamReader2Neo4j() {
|
||||||
|
|
||||||
|
deleteHistoryIfExist();
|
||||||
|
|
||||||
|
String path = "/streamreader2neo4j.json";
|
||||||
|
String jobPath = PathUtil.getAbsolutePathFromClassPath(path);
|
||||||
|
|
||||||
|
ExampleContainer.start(jobPath);
|
||||||
|
|
||||||
|
//根据channel和reader的mock数据，校验结果集是否符合预期
|
||||||
|
verifyWriteResult();
|
||||||
|
}
|
||||||
|
|
||||||
|
private void deleteHistoryIfExist() {
|
||||||
|
String query = "match (n:StreamReader) return n limit 1";
|
||||||
|
String delete = "match (n:StreamReader) delete n";
|
||||||
|
if (neo4jSession.run(query).hasNext()) {
|
||||||
|
neo4jSession.run(delete);
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
private void verifyWriteResult() {
|
||||||
|
int total = CHANNEL * READER_NUM;
|
||||||
|
String query = "match (n:StreamReader) return n";
|
||||||
|
Result run = neo4jSession.run(query);
|
||||||
|
int count = 0;
|
||||||
|
while (run.hasNext()) {
|
||||||
|
Record record = run.next();
|
||||||
|
Node node = record.get("n").asNode();
|
||||||
|
if (node.hasLabel("StreamReader")) {
|
||||||
|
count++;
|
||||||
|
}
|
||||||
|
}
|
||||||
|
Assert.assertEquals(count, total);
|
||||||
|
}
|
||||||
|
@After
|
||||||
|
public void destroy() {
|
||||||
|
if (neo4jSession != null) {
|
||||||
|
neo4jSession.close();
|
||||||
|
}
|
||||||
|
if (neo4jDriver != null) {
|
||||||
|
neo4jDriver.close();
|
||||||
|
}
|
||||||
|
if (container != null) {
|
||||||
|
container.close();
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
private void initConnection() {
|
||||||
|
neo4jDriver =
|
||||||
|
GraphDatabase.driver(
|
||||||
|
CONTAINER_URI,
|
||||||
|
AuthTokens.basic(CONTAINER_NEO4J_USERNAME, CONTAINER_NEO4J_PASSWORD));
|
||||||
|
neo4jSession = neo4jDriver.session(SessionConfig.forDatabase("neo4j"));
|
||||||
|
}
|
||||||
|
}
|
||||||
@ -0,0 +1,51 @@
|
|||||||
|
{
|
||||||
|
"job": {
|
||||||
|
"content": [
|
||||||
|
{
|
||||||
|
"reader": {
|
||||||
|
"name": "streamreader",
|
||||||
|
"parameter": {
|
||||||
|
"sliceRecordCount": 10,
|
||||||
|
"column": [
|
||||||
|
{
|
||||||
|
"type": "string",
|
||||||
|
"value": "StreamReader"
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"type": "string",
|
||||||
|
"value": "1997"
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
|
},
|
||||||
|
"writer": {
|
||||||
|
"name": "neo4jWriter",
|
||||||
|
"parameter": {
|
||||||
|
"uri": "bolt://localhost:7687",
|
||||||
|
"username":"neo4j",
|
||||||
|
"password":"Test@12343",
|
||||||
|
"database":"neo4j",
|
||||||
|
"cypher": "unwind $batch as row CALL apoc.cypher.doIt( 'create (n:`' + row.Label + '`{id:$id})' ,{id: row.id} ) YIELD value RETURN 1 ",
|
||||||
|
"batchDataVariableName": "batch",
|
||||||
|
"batchSize": "3",
|
||||||
|
"properties": [
|
||||||
|
{
|
||||||
|
"name": "Label",
|
||||||
|
"type": "string"
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"name": "id",
|
||||||
|
"type": "STRING"
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
],
|
||||||
|
"setting": {
|
||||||
|
"speed": {
|
||||||
|
"channel": 5
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
37
datax-example/datax-example-streamreader/pom.xml
Normal file
37
datax-example/datax-example-streamreader/pom.xml
Normal file
@ -0,0 +1,37 @@
|
|||||||
|
<?xml version="1.0" encoding="UTF-8"?>
|
||||||
|
<project xmlns="http://maven.apache.org/POM/4.0.0"
|
||||||
|
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
|
||||||
|
xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
|
||||||
|
<modelVersion>4.0.0</modelVersion>
|
||||||
|
<parent>
|
||||||
|
<groupId>com.alibaba.datax</groupId>
|
||||||
|
<artifactId>datax-example</artifactId>
|
||||||
|
<version>0.0.1-SNAPSHOT</version>
|
||||||
|
</parent>
|
||||||
|
|
||||||
|
<artifactId>datax-example-streamreader</artifactId>
|
||||||
|
|
||||||
|
<properties>
|
||||||
|
<maven.compiler.source>8</maven.compiler.source>
|
||||||
|
<maven.compiler.target>8</maven.compiler.target>
|
||||||
|
<project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
|
||||||
|
</properties>
|
||||||
|
<dependencies>
|
||||||
|
<dependency>
|
||||||
|
<groupId>com.alibaba.datax</groupId>
|
||||||
|
<artifactId>datax-example-core</artifactId>
|
||||||
|
<version>0.0.1-SNAPSHOT</version>
|
||||||
|
</dependency>
|
||||||
|
<dependency>
|
||||||
|
<groupId>com.alibaba.datax</groupId>
|
||||||
|
<artifactId>streamreader</artifactId>
|
||||||
|
<version>0.0.1-SNAPSHOT</version>
|
||||||
|
</dependency>
|
||||||
|
<dependency>
|
||||||
|
<groupId>com.alibaba.datax</groupId>
|
||||||
|
<artifactId>streamwriter</artifactId>
|
||||||
|
<version>0.0.1-SNAPSHOT</version>
|
||||||
|
</dependency>
|
||||||
|
</dependencies>
|
||||||
|
|
||||||
|
</project>
|
||||||
@ -0,0 +1,19 @@
|
|||||||
|
package com.alibaba.datax.example.streamreader;
|
||||||
|
|
||||||
|
import com.alibaba.datax.example.ExampleContainer;
|
||||||
|
import com.alibaba.datax.example.util.PathUtil;
|
||||||
|
import org.junit.Test;
|
||||||
|
|
||||||
|
/**
|
||||||
|
* {@code Author} FuYouJ
|
||||||
|
* {@code Date} 2023/8/14 20:16
|
||||||
|
*/
|
||||||
|
|
||||||
|
public class StreamReader2StreamWriterTest {
|
||||||
|
@Test
|
||||||
|
public void testStreamReader2StreamWriter() {
|
||||||
|
String path = "/stream2stream.json";
|
||||||
|
String jobPath = PathUtil.getAbsolutePathFromClassPath(path);
|
||||||
|
ExampleContainer.start(jobPath);
|
||||||
|
}
|
||||||
|
}
|
||||||
@ -0,0 +1,36 @@
|
|||||||
|
{
|
||||||
|
"job": {
|
||||||
|
"content": [
|
||||||
|
{
|
||||||
|
"reader": {
|
||||||
|
"name": "streamreader",
|
||||||
|
"parameter": {
|
||||||
|
"sliceRecordCount": 10,
|
||||||
|
"column": [
|
||||||
|
{
|
||||||
|
"type": "long",
|
||||||
|
"value": "10"
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"type": "string",
|
||||||
|
"value": "hello，你好，世界-DataX"
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
|
},
|
||||||
|
"writer": {
|
||||||
|
"name": "streamwriter",
|
||||||
|
"parameter": {
|
||||||
|
"encoding": "UTF-8",
|
||||||
|
"print": true
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
],
|
||||||
|
"setting": {
|
||||||
|
"speed": {
|
||||||
|
"channel": 5
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
107
datax-example/doc/README.md
Normal file
107
datax-example/doc/README.md
Normal file
@ -0,0 +1,107 @@
|
|||||||
|
## [DataX-Example]调试datax插件的模块
|
||||||
|
|
||||||
|
### 为什么要开发这个模块
|
||||||
|
|
||||||
|
一般使用DataX启动数据同步任务是从datax.py 脚本开始，获取程序datax包目录设置到系统变量datax.home里，此后系统核心插件的加载，配置初始化均依赖于变量datax.home,这带来了一些麻烦，以一次本地 DeBug streamreader 插件为例。
|
||||||
|
|
||||||
|
- maven 打包 datax 生成 datax 目录
|
||||||
|
- 在 IDE 中 设置系统环境变量 datax.home，或者在Engine启动类中硬编码设置datax.home。
|
||||||
|
- 修改插件 streamreader 代码
|
||||||
|
- 再次 maven 打包，使JarLoader 能够加载到最新的 streamreader 代码。
|
||||||
|
- 调试代码
|
||||||
|
|
||||||
|
在以上步骤中，打包完全不必要且最耗时,等待打包也最煎熬。
|
||||||
|
|
||||||
|
所以我编写一个新的模块(datax-example)，此模块特用于本地调试和复现 BUG。如果模块顺利编写完成，那么以上流程将被简化至两步。
|
||||||
|
|
||||||
|
- 修改插件 streamreader 代码。
|
||||||
|
- 调试代码
|
||||||
|
|
||||||
|
<img src="img/img01.png" alt="img" style="zoom:40%;" />
|
||||||
|
|
||||||
|
### 目录结构
|
||||||
|
该目录结构演示了如何使用datax-example-core编写测试用例，和校验代码流程。
|
||||||
|
<img src="img/img03.png" alt="img" style="zoom:100%;" />
|
||||||
|
|
||||||
|
### 实现原理
|
||||||
|
|
||||||
|
- 不修改原有的ConfigParer,使用新的ExampleConfigParser,仅用于example模块。他不依赖datax.home,而是依赖ide编译后的target目录
|
||||||
|
- 将ide的target目录作为每个插件的目录类加载目录。
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
### 如何使用
|
||||||
|
1.修改插件的pom文件，做如下改动。以streamreader为例。<br/>
|
||||||
|
改动前
|
||||||
|
```xml
|
||||||
|
<build>
|
||||||
|
<plugins>
|
||||||
|
<!-- compiler plugin -->
|
||||||
|
<plugin>
|
||||||
|
<artifactId>maven-compiler-plugin</artifactId>
|
||||||
|
<configuration>
|
||||||
|
<source>${jdk-version}</source>
|
||||||
|
<target>${jdk-version}</target>
|
||||||
|
<encoding>${project-sourceEncoding}</encoding>
|
||||||
|
</configuration>
|
||||||
|
</plugin>
|
||||||
|
</plugins>
|
||||||
|
</build>
|
||||||
|
```
|
||||||
|
改动后
|
||||||
|
```xml
|
||||||
|
<build>
|
||||||
|
<resources>
|
||||||
|
<!--将resource目录也输出到target-->
|
||||||
|
<resource>
|
||||||
|
<directory>src/main/resources</directory>
|
||||||
|
<includes>
|
||||||
|
<include>**/*.*</include>
|
||||||
|
</includes>
|
||||||
|
<filtering>true</filtering>
|
||||||
|
</resource>
|
||||||
|
</resources>
|
||||||
|
<plugins>
|
||||||
|
<!-- compiler plugin -->
|
||||||
|
<plugin>
|
||||||
|
<artifactId>maven-compiler-plugin</artifactId>
|
||||||
|
<configuration>
|
||||||
|
<source>${jdk-version}</source>
|
||||||
|
<target>${jdk-version}</target>
|
||||||
|
<encoding>${project-sourceEncoding}</encoding>
|
||||||
|
</configuration>
|
||||||
|
</plugin>
|
||||||
|
</plugins>
|
||||||
|
</build>
|
||||||
|
```
|
||||||
|
#### 在测试模块模块使用
|
||||||
|
参考datax-example/datax-example-streamreader的StreamReader2StreamWriterTest.java
|
||||||
|
```java
|
||||||
|
public class StreamReader2StreamWriterTest {
|
||||||
|
@Test
|
||||||
|
public void testStreamReader2StreamWriter() {
|
||||||
|
String path = "/stream2stream.json";
|
||||||
|
String jobPath = PathUtil.getAbsolutePathFromClassPath(path);
|
||||||
|
ExampleContainer.start(jobPath);
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
```
|
||||||
|
参考datax-example/datax-example-neo4j的StreamReader2Neo4jWriterTest
|
||||||
|
```java
|
||||||
|
public class StreamReader2Neo4jWriterTest{
|
||||||
|
@Test
|
||||||
|
public void streamReader2Neo4j() {
|
||||||
|
|
||||||
|
deleteHistoryIfExist();
|
||||||
|
|
||||||
|
String path = "/streamreader2neo4j.json";
|
||||||
|
String jobPath = PathUtil.getAbsolutePathFromClassPath(path);
|
||||||
|
|
||||||
|
ExampleContainer.start(jobPath);
|
||||||
|
|
||||||
|
//根据channel和reader的mock数据，校验结果集是否符合预期
|
||||||
|
verifyWriteResult();
|
||||||
|
}
|
||||||
|
}
|
||||||
|
```
|
||||||
BIN
datax-example/doc/img/img01.png
Normal file
BIN
datax-example/doc/img/img01.png
Normal file
Binary file not shown.
|
After Width: | Height: | Size: 71 KiB |
BIN
datax-example/doc/img/img02.png
Normal file
BIN
datax-example/doc/img/img02.png
Normal file
Binary file not shown.
|
After Width: | Height: | Size: 66 KiB |
BIN
datax-example/doc/img/img03.png
Normal file
BIN
datax-example/doc/img/img03.png
Normal file
Binary file not shown.
|
After Width: | Height: | Size: 43 KiB |
68
datax-example/pom.xml
Normal file
68
datax-example/pom.xml
Normal file
@ -0,0 +1,68 @@
|
|||||||
|
<?xml version="1.0" encoding="UTF-8"?>
|
||||||
|
<project xmlns="http://maven.apache.org/POM/4.0.0"
|
||||||
|
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
|
||||||
|
xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
|
||||||
|
<modelVersion>4.0.0</modelVersion>
|
||||||
|
<parent>
|
||||||
|
<groupId>com.alibaba.datax</groupId>
|
||||||
|
<artifactId>datax-all</artifactId>
|
||||||
|
<version>0.0.1-SNAPSHOT</version>
|
||||||
|
</parent>
|
||||||
|
|
||||||
|
<artifactId>datax-example</artifactId>
|
||||||
|
<packaging>pom</packaging>
|
||||||
|
<modules>
|
||||||
|
<module>datax-example-core</module>
|
||||||
|
<module>datax-example-streamreader</module>
|
||||||
|
<module>datax-example-neo4j</module>
|
||||||
|
</modules>
|
||||||
|
|
||||||
|
<properties>
|
||||||
|
<maven.compiler.source>8</maven.compiler.source>
|
||||||
|
<maven.compiler.target>8</maven.compiler.target>
|
||||||
|
<project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
|
||||||
|
<junit4.version>4.13.2</junit4.version>
|
||||||
|
</properties>
|
||||||
|
<dependencies>
|
||||||
|
<dependency>
|
||||||
|
<groupId>com.alibaba.datax</groupId>
|
||||||
|
<artifactId>datax-common</artifactId>
|
||||||
|
<version>0.0.1-SNAPSHOT</version>
|
||||||
|
</dependency>
|
||||||
|
<dependency>
|
||||||
|
<groupId>com.alibaba.datax</groupId>
|
||||||
|
<artifactId>datax-core</artifactId>
|
||||||
|
<version>0.0.1-SNAPSHOT</version>
|
||||||
|
</dependency>
|
||||||
|
<dependency>
|
||||||
|
<groupId>junit</groupId>
|
||||||
|
<artifactId>junit</artifactId>
|
||||||
|
<version>${junit4.version}</version>
|
||||||
|
<scope>test</scope>
|
||||||
|
</dependency>
|
||||||
|
</dependencies>
|
||||||
|
|
||||||
|
<build>
|
||||||
|
<resources>
|
||||||
|
<resource>
|
||||||
|
<directory>src/main/resources</directory>
|
||||||
|
<includes>
|
||||||
|
<include>**/*.*</include>
|
||||||
|
</includes>
|
||||||
|
<filtering>true</filtering>
|
||||||
|
</resource>
|
||||||
|
</resources>
|
||||||
|
<plugins>
|
||||||
|
<!-- compiler plugin -->
|
||||||
|
<plugin>
|
||||||
|
<artifactId>maven-compiler-plugin</artifactId>
|
||||||
|
<configuration>
|
||||||
|
<source>${jdk-version}</source>
|
||||||
|
<target>${jdk-version}</target>
|
||||||
|
<encoding>${project-sourceEncoding}</encoding>
|
||||||
|
</configuration>
|
||||||
|
</plugin>
|
||||||
|
</plugins>
|
||||||
|
</build>
|
||||||
|
|
||||||
|
</project>
|
||||||
@ -447,6 +447,9 @@ DataX的内部类型在实现上会选用不同的java类型：
|
|||||||
3. 用户在插件中在`reader`/`writer`配置的`name`字段指定插件名字。框架根据插件的类型（`reader`/`writer`）和插件名称去插件的路径下扫描所有的jar，加入`classpath`。
|
3. 用户在插件中在`reader`/`writer`配置的`name`字段指定插件名字。框架根据插件的类型（`reader`/`writer`）和插件名称去插件的路径下扫描所有的jar，加入`classpath`。
|
||||||
4. 根据插件配置中定义的入口类，框架通过反射实例化对应的`Job`和`Task`对象。
|
4. 根据插件配置中定义的入口类，框架通过反射实例化对应的`Job`和`Task`对象。
|
||||||
|
|
||||||
|
### 编写测试用例
|
||||||
|
1. 在datax-example工程下新建新的插件测试模块,调用`ExampleContainer.start(jobPath)`方法来检测你的代码逻辑是否正确。[datax-example使用](https://github.com/alibaba/DataX/blob/master/datax-example/doc/README.md)
|
||||||
|
|
||||||
|
|
||||||
## 三、Last but not Least
|
## 三、Last but not Least
|
||||||
|
|
||||||
|
|||||||
@ -45,7 +45,7 @@
|
|||||||
<dependency>
|
<dependency>
|
||||||
<groupId>com.jcraft</groupId>
|
<groupId>com.jcraft</groupId>
|
||||||
<artifactId>jsch</artifactId>
|
<artifactId>jsch</artifactId>
|
||||||
<version>0.1.51</version>
|
<version>0.1.54</version>
|
||||||
</dependency>
|
</dependency>
|
||||||
<dependency>
|
<dependency>
|
||||||
<groupId>commons-net</groupId>
|
<groupId>commons-net</groupId>
|
||||||
|
|||||||
@ -64,6 +64,8 @@ public class SftpHelper extends FtpHelper {
|
|||||||
String message = String.format("请确认连接ftp服务器端口是否正确，错误的端口: [%s] ", port);
|
String message = String.format("请确认连接ftp服务器端口是否正确，错误的端口: [%s] ", port);
|
||||||
LOG.error(message);
|
LOG.error(message);
|
||||||
throw DataXException.asDataXException(FtpReaderErrorCode.FAIL_LOGIN, message, e);
|
throw DataXException.asDataXException(FtpReaderErrorCode.FAIL_LOGIN, message, e);
|
||||||
|
}else{
|
||||||
|
throw DataXException.asDataXException(FtpReaderErrorCode.COMMAND_FTP_IO_EXCEPTION, "", e);
|
||||||
}
|
}
|
||||||
}else {
|
}else {
|
||||||
if("Auth fail".equals(e.getMessage())){
|
if("Auth fail".equals(e.getMessage())){
|
||||||
|
|||||||
@ -45,7 +45,7 @@
|
|||||||
<dependency>
|
<dependency>
|
||||||
<groupId>com.jcraft</groupId>
|
<groupId>com.jcraft</groupId>
|
||||||
<artifactId>jsch</artifactId>
|
<artifactId>jsch</artifactId>
|
||||||
<version>0.1.51</version>
|
<version>0.1.54</version>
|
||||||
</dependency>
|
</dependency>
|
||||||
<dependency>
|
<dependency>
|
||||||
<groupId>commons-net</groupId>
|
<groupId>commons-net</groupId>
|
||||||
|
|||||||
297
gaussdbreader/doc/gaussdbreader.md
Normal file
297
gaussdbreader/doc/gaussdbreader.md
Normal file
@ -0,0 +1,297 @@
|
|||||||
|
|
||||||
|
# GaussDbReader 插件文档
|
||||||
|
|
||||||
|
|
||||||
|
___
|
||||||
|
|
||||||
|
|
||||||
|
## 1 快速介绍
|
||||||
|
|
||||||
|
GaussDbReader插件实现了从GaussDB读取数据。在底层实现上，GaussDbReader通过JDBC连接远程GaussDB数据库，并执行相应的sql语句将数据从GaussDB库中SELECT出来。
|
||||||
|
|
||||||
|
## 2 实现原理
|
||||||
|
|
||||||
|
简而言之，GaussDbReader通过JDBC连接器连接到远程的GaussDB数据库，并根据用户配置的信息生成查询SELECT SQL语句并发送到远程GaussDB数据库，并将该SQL执行返回结果使用DataX自定义的数据类型拼装为抽象的数据集，并传递给下游Writer处理。
|
||||||
|
|
||||||
|
对于用户配置Table、Column、Where的信息，GaussDbReader将其拼接为SQL语句发送到GaussDB数据库；对于用户配置querySql信息，GaussDbReader直接将其发送到GaussDB数据库。
|
||||||
|
|
||||||
|
|
||||||
|
## 3 功能说明
|
||||||
|
|
||||||
|
### 3.1 配置样例
|
||||||
|
|
||||||
|
* 配置一个从GaussDB数据库同步抽取数据到本地的作业:
|
||||||
|
|
||||||
|
```
|
||||||
|
{
|
||||||
|
"job": {
|
||||||
|
"setting": {
|
||||||
|
"speed": {
|
||||||
|
//设置传输速度，单位为byte/s，DataX运行会尽可能达到该速度但是不超过它.
|
||||||
|
"byte": 1048576
|
||||||
|
},
|
||||||
|
//出错限制
|
||||||
|
"errorLimit": {
|
||||||
|
//出错的record条数上限，当大于该值即报错。
|
||||||
|
"record": 0,
|
||||||
|
//出错的record百分比上限 1.0表示100%，0.02表示2%
|
||||||
|
"percentage": 0.02
|
||||||
|
}
|
||||||
|
},
|
||||||
|
"content": [
|
||||||
|
{
|
||||||
|
"reader": {
|
||||||
|
"name": "gaussdbreader",
|
||||||
|
"parameter": {
|
||||||
|
// 数据库连接用户名
|
||||||
|
"username": "xx",
|
||||||
|
// 数据库连接密码
|
||||||
|
"password": "xx",
|
||||||
|
"column": [
|
||||||
|
"id","name"
|
||||||
|
],
|
||||||
|
//切分主键
|
||||||
|
"splitPk": "id",
|
||||||
|
"connection": [
|
||||||
|
{
|
||||||
|
"table": [
|
||||||
|
"table"
|
||||||
|
],
|
||||||
|
"jdbcUrl": [
|
||||||
|
"jdbc:opengauss://host:port/database"
|
||||||
|
]
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
|
},
|
||||||
|
"writer": {
|
||||||
|
//writer类型
|
||||||
|
"name": "streamwriter",
|
||||||
|
//是否打印内容
|
||||||
|
"parameter": {
|
||||||
|
"print":true,
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
```
|
||||||
|
|
||||||
|
* 配置一个自定义SQL的数据库同步任务到本地内容的作业：
|
||||||
|
|
||||||
|
```json
|
||||||
|
{
|
||||||
|
"job": {
|
||||||
|
"setting": {
|
||||||
|
"speed": 1048576
|
||||||
|
},
|
||||||
|
"content": [
|
||||||
|
{
|
||||||
|
"reader": {
|
||||||
|
"name": "gaussdbreader",
|
||||||
|
"parameter": {
|
||||||
|
"username": "xx",
|
||||||
|
"password": "xx",
|
||||||
|
"where": "",
|
||||||
|
"connection": [
|
||||||
|
{
|
||||||
|
"querySql": [
|
||||||
|
"select db_id,on_line_flag from db_info where db_id < 10;"
|
||||||
|
],
|
||||||
|
"jdbcUrl": [
|
||||||
|
"jdbc:opengauss://host:port/database", "jdbc:opengauss://host:port/database"
|
||||||
|
]
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
|
},
|
||||||
|
"writer": {
|
||||||
|
"name": "streamwriter",
|
||||||
|
"parameter": {
|
||||||
|
"print": false,
|
||||||
|
"encoding": "UTF-8"
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
|
}
|
||||||
|
```
|
||||||
|
|
||||||
|
|
||||||
|
### 3.2 参数说明
|
||||||
|
|
||||||
|
* **jdbcUrl**
|
||||||
|
|
||||||
|
* 描述：描述的是到对端数据库的JDBC连接信息，使用JSON的数组描述，并支持一个库填写多个连接地址。之所以使用JSON数组描述连接信息，是因为阿里集团内部支持多个IP探测，如果配置了多个，GaussDbReader可以依次探测ip的可连接性，直到选择一个合法的IP。如果全部连接失败，GaussDbReader报错。 注意，jdbcUrl必须包含在connection配置单元中。对于阿里集团外部使用情况，JSON数组填写一个JDBC连接即可。
|
||||||
|
|
||||||
|
jdbcUrl按照GaussDB官方规范，并可以填写连接附件控制信息。具体请参看[GaussDB官方文档](https://docs.opengauss.org/zh/docs/3.1.0/docs/Developerguide/java-sql-Connection.html)。
|
||||||
|
|
||||||
|
* 必选：是 <br />
|
||||||
|
|
||||||
|
* 默认值：无 <br />
|
||||||
|
|
||||||
|
* **username**
|
||||||
|
|
||||||
|
* 描述：数据源的用户名 <br />
|
||||||
|
|
||||||
|
* 必选：是 <br />
|
||||||
|
|
||||||
|
* 默认值：无 <br />
|
||||||
|
|
||||||
|
* **password**
|
||||||
|
|
||||||
|
* 描述：数据源指定用户名的密码 <br />
|
||||||
|
|
||||||
|
* 必选：是 <br />
|
||||||
|
|
||||||
|
* 默认值：无 <br />
|
||||||
|
|
||||||
|
* **table**
|
||||||
|
|
||||||
|
* 描述：所选取的需要同步的表。使用JSON的数组描述，因此支持多张表同时抽取。当配置为多张表时，用户自己需保证多张表是同一schema结构，GaussDbReader不予检查表是否同一逻辑表。注意，table必须包含在connection配置单元中。<br />
|
||||||
|
|
||||||
|
* 必选：是 <br />
|
||||||
|
|
||||||
|
* 默认值：无 <br />
|
||||||
|
|
||||||
|
* **column**
|
||||||
|
|
||||||
|
* 描述：所配置的表中需要同步的列名集合，使用JSON的数组描述字段信息。用户使用\*代表默认使用所有列配置，例如['\*']。
|
||||||
|
|
||||||
|
支持列裁剪，即列可以挑选部分列进行导出。
|
||||||
|
|
||||||
|
支持列换序，即列可以不按照表schema信息进行导出。
|
||||||
|
|
||||||
|
支持常量配置，用户需要按照GaussDB语法格式:
|
||||||
|
["id", "'hello'::varchar", "true", "2.5::real", "power(2,3)"]
|
||||||
|
id为普通列名，'hello'::varchar为字符串常量，true为布尔值，2.5为浮点数, power(2,3)为函数。
|
||||||
|
|
||||||
|
**column必须用户显示指定同步的列集合，不允许为空！**
|
||||||
|
|
||||||
|
* 必选：是 <br />
|
||||||
|
|
||||||
|
* 默认值：无 <br />
|
||||||
|
|
||||||
|
* **splitPk**
|
||||||
|
|
||||||
|
* 描述：GaussDbReader进行数据抽取时，如果指定splitPk，表示用户希望使用splitPk代表的字段进行数据分片，DataX因此会启动并发任务进行数据同步，这样可以大大提高数据同步的效能。
|
||||||
|
|
||||||
|
推荐splitPk用户使用表主键，因为表主键通常情况下比较均匀，因此切分出来的分片也不容易出现数据热点。
|
||||||
|
|
||||||
|
目前splitPk仅支持整形数据切分，`不支持浮点、字符串型、日期等其他类型`。如果用户指定其他非支持类型，GaussDbReader将报错！
|
||||||
|
|
||||||
|
splitPk设置为空，底层将视作用户不允许对单表进行切分，因此使用单通道进行抽取。
|
||||||
|
|
||||||
|
* 必选：否 <br />
|
||||||
|
|
||||||
|
* 默认值：空 <br />
|
||||||
|
|
||||||
|
* **where**
|
||||||
|
|
||||||
|
* 描述：筛选条件，GaussDbReader根据指定的column、table、where条件拼接SQL，并根据这个SQL进行数据抽取。在实际业务场景中，往往会选择当天的数据进行同步，可以将where条件指定为gmt_create > $bizdate 。注意：不可以将where条件指定为limit 10，limit不是SQL的合法where子句。<br />
|
||||||
|
|
||||||
|
where条件可以有效地进行业务增量同步。 where条件不配置或者为空，视作全表同步数据。
|
||||||
|
|
||||||
|
* 必选：否 <br />
|
||||||
|
|
||||||
|
* 默认值：无 <br />
|
||||||
|
|
||||||
|
* **querySql**
|
||||||
|
|
||||||
|
* 描述：在有些业务场景下，where这一配置项不足以描述所筛选的条件，用户可以通过该配置型来自定义筛选SQL。当用户配置了这一项之后，DataX系统就会忽略table，column这些配置型，直接使用这个配置项的内容对数据进行筛选，例如需要进行多表join后同步数据，使用select a,b from table_a join table_b on table_a.id = table_b.id <br />
|
||||||
|
|
||||||
|
`当用户配置querySql时，GaussDbReader直接忽略table、column、where条件的配置`。
|
||||||
|
|
||||||
|
* 必选：否 <br />
|
||||||
|
|
||||||
|
* 默认值：无 <br />
|
||||||
|
|
||||||
|
* **fetchSize**
|
||||||
|
|
||||||
|
* 描述：该配置项定义了插件和数据库服务器端每次批量数据获取条数，该值决定了DataX和服务器端的网络交互次数，能够较大的提升数据抽取性能。<br />
|
||||||
|
|
||||||
|
`注意，该值过大(>2048)可能造成DataX进程OOM。`。
|
||||||
|
|
||||||
|
* 必选：否 <br />
|
||||||
|
|
||||||
|
* 默认值：1024 <br />
|
||||||
|
|
||||||
|
|
||||||
|
### 3.3 类型转换
|
||||||
|
|
||||||
|
目前GaussDbReader支持大部分GaussDB类型，但也存在部分个别类型没有支持的情况，请注意检查你的类型。
|
||||||
|
|
||||||
|
下面列出GaussDbReader针对GaussDB类型转换列表:
|
||||||
|
|
||||||
|
|
||||||
|
| DataX 内部类型| GaussDB 数据类型 |
|
||||||
|
| -------- | ----- |
|
||||||
|
| Long |bigint, bigserial, integer, smallint, serial |
|
||||||
|
| Double |double precision, money, numeric, real |
|
||||||
|
| String |varchar, char, text, bit, inet|
|
||||||
|
| Date |date, time, timestamp |
|
||||||
|
| Boolean |bool|
|
||||||
|
| Bytes |bytea|
|
||||||
|
|
||||||
|
请注意:
|
||||||
|
|
||||||
|
* `除上述罗列字段类型外，其他类型均不支持; money,inet,bit需用户使用a_inet::varchar类似的语法转换`。
|
||||||
|
|
||||||
|
## 4 性能报告
|
||||||
|
|
||||||
|
### 4.1 环境准备
|
||||||
|
|
||||||
|
#### 4.1.1 数据特征
|
||||||
|
建表语句：
|
||||||
|
|
||||||
|
create table pref_test(
|
||||||
|
id serial,
|
||||||
|
a_bigint bigint,
|
||||||
|
a_bit bit(10),
|
||||||
|
a_boolean boolean,
|
||||||
|
a_char character(5),
|
||||||
|
a_date date,
|
||||||
|
a_double double precision,
|
||||||
|
a_integer integer,
|
||||||
|
a_money money,
|
||||||
|
a_num numeric(10,2),
|
||||||
|
a_real real,
|
||||||
|
a_smallint smallint,
|
||||||
|
a_text text,
|
||||||
|
a_time time,
|
||||||
|
a_timestamp timestamp
|
||||||
|
)
|
||||||
|
|
||||||
|
#### 4.1.2 机器参数
|
||||||
|
|
||||||
|
* 执行DataX的机器参数为:
|
||||||
|
1. cpu: 16核 Intel(R) Xeon(R) CPU E5620 @ 2.40GHz
|
||||||
|
2. mem: MemTotal: 24676836kB MemFree: 6365080kB
|
||||||
|
3. net: 百兆双网卡
|
||||||
|
|
||||||
|
* GaussDB数据库机器参数为:
|
||||||
|
D12 24逻辑核 192G内存 12*480G SSD 阵列
|
||||||
|
|
||||||
|
|
||||||
|
### 4.2 测试报告
|
||||||
|
|
||||||
|
#### 4.2.1 单表测试报告
|
||||||
|
|
||||||
|
|
||||||
|
| 通道数 | 是否按照主键切分 | DataX速度(Rec/s) | DataX流量(MB/s) | DataX机器运行负载 |
|
||||||
|
|--------|--------| --------|--------|--------|
|
||||||
|
|1| 否 | 10211 | 0.63 | 0.2 |
|
||||||
|
|1| 是 | 10211 | 0.63 | 0.2 |
|
||||||
|
|4| 否 | 10211 | 0.63 | 0.2 |
|
||||||
|
|4| 是 | 40000 | 2.48 | 0.5 |
|
||||||
|
|8| 否 | 10211 | 0.63 | 0.2 |
|
||||||
|
|8| 是 | 78048 | 4.84 | 0.8 |
|
||||||
|
|
||||||
|
|
||||||
|
说明：
|
||||||
|
|
||||||
|
1. 这里的单表，主键类型为 serial，数据分布均匀。
|
||||||
|
2. 对单表如果没有按照主键切分，那么配置通道个数不会提升速度，效果与1个通道一样。
|
||||||
86
gaussdbreader/pom.xml
Normal file
86
gaussdbreader/pom.xml
Normal file
@ -0,0 +1,86 @@
|
|||||||
|
<?xml version="1.0" encoding="UTF-8"?>
|
||||||
|
<project xmlns="http://maven.apache.org/POM/4.0.0"
|
||||||
|
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
|
||||||
|
xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
|
||||||
|
<parent>
|
||||||
|
<artifactId>datax-all</artifactId>
|
||||||
|
<groupId>com.alibaba.datax</groupId>
|
||||||
|
<version>0.0.1-SNAPSHOT</version>
|
||||||
|
</parent>
|
||||||
|
<modelVersion>4.0.0</modelVersion>
|
||||||
|
|
||||||
|
<artifactId>gaussdbreader</artifactId>
|
||||||
|
<name>gaussdbreader</name>
|
||||||
|
<packaging>jar</packaging>
|
||||||
|
|
||||||
|
<dependencies>
|
||||||
|
<dependency>
|
||||||
|
<groupId>com.alibaba.datax</groupId>
|
||||||
|
<artifactId>datax-common</artifactId>
|
||||||
|
<version>${datax-project-version}</version>
|
||||||
|
<exclusions>
|
||||||
|
<exclusion>
|
||||||
|
<artifactId>slf4j-log4j12</artifactId>
|
||||||
|
<groupId>org.slf4j</groupId>
|
||||||
|
</exclusion>
|
||||||
|
</exclusions>
|
||||||
|
</dependency>
|
||||||
|
|
||||||
|
<dependency>
|
||||||
|
<groupId>org.slf4j</groupId>
|
||||||
|
<artifactId>slf4j-api</artifactId>
|
||||||
|
</dependency>
|
||||||
|
|
||||||
|
<dependency>
|
||||||
|
<groupId>ch.qos.logback</groupId>
|
||||||
|
<artifactId>logback-classic</artifactId>
|
||||||
|
</dependency>
|
||||||
|
|
||||||
|
<dependency>
|
||||||
|
<groupId>com.alibaba.datax</groupId>
|
||||||
|
<artifactId>plugin-rdbms-util</artifactId>
|
||||||
|
<version>${datax-project-version}</version>
|
||||||
|
</dependency>
|
||||||
|
|
||||||
|
<dependency>
|
||||||
|
<groupId>org.opengauss</groupId>
|
||||||
|
<artifactId>opengauss-jdbc</artifactId>
|
||||||
|
<version>3.0.0</version>
|
||||||
|
</dependency>
|
||||||
|
|
||||||
|
</dependencies>
|
||||||
|
|
||||||
|
<build>
|
||||||
|
<plugins>
|
||||||
|
<!-- compiler plugin -->
|
||||||
|
<plugin>
|
||||||
|
<artifactId>maven-compiler-plugin</artifactId>
|
||||||
|
<configuration>
|
||||||
|
<source>${jdk-version}</source>
|
||||||
|
<target>${jdk-version}</target>
|
||||||
|
<encoding>${project-sourceEncoding}</encoding>
|
||||||
|
</configuration>
|
||||||
|
</plugin>
|
||||||
|
<!-- assembly plugin -->
|
||||||
|
<plugin>
|
||||||
|
<artifactId>maven-assembly-plugin</artifactId>
|
||||||
|
<configuration>
|
||||||
|
<descriptors>
|
||||||
|
<descriptor>src/main/assembly/package.xml</descriptor>
|
||||||
|
</descriptors>
|
||||||
|
<finalName>datax</finalName>
|
||||||
|
</configuration>
|
||||||
|
<executions>
|
||||||
|
<execution>
|
||||||
|
<id>dwzip</id>
|
||||||
|
<phase>package</phase>
|
||||||
|
<goals>
|
||||||
|
<goal>single</goal>
|
||||||
|
</goals>
|
||||||
|
</execution>
|
||||||
|
</executions>
|
||||||
|
</plugin>
|
||||||
|
</plugins>
|
||||||
|
</build>
|
||||||
|
|
||||||
|
</project>
|
||||||
35
gaussdbreader/src/main/assembly/package.xml
Executable file
35
gaussdbreader/src/main/assembly/package.xml
Executable file
@ -0,0 +1,35 @@
|
|||||||
|
<assembly
|
||||||
|
xmlns="http://maven.apache.org/plugins/maven-assembly-plugin/assembly/1.1.0"
|
||||||
|
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
|
||||||
|
xsi:schemaLocation="http://maven.apache.org/plugins/maven-assembly-plugin/assembly/1.1.0 http://maven.apache.org/xsd/assembly-1.1.0.xsd">
|
||||||
|
<id></id>
|
||||||
|
<formats>
|
||||||
|
<format>dir</format>
|
||||||
|
</formats>
|
||||||
|
<includeBaseDirectory>false</includeBaseDirectory>
|
||||||
|
<fileSets>
|
||||||
|
<fileSet>
|
||||||
|
<directory>src/main/resources</directory>
|
||||||
|
<includes>
|
||||||
|
<include>plugin.json</include>
|
||||||
|
<include>plugin_job_template.json</include>
|
||||||
|
</includes>
|
||||||
|
<outputDirectory>plugin/reader/gaussdbreader</outputDirectory>
|
||||||
|
</fileSet>
|
||||||
|
<fileSet>
|
||||||
|
<directory>target/</directory>
|
||||||
|
<includes>
|
||||||
|
<include>gaussdbreader-0.0.1-SNAPSHOT.jar</include>
|
||||||
|
</includes>
|
||||||
|
<outputDirectory>plugin/reader/gaussdbreader</outputDirectory>
|
||||||
|
</fileSet>
|
||||||
|
</fileSets>
|
||||||
|
|
||||||
|
<dependencySets>
|
||||||
|
<dependencySet>
|
||||||
|
<useProjectArtifact>false</useProjectArtifact>
|
||||||
|
<outputDirectory>plugin/reader/gaussdbreader/libs</outputDirectory>
|
||||||
|
<scope>runtime</scope>
|
||||||
|
</dependencySet>
|
||||||
|
</dependencySets>
|
||||||
|
</assembly>
|
||||||
@ -0,0 +1,7 @@
|
|||||||
|
package com.alibaba.datax.plugin.reader.gaussdbreader;
|
||||||
|
|
||||||
|
public class Constant {
|
||||||
|
|
||||||
|
public static final int DEFAULT_FETCH_SIZE = 1000;
|
||||||
|
|
||||||
|
}
|
||||||
@ -0,0 +1,86 @@
|
|||||||
|
package com.alibaba.datax.plugin.reader.gaussdbreader;
|
||||||
|
|
||||||
|
import com.alibaba.datax.common.exception.DataXException;
|
||||||
|
import com.alibaba.datax.common.plugin.RecordSender;
|
||||||
|
import com.alibaba.datax.common.spi.Reader;
|
||||||
|
import com.alibaba.datax.common.util.Configuration;
|
||||||
|
import com.alibaba.datax.plugin.rdbms.reader.CommonRdbmsReader;
|
||||||
|
import com.alibaba.datax.plugin.rdbms.util.DBUtilErrorCode;
|
||||||
|
import com.alibaba.datax.plugin.rdbms.util.DataBaseType;
|
||||||
|
|
||||||
|
import java.util.List;
|
||||||
|
|
||||||
|
public class GaussDbReader extends Reader {
|
||||||
|
|
||||||
|
private static final DataBaseType DATABASE_TYPE = DataBaseType.GaussDB;
|
||||||
|
|
||||||
|
public static class Job extends Reader.Job {
|
||||||
|
|
||||||
|
private Configuration originalConfig;
|
||||||
|
private CommonRdbmsReader.Job commonRdbmsReaderMaster;
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public void init() {
|
||||||
|
this.originalConfig = super.getPluginJobConf();
|
||||||
|
int fetchSize = this.originalConfig.getInt(com.alibaba.datax.plugin.rdbms.reader.Constant.FETCH_SIZE,
|
||||||
|
Constant.DEFAULT_FETCH_SIZE);
|
||||||
|
if (fetchSize < 1) {
|
||||||
|
throw DataXException.asDataXException(DBUtilErrorCode.REQUIRED_VALUE,
|
||||||
|
String.format("您配置的fetchSize有误，根据DataX的设计，fetchSize : [%d] 设置值不能小于 1.", fetchSize));
|
||||||
|
}
|
||||||
|
this.originalConfig.set(com.alibaba.datax.plugin.rdbms.reader.Constant.FETCH_SIZE, fetchSize);
|
||||||
|
|
||||||
|
this.commonRdbmsReaderMaster = new CommonRdbmsReader.Job(DATABASE_TYPE);
|
||||||
|
this.commonRdbmsReaderMaster.init(this.originalConfig);
|
||||||
|
}
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public List<Configuration> split(int adviceNumber) {
|
||||||
|
return this.commonRdbmsReaderMaster.split(this.originalConfig, adviceNumber);
|
||||||
|
}
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public void post() {
|
||||||
|
this.commonRdbmsReaderMaster.post(this.originalConfig);
|
||||||
|
}
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public void destroy() {
|
||||||
|
this.commonRdbmsReaderMaster.destroy(this.originalConfig);
|
||||||
|
}
|
||||||
|
|
||||||
|
}
|
||||||
|
|
||||||
|
public static class Task extends Reader.Task {
|
||||||
|
|
||||||
|
private Configuration readerSliceConfig;
|
||||||
|
private CommonRdbmsReader.Task commonRdbmsReaderSlave;
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public void init() {
|
||||||
|
this.readerSliceConfig = super.getPluginJobConf();
|
||||||
|
this.commonRdbmsReaderSlave = new CommonRdbmsReader.Task(DATABASE_TYPE,super.getTaskGroupId(), super.getTaskId());
|
||||||
|
this.commonRdbmsReaderSlave.init(this.readerSliceConfig);
|
||||||
|
}
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public void startRead(RecordSender recordSender) {
|
||||||
|
int fetchSize = this.readerSliceConfig.getInt(com.alibaba.datax.plugin.rdbms.reader.Constant.FETCH_SIZE);
|
||||||
|
|
||||||
|
this.commonRdbmsReaderSlave.startRead(this.readerSliceConfig, recordSender,
|
||||||
|
super.getTaskPluginCollector(), fetchSize);
|
||||||
|
}
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public void post() {
|
||||||
|
this.commonRdbmsReaderSlave.post(this.readerSliceConfig);
|
||||||
|
}
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public void destroy() {
|
||||||
|
this.commonRdbmsReaderSlave.destroy(this.readerSliceConfig);
|
||||||
|
}
|
||||||
|
|
||||||
|
}
|
||||||
|
|
||||||
|
}
|
||||||
6
gaussdbreader/src/main/resources/plugin.json
Executable file
6
gaussdbreader/src/main/resources/plugin.json
Executable file
@ -0,0 +1,6 @@
|
|||||||
|
{
|
||||||
|
"name": "gaussdbreader",
|
||||||
|
"class": "com.alibaba.datax.plugin.reader.gaussdbreader.GaussDbReader",
|
||||||
|
"description": "useScene: prod. mechanism: Jdbc connection using the database, execute select sql, retrieve data from the ResultSet. warn: The more you know about the database, the less problems you encounter.",
|
||||||
|
"developer": "alibaba"
|
||||||
|
}
|
||||||
13
gaussdbreader/src/main/resources/plugin_job_template.json
Normal file
13
gaussdbreader/src/main/resources/plugin_job_template.json
Normal file
@ -0,0 +1,13 @@
|
|||||||
|
{
|
||||||
|
"name": "gaussdbreader",
|
||||||
|
"parameter": {
|
||||||
|
"username": "",
|
||||||
|
"password": "",
|
||||||
|
"connection": [
|
||||||
|
{
|
||||||
|
"table": [],
|
||||||
|
"jdbcUrl": []
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
|
}
|
||||||
267
gaussdbwriter/doc/gaussdbwriter.md
Normal file
267
gaussdbwriter/doc/gaussdbwriter.md
Normal file
@ -0,0 +1,267 @@
|
|||||||
|
# DataX GaussDbWriter
|
||||||
|
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
|
||||||
|
## 1 快速介绍
|
||||||
|
|
||||||
|
GaussDbWriter插件实现了写入数据到 GaussDB主库目的表的功能。在底层实现上，GaussDbWriter通过JDBC连接远程 GaussDB 数据库，并执行相应的 insert into ... sql 语句将数据写入 GaussDB，内部会分批次提交入库。
|
||||||
|
|
||||||
|
GaussDbWriter面向ETL开发工程师，他们使用GaussDbWriter从数仓导入数据到GaussDB。同时 GaussDbWriter亦可以作为数据迁移工具为DBA等用户提供服务。
|
||||||
|
|
||||||
|
|
||||||
|
## 2 实现原理
|
||||||
|
|
||||||
|
GaussDbWriter通过 DataX 框架获取 Reader 生成的协议数据，根据你配置生成相应的SQL插入语句
|
||||||
|
|
||||||
|
|
||||||
|
* `insert into...`(当主键/唯一性索引冲突时会写不进去冲突的行)
|
||||||
|
|
||||||
|
<br />
|
||||||
|
|
||||||
|
注意：
|
||||||
|
1. 目的表所在数据库必须是主库才能写入数据；整个任务至少需具备 insert into...的权限，是否需要其他权限，取决于你任务配置中在 preSql 和 postSql 中指定的语句。
|
||||||
|
2. GaussDbWriter和MysqlWriter不同，不支持配置writeMode参数。
|
||||||
|
|
||||||
|
|
||||||
|
## 3 功能说明
|
||||||
|
|
||||||
|
### 3.1 配置样例
|
||||||
|
|
||||||
|
* 这里使用一份从内存产生到 GaussDbWriter导入的数据。
|
||||||
|
|
||||||
|
```json
|
||||||
|
{
|
||||||
|
"job": {
|
||||||
|
"setting": {
|
||||||
|
"speed": {
|
||||||
|
"channel": 1
|
||||||
|
}
|
||||||
|
},
|
||||||
|
"content": [
|
||||||
|
{
|
||||||
|
"reader": {
|
||||||
|
"name": "streamreader",
|
||||||
|
"parameter": {
|
||||||
|
"column" : [
|
||||||
|
{
|
||||||
|
"value": "DataX",
|
||||||
|
"type": "string"
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"value": 19880808,
|
||||||
|
"type": "long"
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"value": "1988-08-08 08:08:08",
|
||||||
|
"type": "date"
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"value": true,
|
||||||
|
"type": "bool"
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"value": "test",
|
||||||
|
"type": "bytes"
|
||||||
|
}
|
||||||
|
],
|
||||||
|
"sliceRecordCount": 1000
|
||||||
|
}
|
||||||
|
},
|
||||||
|
"writer": {
|
||||||
|
"name": "gaussdbwriter",
|
||||||
|
"parameter": {
|
||||||
|
"username": "xx",
|
||||||
|
"password": "xx",
|
||||||
|
"column": [
|
||||||
|
"id",
|
||||||
|
"name"
|
||||||
|
],
|
||||||
|
"preSql": [
|
||||||
|
"delete from test"
|
||||||
|
],
|
||||||
|
"connection": [
|
||||||
|
{
|
||||||
|
"jdbcUrl": "jdbc:opengauss://127.0.0.1:3002/datax",
|
||||||
|
"table": [
|
||||||
|
"test"
|
||||||
|
]
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
```
|
||||||
|
|
||||||
|
|
||||||
|
### 3.2 参数说明
|
||||||
|
|
||||||
|
* **jdbcUrl**
|
||||||
|
|
||||||
|
* 描述：目的数据库的 JDBC 连接信息 ,jdbcUrl必须包含在connection配置单元中。
|
||||||
|
|
||||||
|
注意：1、在一个数据库上只能配置一个值。
|
||||||
|
2、jdbcUrl按照GaussDB官方规范，并可以填写连接附加参数信息。具体请参看GaussDB官方文档或者咨询对应 DBA。
|
||||||
|
|
||||||
|
|
||||||
|
* 必选：是 <br />
|
||||||
|
|
||||||
|
* 默认值：无 <br />
|
||||||
|
|
||||||
|
* **username**
|
||||||
|
|
||||||
|
* 描述：目的数据库的用户名 <br />
|
||||||
|
|
||||||
|
* 必选：是 <br />
|
||||||
|
|
||||||
|
* 默认值：无 <br />
|
||||||
|
|
||||||
|
* **password**
|
||||||
|
|
||||||
|
* 描述：目的数据库的密码 <br />
|
||||||
|
|
||||||
|
* 必选：是 <br />
|
||||||
|
|
||||||
|
* 默认值：无 <br />
|
||||||
|
|
||||||
|
* **table**
|
||||||
|
|
||||||
|
* 描述：目的表的表名称。支持写入一个或者多个表。当配置为多张表时，必须确保所有表结构保持一致。
|
||||||
|
|
||||||
|
注意：table 和 jdbcUrl 必须包含在 connection 配置单元中
|
||||||
|
|
||||||
|
* 必选：是 <br />
|
||||||
|
|
||||||
|
* 默认值：无 <br />
|
||||||
|
|
||||||
|
* **column**
|
||||||
|
|
||||||
|
* 描述：目的表需要写入数据的字段,字段之间用英文逗号分隔。例如: "column": ["id","name","age"]。如果要依次写入全部列，使用\*表示, 例如: "column": ["\*"]
|
||||||
|
|
||||||
|
注意：1、我们强烈不推荐你这样配置，因为当你目的表字段个数、类型等有改动时，你的任务可能运行不正确或者失败
|
||||||
|
2、此处 column 不能配置任何常量值
|
||||||
|
|
||||||
|
* 必选：是 <br />
|
||||||
|
|
||||||
|
* 默认值：否 <br />
|
||||||
|
|
||||||
|
* **preSql**
|
||||||
|
|
||||||
|
* 描述：写入数据到目的表前，会先执行这里的标准语句。如果 Sql 中有你需要操作到的表名称，请使用 `@table` 表示，这样在实际执行 Sql 语句时，会对变量按照实际表名称进行替换。比如你的任务是要写入到目的端的100个同构分表(表名称为:datax_00,datax01, ... datax_98,datax_99)，并且你希望导入数据前，先对表中数据进行删除操作，那么你可以这样配置：`"preSql":["delete from @table"]`，效果是：在执行到每个表写入数据前，会先执行对应的 delete from 对应表名称 <br />
|
||||||
|
|
||||||
|
* 必选：否 <br />
|
||||||
|
|
||||||
|
* 默认值：无 <br />
|
||||||
|
|
||||||
|
* **postSql**
|
||||||
|
|
||||||
|
* 描述：写入数据到目的表后，会执行这里的标准语句。（原理同 preSql ） <br />
|
||||||
|
|
||||||
|
* 必选：否 <br />
|
||||||
|
|
||||||
|
* 默认值：无 <br />
|
||||||
|
|
||||||
|
* **batchSize**
|
||||||
|
|
||||||
|
* 描述：一次性批量提交的记录数大小，该值可以极大减少DataX与GaussDB的网络交互次数，并提升整体吞吐量。但是该值设置过大可能会造成DataX运行进程OOM情况。<br />
|
||||||
|
|
||||||
|
* 必选：否 <br />
|
||||||
|
|
||||||
|
* 默认值：1024 <br />
|
||||||
|
|
||||||
|
### 3.3 类型转换
|
||||||
|
|
||||||
|
目前 GaussDbWriter支持大部分 GaussDB类型，但也存在部分没有支持的情况，请注意检查你的类型。
|
||||||
|
|
||||||
|
下面列出 GaussDbWriter针对 GaussDB类型转换列表:
|
||||||
|
|
||||||
|
| DataX 内部类型| GaussDB 数据类型 |
|
||||||
|
| -------- | ----- |
|
||||||
|
| Long |bigint, bigserial, integer, smallint, serial |
|
||||||
|
| Double |double precision, money, numeric, real |
|
||||||
|
| String |varchar, char, text, bit|
|
||||||
|
| Date |date, time, timestamp |
|
||||||
|
| Boolean |bool|
|
||||||
|
| Bytes |bytea|
|
||||||
|
|
||||||
|
## 4 性能报告
|
||||||
|
|
||||||
|
### 4.1 环境准备
|
||||||
|
|
||||||
|
#### 4.1.1 数据特征
|
||||||
|
建表语句：
|
||||||
|
|
||||||
|
create table pref_test(
|
||||||
|
id serial,
|
||||||
|
a_bigint bigint,
|
||||||
|
a_bit bit(10),
|
||||||
|
a_boolean boolean,
|
||||||
|
a_char character(5),
|
||||||
|
a_date date,
|
||||||
|
a_double double precision,
|
||||||
|
a_integer integer,
|
||||||
|
a_money money,
|
||||||
|
a_num numeric(10,2),
|
||||||
|
a_real real,
|
||||||
|
a_smallint smallint,
|
||||||
|
a_text text,
|
||||||
|
a_time time,
|
||||||
|
a_timestamp timestamp
|
||||||
|
)
|
||||||
|
|
||||||
|
#### 4.1.2 机器参数
|
||||||
|
|
||||||
|
* 执行DataX的机器参数为:
|
||||||
|
1. cpu: 16核 Intel(R) Xeon(R) CPU E5620 @ 2.40GHz
|
||||||
|
2. mem: MemTotal: 24676836kB MemFree: 6365080kB
|
||||||
|
3. net: 百兆双网卡
|
||||||
|
|
||||||
|
* GaussDB数据库机器参数为:
|
||||||
|
D12 24逻辑核 192G内存 12*480G SSD 阵列
|
||||||
|
|
||||||
|
|
||||||
|
### 4.2 测试报告
|
||||||
|
|
||||||
|
#### 4.2.1 单表测试报告
|
||||||
|
|
||||||
|
| 通道数| 批量提交batchSize | DataX速度(Rec/s)| DataX流量(M/s) | DataX机器运行负载
|
||||||
|
|--------|--------| --------|--------|--------|--------|
|
||||||
|
|1| 128 | 9259 | 0.55 | 0.3
|
||||||
|
|1| 512 | 10869 | 0.653 | 0.3
|
||||||
|
|1| 2048 | 9803 | 0.589 | 0.8
|
||||||
|
|4| 128 | 30303 | 1.82 | 1
|
||||||
|
|4| 512 | 36363 | 2.18 | 1
|
||||||
|
|4| 2048 | 36363 | 2.18 | 1
|
||||||
|
|8| 128 | 57142 | 3.43 | 2
|
||||||
|
|8| 512 | 66666 | 4.01 | 1.5
|
||||||
|
|8| 2048 | 66666 | 4.01 | 1.1
|
||||||
|
|16| 128 | 88888 | 5.34 | 1.8
|
||||||
|
|16| 2048 | 94117 | 5.65 | 2.5
|
||||||
|
|32| 512 | 76190 | 4.58 | 3
|
||||||
|
|
||||||
|
#### 4.2.2 性能测试小结
|
||||||
|
1. `channel数对性能影响很大`
|
||||||
|
2. `通常不建议写入数据库时，通道个数 > 32`
|
||||||
|
|
||||||
|
|
||||||
|
## FAQ
|
||||||
|
|
||||||
|
***
|
||||||
|
|
||||||
|
**Q: GaussDbWriter 执行 postSql 语句报错，那么数据导入到目标数据库了吗?**
|
||||||
|
|
||||||
|
A: DataX 导入过程存在三块逻辑，pre 操作、导入操作、post 操作，其中任意一环报错，DataX 作业报错。由于 DataX 不能保证在同一个事务完成上述几个操作，因此有可能数据已经落入到目标端。
|
||||||
|
|
||||||
|
***
|
||||||
|
|
||||||
|
**Q: 按照上述说法，那么有部分脏数据导入数据库，如果影响到线上数据库怎么办?**
|
||||||
|
|
||||||
|
A: 目前有两种解法，第一种配置 pre 语句，该 sql 可以清理当天导入数据， DataX 每次导入时候可以把上次清理干净并导入完整数据。
|
||||||
|
第二种，向临时表导入数据，完成后再 rename 到线上表。
|
||||||
|
|
||||||
|
***
|
||||||
86
gaussdbwriter/pom.xml
Normal file
86
gaussdbwriter/pom.xml
Normal file
@ -0,0 +1,86 @@
|
|||||||
|
<?xml version="1.0" encoding="UTF-8"?>
|
||||||
|
<project xmlns="http://maven.apache.org/POM/4.0.0"
|
||||||
|
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
|
||||||
|
xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
|
||||||
|
<parent>
|
||||||
|
<artifactId>datax-all</artifactId>
|
||||||
|
<groupId>com.alibaba.datax</groupId>
|
||||||
|
<version>0.0.1-SNAPSHOT</version>
|
||||||
|
</parent>
|
||||||
|
<modelVersion>4.0.0</modelVersion>
|
||||||
|
|
||||||
|
<artifactId>gaussdbwriter</artifactId>
|
||||||
|
<name>gaussdbwriter</name>
|
||||||
|
<packaging>jar</packaging>
|
||||||
|
|
||||||
|
<dependencies>
|
||||||
|
<dependency>
|
||||||
|
<groupId>com.alibaba.datax</groupId>
|
||||||
|
<artifactId>datax-common</artifactId>
|
||||||
|
<version>${datax-project-version}</version>
|
||||||
|
<exclusions>
|
||||||
|
<exclusion>
|
||||||
|
<artifactId>slf4j-log4j12</artifactId>
|
||||||
|
<groupId>org.slf4j</groupId>
|
||||||
|
</exclusion>
|
||||||
|
</exclusions>
|
||||||
|
</dependency>
|
||||||
|
|
||||||
|
<dependency>
|
||||||
|
<groupId>org.slf4j</groupId>
|
||||||
|
<artifactId>slf4j-api</artifactId>
|
||||||
|
</dependency>
|
||||||
|
|
||||||
|
<dependency>
|
||||||
|
<groupId>ch.qos.logback</groupId>
|
||||||
|
<artifactId>logback-classic</artifactId>
|
||||||
|
</dependency>
|
||||||
|
|
||||||
|
<dependency>
|
||||||
|
<groupId>com.alibaba.datax</groupId>
|
||||||
|
<artifactId>plugin-rdbms-util</artifactId>
|
||||||
|
<version>${datax-project-version}</version>
|
||||||
|
</dependency>
|
||||||
|
|
||||||
|
<dependency>
|
||||||
|
<groupId>org.opengauss</groupId>
|
||||||
|
<artifactId>opengauss-jdbc</artifactId>
|
||||||
|
<version>3.0.0</version>
|
||||||
|
</dependency>
|
||||||
|
|
||||||
|
</dependencies>
|
||||||
|
|
||||||
|
<build>
|
||||||
|
<plugins>
|
||||||
|
<!-- compiler plugin -->
|
||||||
|
<plugin>
|
||||||
|
<artifactId>maven-compiler-plugin</artifactId>
|
||||||
|
<configuration>
|
||||||
|
<source>${jdk-version}</source>
|
||||||
|
<target>${jdk-version}</target>
|
||||||
|
<encoding>${project-sourceEncoding}</encoding>
|
||||||
|
</configuration>
|
||||||
|
</plugin>
|
||||||
|
<!-- assembly plugin -->
|
||||||
|
<plugin>
|
||||||
|
<artifactId>maven-assembly-plugin</artifactId>
|
||||||
|
<configuration>
|
||||||
|
<descriptors>
|
||||||
|
<descriptor>src/main/assembly/package.xml</descriptor>
|
||||||
|
</descriptors>
|
||||||
|
<finalName>datax</finalName>
|
||||||
|
</configuration>
|
||||||
|
<executions>
|
||||||
|
<execution>
|
||||||
|
<id>dwzip</id>
|
||||||
|
<phase>package</phase>
|
||||||
|
<goals>
|
||||||
|
<goal>single</goal>
|
||||||
|
</goals>
|
||||||
|
</execution>
|
||||||
|
</executions>
|
||||||
|
</plugin>
|
||||||
|
</plugins>
|
||||||
|
</build>
|
||||||
|
|
||||||
|
</project>
|
||||||
35
gaussdbwriter/src/main/assembly/package.xml
Executable file
35
gaussdbwriter/src/main/assembly/package.xml
Executable file
@ -0,0 +1,35 @@
|
|||||||
|
<assembly
|
||||||
|
xmlns="http://maven.apache.org/plugins/maven-assembly-plugin/assembly/1.1.0"
|
||||||
|
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
|
||||||
|
xsi:schemaLocation="http://maven.apache.org/plugins/maven-assembly-plugin/assembly/1.1.0 http://maven.apache.org/xsd/assembly-1.1.0.xsd">
|
||||||
|
<id></id>
|
||||||
|
<formats>
|
||||||
|
<format>dir</format>
|
||||||
|
</formats>
|
||||||
|
<includeBaseDirectory>false</includeBaseDirectory>
|
||||||
|
<fileSets>
|
||||||
|
<fileSet>
|
||||||
|
<directory>src/main/resources</directory>
|
||||||
|
<includes>
|
||||||
|
<include>plugin.json</include>
|
||||||
|
<include>plugin_job_template.json</include>
|
||||||
|
</includes>
|
||||||
|
<outputDirectory>plugin/writer/gaussdbwriter</outputDirectory>
|
||||||
|
</fileSet>
|
||||||
|
<fileSet>
|
||||||
|
<directory>target/</directory>
|
||||||
|
<includes>
|
||||||
|
<include>gaussdbwriter-0.0.1-SNAPSHOT.jar</include>
|
||||||
|
</includes>
|
||||||
|
<outputDirectory>plugin/writer/gaussdbwriter</outputDirectory>
|
||||||
|
</fileSet>
|
||||||
|
</fileSets>
|
||||||
|
|
||||||
|
<dependencySets>
|
||||||
|
<dependencySet>
|
||||||
|
<useProjectArtifact>false</useProjectArtifact>
|
||||||
|
<outputDirectory>plugin/writer/gaussdbwriter/libs</outputDirectory>
|
||||||
|
<scope>runtime</scope>
|
||||||
|
</dependencySet>
|
||||||
|
</dependencySets>
|
||||||
|
</assembly>
|
||||||
@ -0,0 +1,103 @@
|
|||||||
|
package com.alibaba.datax.plugin.reader.gaussdbwriter;
|
||||||
|
|
||||||
|
import com.alibaba.datax.common.exception.DataXException;
|
||||||
|
import com.alibaba.datax.common.plugin.RecordReceiver;
|
||||||
|
import com.alibaba.datax.common.spi.Writer;
|
||||||
|
import com.alibaba.datax.common.util.Configuration;
|
||||||
|
import com.alibaba.datax.plugin.rdbms.util.DBUtilErrorCode;
|
||||||
|
import com.alibaba.datax.plugin.rdbms.util.DataBaseType;
|
||||||
|
import com.alibaba.datax.plugin.rdbms.writer.CommonRdbmsWriter;
|
||||||
|
import com.alibaba.datax.plugin.rdbms.writer.Key;
|
||||||
|
|
||||||
|
import java.util.List;
|
||||||
|
|
||||||
|
public class GaussDbWriter extends Writer {
|
||||||
|
|
||||||
|
private static final DataBaseType DATABASE_TYPE = DataBaseType.GaussDB;
|
||||||
|
|
||||||
|
public static class Job extends Writer.Job {
|
||||||
|
private Configuration originalConfig = null;
|
||||||
|
private CommonRdbmsWriter.Job commonRdbmsWriterMaster;
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public void init() {
|
||||||
|
this.originalConfig = super.getPluginJobConf();
|
||||||
|
|
||||||
|
// warn：not like mysql, GaussDB only support insert mode, don't use
|
||||||
|
String writeMode = this.originalConfig.getString(Key.WRITE_MODE);
|
||||||
|
if (null != writeMode) {
|
||||||
|
throw DataXException.asDataXException(DBUtilErrorCode.CONF_ERROR,
|
||||||
|
String.format("写入模式(writeMode)配置有误. 因为GaussDB不支持配置参数项 writeMode: %s, GaussDB仅使用insert sql 插入数据. 请检查您的配置并作出修改.", writeMode));
|
||||||
|
}
|
||||||
|
|
||||||
|
this.commonRdbmsWriterMaster = new CommonRdbmsWriter.Job(DATABASE_TYPE);
|
||||||
|
this.commonRdbmsWriterMaster.init(this.originalConfig);
|
||||||
|
}
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public void prepare() {
|
||||||
|
this.commonRdbmsWriterMaster.prepare(this.originalConfig);
|
||||||
|
}
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public List<Configuration> split(int mandatoryNumber) {
|
||||||
|
return this.commonRdbmsWriterMaster.split(this.originalConfig, mandatoryNumber);
|
||||||
|
}
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public void post() {
|
||||||
|
this.commonRdbmsWriterMaster.post(this.originalConfig);
|
||||||
|
}
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public void destroy() {
|
||||||
|
this.commonRdbmsWriterMaster.destroy(this.originalConfig);
|
||||||
|
}
|
||||||
|
|
||||||
|
}
|
||||||
|
|
||||||
|
public static class Task extends Writer.Task {
|
||||||
|
private Configuration writerSliceConfig;
|
||||||
|
private CommonRdbmsWriter.Task commonRdbmsWriterSlave;
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public void init() {
|
||||||
|
this.writerSliceConfig = super.getPluginJobConf();
|
||||||
|
this.commonRdbmsWriterSlave = new CommonRdbmsWriter.Task(DATABASE_TYPE){
|
||||||
|
@Override
|
||||||
|
public String calcValueHolder(String columnType){
|
||||||
|
if("serial".equalsIgnoreCase(columnType)){
|
||||||
|
return "?::int";
|
||||||
|
}else if("bigserial".equalsIgnoreCase(columnType)){
|
||||||
|
return "?::int8";
|
||||||
|
}else if("bit".equalsIgnoreCase(columnType)){
|
||||||
|
return "?::bit varying";
|
||||||
|
}
|
||||||
|
return "?::" + columnType;
|
||||||
|
}
|
||||||
|
};
|
||||||
|
this.commonRdbmsWriterSlave.init(this.writerSliceConfig);
|
||||||
|
}
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public void prepare() {
|
||||||
|
this.commonRdbmsWriterSlave.prepare(this.writerSliceConfig);
|
||||||
|
}
|
||||||
|
|
||||||
|
public void startWrite(RecordReceiver recordReceiver) {
|
||||||
|
this.commonRdbmsWriterSlave.startWrite(recordReceiver, this.writerSliceConfig, super.getTaskPluginCollector());
|
||||||
|
}
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public void post() {
|
||||||
|
this.commonRdbmsWriterSlave.post(this.writerSliceConfig);
|
||||||
|
}
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public void destroy() {
|
||||||
|
this.commonRdbmsWriterSlave.destroy(this.writerSliceConfig);
|
||||||
|
}
|
||||||
|
|
||||||
|
}
|
||||||
|
|
||||||
|
}
|
||||||
6
gaussdbwriter/src/main/resources/plugin.json
Executable file
6
gaussdbwriter/src/main/resources/plugin.json
Executable file
@ -0,0 +1,6 @@
|
|||||||
|
{
|
||||||
|
"name": "gaussdbwriter",
|
||||||
|
"class": "com.alibaba.datax.plugin.writer.gaussdbwriter.GaussDbWriter",
|
||||||
|
"description": "useScene: prod. mechanism: Jdbc connection using the database, execute insert sql. warn: The more you know about the database, the less problems you encounter.",
|
||||||
|
"developer": "alibaba"
|
||||||
|
}
|
||||||
16
gaussdbwriter/src/main/resources/plugin_job_template.json
Normal file
16
gaussdbwriter/src/main/resources/plugin_job_template.json
Normal file
@ -0,0 +1,16 @@
|
|||||||
|
{
|
||||||
|
"name": "gaussdbwriter",
|
||||||
|
"parameter": {
|
||||||
|
"username": "",
|
||||||
|
"password": "",
|
||||||
|
"column": [],
|
||||||
|
"connection": [
|
||||||
|
{
|
||||||
|
"jdbcUrl": "",
|
||||||
|
"table": []
|
||||||
|
}
|
||||||
|
],
|
||||||
|
"preSql": [],
|
||||||
|
"postSql": []
|
||||||
|
}
|
||||||
|
}
|
||||||
@ -61,11 +61,15 @@ hbase11xsqlreader插件实现了从Phoenix(HBase SQL)读取数据。在底层实
|
|||||||
"hbaseConfig": {
|
"hbaseConfig": {
|
||||||
"hbase.zookeeper.quorum": "hb-proxy-xxx-002.hbase.rds.aliyuncs.com,hb-proxy-xxx-001.hbase.rds.aliyuncs.com,hb-proxy-xxx-003.hbase.rds.aliyuncs.com"
|
"hbase.zookeeper.quorum": "hb-proxy-xxx-002.hbase.rds.aliyuncs.com,hb-proxy-xxx-001.hbase.rds.aliyuncs.com,hb-proxy-xxx-003.hbase.rds.aliyuncs.com"
|
||||||
},
|
},
|
||||||
|
//填写要读取的phoenix的命名空间
|
||||||
|
"schema": "TAG",
|
||||||
//填写要读取的phoenix的表名
|
//填写要读取的phoenix的表名
|
||||||
"table": "US_POPULATION",
|
"table": "US_POPULATION",
|
||||||
//填写要读取的列名，不填读取所有列
|
//填写要读取的列名，不填读取所有列
|
||||||
"column": [
|
"column": [
|
||||||
]
|
],
|
||||||
|
//查询条件
|
||||||
|
"where": "id="
|
||||||
}
|
}
|
||||||
},
|
},
|
||||||
"writer": {
|
"writer": {
|
||||||
@ -92,11 +96,18 @@ hbase11xsqlreader插件实现了从Phoenix(HBase SQL)读取数据。在底层实
|
|||||||
|
|
||||||
* 必选：是 <br />
|
* 必选：是 <br />
|
||||||
|
|
||||||
|
* 默认值：无 <br />
|
||||||
|
* **schema**
|
||||||
|
|
||||||
|
* 描述：编写Phoenix中的namespace，该值设置为''
|
||||||
|
|
||||||
|
* 必选：是 <br />
|
||||||
|
|
||||||
* 默认值：无 <br />
|
* 默认值：无 <br />
|
||||||
|
|
||||||
* **table**
|
* **table**
|
||||||
|
|
||||||
* 描述：编写Phoenix中的表名,如果有namespace，该值设置为'namespace.tablename'
|
* 描述：编写Phoenix中的表名，该值设置为'tablename'
|
||||||
|
|
||||||
* 必选：是 <br />
|
* 必选：是 <br />
|
||||||
|
|
||||||
@ -109,7 +120,13 @@ hbase11xsqlreader插件实现了从Phoenix(HBase SQL)读取数据。在底层实
|
|||||||
* 必选：是 <br />
|
* 必选：是 <br />
|
||||||
|
|
||||||
* 默认值：无 <br />
|
* 默认值：无 <br />
|
||||||
|
* **where**
|
||||||
|
|
||||||
|
* 描述：填写需要从phoenix表中读取条件判断。
|
||||||
|
|
||||||
|
* 可选：是 <br />
|
||||||
|
|
||||||
|
* 默认值：无 <br />
|
||||||
|
|
||||||
### 3.3 类型转换
|
### 3.3 类型转换
|
||||||
|
|
||||||
@ -172,11 +189,14 @@ hbase11xsqlreader插件实现了从Phoenix(HBase SQL)读取数据。在底层实
|
|||||||
"hbaseConfig": {
|
"hbaseConfig": {
|
||||||
"hbase.zookeeper.quorum": "hb-proxy-xxx-002.hbase.rds.aliyuncs.com,hb-proxy-xxx-001.hbase.rds.aliyuncs.com,hb-proxy-xxx-003.hbase.rds.aliyuncs.com"
|
"hbase.zookeeper.quorum": "hb-proxy-xxx-002.hbase.rds.aliyuncs.com,hb-proxy-xxx-001.hbase.rds.aliyuncs.com,hb-proxy-xxx-003.hbase.rds.aliyuncs.com"
|
||||||
},
|
},
|
||||||
|
"schema": "TAG",
|
||||||
//填写要读取的phoenix的表名
|
//填写要读取的phoenix的表名
|
||||||
"table": "US_POPULATION",
|
"table": "US_POPULATION",
|
||||||
//填写要读取的列名，不填读取所有列
|
//填写要读取的列名，不填读取所有列
|
||||||
"column": [
|
"column": [
|
||||||
]
|
],
|
||||||
|
//查询条件
|
||||||
|
"where": "id="
|
||||||
}
|
}
|
||||||
},
|
},
|
||||||
"writer": {
|
"writer": {
|
||||||
@ -204,7 +224,13 @@ hbase11xsqlreader插件实现了从Phoenix(HBase SQL)读取数据。在底层实
|
|||||||
* 必选：是 <br />
|
* 必选：是 <br />
|
||||||
|
|
||||||
* 默认值：无 <br />
|
* 默认值：无 <br />
|
||||||
|
* **schema**
|
||||||
|
|
||||||
|
* 描述：编写Phoenix中的namespace，该值设置为''
|
||||||
|
|
||||||
|
* 必选：是 <br />
|
||||||
|
|
||||||
|
* 默认值：无 <br />
|
||||||
* **table**
|
* **table**
|
||||||
|
|
||||||
* 描述：编写Phoenix中的表名,如果有namespace，该值设置为'namespace.tablename'
|
* 描述：编写Phoenix中的表名,如果有namespace，该值设置为'namespace.tablename'
|
||||||
@ -220,7 +246,13 @@ hbase11xsqlreader插件实现了从Phoenix(HBase SQL)读取数据。在底层实
|
|||||||
* 必选：是 <br />
|
* 必选：是 <br />
|
||||||
|
|
||||||
* 默认值：无 <br />
|
* 默认值：无 <br />
|
||||||
|
* **where**
|
||||||
|
|
||||||
|
* 描述：填写需要从phoenix表中读取条件判断。
|
||||||
|
|
||||||
|
* 可选：是 <br />
|
||||||
|
|
||||||
|
* 默认值：无 <br />
|
||||||
|
|
||||||
### 3.3 类型转换
|
### 3.3 类型转换
|
||||||
|
|
||||||
|
|||||||
@ -26,9 +26,7 @@ import java.io.IOException;
|
|||||||
import java.sql.Connection;
|
import java.sql.Connection;
|
||||||
import java.sql.DriverManager;
|
import java.sql.DriverManager;
|
||||||
import java.sql.SQLException;
|
import java.sql.SQLException;
|
||||||
import java.util.ArrayList;
|
import java.util.*;
|
||||||
import java.util.List;
|
|
||||||
import java.util.Map;
|
|
||||||
|
|
||||||
|
|
||||||
public class HbaseSQLHelper {
|
public class HbaseSQLHelper {
|
||||||
@ -50,11 +48,15 @@ public class HbaseSQLHelper {
|
|||||||
String zkUrl = readerConfig.getZkUrl();
|
String zkUrl = readerConfig.getZkUrl();
|
||||||
|
|
||||||
PhoenixConfigurationUtil.setInputClass(conf, PhoenixRecordWritable.class);
|
PhoenixConfigurationUtil.setInputClass(conf, PhoenixRecordWritable.class);
|
||||||
PhoenixConfigurationUtil.setInputTableName(conf, table);
|
|
||||||
|
PhoenixConfigurationUtil.setInputTableName(conf, readerConfig.getSchema()+"."+table);
|
||||||
|
|
||||||
if (!columns.isEmpty()) {
|
if (!columns.isEmpty()) {
|
||||||
PhoenixConfigurationUtil.setSelectColumnNames(conf, columns.toArray(new String[columns.size()]));
|
PhoenixConfigurationUtil.setSelectColumnNames(conf, columns.toArray(new String[columns.size()]));
|
||||||
}
|
}
|
||||||
|
if(Objects.nonNull(readerConfig.getWhere())){
|
||||||
|
PhoenixConfigurationUtil.setInputTableConditions(conf,readerConfig.getWhere());
|
||||||
|
}
|
||||||
PhoenixEmbeddedDriver.ConnectionInfo info = null;
|
PhoenixEmbeddedDriver.ConnectionInfo info = null;
|
||||||
try {
|
try {
|
||||||
info = PhoenixEmbeddedDriver.ConnectionInfo.create(zkUrl);
|
info = PhoenixEmbeddedDriver.ConnectionInfo.create(zkUrl);
|
||||||
@ -67,15 +69,19 @@ public class HbaseSQLHelper {
|
|||||||
conf.setInt(HConstants.ZOOKEEPER_CLIENT_PORT, info.getPort());
|
conf.setInt(HConstants.ZOOKEEPER_CLIENT_PORT, info.getPort());
|
||||||
if (info.getRootNode() != null)
|
if (info.getRootNode() != null)
|
||||||
conf.set(HConstants.ZOOKEEPER_ZNODE_PARENT, info.getRootNode());
|
conf.set(HConstants.ZOOKEEPER_ZNODE_PARENT, info.getRootNode());
|
||||||
|
conf.set(Key.NAME_SPACE_MAPPING_ENABLED,"true");
|
||||||
|
conf.set(Key.SYSTEM_TABLES_TO_NAMESPACE,"true");
|
||||||
return conf;
|
return conf;
|
||||||
}
|
}
|
||||||
|
|
||||||
public static List<String> getPColumnNames(String connectionString, String tableName) throws SQLException {
|
public static List<String> getPColumnNames(String connectionString, String tableName,String schema) throws SQLException {
|
||||||
Connection con =
|
Properties pro = new Properties();
|
||||||
DriverManager.getConnection(connectionString);
|
pro.put(Key.NAME_SPACE_MAPPING_ENABLED, true);
|
||||||
|
pro.put(Key.SYSTEM_TABLES_TO_NAMESPACE, true);
|
||||||
|
Connection con = DriverManager.getConnection(connectionString,pro);
|
||||||
PhoenixConnection phoenixConnection = con.unwrap(PhoenixConnection.class);
|
PhoenixConnection phoenixConnection = con.unwrap(PhoenixConnection.class);
|
||||||
MetaDataClient metaDataClient = new MetaDataClient(phoenixConnection);
|
MetaDataClient metaDataClient = new MetaDataClient(phoenixConnection);
|
||||||
PTable table = metaDataClient.updateCache("", tableName).getTable();
|
PTable table = metaDataClient.updateCache(schema, tableName).getTable();
|
||||||
List<String> columnNames = new ArrayList<String>();
|
List<String> columnNames = new ArrayList<String>();
|
||||||
for (PColumn pColumn : table.getColumns()) {
|
for (PColumn pColumn : table.getColumns()) {
|
||||||
if (!pColumn.getName().getString().equals(SaltingUtil.SALTING_COLUMN_NAME))
|
if (!pColumn.getName().getString().equals(SaltingUtil.SALTING_COLUMN_NAME))
|
||||||
|
|||||||
@ -9,6 +9,7 @@ import org.slf4j.LoggerFactory;
|
|||||||
|
|
||||||
import java.sql.SQLException;
|
import java.sql.SQLException;
|
||||||
import java.util.List;
|
import java.util.List;
|
||||||
|
import java.util.StringJoiner;
|
||||||
|
|
||||||
public class HbaseSQLReaderConfig {
|
public class HbaseSQLReaderConfig {
|
||||||
private final static Logger LOG = LoggerFactory.getLogger(HbaseSQLReaderConfig.class);
|
private final static Logger LOG = LoggerFactory.getLogger(HbaseSQLReaderConfig.class);
|
||||||
@ -27,6 +28,9 @@ public class HbaseSQLReaderConfig {
|
|||||||
private String tableName;
|
private String tableName;
|
||||||
private List<String> columns; // 目的表的所有列的列名，包括主键和非主键，不包括时间列
|
private List<String> columns; // 目的表的所有列的列名，包括主键和非主键，不包括时间列
|
||||||
|
|
||||||
|
private String where;//条件
|
||||||
|
|
||||||
|
private String schema;//
|
||||||
/**
|
/**
|
||||||
* @return 获取原始的datax配置
|
* @return 获取原始的datax配置
|
||||||
*/
|
*/
|
||||||
@ -96,22 +100,27 @@ public class HbaseSQLReaderConfig {
|
|||||||
}
|
}
|
||||||
String zkQuorum = zkCfg.getFirst();
|
String zkQuorum = zkCfg.getFirst();
|
||||||
String znode = zkCfg.getSecond();
|
String znode = zkCfg.getSecond();
|
||||||
|
|
||||||
if (zkQuorum == null || zkQuorum.isEmpty()) {
|
if (zkQuorum == null || zkQuorum.isEmpty()) {
|
||||||
throw DataXException.asDataXException(
|
throw DataXException.asDataXException(
|
||||||
HbaseSQLReaderErrorCode.ILLEGAL_VALUE, "HBase的hbase.zookeeper.quorum配置不能为空" );
|
HbaseSQLReaderErrorCode.ILLEGAL_VALUE, "HBase的hbase.zookeeper.quorum配置不能为空" );
|
||||||
}
|
}
|
||||||
// 生成sql使用的连接字符串， 格式： jdbc:hbase:zk_quorum:2181:/znode_parent
|
// 生成sql使用的连接字符串， 格式： jdbc:hbase:zk_quorum:2181:/znode_parent
|
||||||
cfg.connectionString = "jdbc:phoenix:" + zkQuorum;
|
StringBuilder connectionString=new StringBuilder("jdbc:phoenix:");
|
||||||
cfg.zkUrl = zkQuorum + ":2181";
|
connectionString.append(zkQuorum);
|
||||||
|
cfg.connectionString = connectionString.toString();
|
||||||
|
StringBuilder zkUrl =new StringBuilder(zkQuorum);
|
||||||
|
cfg.zkUrl = zkUrl.append(":2181").toString();
|
||||||
if (!znode.isEmpty()) {
|
if (!znode.isEmpty()) {
|
||||||
cfg.connectionString += cfg.connectionString + ":" + znode;
|
cfg.connectionString = connectionString.append(":").append(znode).toString();
|
||||||
cfg.zkUrl += cfg.zkUrl + ":" + znode;
|
cfg.zkUrl=zkUrl.append(":").append(znode).toString();
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
private static void parseTableConfig(HbaseSQLReaderConfig cfg, Configuration dataxCfg) {
|
private static void parseTableConfig(HbaseSQLReaderConfig cfg, Configuration dataxCfg) {
|
||||||
// 解析并检查表名
|
// 解析并检查表名
|
||||||
cfg.tableName = dataxCfg.getString(Key.TABLE);
|
cfg.tableName = dataxCfg.getString(Key.TABLE);
|
||||||
|
cfg.schema = dataxCfg.getString(Key.SCHEMA);
|
||||||
if (cfg.tableName == null || cfg.tableName.isEmpty()) {
|
if (cfg.tableName == null || cfg.tableName.isEmpty()) {
|
||||||
throw DataXException.asDataXException(
|
throw DataXException.asDataXException(
|
||||||
HbaseSQLReaderErrorCode.ILLEGAL_VALUE, "HBase的tableName配置不能为空,请检查并修改配置." );
|
HbaseSQLReaderErrorCode.ILLEGAL_VALUE, "HBase的tableName配置不能为空,请检查并修改配置." );
|
||||||
@ -124,13 +133,14 @@ public class HbaseSQLReaderConfig {
|
|||||||
HbaseSQLReaderErrorCode.ILLEGAL_VALUE, "您配置的tableName含有非法字符{0}，请检查您的配置.");
|
HbaseSQLReaderErrorCode.ILLEGAL_VALUE, "您配置的tableName含有非法字符{0}，请检查您的配置.");
|
||||||
} else if (cfg.columns.isEmpty()) {
|
} else if (cfg.columns.isEmpty()) {
|
||||||
try {
|
try {
|
||||||
cfg.columns = HbaseSQLHelper.getPColumnNames(cfg.connectionString, cfg.tableName);
|
cfg.columns = HbaseSQLHelper.getPColumnNames(cfg.connectionString, cfg.tableName,cfg.schema);
|
||||||
dataxCfg.set(Key.COLUMN, cfg.columns);
|
dataxCfg.set(Key.COLUMN, cfg.columns);
|
||||||
} catch (SQLException e) {
|
} catch (SQLException e) {
|
||||||
throw DataXException.asDataXException(
|
throw DataXException.asDataXException(
|
||||||
HbaseSQLReaderErrorCode.GET_PHOENIX_COLUMN_ERROR, "HBase的columns配置不能为空,请添加目标表的列名配置." + e.getMessage(), e);
|
HbaseSQLReaderErrorCode.GET_PHOENIX_COLUMN_ERROR, "HBase的columns配置不能为空,请添加目标表的列名配置." + e.getMessage(), e);
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
cfg.where=dataxCfg.getString(Key.WHERE);
|
||||||
}
|
}
|
||||||
|
|
||||||
@Override
|
@Override
|
||||||
@ -151,6 +161,8 @@ public class HbaseSQLReaderConfig {
|
|||||||
ret.append(",");
|
ret.append(",");
|
||||||
}
|
}
|
||||||
ret.setLength(ret.length() - 1);
|
ret.setLength(ret.length() - 1);
|
||||||
|
ret.append("[where=]").append(getWhere());
|
||||||
|
ret.append("[schema=]").append(getSchema());
|
||||||
ret.append("\n");
|
ret.append("\n");
|
||||||
|
|
||||||
return ret.toString();
|
return ret.toString();
|
||||||
@ -161,4 +173,20 @@ public class HbaseSQLReaderConfig {
|
|||||||
*/
|
*/
|
||||||
private HbaseSQLReaderConfig() {
|
private HbaseSQLReaderConfig() {
|
||||||
}
|
}
|
||||||
|
|
||||||
|
public String getWhere() {
|
||||||
|
return where;
|
||||||
|
}
|
||||||
|
|
||||||
|
public void setWhere(String where) {
|
||||||
|
this.where = where;
|
||||||
|
}
|
||||||
|
|
||||||
|
public String getSchema() {
|
||||||
|
return schema;
|
||||||
|
}
|
||||||
|
|
||||||
|
public void setSchema(String schema) {
|
||||||
|
this.schema = schema;
|
||||||
|
}
|
||||||
}
|
}
|
||||||
|
|||||||
@ -19,10 +19,8 @@ import org.slf4j.LoggerFactory;
|
|||||||
import java.io.IOException;
|
import java.io.IOException;
|
||||||
import java.math.BigDecimal;
|
import java.math.BigDecimal;
|
||||||
import java.sql.*;
|
import java.sql.*;
|
||||||
import java.util.HashMap;
|
import java.sql.Date;
|
||||||
import java.util.LinkedHashMap;
|
import java.util.*;
|
||||||
import java.util.List;
|
|
||||||
import java.util.Map;
|
|
||||||
|
|
||||||
/**
|
/**
|
||||||
* Created by admin on 1/3/18.
|
* Created by admin on 1/3/18.
|
||||||
@ -42,11 +40,14 @@ public class HbaseSQLReaderTask {
|
|||||||
}
|
}
|
||||||
|
|
||||||
private void getPColumns() throws SQLException {
|
private void getPColumns() throws SQLException {
|
||||||
|
Properties pro = new Properties();
|
||||||
|
pro.put(Key.NAME_SPACE_MAPPING_ENABLED, true);
|
||||||
|
pro.put(Key.SYSTEM_TABLES_TO_NAMESPACE, true);
|
||||||
Connection con =
|
Connection con =
|
||||||
DriverManager.getConnection(this.readerConfig.getConnectionString());
|
DriverManager.getConnection(this.readerConfig.getConnectionString(),pro);
|
||||||
PhoenixConnection phoenixConnection = con.unwrap(PhoenixConnection.class);
|
PhoenixConnection phoenixConnection = con.unwrap(PhoenixConnection.class);
|
||||||
MetaDataClient metaDataClient = new MetaDataClient(phoenixConnection);
|
MetaDataClient metaDataClient = new MetaDataClient(phoenixConnection);
|
||||||
PTable table = metaDataClient.updateCache("", this.readerConfig.getTableName()).getTable();
|
PTable table = metaDataClient.updateCache(this.readerConfig.getSchema(), this.readerConfig.getTableName()).getTable();
|
||||||
List<String> columnNames = this.readerConfig.getColumns();
|
List<String> columnNames = this.readerConfig.getColumns();
|
||||||
for (PColumn pColumn : table.getColumns()) {
|
for (PColumn pColumn : table.getColumns()) {
|
||||||
if (columnNames.contains(pColumn.getName().getString())) {
|
if (columnNames.contains(pColumn.getName().getString())) {
|
||||||
|
|||||||
@ -24,5 +24,18 @@ public final class Key {
|
|||||||
* 【必选】列配置
|
* 【必选】列配置
|
||||||
*/
|
*/
|
||||||
public final static String COLUMN = "column";
|
public final static String COLUMN = "column";
|
||||||
|
/**
|
||||||
|
*
|
||||||
|
*/
|
||||||
|
public static final String WHERE = "where";
|
||||||
|
|
||||||
|
/**
|
||||||
|
* 【可选】Phoenix表所属schema，默认为空
|
||||||
|
*/
|
||||||
|
public static final String SCHEMA = "schema";
|
||||||
|
|
||||||
|
public static final String NAME_SPACE_MAPPING_ENABLED = "phoenix.schema.isNamespaceMappingEnabled";
|
||||||
|
|
||||||
|
public static final String SYSTEM_TABLES_TO_NAMESPACE = "phoenix.schema.mapSystemTablesToNamespace";
|
||||||
|
|
||||||
}
|
}
|
||||||
|
|||||||
@ -1,5 +1,6 @@
|
|||||||
<?xml version="1.0" encoding="UTF-8"?>
|
<?xml version="1.0" encoding="UTF-8"?>
|
||||||
<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
|
<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
|
||||||
|
xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
|
||||||
<parent>
|
<parent>
|
||||||
<artifactId>datax-all</artifactId>
|
<artifactId>datax-all</artifactId>
|
||||||
<groupId>com.alibaba.datax</groupId>
|
<groupId>com.alibaba.datax</groupId>
|
||||||
@ -111,6 +112,42 @@
|
|||||||
<version>${datax-project-version}</version>
|
<version>${datax-project-version}</version>
|
||||||
</dependency>
|
</dependency>
|
||||||
|
|
||||||
|
<dependency>
|
||||||
|
<groupId>org.apache.parquet</groupId>
|
||||||
|
<artifactId>parquet-column</artifactId>
|
||||||
|
<version>1.12.0</version>
|
||||||
|
</dependency>
|
||||||
|
<dependency>
|
||||||
|
<groupId>org.apache.parquet</groupId>
|
||||||
|
<artifactId>parquet-avro</artifactId>
|
||||||
|
<version>1.12.0</version>
|
||||||
|
</dependency>
|
||||||
|
<dependency>
|
||||||
|
<groupId>org.apache.parquet</groupId>
|
||||||
|
<artifactId>parquet-common</artifactId>
|
||||||
|
<version>1.12.0</version>
|
||||||
|
</dependency>
|
||||||
|
<dependency>
|
||||||
|
<groupId>org.apache.parquet</groupId>
|
||||||
|
<artifactId>parquet-format</artifactId>
|
||||||
|
<version>2.3.0</version>
|
||||||
|
</dependency>
|
||||||
|
<dependency>
|
||||||
|
<groupId>org.apache.parquet</groupId>
|
||||||
|
<artifactId>parquet-jackson</artifactId>
|
||||||
|
<version>1.12.0</version>
|
||||||
|
</dependency>
|
||||||
|
<dependency>
|
||||||
|
<groupId>org.apache.parquet</groupId>
|
||||||
|
<artifactId>parquet-encoding</artifactId>
|
||||||
|
<version>1.12.0</version>
|
||||||
|
</dependency>
|
||||||
|
<dependency>
|
||||||
|
<groupId>org.apache.parquet</groupId>
|
||||||
|
<artifactId>parquet-hadoop</artifactId>
|
||||||
|
<version>1.12.0</version>
|
||||||
|
</dependency>
|
||||||
|
|
||||||
</dependencies>
|
</dependencies>
|
||||||
|
|
||||||
<build>
|
<build>
|
||||||
|
|||||||
@ -37,6 +37,28 @@
|
|||||||
<!--</includes>-->
|
<!--</includes>-->
|
||||||
<!--<outputDirectory>plugin/reader/hdfsreader/libs</outputDirectory>-->
|
<!--<outputDirectory>plugin/reader/hdfsreader/libs</outputDirectory>-->
|
||||||
<!--</fileSet>-->
|
<!--</fileSet>-->
|
||||||
|
<!--<fileSet>-->
|
||||||
|
<!--<directory>src/main/libs</directory>-->
|
||||||
|
<!--<includes>-->
|
||||||
|
<!--<include>*.*</include>-->
|
||||||
|
<!--</includes>-->
|
||||||
|
<!--<outputDirectory>plugin/reader/hdfsreader/libs</outputDirectory>-->
|
||||||
|
<!--</fileSet>-->
|
||||||
|
|
||||||
|
<fileSet>
|
||||||
|
<directory>src/main/libs</directory>
|
||||||
|
<includes>
|
||||||
|
<include>*.*</include>
|
||||||
|
</includes>
|
||||||
|
<outputDirectory>plugin/reader/ossreader/libs</outputDirectory>
|
||||||
|
</fileSet>
|
||||||
|
<fileSet>
|
||||||
|
<directory>src/main/libs</directory>
|
||||||
|
<includes>
|
||||||
|
<include>*.*</include>
|
||||||
|
</includes>
|
||||||
|
<outputDirectory>plugin/reader/hivereader/libs</outputDirectory>
|
||||||
|
</fileSet>
|
||||||
</fileSets>
|
</fileSets>
|
||||||
|
|
||||||
<dependencySets>
|
<dependencySets>
|
||||||
|
|||||||
@ -10,4 +10,5 @@ public class Constant {
|
|||||||
public static final String CSV = "CSV";
|
public static final String CSV = "CSV";
|
||||||
public static final String SEQ = "SEQ";
|
public static final String SEQ = "SEQ";
|
||||||
public static final String RC = "RC";
|
public static final String RC = "RC";
|
||||||
|
public static final String PARQUET = "PARQUET";
|
||||||
}
|
}
|
||||||
|
|||||||
@ -9,12 +9,16 @@ import com.alibaba.datax.plugin.unstructuredstorage.reader.ColumnEntry;
|
|||||||
import com.alibaba.datax.plugin.unstructuredstorage.reader.UnstructuredStorageReaderErrorCode;
|
import com.alibaba.datax.plugin.unstructuredstorage.reader.UnstructuredStorageReaderErrorCode;
|
||||||
import com.alibaba.datax.plugin.unstructuredstorage.reader.UnstructuredStorageReaderUtil;
|
import com.alibaba.datax.plugin.unstructuredstorage.reader.UnstructuredStorageReaderUtil;
|
||||||
import com.alibaba.fastjson2.JSON;
|
import com.alibaba.fastjson2.JSON;
|
||||||
|
import com.alibaba.fastjson2.JSONArray;
|
||||||
import com.alibaba.fastjson2.JSONObject;
|
import com.alibaba.fastjson2.JSONObject;
|
||||||
|
import org.apache.commons.lang3.BooleanUtils;
|
||||||
import org.apache.commons.lang3.StringUtils;
|
import org.apache.commons.lang3.StringUtils;
|
||||||
|
import org.apache.commons.lang3.exception.ExceptionUtils;
|
||||||
import org.apache.hadoop.fs.FSDataInputStream;
|
import org.apache.hadoop.fs.FSDataInputStream;
|
||||||
import org.apache.hadoop.fs.FileStatus;
|
import org.apache.hadoop.fs.FileStatus;
|
||||||
import org.apache.hadoop.fs.FileSystem;
|
import org.apache.hadoop.fs.FileSystem;
|
||||||
import org.apache.hadoop.fs.Path;
|
import org.apache.hadoop.fs.Path;
|
||||||
|
import org.apache.hadoop.hive.common.type.HiveDecimal;
|
||||||
import org.apache.hadoop.hive.ql.io.RCFile;
|
import org.apache.hadoop.hive.ql.io.RCFile;
|
||||||
import org.apache.hadoop.hive.ql.io.RCFileRecordReader;
|
import org.apache.hadoop.hive.ql.io.RCFileRecordReader;
|
||||||
import org.apache.hadoop.hive.ql.io.orc.OrcFile;
|
import org.apache.hadoop.hive.ql.io.orc.OrcFile;
|
||||||
@ -29,14 +33,30 @@ import org.apache.hadoop.io.*;
|
|||||||
import org.apache.hadoop.mapred.*;
|
import org.apache.hadoop.mapred.*;
|
||||||
import org.apache.hadoop.security.UserGroupInformation;
|
import org.apache.hadoop.security.UserGroupInformation;
|
||||||
import org.apache.hadoop.util.ReflectionUtils;
|
import org.apache.hadoop.util.ReflectionUtils;
|
||||||
|
import org.apache.parquet.example.data.Group;
|
||||||
|
import org.apache.parquet.hadoop.ParquetReader;
|
||||||
|
import org.apache.parquet.hadoop.example.GroupReadSupport;
|
||||||
|
import org.apache.parquet.hadoop.util.HadoopInputFile;
|
||||||
|
import org.apache.parquet.io.api.Binary;
|
||||||
|
import org.apache.parquet.schema.MessageType;
|
||||||
|
import org.apache.parquet.schema.MessageTypeParser;
|
||||||
|
import org.apache.parquet.schema.PrimitiveType;
|
||||||
import org.slf4j.Logger;
|
import org.slf4j.Logger;
|
||||||
import org.slf4j.LoggerFactory;
|
import org.slf4j.LoggerFactory;
|
||||||
|
|
||||||
import java.io.IOException;
|
import java.io.IOException;
|
||||||
import java.io.InputStream;
|
import java.io.InputStream;
|
||||||
import java.nio.ByteBuffer;
|
import java.nio.ByteBuffer;
|
||||||
|
import java.nio.ByteOrder;
|
||||||
|
import java.sql.Timestamp;
|
||||||
import java.text.SimpleDateFormat;
|
import java.text.SimpleDateFormat;
|
||||||
|
import java.time.LocalDate;
|
||||||
|
import java.time.LocalDateTime;
|
||||||
|
import java.time.LocalTime;
|
||||||
import java.util.*;
|
import java.util.*;
|
||||||
|
import java.util.concurrent.ConcurrentHashMap;
|
||||||
|
import java.util.concurrent.TimeUnit;
|
||||||
|
import java.util.stream.Collectors;
|
||||||
|
|
||||||
/**
|
/**
|
||||||
* Created by mingya.wmy on 2015/8/12.
|
* Created by mingya.wmy on 2015/8/12.
|
||||||
@ -56,6 +76,10 @@ public class DFSUtil {
|
|||||||
public static final String HDFS_DEFAULTFS_KEY = "fs.defaultFS";
|
public static final String HDFS_DEFAULTFS_KEY = "fs.defaultFS";
|
||||||
public static final String HADOOP_SECURITY_AUTHENTICATION_KEY = "hadoop.security.authentication";
|
public static final String HADOOP_SECURITY_AUTHENTICATION_KEY = "hadoop.security.authentication";
|
||||||
|
|
||||||
|
private Boolean skipEmptyOrcFile = false;
|
||||||
|
|
||||||
|
private Integer orcFileEmptySize = null;
|
||||||
|
|
||||||
|
|
||||||
public DFSUtil(Configuration taskConfig) {
|
public DFSUtil(Configuration taskConfig) {
|
||||||
hadoopConf = new org.apache.hadoop.conf.Configuration();
|
hadoopConf = new org.apache.hadoop.conf.Configuration();
|
||||||
@ -79,6 +103,7 @@ public class DFSUtil {
|
|||||||
this.hadoopConf.set(HADOOP_SECURITY_AUTHENTICATION_KEY, "kerberos");
|
this.hadoopConf.set(HADOOP_SECURITY_AUTHENTICATION_KEY, "kerberos");
|
||||||
}
|
}
|
||||||
this.kerberosAuthentication(this.kerberosPrincipal, this.kerberosKeytabFilePath);
|
this.kerberosAuthentication(this.kerberosPrincipal, this.kerberosKeytabFilePath);
|
||||||
|
this.skipEmptyOrcFile = taskConfig.getBool(Key.SKIP_EMPTY_ORCFILE, false);
|
||||||
|
|
||||||
LOG.info(String.format("hadoopConfig details:%s", JSON.toJSONString(this.hadoopConf)));
|
LOG.info(String.format("hadoopConfig details:%s", JSON.toJSONString(this.hadoopConf)));
|
||||||
}
|
}
|
||||||
@ -102,10 +127,11 @@ public class DFSUtil {
|
|||||||
* @param srcPaths 路径列表
|
* @param srcPaths 路径列表
|
||||||
* @param specifiedFileType 指定文件类型
|
* @param specifiedFileType 指定文件类型
|
||||||
*/
|
*/
|
||||||
public HashSet<String> getAllFiles(List<String> srcPaths, String specifiedFileType) {
|
public HashSet<String> getAllFiles(List<String> srcPaths, String specifiedFileType, Boolean skipEmptyOrcFile, Integer orcFileEmptySize) {
|
||||||
|
|
||||||
this.specifiedFileType = specifiedFileType;
|
this.specifiedFileType = specifiedFileType;
|
||||||
|
this.skipEmptyOrcFile = skipEmptyOrcFile;
|
||||||
|
this.orcFileEmptySize = orcFileEmptySize;
|
||||||
if (!srcPaths.isEmpty()) {
|
if (!srcPaths.isEmpty()) {
|
||||||
for (String eachPath : srcPaths) {
|
for (String eachPath : srcPaths) {
|
||||||
LOG.info(String.format("get HDFS all files in path = [%s]", eachPath));
|
LOG.info(String.format("get HDFS all files in path = [%s]", eachPath));
|
||||||
@ -127,9 +153,13 @@ public class DFSUtil {
|
|||||||
FileStatus stats[] = hdfs.globStatus(path);
|
FileStatus stats[] = hdfs.globStatus(path);
|
||||||
for (FileStatus f : stats) {
|
for (FileStatus f : stats) {
|
||||||
if (f.isFile()) {
|
if (f.isFile()) {
|
||||||
if (f.getLen() == 0) {
|
long fileLength = f.getLen();
|
||||||
|
if (fileLength == 0) {
|
||||||
String message = String.format("文件[%s]长度为0，将会跳过不作处理！", hdfsPath);
|
String message = String.format("文件[%s]长度为0，将会跳过不作处理！", hdfsPath);
|
||||||
LOG.warn(message);
|
LOG.warn(message);
|
||||||
|
} else if (BooleanUtils.isTrue(this.skipEmptyOrcFile) && this.orcFileEmptySize != null && fileLength <= this.orcFileEmptySize) {
|
||||||
|
String message = String.format("The orc file [%s] is empty, file size: %s, DataX will skip it !", f.getPath().toString(), fileLength);
|
||||||
|
LOG.warn(message);
|
||||||
} else {
|
} else {
|
||||||
addSourceFileByType(f.getPath().toString());
|
addSourceFileByType(f.getPath().toString());
|
||||||
}
|
}
|
||||||
@ -167,7 +197,16 @@ public class DFSUtil {
|
|||||||
LOG.info(String.format("[%s] 是目录, 递归获取该目录下的文件", f.getPath().toString()));
|
LOG.info(String.format("[%s] 是目录, 递归获取该目录下的文件", f.getPath().toString()));
|
||||||
getHDFSAllFilesNORegex(f.getPath().toString(), hdfs);
|
getHDFSAllFilesNORegex(f.getPath().toString(), hdfs);
|
||||||
} else if (f.isFile()) {
|
} else if (f.isFile()) {
|
||||||
|
long fileLength = f.getLen();
|
||||||
|
if (fileLength == 0) {
|
||||||
|
String message = String.format("The file [%s] is empty, DataX will skip it !", f.getPath().toString());
|
||||||
|
LOG.warn(message);
|
||||||
|
continue;
|
||||||
|
} else if (BooleanUtils.isTrue(this.skipEmptyOrcFile) && this.orcFileEmptySize != null && fileLength <= this.orcFileEmptySize) {
|
||||||
|
String message = String.format("The orc file [%s] is empty, file size: %s, DataX will skip it !", f.getPath().toString(), fileLength);
|
||||||
|
LOG.warn(message);
|
||||||
|
continue;
|
||||||
|
}
|
||||||
addSourceFileByType(f.getPath().toString());
|
addSourceFileByType(f.getPath().toString());
|
||||||
} else {
|
} else {
|
||||||
String message = String.format("该路径[%s]文件类型既不是目录也不是文件，插件自动忽略。",
|
String message = String.format("该路径[%s]文件类型既不是目录也不是文件，插件自动忽略。",
|
||||||
@ -332,7 +371,19 @@ public class DFSUtil {
|
|||||||
//Each file as a split
|
//Each file as a split
|
||||||
//TODO multy threads
|
//TODO multy threads
|
||||||
// OrcInputFormat getSplits params numSplits not used, splits size = block numbers
|
// OrcInputFormat getSplits params numSplits not used, splits size = block numbers
|
||||||
InputSplit[] splits = in.getSplits(conf, -1);
|
InputSplit[] splits;
|
||||||
|
try {
|
||||||
|
splits = in.getSplits(conf, 1);
|
||||||
|
} catch (Exception splitException) {
|
||||||
|
if (Boolean.TRUE.equals(this.skipEmptyOrcFile)) {
|
||||||
|
boolean isOrcFileEmptyException = checkIsOrcEmptyFileExecption(splitException);
|
||||||
|
if (isOrcFileEmptyException) {
|
||||||
|
LOG.info("skipEmptyOrcFile: true, \"{}\" is an empty orc file, skip it!", sourceOrcFilePath);
|
||||||
|
return;
|
||||||
|
}
|
||||||
|
}
|
||||||
|
throw splitException;
|
||||||
|
}
|
||||||
for (InputSplit split : splits) {
|
for (InputSplit split : splits) {
|
||||||
{
|
{
|
||||||
RecordReader reader = in.getRecordReader(split, conf, Reporter.NULL);
|
RecordReader reader = in.getRecordReader(split, conf, Reporter.NULL);
|
||||||
@ -349,8 +400,11 @@ public class DFSUtil {
|
|||||||
Object field = inspector.getStructFieldData(value, fields.get(i));
|
Object field = inspector.getStructFieldData(value, fields.get(i));
|
||||||
recordFields.add(field);
|
recordFields.add(field);
|
||||||
}
|
}
|
||||||
|
List<ColumnEntry> hivePartitionColumnEntrys = UnstructuredStorageReaderUtil.getListColumnEntry(readerSliceConfig, com.alibaba.datax.plugin.unstructuredstorage.reader.Key.HIVE_PARTION_COLUMN);
|
||||||
|
ArrayList<Column> hivePartitionColumns = new ArrayList<>();
|
||||||
|
hivePartitionColumns = UnstructuredStorageReaderUtil.getHivePartitionColumns(sourceOrcFilePath, hivePartitionColumnEntrys);
|
||||||
transportOneRecord(column, recordFields, recordSender,
|
transportOneRecord(column, recordFields, recordSender,
|
||||||
taskPluginCollector, isReadAllColumns, nullFormat);
|
taskPluginCollector, isReadAllColumns, nullFormat,hivePartitionColumns);
|
||||||
}
|
}
|
||||||
reader.close();
|
reader.close();
|
||||||
}
|
}
|
||||||
@ -367,8 +421,20 @@ public class DFSUtil {
|
|||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
|
private boolean checkIsOrcEmptyFileExecption(Exception e) {
|
||||||
|
if (e == null) {
|
||||||
|
return false;
|
||||||
|
}
|
||||||
|
|
||||||
|
String fullStackTrace = ExceptionUtils.getStackTrace(e);
|
||||||
|
if (fullStackTrace.contains("org.apache.orc.impl.ReaderImpl.getRawDataSizeOfColumn") && fullStackTrace.contains("Caused by: java.lang.IndexOutOfBoundsException: Index: 1, Size: 1")) {
|
||||||
|
return true;
|
||||||
|
}
|
||||||
|
return false;
|
||||||
|
}
|
||||||
|
|
||||||
private Record transportOneRecord(List<ColumnEntry> columnConfigs, List<Object> recordFields
|
private Record transportOneRecord(List<ColumnEntry> columnConfigs, List<Object> recordFields
|
||||||
, RecordSender recordSender, TaskPluginCollector taskPluginCollector, boolean isReadAllColumns, String nullFormat) {
|
, RecordSender recordSender, TaskPluginCollector taskPluginCollector, boolean isReadAllColumns, String nullFormat, ArrayList<Column> hiveParitionColumns) {
|
||||||
Record record = recordSender.createRecord();
|
Record record = recordSender.createRecord();
|
||||||
Column columnGenerated;
|
Column columnGenerated;
|
||||||
try {
|
try {
|
||||||
@ -555,8 +621,9 @@ public class DFSUtil {
|
|||||||
} else if (StringUtils.equalsIgnoreCase(specifiedFileType, Constant.SEQ)) {
|
} else if (StringUtils.equalsIgnoreCase(specifiedFileType, Constant.SEQ)) {
|
||||||
|
|
||||||
return isSequenceFile(filepath, in);
|
return isSequenceFile(filepath, in);
|
||||||
|
} else if (StringUtils.equalsIgnoreCase(specifiedFileType, Constant.PARQUET)) {
|
||||||
|
return true;
|
||||||
}
|
}
|
||||||
|
|
||||||
} catch (Exception e) {
|
} catch (Exception e) {
|
||||||
String message = String.format("检查文件[%s]类型失败，目前支持ORC,SEQUENCE,RCFile,TEXT,CSV五种格式的文件," +
|
String message = String.format("检查文件[%s]类型失败，目前支持ORC,SEQUENCE,RCFile,TEXT,CSV五种格式的文件," +
|
||||||
"请检查您文件类型和文件是否正确。", filepath);
|
"请检查您文件类型和文件是否正确。", filepath);
|
||||||
@ -693,4 +760,332 @@ public class DFSUtil {
|
|||||||
return false;
|
return false;
|
||||||
}
|
}
|
||||||
|
|
||||||
|
public void parquetFileStartRead(String sourceParquetFilePath, Configuration readerSliceConfig, RecordSender recordSender, TaskPluginCollector taskPluginCollector) {
|
||||||
|
String schemaString = readerSliceConfig.getString(Key.PARQUET_SCHEMA);
|
||||||
|
if (StringUtils.isNotBlank(schemaString)) {
|
||||||
|
LOG.info("You config parquet schema, use it {}", schemaString);
|
||||||
|
} else {
|
||||||
|
schemaString = getParquetSchema(sourceParquetFilePath, hadoopConf);
|
||||||
|
LOG.info("Parquet schema parsed from: {} , schema is {}", sourceParquetFilePath, schemaString);
|
||||||
|
if (StringUtils.isBlank(schemaString)) {
|
||||||
|
throw DataXException.asDataXException("ParquetSchema is required, please check your config");
|
||||||
|
}
|
||||||
|
}
|
||||||
|
MessageType parquetSchema = null;
|
||||||
|
List<org.apache.parquet.schema.Type> parquetTypes = null;
|
||||||
|
Map<String, ParquetMeta> parquetMetaMap = null;
|
||||||
|
int fieldCount = 0;
|
||||||
|
try {
|
||||||
|
parquetSchema = MessageTypeParser.parseMessageType(schemaString);
|
||||||
|
fieldCount = parquetSchema.getFieldCount();
|
||||||
|
parquetTypes = parquetSchema.getFields();
|
||||||
|
parquetMetaMap = ParquetMessageHelper.parseParquetTypes(parquetTypes);
|
||||||
|
} catch (Exception e) {
|
||||||
|
String message = String.format("Error parsing to MessageType via Schema string [%s]", schemaString);
|
||||||
|
LOG.error(message);
|
||||||
|
throw DataXException.asDataXException(HdfsReaderErrorCode.PARSE_MESSAGE_TYPE_FROM_SCHEMA_ERROR, e);
|
||||||
|
}
|
||||||
|
List<ColumnEntry> column = UnstructuredStorageReaderUtil.getListColumnEntry(readerSliceConfig, com.alibaba.datax.plugin.unstructuredstorage.reader.Key.COLUMN);
|
||||||
|
String nullFormat = readerSliceConfig.getString(com.alibaba.datax.plugin.unstructuredstorage.reader.Key.NULL_FORMAT);
|
||||||
|
boolean isUtcTimestamp = readerSliceConfig.getBool(Key.PARQUET_UTC_TIMESTAMP, false);
|
||||||
|
boolean isReadAllColumns = (column == null || column.size() == 0) ? true : false;
|
||||||
|
LOG.info("ReadingAllColums: " + isReadAllColumns);
|
||||||
|
|
||||||
|
/**
|
||||||
|
* 支持 hive 表中间加列场景
|
||||||
|
*
|
||||||
|
* 开关默认 false，在 hive表存在中间加列的场景打开，需要根据 name排序
|
||||||
|
* 不默认打开的原因
|
||||||
|
* 1、存量hdfs任务，只根据 index获取字段，无name字段配置
|
||||||
|
* 2、中间加列场景比较少
|
||||||
|
* 3、存量任务可能存在列错位的问题，不能随意纠正
|
||||||
|
*/
|
||||||
|
boolean supportAddMiddleColumn = readerSliceConfig.getBool(Key.SUPPORT_ADD_MIDDLE_COLUMN, false);
|
||||||
|
|
||||||
|
boolean printNullValueException = readerSliceConfig.getBool("printNullValueException", false);
|
||||||
|
List<Integer> ignoreIndex = readerSliceConfig.getList("ignoreIndex", new ArrayList<Integer>(), Integer.class);
|
||||||
|
|
||||||
|
JobConf conf = new JobConf(hadoopConf);
|
||||||
|
ParquetReader<Group> reader = null;
|
||||||
|
try {
|
||||||
|
Path parquetFilePath = new Path(sourceParquetFilePath);
|
||||||
|
GroupReadSupport readSupport = new GroupReadSupport();
|
||||||
|
readSupport.init(conf, null, parquetSchema);
|
||||||
|
// 这里初始化parquetReader的时候，会getFileSystem，如果是HA集群，期间会根据hadoopConfig中区加载failover类，这里初始化builder带上conf
|
||||||
|
ParquetReader.Builder parquetReaderBuilder = ParquetReader.builder(readSupport, parquetFilePath);
|
||||||
|
parquetReaderBuilder.withConf(hadoopConf);
|
||||||
|
reader = parquetReaderBuilder.build();
|
||||||
|
Group g = null;
|
||||||
|
|
||||||
|
// 从文件名中解析分区信息
|
||||||
|
List<ColumnEntry> hivePartitionColumnEntrys = UnstructuredStorageReaderUtil.getListColumnEntry(readerSliceConfig, com.alibaba.datax.plugin.unstructuredstorage.reader.Key.HIVE_PARTION_COLUMN);
|
||||||
|
ArrayList<Column> hivePartitionColumns = new ArrayList<>();
|
||||||
|
hivePartitionColumns = UnstructuredStorageReaderUtil.getHivePartitionColumns(sourceParquetFilePath, hivePartitionColumnEntrys);
|
||||||
|
List<String> schemaFieldList = null;
|
||||||
|
Map<Integer, String> colNameIndexMap = null;
|
||||||
|
Map<Integer, Integer> indexMap = null;
|
||||||
|
if (supportAddMiddleColumn) {
|
||||||
|
boolean nonName = column.stream().anyMatch(columnEntry -> StringUtils.isEmpty(columnEntry.getName()));
|
||||||
|
if (nonName) {
|
||||||
|
throw new DataXException("You configured column item without name, please correct it");
|
||||||
|
}
|
||||||
|
List<org.apache.parquet.schema.Type> parquetFileFields = getParquetFileFields(parquetFilePath, hadoopConf);
|
||||||
|
schemaFieldList = parquetFileFields.stream().map(org.apache.parquet.schema.Type::getName).collect(Collectors.toList());
|
||||||
|
colNameIndexMap = new ConcurrentHashMap<>();
|
||||||
|
Map<Integer, String> finalColNameIndexMap = colNameIndexMap;
|
||||||
|
column.forEach(columnEntry -> finalColNameIndexMap.put(columnEntry.getIndex(), columnEntry.getName()));
|
||||||
|
Iterator<Map.Entry<Integer, String>> iterator = finalColNameIndexMap.entrySet().iterator();
|
||||||
|
while (iterator.hasNext()) {
|
||||||
|
Map.Entry<Integer, String> next = iterator.next();
|
||||||
|
if (!schemaFieldList.contains(next.getValue())) {
|
||||||
|
finalColNameIndexMap.remove((next.getKey()));
|
||||||
|
}
|
||||||
|
}
|
||||||
|
LOG.info("SupportAddMiddleColumn is true, fields from parquet file is {}, " +
|
||||||
|
"colNameIndexMap is {}", JSON.toJSONString(schemaFieldList), JSON.toJSONString(colNameIndexMap));
|
||||||
|
fieldCount = column.size();
|
||||||
|
indexMap = new HashMap<>();
|
||||||
|
for (int j = 0; j < fieldCount; j++) {
|
||||||
|
if (colNameIndexMap.containsKey(j)) {
|
||||||
|
int index = findIndex(schemaFieldList, findEleInMap(colNameIndexMap, j));
|
||||||
|
indexMap.put(j, index);
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
while ((g = reader.read()) != null) {
|
||||||
|
List<Object> formattedRecord = new ArrayList<Object>(fieldCount);
|
||||||
|
try {
|
||||||
|
for (int j = 0; j < fieldCount; j++) {
|
||||||
|
Object data = null;
|
||||||
|
try {
|
||||||
|
if (null != ignoreIndex && !ignoreIndex.isEmpty() && ignoreIndex.contains(j)) {
|
||||||
|
data = null;
|
||||||
|
} else {
|
||||||
|
if (supportAddMiddleColumn) {
|
||||||
|
if (!colNameIndexMap.containsKey(j)) {
|
||||||
|
formattedRecord.add(null);
|
||||||
|
continue;
|
||||||
|
} else {
|
||||||
|
data = DFSUtil.this.readFields(g, parquetTypes.get(indexMap.get(j)), indexMap.get(j), parquetMetaMap, isUtcTimestamp);
|
||||||
|
}
|
||||||
|
} else {
|
||||||
|
data = DFSUtil.this.readFields(g, parquetTypes.get(j), j, parquetMetaMap, isUtcTimestamp);
|
||||||
|
}
|
||||||
|
}
|
||||||
|
} catch (RuntimeException e) {
|
||||||
|
if (printNullValueException) {
|
||||||
|
LOG.warn(e.getMessage());
|
||||||
|
}
|
||||||
|
}
|
||||||
|
formattedRecord.add(data);
|
||||||
|
}
|
||||||
|
transportOneRecord(column, formattedRecord, recordSender, taskPluginCollector, isReadAllColumns, nullFormat, hivePartitionColumns);
|
||||||
|
} catch (Exception e) {
|
||||||
|
throw DataXException.asDataXException(HdfsReaderErrorCode.READ_PARQUET_ERROR, e);
|
||||||
|
}
|
||||||
|
}
|
||||||
|
} catch (Exception e) {
|
||||||
|
throw DataXException.asDataXException(HdfsReaderErrorCode.READ_PARQUET_ERROR, e);
|
||||||
|
} finally {
|
||||||
|
org.apache.commons.io.IOUtils.closeQuietly(reader);
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
private String findEleInMap(Map<Integer, String> map, Integer key) {
|
||||||
|
Iterator<Map.Entry<Integer, String>> iterator = map.entrySet().iterator();
|
||||||
|
while (iterator.hasNext()) {
|
||||||
|
Map.Entry<Integer, String> next = iterator.next();
|
||||||
|
if (key.equals(next.getKey())) {
|
||||||
|
return next.getValue();
|
||||||
|
}
|
||||||
|
}
|
||||||
|
return null;
|
||||||
|
}
|
||||||
|
|
||||||
|
private int findIndex(List<String> schemaFieldList, String colName) {
|
||||||
|
for (int i = 0; i < schemaFieldList.size(); i++) {
|
||||||
|
if (schemaFieldList.get(i).equals(colName)) {
|
||||||
|
return i;
|
||||||
|
}
|
||||||
|
}
|
||||||
|
return -1;
|
||||||
|
}
|
||||||
|
|
||||||
|
private List<org.apache.parquet.schema.Type> getParquetFileFields(Path filePath, org.apache.hadoop.conf.Configuration configuration) {
|
||||||
|
try (org.apache.parquet.hadoop.ParquetFileReader reader = org.apache.parquet.hadoop.ParquetFileReader.open(HadoopInputFile.fromPath(filePath, configuration))) {
|
||||||
|
org.apache.parquet.schema.MessageType schema = reader.getFooter().getFileMetaData().getSchema();
|
||||||
|
List<org.apache.parquet.schema.Type> fields = schema.getFields();
|
||||||
|
return fields;
|
||||||
|
} catch (IOException e) {
|
||||||
|
LOG.error("Fetch parquet field error", e);
|
||||||
|
throw new DataXException(String.format("Fetch parquet field error, msg is %s", e.getMessage()));
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
private String getParquetSchema(String sourceParquetFilePath, org.apache.hadoop.conf.Configuration hadoopConf) {
|
||||||
|
GroupReadSupport readSupport = new GroupReadSupport();
|
||||||
|
ParquetReader.Builder parquetReaderBuilder = ParquetReader.builder(readSupport, new Path(sourceParquetFilePath));
|
||||||
|
ParquetReader<Group> reader = null;
|
||||||
|
try {
|
||||||
|
parquetReaderBuilder.withConf(hadoopConf);
|
||||||
|
reader = parquetReaderBuilder.build();
|
||||||
|
Group g = null;
|
||||||
|
if ((g = reader.read()) != null) {
|
||||||
|
return g.getType().toString();
|
||||||
|
}
|
||||||
|
} catch (Throwable e) {
|
||||||
|
LOG.error("Inner error, getParquetSchema failed, message is {}", e.getMessage());
|
||||||
|
} finally {
|
||||||
|
org.apache.commons.io.IOUtils.closeQuietly(reader);
|
||||||
|
}
|
||||||
|
return null;
|
||||||
|
}
|
||||||
|
|
||||||
|
/**
|
||||||
|
* parquet 相关
|
||||||
|
*/
|
||||||
|
private static final int JULIAN_EPOCH_OFFSET_DAYS = 2440588;
|
||||||
|
private static final long MILLIS_IN_DAY = TimeUnit.DAYS.toMillis(1);
|
||||||
|
private static final long NANOS_PER_MILLISECOND = TimeUnit.MILLISECONDS.toNanos(1);
|
||||||
|
|
||||||
|
private long julianDayToMillis(int julianDay) {
|
||||||
|
return (julianDay - JULIAN_EPOCH_OFFSET_DAYS) * MILLIS_IN_DAY;
|
||||||
|
}
|
||||||
|
|
||||||
|
private org.apache.parquet.schema.OriginalType getOriginalType(org.apache.parquet.schema.Type type, Map<String, ParquetMeta> parquetMetaMap) {
|
||||||
|
ParquetMeta meta = parquetMetaMap.get(type.getName());
|
||||||
|
return meta.getOriginalType();
|
||||||
|
}
|
||||||
|
|
||||||
|
private org.apache.parquet.schema.PrimitiveType asPrimitiveType(org.apache.parquet.schema.Type type, Map<String, ParquetMeta> parquetMetaMap) {
|
||||||
|
ParquetMeta meta = parquetMetaMap.get(type.getName());
|
||||||
|
return meta.getPrimitiveType();
|
||||||
|
}
|
||||||
|
|
||||||
|
private Object readFields(Group g, org.apache.parquet.schema.Type type, int index, Map<String, ParquetMeta> parquetMetaMap, boolean isUtcTimestamp) {
|
||||||
|
if (this.getOriginalType(type, parquetMetaMap) == org.apache.parquet.schema.OriginalType.MAP) {
|
||||||
|
Group groupData = g.getGroup(index, 0);
|
||||||
|
List<org.apache.parquet.schema.Type> parquetTypes = groupData.getType().getFields();
|
||||||
|
JSONObject data = new JSONObject();
|
||||||
|
for (int i = 0; i < parquetTypes.size(); i++) {
|
||||||
|
int j = groupData.getFieldRepetitionCount(i);
|
||||||
|
// map key value 的对数
|
||||||
|
for (int k = 0; k < j; k++) {
|
||||||
|
Group groupDataK = groupData.getGroup(0, k);
|
||||||
|
List<org.apache.parquet.schema.Type> parquetTypesK = groupDataK.getType().getFields();
|
||||||
|
if (2 != parquetTypesK.size()) {
|
||||||
|
// warn: 不是key value成对出现
|
||||||
|
throw new RuntimeException(String.format("bad parquet map type: %s", groupData.getValueToString(index, 0)));
|
||||||
|
}
|
||||||
|
Object subDataKey = this.readFields(groupDataK, parquetTypesK.get(0), 0, parquetMetaMap, isUtcTimestamp);
|
||||||
|
Object subDataValue = this.readFields(groupDataK, parquetTypesK.get(1), 1, parquetMetaMap, isUtcTimestamp);
|
||||||
|
if (StringUtils.equalsIgnoreCase("key", parquetTypesK.get(0).getName())) {
|
||||||
|
((JSONObject) data).put(subDataKey.toString(), subDataValue);
|
||||||
|
} else {
|
||||||
|
((JSONObject) data).put(subDataValue.toString(), subDataKey);
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
return data;
|
||||||
|
} else if (this.getOriginalType(type, parquetMetaMap) == org.apache.parquet.schema.OriginalType.MAP_KEY_VALUE) {
|
||||||
|
Group groupData = g.getGroup(index, 0);
|
||||||
|
List<org.apache.parquet.schema.Type> parquetTypes = groupData.getType().getFields();
|
||||||
|
JSONObject data = new JSONObject();
|
||||||
|
for (int i = 0; i < parquetTypes.size(); i++) {
|
||||||
|
int j = groupData.getFieldRepetitionCount(i);
|
||||||
|
// map key value 的对数
|
||||||
|
for (int k = 0; k < j; k++) {
|
||||||
|
Group groupDataK = groupData.getGroup(0, k);
|
||||||
|
List<org.apache.parquet.schema.Type> parquetTypesK = groupDataK.getType().getFields();
|
||||||
|
if (2 != parquetTypesK.size()) {
|
||||||
|
// warn: 不是key value成对出现
|
||||||
|
throw new RuntimeException(String.format("bad parquet map type: %s", groupData.getValueToString(index, 0)));
|
||||||
|
}
|
||||||
|
Object subDataKey = this.readFields(groupDataK, parquetTypesK.get(0), 0, parquetMetaMap, isUtcTimestamp);
|
||||||
|
Object subDataValue = this.readFields(groupDataK, parquetTypesK.get(1), 1, parquetMetaMap, isUtcTimestamp);
|
||||||
|
if (StringUtils.equalsIgnoreCase("key", parquetTypesK.get(0).getName())) {
|
||||||
|
((JSONObject) data).put(subDataKey.toString(), subDataValue);
|
||||||
|
} else {
|
||||||
|
((JSONObject) data).put(subDataValue.toString(), subDataKey);
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
return data;
|
||||||
|
} else if (this.getOriginalType(type, parquetMetaMap) == org.apache.parquet.schema.OriginalType.LIST) {
|
||||||
|
Group groupData = g.getGroup(index, 0);
|
||||||
|
List<org.apache.parquet.schema.Type> parquetTypes = groupData.getType().getFields();
|
||||||
|
JSONArray data = new JSONArray();
|
||||||
|
for (int i = 0; i < parquetTypes.size(); i++) {
|
||||||
|
Object subData = this.readFields(groupData, parquetTypes.get(i), i, parquetMetaMap, isUtcTimestamp);
|
||||||
|
data.add(subData);
|
||||||
|
}
|
||||||
|
return data;
|
||||||
|
} else if (this.getOriginalType(type, parquetMetaMap) == org.apache.parquet.schema.OriginalType.DECIMAL) {
|
||||||
|
Binary binaryDate = g.getBinary(index, 0);
|
||||||
|
if (null == binaryDate) {
|
||||||
|
return null;
|
||||||
|
} else {
|
||||||
|
org.apache.hadoop.hive.serde2.io.HiveDecimalWritable decimalWritable = new org.apache.hadoop.hive.serde2.io.HiveDecimalWritable(binaryDate.getBytes(), this.asPrimitiveType(type, parquetMetaMap).getDecimalMetadata().getScale());
|
||||||
|
// g.getType().getFields().get(1).asPrimitiveType().getDecimalMetadata().getScale()
|
||||||
|
HiveDecimal hiveDecimal = decimalWritable.getHiveDecimal();
|
||||||
|
if (null == hiveDecimal) {
|
||||||
|
return null;
|
||||||
|
} else {
|
||||||
|
return hiveDecimal.bigDecimalValue();
|
||||||
|
}
|
||||||
|
// return decimalWritable.doubleValue();
|
||||||
|
}
|
||||||
|
} else if (this.getOriginalType(type, parquetMetaMap) == org.apache.parquet.schema.OriginalType.DATE) {
|
||||||
|
return java.sql.Date.valueOf(LocalDate.ofEpochDay(g.getInteger(index, 0)));
|
||||||
|
} else if (this.getOriginalType(type, parquetMetaMap) == org.apache.parquet.schema.OriginalType.UTF8) {
|
||||||
|
return g.getValueToString(index, 0);
|
||||||
|
} else {
|
||||||
|
if (type.isPrimitive()) {
|
||||||
|
PrimitiveType.PrimitiveTypeName primitiveTypeName = this.asPrimitiveType(type, parquetMetaMap).getPrimitiveTypeName();
|
||||||
|
if (PrimitiveType.PrimitiveTypeName.BINARY == primitiveTypeName) {
|
||||||
|
return g.getValueToString(index, 0);
|
||||||
|
} else if (PrimitiveType.PrimitiveTypeName.BOOLEAN == primitiveTypeName) {
|
||||||
|
return g.getValueToString(index, 0);
|
||||||
|
} else if (PrimitiveType.PrimitiveTypeName.DOUBLE == primitiveTypeName) {
|
||||||
|
return g.getValueToString(index, 0);
|
||||||
|
} else if (PrimitiveType.PrimitiveTypeName.FIXED_LEN_BYTE_ARRAY == primitiveTypeName) {
|
||||||
|
return g.getValueToString(index, 0);
|
||||||
|
} else if (PrimitiveType.PrimitiveTypeName.FLOAT == primitiveTypeName) {
|
||||||
|
return g.getValueToString(index, 0);
|
||||||
|
} else if (PrimitiveType.PrimitiveTypeName.INT32 == primitiveTypeName) {
|
||||||
|
return g.getValueToString(index, 0);
|
||||||
|
} else if (PrimitiveType.PrimitiveTypeName.INT64 == primitiveTypeName) {
|
||||||
|
return g.getValueToString(index, 0);
|
||||||
|
} else if (PrimitiveType.PrimitiveTypeName.INT96 == primitiveTypeName) {
|
||||||
|
Binary dataInt96 = g.getInt96(index, 0);
|
||||||
|
if (null == dataInt96) {
|
||||||
|
return null;
|
||||||
|
} else {
|
||||||
|
ByteBuffer buf = dataInt96.toByteBuffer();
|
||||||
|
buf.order(ByteOrder.LITTLE_ENDIAN);
|
||||||
|
long timeOfDayNanos = buf.getLong();
|
||||||
|
int julianDay = buf.getInt();
|
||||||
|
if (isUtcTimestamp) {
|
||||||
|
// UTC
|
||||||
|
LocalDate localDate = LocalDate.ofEpochDay(julianDay - JULIAN_EPOCH_OFFSET_DAYS);
|
||||||
|
LocalTime localTime = LocalTime.ofNanoOfDay(timeOfDayNanos);
|
||||||
|
return Timestamp.valueOf(LocalDateTime.of(localDate, localTime));
|
||||||
|
} else {
|
||||||
|
// local time
|
||||||
|
long mills = julianDayToMillis(julianDay) + (timeOfDayNanos / NANOS_PER_MILLISECOND);
|
||||||
|
Timestamp timestamp = new Timestamp(mills);
|
||||||
|
timestamp.setNanos((int) (timeOfDayNanos % TimeUnit.SECONDS.toNanos(1)));
|
||||||
|
return timestamp;
|
||||||
|
}
|
||||||
|
}
|
||||||
|
} else {
|
||||||
|
return g.getValueToString(index, 0);
|
||||||
|
}
|
||||||
|
} else {
|
||||||
|
return g.getValueToString(index, 0);
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
|
||||||
}
|
}
|
||||||
|
|||||||
@ -0,0 +1,21 @@
|
|||||||
|
package com.alibaba.datax.plugin.reader.hdfsreader;
|
||||||
|
|
||||||
|
import org.apache.hadoop.fs.Path;
|
||||||
|
import org.apache.hadoop.fs.PathFilter;
|
||||||
|
|
||||||
|
/**
|
||||||
|
* Created by wmy on 16/11/29.
|
||||||
|
*/
|
||||||
|
public class HdfsPathFilter implements PathFilter {
|
||||||
|
|
||||||
|
private String regex = null;
|
||||||
|
|
||||||
|
public HdfsPathFilter(String regex) {
|
||||||
|
this.regex = regex;
|
||||||
|
}
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public boolean accept(Path path) {
|
||||||
|
return regex != null ? path.getName().matches(regex) : true;
|
||||||
|
}
|
||||||
|
}
|
||||||
@ -41,6 +41,8 @@ public class HdfsReader extends Reader {
|
|||||||
private String specifiedFileType = null;
|
private String specifiedFileType = null;
|
||||||
private DFSUtil dfsUtil = null;
|
private DFSUtil dfsUtil = null;
|
||||||
private List<String> path = null;
|
private List<String> path = null;
|
||||||
|
private boolean skipEmptyOrcFile = false;
|
||||||
|
private Integer orcFileEmptySize = null;
|
||||||
|
|
||||||
@Override
|
@Override
|
||||||
public void init() {
|
public void init() {
|
||||||
@ -81,9 +83,10 @@ public class HdfsReader extends Reader {
|
|||||||
!specifiedFileType.equalsIgnoreCase(Constant.TEXT) &&
|
!specifiedFileType.equalsIgnoreCase(Constant.TEXT) &&
|
||||||
!specifiedFileType.equalsIgnoreCase(Constant.CSV) &&
|
!specifiedFileType.equalsIgnoreCase(Constant.CSV) &&
|
||||||
!specifiedFileType.equalsIgnoreCase(Constant.SEQ) &&
|
!specifiedFileType.equalsIgnoreCase(Constant.SEQ) &&
|
||||||
!specifiedFileType.equalsIgnoreCase(Constant.RC)){
|
!specifiedFileType.equalsIgnoreCase(Constant.RC) &&
|
||||||
String message = "HdfsReader插件目前支持ORC, TEXT, CSV, SEQUENCE, RC五种格式的文件," +
|
!specifiedFileType.equalsIgnoreCase(Constant.PARQUET)){
|
||||||
"请将fileType选项的值配置为ORC, TEXT, CSV, SEQUENCE 或者 RC";
|
String message = "HdfsReader插件目前支持ORC, TEXT, CSV, SEQUENCE, RC, PARQUET 六种格式的文件," +
|
||||||
|
"请将fileType选项的值配置为ORC, TEXT, CSV, SEQUENCE,RC 和 PARQUET";
|
||||||
throw DataXException.asDataXException(HdfsReaderErrorCode.FILE_TYPE_ERROR, message);
|
throw DataXException.asDataXException(HdfsReaderErrorCode.FILE_TYPE_ERROR, message);
|
||||||
}
|
}
|
||||||
|
|
||||||
@ -115,6 +118,16 @@ public class HdfsReader extends Reader {
|
|||||||
UnstructuredStorageReaderUtil.validateCompress(this.readerOriginConfig);
|
UnstructuredStorageReaderUtil.validateCompress(this.readerOriginConfig);
|
||||||
UnstructuredStorageReaderUtil.validateCsvReaderConfig(this.readerOriginConfig);
|
UnstructuredStorageReaderUtil.validateCsvReaderConfig(this.readerOriginConfig);
|
||||||
}
|
}
|
||||||
|
if (this.specifiedFileType.equalsIgnoreCase(Constant.ORC)) {
|
||||||
|
skipEmptyOrcFile = this.readerOriginConfig.getBool(Key.SKIP_EMPTY_ORCFILE, false);
|
||||||
|
orcFileEmptySize = this.readerOriginConfig.getInt(Key.ORCFILE_EMPTYSIZE);
|
||||||
|
//将orcFileEmptySize必填项检查去掉，仅需要配置skipEmptyOrcFile即可，考虑历史任务兼容性(For中华保险)，保留orcFileEmptySize参数配置
|
||||||
|
//if (skipEmptyOrcFile && orcFileEmptySize == null) {
|
||||||
|
// throw new IllegalArgumentException("When \"skipEmptyOrcFile\" is configured, "
|
||||||
|
// + "parameter \"orcFileEmptySize\" cannot be null.");
|
||||||
|
//}
|
||||||
|
}
|
||||||
|
LOG.info("skipEmptyOrcFile: {}, orcFileEmptySize: {}", skipEmptyOrcFile, orcFileEmptySize);
|
||||||
|
|
||||||
}
|
}
|
||||||
|
|
||||||
@ -166,7 +179,7 @@ public class HdfsReader extends Reader {
|
|||||||
@Override
|
@Override
|
||||||
public void prepare() {
|
public void prepare() {
|
||||||
LOG.info("prepare(), start to getAllFiles...");
|
LOG.info("prepare(), start to getAllFiles...");
|
||||||
this.sourceFiles = dfsUtil.getAllFiles(path, specifiedFileType);
|
this.sourceFiles = dfsUtil.getAllFiles(path, specifiedFileType,skipEmptyOrcFile, orcFileEmptySize);
|
||||||
LOG.info(String.format("您即将读取的文件数为: [%s], 列表为: [%s]",
|
LOG.info(String.format("您即将读取的文件数为: [%s], 列表为: [%s]",
|
||||||
this.sourceFiles.size(),
|
this.sourceFiles.size(),
|
||||||
StringUtils.join(this.sourceFiles, ",")));
|
StringUtils.join(this.sourceFiles, ",")));
|
||||||
@ -273,7 +286,9 @@ public class HdfsReader extends Reader {
|
|||||||
}else if(specifiedFileType.equalsIgnoreCase(Constant.RC)){
|
}else if(specifiedFileType.equalsIgnoreCase(Constant.RC)){
|
||||||
|
|
||||||
dfsUtil.rcFileStartRead(sourceFile, this.taskConfig, recordSender, this.getTaskPluginCollector());
|
dfsUtil.rcFileStartRead(sourceFile, this.taskConfig, recordSender, this.getTaskPluginCollector());
|
||||||
}else {
|
} else if (specifiedFileType.equalsIgnoreCase(Constant.PARQUET)) {
|
||||||
|
dfsUtil.parquetFileStartRead(sourceFile, this.taskConfig, recordSender, this.getTaskPluginCollector());
|
||||||
|
} else {
|
||||||
|
|
||||||
String message = "HdfsReader插件目前支持ORC, TEXT, CSV, SEQUENCE, RC五种格式的文件," +
|
String message = "HdfsReader插件目前支持ORC, TEXT, CSV, SEQUENCE, RC五种格式的文件," +
|
||||||
"请将fileType选项的值配置为ORC, TEXT, CSV, SEQUENCE 或者 RC";
|
"请将fileType选项的值配置为ORC, TEXT, CSV, SEQUENCE 或者 RC";
|
||||||
|
|||||||
@ -19,7 +19,12 @@ public enum HdfsReaderErrorCode implements ErrorCode {
|
|||||||
FILE_TYPE_UNSUPPORT("HdfsReader-12", "文件类型目前不支持"),
|
FILE_TYPE_UNSUPPORT("HdfsReader-12", "文件类型目前不支持"),
|
||||||
KERBEROS_LOGIN_ERROR("HdfsReader-13", "KERBEROS认证失败"),
|
KERBEROS_LOGIN_ERROR("HdfsReader-13", "KERBEROS认证失败"),
|
||||||
READ_SEQUENCEFILE_ERROR("HdfsReader-14", "读取SequenceFile文件出错"),
|
READ_SEQUENCEFILE_ERROR("HdfsReader-14", "读取SequenceFile文件出错"),
|
||||||
READ_RCFILE_ERROR("HdfsReader-15", "读取RCFile文件出错"),;
|
READ_RCFILE_ERROR("HdfsReader-15", "读取RCFile文件出错"),
|
||||||
|
INIT_RCFILE_SERDE_ERROR("HdfsReader-16", "Deserialize RCFile, initialization failed!"),
|
||||||
|
PARSE_MESSAGE_TYPE_FROM_SCHEMA_ERROR("HdfsReader-17", "Error parsing ParquetSchema"),
|
||||||
|
INVALID_PARQUET_SCHEMA("HdfsReader-18", "ParquetSchema is invalid"),
|
||||||
|
READ_PARQUET_ERROR("HdfsReader-19", "Error reading Parquet file"),
|
||||||
|
CONNECT_HDFS_IO_ERROR("HdfsReader-20", "I/O exception in establishing connection with HDFS");
|
||||||
|
|
||||||
private final String code;
|
private final String code;
|
||||||
private final String description;
|
private final String description;
|
||||||
|
|||||||
@ -7,9 +7,60 @@ public final class Key {
|
|||||||
*/
|
*/
|
||||||
public final static String PATH = "path";
|
public final static String PATH = "path";
|
||||||
public final static String DEFAULT_FS = "defaultFS";
|
public final static String DEFAULT_FS = "defaultFS";
|
||||||
|
public final static String HIVE_VERSION = "hiveVersion";
|
||||||
public static final String FILETYPE = "fileType";
|
public static final String FILETYPE = "fileType";
|
||||||
public static final String HADOOP_CONFIG = "hadoopConfig";
|
public static final String HADOOP_CONFIG = "hadoopConfig";
|
||||||
public static final String HAVE_KERBEROS = "haveKerberos";
|
public static final String HAVE_KERBEROS = "haveKerberos";
|
||||||
public static final String KERBEROS_KEYTAB_FILE_PATH = "kerberosKeytabFilePath";
|
public static final String KERBEROS_KEYTAB_FILE_PATH = "kerberosKeytabFilePath";
|
||||||
|
public static final String KERBEROS_CONF_FILE_PATH = "kerberosConfFilePath";
|
||||||
public static final String KERBEROS_PRINCIPAL = "kerberosPrincipal";
|
public static final String KERBEROS_PRINCIPAL = "kerberosPrincipal";
|
||||||
|
public static final String PATH_FILTER = "pathFilter";
|
||||||
|
public static final String PARQUET_SCHEMA = "parquetSchema";
|
||||||
|
/**
|
||||||
|
* hive 3.x 或 cdh高版本，使用UTC时区存储时间戳，如果发现时区偏移，该配置项要配置成 true
|
||||||
|
*/
|
||||||
|
public static final String PARQUET_UTC_TIMESTAMP = "parquetUtcTimestamp";
|
||||||
|
public static final String SUCCESS_ON_NO_FILE = "successOnNoFile";
|
||||||
|
public static final String PROTECTION = "protection";
|
||||||
|
|
||||||
|
/**
|
||||||
|
* 用于显示地指定hdfs客户端的用户名
|
||||||
|
*/
|
||||||
|
public static final String HDFS_USERNAME = "hdfsUsername";
|
||||||
|
|
||||||
|
/**
|
||||||
|
* ORC FILE空文件大小
|
||||||
|
*/
|
||||||
|
public static final String ORCFILE_EMPTYSIZE = "orcFileEmptySize";
|
||||||
|
|
||||||
|
/**
|
||||||
|
* 是否跳过空的OrcFile
|
||||||
|
*/
|
||||||
|
public static final String SKIP_EMPTY_ORCFILE = "skipEmptyOrcFile";
|
||||||
|
|
||||||
|
/**
|
||||||
|
* 是否跳过 orc meta 信息
|
||||||
|
*/
|
||||||
|
|
||||||
|
public static final String SKIP_ORC_META = "skipOrcMetaInfo";
|
||||||
|
/**
|
||||||
|
* 过滤_或者.开头的文件
|
||||||
|
*/
|
||||||
|
public static final String REGEX_PATTERN = "^.*[/][^._].*";
|
||||||
|
|
||||||
|
public static final String FILTER_TAG_FILE = "filterTagFile";
|
||||||
|
|
||||||
|
// high level params refs https://github.com/aliyun/alibabacloud-jindodata/blob/master/docs/user/4.x/4.4.0/oss/configuration/jindosdk_configuration_list.md
|
||||||
|
// <!-- oss 并发下载任务队列大小 -->
|
||||||
|
public static final String FS_OSS_DOWNLOAD_QUEUE_SIZE = "ossDownloadQueueSize";
|
||||||
|
|
||||||
|
// <!-- 进程内 oss 最大并发下载任务数 -->
|
||||||
|
public static final String FS_OSS_DOWNLOAD_THREAD_CONCURRENCY = "ossDownloadThreadConcurrency";
|
||||||
|
|
||||||
|
public static final String FS_OSS_READ_READAHEAD_BUFFER_COUNT = "ossDownloadBufferCount";
|
||||||
|
|
||||||
|
public static final String FILE_SYSTEM_TYPE = "fileSystemType";
|
||||||
|
public static final String CDH_3_X_HIVE_VERSION = "3.1.3-cdh";
|
||||||
|
|
||||||
|
public static final String SUPPORT_ADD_MIDDLE_COLUMN = "supportAddMiddleColumn";
|
||||||
}
|
}
|
||||||
|
|||||||
@ -0,0 +1,33 @@
|
|||||||
|
package com.alibaba.datax.plugin.reader.hdfsreader;
|
||||||
|
|
||||||
|
import org.apache.parquet.schema.OriginalType;
|
||||||
|
import org.apache.parquet.schema.PrimitiveType;
|
||||||
|
|
||||||
|
import java.util.HashMap;
|
||||||
|
import java.util.List;
|
||||||
|
import java.util.Map;
|
||||||
|
|
||||||
|
/**
|
||||||
|
* @author jitongchen
|
||||||
|
* @date 2023/9/7 10:20 AM
|
||||||
|
*/
|
||||||
|
public class ParquetMessageHelper {
|
||||||
|
public static Map<String, ParquetMeta> parseParquetTypes(List<org.apache.parquet.schema.Type> parquetTypes) {
|
||||||
|
int fieldCount = parquetTypes.size();
|
||||||
|
Map<String, ParquetMeta> parquetMetaMap = new HashMap<String, ParquetMeta>();
|
||||||
|
for (int i = 0; i < fieldCount; i++) {
|
||||||
|
org.apache.parquet.schema.Type type = parquetTypes.get(i);
|
||||||
|
String name = type.getName();
|
||||||
|
ParquetMeta parquetMeta = new ParquetMeta();
|
||||||
|
parquetMeta.setName(name);
|
||||||
|
OriginalType originalType = type.getOriginalType();
|
||||||
|
parquetMeta.setOriginalType(originalType);
|
||||||
|
if (type.isPrimitive()) {
|
||||||
|
PrimitiveType primitiveType = type.asPrimitiveType();
|
||||||
|
parquetMeta.setPrimitiveType(primitiveType);
|
||||||
|
}
|
||||||
|
parquetMetaMap.put(name, parquetMeta);
|
||||||
|
}
|
||||||
|
return parquetMetaMap;
|
||||||
|
}
|
||||||
|
}
|
||||||
@ -0,0 +1,38 @@
|
|||||||
|
package com.alibaba.datax.plugin.reader.hdfsreader;
|
||||||
|
|
||||||
|
import org.apache.parquet.schema.OriginalType;
|
||||||
|
import org.apache.parquet.schema.PrimitiveType;
|
||||||
|
|
||||||
|
/**
|
||||||
|
* @author jitongchen
|
||||||
|
* @date 2023/9/7 10:20 AM
|
||||||
|
*/
|
||||||
|
public class ParquetMeta {
|
||||||
|
private String name;
|
||||||
|
private OriginalType originalType;
|
||||||
|
private PrimitiveType primitiveType;
|
||||||
|
|
||||||
|
public String getName() {
|
||||||
|
return name;
|
||||||
|
}
|
||||||
|
|
||||||
|
public void setName(String name) {
|
||||||
|
this.name = name;
|
||||||
|
}
|
||||||
|
|
||||||
|
public OriginalType getOriginalType() {
|
||||||
|
return originalType;
|
||||||
|
}
|
||||||
|
|
||||||
|
public void setOriginalType(OriginalType originalType) {
|
||||||
|
this.originalType = originalType;
|
||||||
|
}
|
||||||
|
|
||||||
|
public PrimitiveType getPrimitiveType() {
|
||||||
|
return primitiveType;
|
||||||
|
}
|
||||||
|
|
||||||
|
public void setPrimitiveType(PrimitiveType primitiveType) {
|
||||||
|
this.primitiveType = primitiveType;
|
||||||
|
}
|
||||||
|
}
|
||||||
@ -27,9 +27,8 @@ import org.apache.hadoop.mapred.*;
|
|||||||
import org.apache.hadoop.security.UserGroupInformation;
|
import org.apache.hadoop.security.UserGroupInformation;
|
||||||
import org.slf4j.Logger;
|
import org.slf4j.Logger;
|
||||||
import org.slf4j.LoggerFactory;
|
import org.slf4j.LoggerFactory;
|
||||||
import parquet.schema.OriginalType;
|
import parquet.hadoop.metadata.CompressionCodecName;
|
||||||
import parquet.schema.PrimitiveType;
|
import parquet.schema.*;
|
||||||
import parquet.schema.Types;
|
|
||||||
|
|
||||||
import java.io.IOException;
|
import java.io.IOException;
|
||||||
import java.text.SimpleDateFormat;
|
import java.text.SimpleDateFormat;
|
||||||
@ -626,4 +625,61 @@ public class HdfsHelper {
|
|||||||
}
|
}
|
||||||
return typeBuilder.named("m").toString();
|
return typeBuilder.named("m").toString();
|
||||||
}
|
}
|
||||||
|
|
||||||
|
public void parquetFileStartWrite(RecordReceiver lineReceiver, Configuration config, String fileName, TaskPluginCollector taskPluginCollector, Configuration taskConfig) {
|
||||||
|
MessageType messageType = null;
|
||||||
|
ParquetFileProccessor proccessor = null;
|
||||||
|
Path outputPath = new Path(fileName);
|
||||||
|
String schema = config.getString(Key.PARQUET_SCHEMA);
|
||||||
|
try {
|
||||||
|
messageType = MessageTypeParser.parseMessageType(schema);
|
||||||
|
} catch (Exception e) {
|
||||||
|
String message = String.format("Error parsing the Schema string [%s] into MessageType", schema);
|
||||||
|
LOG.error(message);
|
||||||
|
throw DataXException.asDataXException(HdfsWriterErrorCode.PARSE_MESSAGE_TYPE_FROM_SCHEMA_ERROR, e);
|
||||||
|
}
|
||||||
|
|
||||||
|
// determine the compression codec
|
||||||
|
String compress = config.getString(Key.COMPRESS, null);
|
||||||
|
// be compatible with the old NONE
|
||||||
|
if ("NONE".equalsIgnoreCase(compress)) {
|
||||||
|
compress = "UNCOMPRESSED";
|
||||||
|
}
|
||||||
|
CompressionCodecName compressionCodecName = CompressionCodecName.fromConf(compress);
|
||||||
|
LOG.info("The compression codec used for parquet writing is: {}", compressionCodecName, compress);
|
||||||
|
try {
|
||||||
|
proccessor = new ParquetFileProccessor(outputPath, messageType, compressionCodecName, false, taskConfig, taskPluginCollector, hadoopConf);
|
||||||
|
} catch (Exception e) {
|
||||||
|
String message = String.format("Initializing ParquetFileProccessor based on Schema[%s] failed.", schema);
|
||||||
|
LOG.error(message);
|
||||||
|
throw DataXException.asDataXException(HdfsWriterErrorCode.INIT_PROCCESSOR_FAILURE, e);
|
||||||
|
}
|
||||||
|
SimpleDateFormat dateFormat = new SimpleDateFormat("yyyyMMddHHmm");
|
||||||
|
String attempt = "attempt_" + dateFormat.format(new Date()) + "_0001_m_000000_0";
|
||||||
|
conf.set(JobContext.TASK_ATTEMPT_ID, attempt);
|
||||||
|
FileOutputFormat outFormat = new TextOutputFormat();
|
||||||
|
outFormat.setOutputPath(conf, outputPath);
|
||||||
|
outFormat.setWorkOutputPath(conf, outputPath);
|
||||||
|
try {
|
||||||
|
Record record = null;
|
||||||
|
while ((record = lineReceiver.getFromReader()) != null) {
|
||||||
|
proccessor.write(record);
|
||||||
|
}
|
||||||
|
} catch (Exception e) {
|
||||||
|
String message = String.format("An exception occurred while writing the file file [%s]", fileName);
|
||||||
|
LOG.error(message);
|
||||||
|
Path path = new Path(fileName);
|
||||||
|
deleteDir(path.getParent());
|
||||||
|
throw DataXException.asDataXException(HdfsWriterErrorCode.Write_FILE_IO_ERROR, e);
|
||||||
|
} finally {
|
||||||
|
if (proccessor != null) {
|
||||||
|
try {
|
||||||
|
proccessor.close();
|
||||||
|
} catch (IOException e) {
|
||||||
|
LOG.error(e.getMessage(), e);
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
}
|
}
|
||||||
|
|||||||
@ -53,8 +53,8 @@ public class HdfsWriter extends Writer {
|
|||||||
this.defaultFS = this.writerSliceConfig.getNecessaryValue(Key.DEFAULT_FS, HdfsWriterErrorCode.REQUIRED_VALUE);
|
this.defaultFS = this.writerSliceConfig.getNecessaryValue(Key.DEFAULT_FS, HdfsWriterErrorCode.REQUIRED_VALUE);
|
||||||
//fileType check
|
//fileType check
|
||||||
this.fileType = this.writerSliceConfig.getNecessaryValue(Key.FILE_TYPE, HdfsWriterErrorCode.REQUIRED_VALUE);
|
this.fileType = this.writerSliceConfig.getNecessaryValue(Key.FILE_TYPE, HdfsWriterErrorCode.REQUIRED_VALUE);
|
||||||
if( !fileType.equalsIgnoreCase("ORC") && !fileType.equalsIgnoreCase("TEXT")){
|
if (!fileType.equalsIgnoreCase("ORC") && !fileType.equalsIgnoreCase("TEXT") && !fileType.equalsIgnoreCase("PARQUET")) {
|
||||||
String message = "HdfsWriter插件目前只支持ORC和TEXT两种格式的文件,请将filetype选项的值配置为ORC或者TEXT";
|
String message = "HdfsWriter插件目前只支持ORC、TEXT、PARQUET三种格式的文件,请将filetype选项的值配置为ORC、TEXT或PARQUET";
|
||||||
throw DataXException.asDataXException(HdfsWriterErrorCode.ILLEGAL_VALUE, message);
|
throw DataXException.asDataXException(HdfsWriterErrorCode.ILLEGAL_VALUE, message);
|
||||||
}
|
}
|
||||||
//path
|
//path
|
||||||
@ -415,6 +415,9 @@ public class HdfsWriter extends Writer {
|
|||||||
//写ORC FILE
|
//写ORC FILE
|
||||||
hdfsHelper.orcFileStartWrite(lineReceiver,this.writerSliceConfig, this.fileName,
|
hdfsHelper.orcFileStartWrite(lineReceiver,this.writerSliceConfig, this.fileName,
|
||||||
this.getTaskPluginCollector());
|
this.getTaskPluginCollector());
|
||||||
|
} else if (fileType.equalsIgnoreCase("PARQUET")) {
|
||||||
|
//写PARQUET FILE
|
||||||
|
hdfsHelper.parquetFileStartWrite(lineReceiver, this.writerSliceConfig, this.fileName, this.getTaskPluginCollector(), this.writerSliceConfig);
|
||||||
}
|
}
|
||||||
|
|
||||||
LOG.info("end do write");
|
LOG.info("end do write");
|
||||||
|
|||||||
@ -16,7 +16,11 @@ public enum HdfsWriterErrorCode implements ErrorCode {
|
|||||||
CONNECT_HDFS_IO_ERROR("HdfsWriter-06", "与HDFS建立连接时出现IO异常."),
|
CONNECT_HDFS_IO_ERROR("HdfsWriter-06", "与HDFS建立连接时出现IO异常."),
|
||||||
COLUMN_REQUIRED_VALUE("HdfsWriter-07", "您column配置中缺失了必须填写的参数值."),
|
COLUMN_REQUIRED_VALUE("HdfsWriter-07", "您column配置中缺失了必须填写的参数值."),
|
||||||
HDFS_RENAME_FILE_ERROR("HdfsWriter-08", "将文件移动到配置路径失败."),
|
HDFS_RENAME_FILE_ERROR("HdfsWriter-08", "将文件移动到配置路径失败."),
|
||||||
KERBEROS_LOGIN_ERROR("HdfsWriter-09", "KERBEROS认证失败");
|
KERBEROS_LOGIN_ERROR("HdfsWriter-09", "KERBEROS认证失败"),
|
||||||
|
PARSE_MESSAGE_TYPE_FROM_SCHEMA_ERROR("HdfsWriter-10", "Parse parquet schema error"),
|
||||||
|
|
||||||
|
INIT_PROCCESSOR_FAILURE("HdfsWriter-11", "Init processor failed");
|
||||||
|
|
||||||
|
|
||||||
private final String code;
|
private final String code;
|
||||||
private final String description;
|
private final String description;
|
||||||
|
|||||||
@ -46,4 +46,32 @@ public class Key {
|
|||||||
|
|
||||||
public static final String PARQUET_SCHEMA = "parquetSchema";
|
public static final String PARQUET_SCHEMA = "parquetSchema";
|
||||||
public static final String PARQUET_MERGE_RESULT = "parquetMergeResult";
|
public static final String PARQUET_MERGE_RESULT = "parquetMergeResult";
|
||||||
|
|
||||||
|
/**
|
||||||
|
* hive 3.x 或 cdh高版本，使用UTC时区存储时间戳，如果发现时区偏移，该配置项要配置成 true
|
||||||
|
*/
|
||||||
|
public static final String PARQUET_UTC_TIMESTAMP = "parquetUtcTimestamp";
|
||||||
|
|
||||||
|
// Kerberos
|
||||||
|
public static final String KERBEROS_CONF_FILE_PATH = "kerberosConfFilePath";
|
||||||
|
|
||||||
|
// PanguFS
|
||||||
|
public final static String PANGU_FS_CONFIG = "panguFSConfig";
|
||||||
|
public final static String PANGU_FS_CONFIG_NUWA_CLUSTER = "nuwaCluster";
|
||||||
|
public final static String PANGU_FS_CONFIG_NUWA_SERVERS = "nuwaServers";
|
||||||
|
public final static String PANGU_FS_CONFIG_NUWA_PROXIES = "nuwaProxies";
|
||||||
|
public final static String PANGU_FS_CONFIG_CAPABILITY = "capability";
|
||||||
|
|
||||||
|
|
||||||
|
public static final String FS_OSS_UPLOAD_THREAD_CONCURRENCY = "ossUploadConcurrency";
|
||||||
|
// <!-- oss 并发上传任务队列大小 -->
|
||||||
|
public static final String FS_OSS_UPLOAD_QUEUE_SIZE = "ossUploadQueueSize";
|
||||||
|
// <!-- 进程内 oss 最大并发上传任务数 -->
|
||||||
|
public static final String FS_OSS_UPLOAD_MAX_PENDING_TASKS_PER_STREAM = "ossUploadMaxPendingTasksPerStream";
|
||||||
|
|
||||||
|
public static final String FS_OSS_BLOCKLET_SIZE_MB = "ossBlockSize";
|
||||||
|
|
||||||
|
public static final String FILE_SYSTEM_TYPE = "fileSystemType";
|
||||||
|
public static final String ENABLE_COLUMN_EXCHANGE = "enableColumnExchange";
|
||||||
|
public static final String SUPPORT_HIVE_DATETIME = "supportHiveDateTime";
|
||||||
}
|
}
|
||||||
|
|||||||
@ -0,0 +1,30 @@
|
|||||||
|
package com.alibaba.datax.plugin.writer.hdfswriter;
|
||||||
|
|
||||||
|
import com.alibaba.datax.common.element.Record;
|
||||||
|
import com.alibaba.datax.common.plugin.TaskPluginCollector;
|
||||||
|
import com.alibaba.datax.common.util.Configuration;
|
||||||
|
import org.apache.hadoop.fs.Path;
|
||||||
|
import parquet.hadoop.ParquetWriter;
|
||||||
|
import parquet.hadoop.metadata.CompressionCodecName;
|
||||||
|
import parquet.schema.MessageType;
|
||||||
|
|
||||||
|
import java.io.IOException;
|
||||||
|
|
||||||
|
/**
|
||||||
|
* @author jitongchen
|
||||||
|
* @date 2023/9/7 9:41 AM
|
||||||
|
*/
|
||||||
|
public class ParquetFileProccessor extends ParquetWriter<Record> {
|
||||||
|
|
||||||
|
public ParquetFileProccessor(Path file, MessageType schema, boolean enableDictionary, Configuration taskConfig, TaskPluginCollector taskPluginCollector, org.apache.hadoop.conf.Configuration configuration) throws IOException {
|
||||||
|
this(file, schema, CompressionCodecName.UNCOMPRESSED, enableDictionary, taskConfig, taskPluginCollector, configuration);
|
||||||
|
}
|
||||||
|
|
||||||
|
public ParquetFileProccessor(Path file, MessageType schema, CompressionCodecName codecName, boolean enableDictionary, Configuration taskConfig, TaskPluginCollector taskPluginCollector) throws IOException {
|
||||||
|
super(file, new ParquetFileSupport(schema, taskConfig, taskPluginCollector), codecName, DEFAULT_BLOCK_SIZE, DEFAULT_PAGE_SIZE, DEFAULT_PAGE_SIZE, enableDictionary, false, DEFAULT_WRITER_VERSION);
|
||||||
|
}
|
||||||
|
|
||||||
|
public ParquetFileProccessor(Path file, MessageType schema, CompressionCodecName codecName, boolean enableDictionary, Configuration taskConfig, TaskPluginCollector taskPluginCollector, org.apache.hadoop.conf.Configuration configuration) throws IOException {
|
||||||
|
super(file, new ParquetFileSupport(schema, taskConfig, taskPluginCollector), codecName, DEFAULT_BLOCK_SIZE, DEFAULT_PAGE_SIZE, DEFAULT_PAGE_SIZE, enableDictionary, false, DEFAULT_WRITER_VERSION, configuration);
|
||||||
|
}
|
||||||
|
}
|
||||||
@ -0,0 +1,642 @@
|
|||||||
|
package com.alibaba.datax.plugin.writer.hdfswriter;
|
||||||
|
|
||||||
|
import com.alibaba.datax.common.element.*;
|
||||||
|
import com.alibaba.datax.common.plugin.TaskPluginCollector;
|
||||||
|
import com.alibaba.datax.common.util.LimitLogger;
|
||||||
|
import com.alibaba.fastjson2.JSON;
|
||||||
|
import com.alibaba.fastjson2.JSONArray;
|
||||||
|
import com.alibaba.fastjson2.JSONObject;
|
||||||
|
import org.apache.commons.lang3.StringUtils;
|
||||||
|
import org.apache.hadoop.conf.Configuration;
|
||||||
|
import org.apache.hadoop.hive.ql.io.parquet.serde.ParquetHiveSerDe;
|
||||||
|
import org.slf4j.Logger;
|
||||||
|
import org.slf4j.LoggerFactory;
|
||||||
|
import parquet.column.ColumnDescriptor;
|
||||||
|
import parquet.hadoop.api.WriteSupport;
|
||||||
|
import parquet.io.api.Binary;
|
||||||
|
import parquet.io.api.RecordConsumer;
|
||||||
|
import parquet.schema.*;
|
||||||
|
|
||||||
|
import java.math.BigDecimal;
|
||||||
|
import java.math.RoundingMode;
|
||||||
|
import java.nio.ByteBuffer;
|
||||||
|
import java.nio.ByteOrder;
|
||||||
|
import java.sql.Timestamp;
|
||||||
|
import java.text.SimpleDateFormat;
|
||||||
|
import java.time.LocalDateTime;
|
||||||
|
import java.time.OffsetDateTime;
|
||||||
|
import java.time.ZoneOffset;
|
||||||
|
import java.time.temporal.ChronoField;
|
||||||
|
import java.util.Arrays;
|
||||||
|
import java.util.Date;
|
||||||
|
import java.util.HashMap;
|
||||||
|
import java.util.List;
|
||||||
|
import java.util.concurrent.TimeUnit;
|
||||||
|
|
||||||
|
/**
|
||||||
|
* @author jitongchen
|
||||||
|
* @date 2023/9/7 9:41 AM
|
||||||
|
*/
|
||||||
|
public class ParquetFileSupport extends WriteSupport<Record> {
|
||||||
|
public static final Logger LOGGER = LoggerFactory.getLogger(ParquetFileSupport.class);
|
||||||
|
private MessageType schema;
|
||||||
|
private List<ColumnDescriptor> columns;
|
||||||
|
private RecordConsumer recordConsumer;
|
||||||
|
private boolean useRawDataTransf = true;
|
||||||
|
private boolean printStackTrace = true;
|
||||||
|
|
||||||
|
// 不通类型的nullFormat
|
||||||
|
private String nullFormat;
|
||||||
|
|
||||||
|
private String dateFormat;
|
||||||
|
private boolean isUtcTimestamp;
|
||||||
|
private SimpleDateFormat dateParse;
|
||||||
|
private Binary binaryForNull;
|
||||||
|
private TaskPluginCollector taskPluginCollector;
|
||||||
|
private String dataxParquetMode;
|
||||||
|
|
||||||
|
public ParquetFileSupport(MessageType schema, com.alibaba.datax.common.util.Configuration taskConfig, TaskPluginCollector taskPluginCollector) {
|
||||||
|
this.schema = schema;
|
||||||
|
this.columns = schema.getColumns();
|
||||||
|
this.useRawDataTransf = taskConfig.getBool(Key.PARQUET_FILE_USE_RAW_DATA_TRANSF, true);
|
||||||
|
|
||||||
|
// 不通类型的nullFormat
|
||||||
|
this.nullFormat = taskConfig.getString(Key.NULL_FORMAT, Constant.DEFAULT_NULL_FORMAT);
|
||||||
|
this.binaryForNull = Binary.fromString(this.nullFormat);
|
||||||
|
|
||||||
|
this.dateFormat = taskConfig.getString(Key.DATE_FORMAT, null);
|
||||||
|
if (StringUtils.isNotBlank(this.dateFormat)) {
|
||||||
|
this.dateParse = new SimpleDateFormat(dateFormat);
|
||||||
|
}
|
||||||
|
|
||||||
|
this.isUtcTimestamp = taskConfig.getBool(Key.PARQUET_UTC_TIMESTAMP, false);
|
||||||
|
|
||||||
|
this.taskPluginCollector = taskPluginCollector;
|
||||||
|
if (taskConfig.getKeys().contains("dataxParquetMode")) {
|
||||||
|
this.dataxParquetMode = taskConfig.getString("dataxParquetMode");
|
||||||
|
} else {
|
||||||
|
// 默认值是columns
|
||||||
|
this.dataxParquetMode = "columns";
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public WriteContext init(Configuration configuration) {
|
||||||
|
return new WriteContext(schema, new HashMap<String, String>());
|
||||||
|
}
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public void prepareForWrite(RecordConsumer recordConsumer) {
|
||||||
|
this.recordConsumer = recordConsumer;
|
||||||
|
}
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public void write(Record values) {
|
||||||
|
if (dataxParquetMode.equalsIgnoreCase("fields")) {
|
||||||
|
writeBaseOnFields(values);
|
||||||
|
return;
|
||||||
|
}
|
||||||
|
|
||||||
|
// NOTE: 下面的实现其实是不对的，只是看代码注释貌似有用户已经在用
|
||||||
|
// 所以暂时不动下面的逻辑。
|
||||||
|
// 默认走的就是下面的这条代码路径
|
||||||
|
if (values != null && columns != null && values.getColumnNumber() == columns.size()) {
|
||||||
|
recordConsumer.startMessage();
|
||||||
|
for (int i = 0; i < columns.size(); i++) {
|
||||||
|
Column value = values.getColumn(i);
|
||||||
|
ColumnDescriptor columnDescriptor = columns.get(i);
|
||||||
|
Type type = this.schema.getFields().get(i);
|
||||||
|
if (value != null) {
|
||||||
|
try {
|
||||||
|
if (this.useRawDataTransf) {
|
||||||
|
if (value.getRawData() == null) {
|
||||||
|
continue;
|
||||||
|
}
|
||||||
|
recordConsumer.startField(columnDescriptor.getPath()[0], i);
|
||||||
|
// 原来使用Column->RawData的方法其实是错误的类型转换策略，会将DataX的数据内部表示形象序列化出去
|
||||||
|
// 但是 Parquet 已经有用户使用了，故暂时只是配置项切换
|
||||||
|
String rawData = value.getRawData().toString();
|
||||||
|
switch (columnDescriptor.getType()) {
|
||||||
|
case BOOLEAN:
|
||||||
|
recordConsumer.addBoolean(Boolean.parseBoolean(rawData));
|
||||||
|
break;
|
||||||
|
case FLOAT:
|
||||||
|
recordConsumer.addFloat(Float.parseFloat(rawData));
|
||||||
|
break;
|
||||||
|
case DOUBLE:
|
||||||
|
recordConsumer.addDouble(Double.parseDouble(rawData));
|
||||||
|
break;
|
||||||
|
case INT32:
|
||||||
|
OriginalType originalType = type.getOriginalType();
|
||||||
|
if (originalType != null && StringUtils.equalsIgnoreCase("DATE", originalType.name())) {
|
||||||
|
int realVal = (int) (new java.sql.Date(Long.parseLong(rawData)).toLocalDate().toEpochDay());
|
||||||
|
recordConsumer.addInteger(realVal);
|
||||||
|
} else {
|
||||||
|
recordConsumer.addInteger(Integer.parseInt(rawData));
|
||||||
|
}
|
||||||
|
break;
|
||||||
|
case INT64:
|
||||||
|
recordConsumer.addLong(Long.valueOf(rawData));
|
||||||
|
break;
|
||||||
|
case INT96:
|
||||||
|
recordConsumer.addBinary(timestampColToBinary(value));
|
||||||
|
break;
|
||||||
|
case BINARY:
|
||||||
|
recordConsumer.addBinary(Binary.fromString(rawData));
|
||||||
|
break;
|
||||||
|
case FIXED_LEN_BYTE_ARRAY:
|
||||||
|
PrimitiveType primitiveType = type.asPrimitiveType();
|
||||||
|
if (primitiveType.getDecimalMetadata() != null) {
|
||||||
|
// decimal
|
||||||
|
recordConsumer.addBinary(decimalToBinary(value, primitiveType.getDecimalMetadata().getPrecision(), primitiveType.getDecimalMetadata().getScale()));
|
||||||
|
break;
|
||||||
|
}
|
||||||
|
/* fall through */
|
||||||
|
default:
|
||||||
|
recordConsumer.addBinary(Binary.fromString(rawData));
|
||||||
|
break;
|
||||||
|
}
|
||||||
|
|
||||||
|
recordConsumer.endField(columnDescriptor.getPath()[0], i);
|
||||||
|
} else {
|
||||||
|
boolean isNull = null == value.getRawData();
|
||||||
|
|
||||||
|
if (!isNull) {
|
||||||
|
recordConsumer.startField(columnDescriptor.getPath()[0], i);
|
||||||
|
|
||||||
|
// no skip: empty fields are illegal, the field should be ommited completely instead
|
||||||
|
switch (columnDescriptor.getType()) {
|
||||||
|
case BOOLEAN:
|
||||||
|
recordConsumer.addBoolean(value.asBoolean());
|
||||||
|
break;
|
||||||
|
case FLOAT:
|
||||||
|
recordConsumer.addFloat(value.asDouble().floatValue());
|
||||||
|
break;
|
||||||
|
case DOUBLE:
|
||||||
|
recordConsumer.addDouble(value.asDouble());
|
||||||
|
break;
|
||||||
|
case INT32:
|
||||||
|
OriginalType originalType = type.getOriginalType();
|
||||||
|
if (originalType != null && StringUtils.equalsIgnoreCase("DATE", originalType.name())) {
|
||||||
|
int realVal = (int) (new java.sql.Date(value.asLong()).toLocalDate().toEpochDay());
|
||||||
|
recordConsumer.addInteger(realVal);
|
||||||
|
} else {
|
||||||
|
recordConsumer.addInteger(value.asLong().intValue());
|
||||||
|
}
|
||||||
|
break;
|
||||||
|
case INT64:
|
||||||
|
recordConsumer.addLong(value.asLong());
|
||||||
|
break;
|
||||||
|
case INT96:
|
||||||
|
recordConsumer.addBinary(timestampColToBinary(value));
|
||||||
|
break;
|
||||||
|
case BINARY:
|
||||||
|
String valueAsString2Write = null;
|
||||||
|
if (Column.Type.DATE == value.getType() && null != this.dateParse) {
|
||||||
|
valueAsString2Write = dateParse.format(value.asDate());
|
||||||
|
} else {
|
||||||
|
valueAsString2Write = value.asString();
|
||||||
|
}
|
||||||
|
recordConsumer.addBinary(Binary.fromString(valueAsString2Write));
|
||||||
|
break;
|
||||||
|
case FIXED_LEN_BYTE_ARRAY:
|
||||||
|
PrimitiveType primitiveType = type.asPrimitiveType();
|
||||||
|
if (primitiveType.getDecimalMetadata() != null) {
|
||||||
|
// decimal
|
||||||
|
recordConsumer.addBinary(decimalToBinary(value, primitiveType.getDecimalMetadata().getPrecision(), primitiveType.getDecimalMetadata().getScale()));
|
||||||
|
break;
|
||||||
|
}
|
||||||
|
/* fall through */
|
||||||
|
default:
|
||||||
|
recordConsumer.addBinary(Binary.fromString(value.asString()));
|
||||||
|
break;
|
||||||
|
}
|
||||||
|
recordConsumer.endField(columnDescriptor.getPath()[0], i);
|
||||||
|
}
|
||||||
|
}
|
||||||
|
} catch (Exception e) {
|
||||||
|
if (printStackTrace) {
|
||||||
|
printStackTrace = false;
|
||||||
|
LOGGER.warn("write to parquet error: {}", e.getMessage(), e);
|
||||||
|
}
|
||||||
|
// dirty data
|
||||||
|
if (null != this.taskPluginCollector) {
|
||||||
|
// job post 里面的merge taskPluginCollector 为null
|
||||||
|
this.taskPluginCollector.collectDirtyRecord(values, e, e.getMessage());
|
||||||
|
}
|
||||||
|
}
|
||||||
|
} else {
|
||||||
|
recordConsumer.addBinary(this.binaryForNull);
|
||||||
|
}
|
||||||
|
}
|
||||||
|
recordConsumer.endMessage();
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
private Binary decimalToBinary(Column value, int precision, int scale) {
|
||||||
|
BigDecimal bigDecimal = value.asBigDecimal();
|
||||||
|
bigDecimal = bigDecimal.setScale(scale, RoundingMode.HALF_UP);
|
||||||
|
byte[] decimalBytes = bigDecimal.unscaledValue().toByteArray();
|
||||||
|
|
||||||
|
int precToBytes = ParquetHiveSerDe.PRECISION_TO_BYTE_COUNT[precision - 1];
|
||||||
|
if (precToBytes == decimalBytes.length) {
|
||||||
|
// No padding needed.
|
||||||
|
return Binary.fromByteArray(decimalBytes);
|
||||||
|
}
|
||||||
|
|
||||||
|
byte[] tgt = new byte[precToBytes];
|
||||||
|
|
||||||
|
// padding -1 for negative number
|
||||||
|
if (bigDecimal.compareTo(new BigDecimal("0")) < 0) {
|
||||||
|
Arrays.fill(tgt, 0, precToBytes - decimalBytes.length, (byte) -1);
|
||||||
|
}
|
||||||
|
|
||||||
|
System.arraycopy(decimalBytes, 0, tgt, precToBytes - decimalBytes.length, decimalBytes.length);
|
||||||
|
return Binary.fromByteArray(tgt);
|
||||||
|
}
|
||||||
|
|
||||||
|
private static final int JULIAN_EPOCH_OFFSET_DAYS = 2_440_588;
|
||||||
|
private static final long MILLIS_IN_DAY = TimeUnit.DAYS.toMillis(1);
|
||||||
|
private static final long MILLS_PER_SECOND = TimeUnit.SECONDS.toMillis(1);
|
||||||
|
private static final long NANOS_PER_DAY = TimeUnit.DAYS.toNanos(1);
|
||||||
|
private static final long NANOS_PER_SECOND = TimeUnit.SECONDS.toNanos(1);
|
||||||
|
private static final ZoneOffset defaultOffset = OffsetDateTime.now().getOffset();
|
||||||
|
|
||||||
|
/**
|
||||||
|
* int 96 is timestamp in parquet
|
||||||
|
*
|
||||||
|
* @param valueColumn
|
||||||
|
* @return
|
||||||
|
*/
|
||||||
|
private Binary timestampColToBinary(Column valueColumn) {
|
||||||
|
if (valueColumn.getRawData() == null) {
|
||||||
|
return Binary.EMPTY;
|
||||||
|
}
|
||||||
|
long mills;
|
||||||
|
long nanos = 0;
|
||||||
|
if (valueColumn instanceof DateColumn) {
|
||||||
|
DateColumn dateColumn = (DateColumn) valueColumn;
|
||||||
|
mills = dateColumn.asLong();
|
||||||
|
nanos = dateColumn.getNanos();
|
||||||
|
} else {
|
||||||
|
mills = valueColumn.asLong();
|
||||||
|
}
|
||||||
|
int julianDay;
|
||||||
|
long nanosOfDay;
|
||||||
|
if (isUtcTimestamp) {
|
||||||
|
// utc ignore current timezone (task should set timezone same as hive/hdfs)
|
||||||
|
long seconds = mills >= 0 ? mills / MILLS_PER_SECOND : (mills / MILLS_PER_SECOND - 1);
|
||||||
|
LocalDateTime localDateTime = LocalDateTime.ofEpochSecond(seconds, (int) nanos, defaultOffset);
|
||||||
|
julianDay = (int) (localDateTime.getLong(ChronoField.EPOCH_DAY) + JULIAN_EPOCH_OFFSET_DAYS);
|
||||||
|
nanosOfDay = localDateTime.getLong(ChronoField.NANO_OF_DAY);
|
||||||
|
} else {
|
||||||
|
// local date
|
||||||
|
julianDay = (int) ((mills / MILLIS_IN_DAY) + JULIAN_EPOCH_OFFSET_DAYS);
|
||||||
|
if (mills >= 0) {
|
||||||
|
nanosOfDay = ((mills % MILLIS_IN_DAY) / MILLS_PER_SECOND) * NANOS_PER_SECOND + nanos;
|
||||||
|
} else {
|
||||||
|
julianDay--;
|
||||||
|
nanosOfDay = (((mills % MILLIS_IN_DAY) / MILLS_PER_SECOND) - 1) * NANOS_PER_SECOND + nanos;
|
||||||
|
nanosOfDay += NANOS_PER_DAY;
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
ByteBuffer buf = ByteBuffer.allocate(12);
|
||||||
|
buf.order(ByteOrder.LITTLE_ENDIAN);
|
||||||
|
buf.putLong(nanosOfDay);
|
||||||
|
buf.putInt(julianDay);
|
||||||
|
buf.flip();
|
||||||
|
return Binary.fromByteBuffer(buf);
|
||||||
|
}
|
||||||
|
|
||||||
|
private void writeBaseOnFields(Record values) {
|
||||||
|
//LOGGER.info("Writing parquet data using fields mode(The correct mode.)");
|
||||||
|
List<Type> types = this.schema.getFields();
|
||||||
|
|
||||||
|
if (values != null && types != null && values.getColumnNumber() == types.size()) {
|
||||||
|
recordConsumer.startMessage();
|
||||||
|
writeFields(types, values);
|
||||||
|
recordConsumer.endMessage();
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
private void writeFields(List<Type> types, Record values) {
|
||||||
|
for (int i = 0; i < types.size(); i++) {
|
||||||
|
Type type = types.get(i);
|
||||||
|
Column value = values.getColumn(i);
|
||||||
|
if (value != null) {
|
||||||
|
try {
|
||||||
|
if (type.isPrimitive()) {
|
||||||
|
writePrimitiveType(type, value, i);
|
||||||
|
} else {
|
||||||
|
writeGroupType(type, (JSON) JSON.parse(value.asString()), i);
|
||||||
|
}
|
||||||
|
} catch (Exception e) {
|
||||||
|
if (printStackTrace) {
|
||||||
|
printStackTrace = false;
|
||||||
|
LOGGER.warn("write to parquet error: {}", e.getMessage(), e);
|
||||||
|
}
|
||||||
|
// dirty data
|
||||||
|
if (null != this.taskPluginCollector) {
|
||||||
|
// job post 里面的merge taskPluginCollector 为null
|
||||||
|
this.taskPluginCollector.collectDirtyRecord(values, e, e.getMessage());
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
private void writeFields(List<Type> types, JSONObject values) {
|
||||||
|
for (int i = 0; i < types.size(); i++) {
|
||||||
|
Type type = types.get(i);
|
||||||
|
Object value = values.get(type.getName());
|
||||||
|
|
||||||
|
if (value != null) {
|
||||||
|
try {
|
||||||
|
if (type.isPrimitive()) {
|
||||||
|
writePrimitiveType(type, value, i);
|
||||||
|
} else {
|
||||||
|
writeGroupType(type, (JSON) value, i);
|
||||||
|
}
|
||||||
|
} catch (Exception e) {
|
||||||
|
if (printStackTrace) {
|
||||||
|
printStackTrace = false;
|
||||||
|
LOGGER.warn("write to parquet error: {}", e.getMessage(), e);
|
||||||
|
}
|
||||||
|
}
|
||||||
|
} else {
|
||||||
|
recordConsumer.addBinary(this.binaryForNull);
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
private void writeGroupType(Type type, JSON value, int index) {
|
||||||
|
GroupType groupType = type.asGroupType();
|
||||||
|
OriginalType originalType = groupType.getOriginalType();
|
||||||
|
if (originalType != null) {
|
||||||
|
switch (originalType) {
|
||||||
|
case MAP:
|
||||||
|
writeMap(groupType, value, index);
|
||||||
|
break;
|
||||||
|
case LIST:
|
||||||
|
writeList(groupType, value, index);
|
||||||
|
break;
|
||||||
|
default:
|
||||||
|
break;
|
||||||
|
}
|
||||||
|
} else {
|
||||||
|
// struct
|
||||||
|
writeStruct(groupType, value, index);
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
private void writeMap(GroupType groupType, JSON value, int index) {
|
||||||
|
if (value == null) {
|
||||||
|
return;
|
||||||
|
}
|
||||||
|
|
||||||
|
JSONObject json = (JSONObject) value;
|
||||||
|
|
||||||
|
if (json.isEmpty()) {
|
||||||
|
return;
|
||||||
|
}
|
||||||
|
|
||||||
|
recordConsumer.startField(groupType.getName(), index);
|
||||||
|
|
||||||
|
recordConsumer.startGroup();
|
||||||
|
|
||||||
|
// map
|
||||||
|
// key_value start
|
||||||
|
recordConsumer.startField("key_value", 0);
|
||||||
|
recordConsumer.startGroup();
|
||||||
|
|
||||||
|
List<Type> keyValueFields = groupType.getFields().get(0).asGroupType().getFields();
|
||||||
|
Type keyType = keyValueFields.get(0);
|
||||||
|
Type valueType = keyValueFields.get(1);
|
||||||
|
for (String key : json.keySet()) {
|
||||||
|
// key
|
||||||
|
writePrimitiveType(keyType, key, 0);
|
||||||
|
|
||||||
|
// value
|
||||||
|
if (valueType.isPrimitive()) {
|
||||||
|
writePrimitiveType(valueType, json.get(key), 1);
|
||||||
|
} else {
|
||||||
|
writeGroupType(valueType, (JSON) json.get(key), 1);
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
recordConsumer.endGroup();
|
||||||
|
recordConsumer.endField("key_value", 0);
|
||||||
|
// key_value end
|
||||||
|
|
||||||
|
recordConsumer.endGroup();
|
||||||
|
recordConsumer.endField(groupType.getName(), index);
|
||||||
|
}
|
||||||
|
|
||||||
|
private void writeList(GroupType groupType, JSON value, int index) {
|
||||||
|
if (value == null) {
|
||||||
|
return;
|
||||||
|
}
|
||||||
|
|
||||||
|
JSONArray json = (JSONArray) value;
|
||||||
|
|
||||||
|
if (json.isEmpty()) {
|
||||||
|
return;
|
||||||
|
}
|
||||||
|
|
||||||
|
recordConsumer.startField(groupType.getName(), index);
|
||||||
|
// list
|
||||||
|
recordConsumer.startGroup();
|
||||||
|
|
||||||
|
// list start
|
||||||
|
recordConsumer.startField("list", 0);
|
||||||
|
recordConsumer.startGroup();
|
||||||
|
|
||||||
|
Type elementType = groupType.getFields().get(0).asGroupType().getFields().get(0);
|
||||||
|
|
||||||
|
if (elementType.isPrimitive()) {
|
||||||
|
for (Object elementValue : json) {
|
||||||
|
writePrimitiveType(elementType, elementValue, 0);
|
||||||
|
}
|
||||||
|
} else {
|
||||||
|
for (Object elementValue : json) {
|
||||||
|
writeGroupType(elementType, (JSON) elementValue, 0);
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
recordConsumer.endGroup();
|
||||||
|
recordConsumer.endField("list", 0);
|
||||||
|
// list end
|
||||||
|
recordConsumer.endGroup();
|
||||||
|
|
||||||
|
recordConsumer.endField(groupType.getName(), index);
|
||||||
|
}
|
||||||
|
|
||||||
|
private void writeStruct(GroupType groupType, JSON value, int index) {
|
||||||
|
if (value == null) {
|
||||||
|
return;
|
||||||
|
}
|
||||||
|
JSONObject json = (JSONObject) value;
|
||||||
|
if (json.isEmpty()) {
|
||||||
|
return;
|
||||||
|
}
|
||||||
|
|
||||||
|
recordConsumer.startField(groupType.getName(), index);
|
||||||
|
// struct start
|
||||||
|
recordConsumer.startGroup();
|
||||||
|
|
||||||
|
writeFields(groupType.getFields(), json);
|
||||||
|
recordConsumer.endGroup();
|
||||||
|
// struct end
|
||||||
|
recordConsumer.endField(groupType.getName(), index);
|
||||||
|
}
|
||||||
|
|
||||||
|
private void writePrimitiveType(Type type, Object value, int index) {
|
||||||
|
if (value == null) {
|
||||||
|
return;
|
||||||
|
}
|
||||||
|
|
||||||
|
recordConsumer.startField(type.getName(), index);
|
||||||
|
PrimitiveType primitiveType = type.asPrimitiveType();
|
||||||
|
|
||||||
|
switch (primitiveType.getPrimitiveTypeName()) {
|
||||||
|
case BOOLEAN:
|
||||||
|
recordConsumer.addBoolean((Boolean) value);
|
||||||
|
break;
|
||||||
|
case FLOAT:
|
||||||
|
if (value instanceof Float) {
|
||||||
|
recordConsumer.addFloat(((Float) value).floatValue());
|
||||||
|
} else if (value instanceof Double) {
|
||||||
|
recordConsumer.addFloat(((Double) value).floatValue());
|
||||||
|
} else if (value instanceof Long) {
|
||||||
|
recordConsumer.addFloat(((Long) value).floatValue());
|
||||||
|
} else if (value instanceof Integer) {
|
||||||
|
recordConsumer.addFloat(((Integer) value).floatValue());
|
||||||
|
}
|
||||||
|
break;
|
||||||
|
case DOUBLE:
|
||||||
|
if (value instanceof Float) {
|
||||||
|
recordConsumer.addDouble(((Float) value).doubleValue());
|
||||||
|
} else if (value instanceof Double) {
|
||||||
|
recordConsumer.addDouble(((Double) value).doubleValue());
|
||||||
|
} else if (value instanceof Long) {
|
||||||
|
recordConsumer.addDouble(((Long) value).doubleValue());
|
||||||
|
} else if (value instanceof Integer) {
|
||||||
|
recordConsumer.addDouble(((Integer) value).doubleValue());
|
||||||
|
}
|
||||||
|
break;
|
||||||
|
case INT32:
|
||||||
|
if (value instanceof Integer) {
|
||||||
|
recordConsumer.addInteger((Integer) value);
|
||||||
|
} else if (value instanceof Long) {
|
||||||
|
recordConsumer.addInteger(((Long) value).intValue());
|
||||||
|
} else {
|
||||||
|
// 之前代码写的有问题，导致这里丢列了没抛异常，先收集，后续看看有没有任务命中在决定怎么改
|
||||||
|
LimitLogger.limit("dirtyDataHiveWriterParquet", TimeUnit.MINUTES.toMillis(1), () -> LOGGER.warn("dirtyDataHiveWriterParquet {}", String.format("Invalid value: %s(clazz: %s) for field: %s", value, value.getClass(), type.getName())));
|
||||||
|
}
|
||||||
|
break;
|
||||||
|
case INT64:
|
||||||
|
if (value instanceof Integer) {
|
||||||
|
recordConsumer.addLong(((Integer) value).longValue());
|
||||||
|
} else if (value instanceof Long) {
|
||||||
|
recordConsumer.addInteger(((Long) value).intValue());
|
||||||
|
} else {
|
||||||
|
// 之前代码写的有问题，导致这里丢列了没抛异常，先收集，后续看看有没有任务命中在决定怎么改
|
||||||
|
LimitLogger.limit("dirtyDataHiveWriterParquet", TimeUnit.MINUTES.toMillis(1), () -> LOGGER.warn("dirtyDataHiveWriterParquet {}", String.format("Invalid value: %s(clazz: %s) for field: %s", value, value.getClass(), type.getName())));
|
||||||
|
}
|
||||||
|
break;
|
||||||
|
case INT96:
|
||||||
|
if (value instanceof Integer) {
|
||||||
|
recordConsumer.addBinary(timestampColToBinary(new LongColumn((Integer) value)));
|
||||||
|
} else if (value instanceof Long) {
|
||||||
|
recordConsumer.addBinary(timestampColToBinary(new LongColumn((Long) value)));
|
||||||
|
} else if (value instanceof Timestamp) {
|
||||||
|
recordConsumer.addBinary(timestampColToBinary(new DateColumn((Timestamp) value)));
|
||||||
|
} else if (value instanceof Date) {
|
||||||
|
recordConsumer.addBinary(timestampColToBinary(new DateColumn((Date) value)));
|
||||||
|
} else {
|
||||||
|
recordConsumer.addBinary(timestampColToBinary(new StringColumn(value.toString())));
|
||||||
|
}
|
||||||
|
break;
|
||||||
|
case FIXED_LEN_BYTE_ARRAY:
|
||||||
|
if (primitiveType.getDecimalMetadata() != null) {
|
||||||
|
// decimal
|
||||||
|
Column column;
|
||||||
|
if (value instanceof Integer) {
|
||||||
|
column = new LongColumn((Integer) value);
|
||||||
|
} else if (value instanceof Long) {
|
||||||
|
column = new LongColumn((Long) value);
|
||||||
|
} else if (value instanceof Double) {
|
||||||
|
column = new DoubleColumn((Double) value);
|
||||||
|
} else if (value instanceof BigDecimal) {
|
||||||
|
column = new DoubleColumn((BigDecimal) value);
|
||||||
|
} else {
|
||||||
|
column = new StringColumn(value.toString());
|
||||||
|
}
|
||||||
|
recordConsumer.addBinary(decimalToBinary(column, primitiveType.getDecimalMetadata().getPrecision(), primitiveType.getDecimalMetadata().getScale()));
|
||||||
|
break;
|
||||||
|
}
|
||||||
|
/* fall through */
|
||||||
|
case BINARY:
|
||||||
|
default:
|
||||||
|
recordConsumer.addBinary(Binary.fromString((String) value));
|
||||||
|
break;
|
||||||
|
}
|
||||||
|
recordConsumer.endField(type.getName(), index);
|
||||||
|
}
|
||||||
|
|
||||||
|
private void writePrimitiveType(Type type, Column value, int index) {
|
||||||
|
if (value == null || value.getRawData() == null) {
|
||||||
|
return;
|
||||||
|
}
|
||||||
|
|
||||||
|
recordConsumer.startField(type.getName(), index);
|
||||||
|
PrimitiveType primitiveType = type.asPrimitiveType();
|
||||||
|
switch (primitiveType.getPrimitiveTypeName()) {
|
||||||
|
case BOOLEAN:
|
||||||
|
recordConsumer.addBoolean(value.asBoolean());
|
||||||
|
break;
|
||||||
|
case FLOAT:
|
||||||
|
recordConsumer.addFloat(value.asDouble().floatValue());
|
||||||
|
break;
|
||||||
|
case DOUBLE:
|
||||||
|
recordConsumer.addDouble(value.asDouble());
|
||||||
|
break;
|
||||||
|
case INT32:
|
||||||
|
OriginalType originalType = type.getOriginalType();
|
||||||
|
if (OriginalType.DATE.equals(originalType)) {
|
||||||
|
int realVal = (int) (new java.sql.Date(value.asLong()).toLocalDate().toEpochDay());
|
||||||
|
recordConsumer.addInteger(realVal);
|
||||||
|
} else {
|
||||||
|
recordConsumer.addInteger(value.asLong().intValue());
|
||||||
|
}
|
||||||
|
break;
|
||||||
|
case INT64:
|
||||||
|
recordConsumer.addLong(value.asLong());
|
||||||
|
break;
|
||||||
|
case INT96:
|
||||||
|
recordConsumer.addBinary(timestampColToBinary(value));
|
||||||
|
break;
|
||||||
|
case BINARY:
|
||||||
|
String valueAsString2Write = null;
|
||||||
|
if (Column.Type.DATE == value.getType() && null != this.dateParse) {
|
||||||
|
valueAsString2Write = dateParse.format(value.asDate());
|
||||||
|
} else {
|
||||||
|
valueAsString2Write = value.asString();
|
||||||
|
}
|
||||||
|
recordConsumer.addBinary(Binary.fromString(valueAsString2Write));
|
||||||
|
break;
|
||||||
|
case FIXED_LEN_BYTE_ARRAY:
|
||||||
|
if (primitiveType.getDecimalMetadata() != null) {
|
||||||
|
// decimal
|
||||||
|
recordConsumer.addBinary(decimalToBinary(value, primitiveType.getDecimalMetadata().getPrecision(), primitiveType.getDecimalMetadata().getScale()));
|
||||||
|
break;
|
||||||
|
}
|
||||||
|
/* fall through */
|
||||||
|
default:
|
||||||
|
recordConsumer.addBinary(Binary.fromString(value.asString()));
|
||||||
|
break;
|
||||||
|
}
|
||||||
|
recordConsumer.endField(type.getName(), index);
|
||||||
|
}
|
||||||
|
}
|
||||||
193
neo4jwriter/doc/neo4jwriter.md
Normal file
193
neo4jwriter/doc/neo4jwriter.md
Normal file
@ -0,0 +1,193 @@
|
|||||||
|
# DataX neo4jWriter 插件文档
|
||||||
|
|
||||||
|
## 功能简介
|
||||||
|
|
||||||
|
本目前市面上的neo4j 批量导入主要有Cypher Create,Load CSV,第三方或者官方提供的Batch Import。Load CSV支持节点10W级别一下，Batch Import 需要对数据库进行停机。要想实现不停机的数据写入，Cypher是最好的方式。
|
||||||
|
|
||||||
|
## 支持版本
|
||||||
|
|
||||||
|
支持Neo4j 4 和Neo4j 5,如果是Neo4j 3,需要自行将驱动降低至相对应的版本进行编译。
|
||||||
|
|
||||||
|
## 实现原理
|
||||||
|
|
||||||
|
将datax的数据转换成了neo4j驱动能识别的对象，利用 unwind 语法进行批量插入。
|
||||||
|
|
||||||
|
## 如何配置
|
||||||
|
|
||||||
|
### 配置项介绍
|
||||||
|
|
||||||
|
| 配置 | 说明 | 是否必须 | 默认值 | 示例 |
|
||||||
|
|:-------------------------------|--------------------| -------- | ------ | ---------------------------------------------------- |
|
||||||
|
| database | 数据库名字 | 是 | - | neo4j |
|
||||||
|
| uri | 数据库访问链接 | 是 | - | bolt://localhost:7687 |
|
||||||
|
| username | 访问用户名 | 是 | - | neo4j |
|
||||||
|
| password | 访问密码 | 是 | - | neo4j |
|
||||||
|
| bearerToken | 权限相关 | 否 | - | - |
|
||||||
|
| kerberosTicket | 权限相关 | 否 | - | - |
|
||||||
|
| cypher | 同步语句 | 是 | - | unwind $batch as row create(p) set p.name = row.name |
|
||||||
|
| batchDataVariableName | unwind 携带的数据变量名 | | | batch |
|
||||||
|
| properties | 定义neo4j中数据的属性名字和类型 | 是 | - | 见后续案例 |
|
||||||
|
| batchSize | 一批写入数据量 | 否 | 1000 | |
|
||||||
|
| maxTransactionRetryTimeSeconds | 事务运行最长时间 | 否 | 30秒 | 30 |
|
||||||
|
| maxConnectionTimeoutSeconds | 驱动最长链接时间 | 否 | 30秒 | 30 |
|
||||||
|
| retryTimes | 发生错误的重试次数 | 否 | 3次 | 3 |
|
||||||
|
| retrySleepMills | 重试失败后的等待时间 | 否 | 3秒 | 3 |
|
||||||
|
|
||||||
|
### 支持的数据类型
|
||||||
|
> 配置时均忽略大小写
|
||||||
|
```
|
||||||
|
BOOLEAN,
|
||||||
|
STRING,
|
||||||
|
LONG,
|
||||||
|
SHORT,
|
||||||
|
INTEGER,
|
||||||
|
DOUBLE,
|
||||||
|
FLOAT,
|
||||||
|
LOCAL_DATE,
|
||||||
|
LOCAL_TIME,
|
||||||
|
LOCAL_DATE_TIME,
|
||||||
|
LIST,
|
||||||
|
//map类型支持 . 属性表达式取值
|
||||||
|
MAP,
|
||||||
|
CHAR_ARRAY,
|
||||||
|
BYTE_ARRAY,
|
||||||
|
BOOLEAN_ARRAY,
|
||||||
|
STRING_ARRAY,
|
||||||
|
LONG_ARRAY,
|
||||||
|
INT_ARRAY,
|
||||||
|
SHORT_ARRAY,
|
||||||
|
DOUBLE_ARRAY,
|
||||||
|
FLOAT_ARRAY,
|
||||||
|
Object_ARRAY
|
||||||
|
```
|
||||||
|
|
||||||
|
### 写节点
|
||||||
|
|
||||||
|
这里提供了一个写节点包含很多类型属性的例子。你可以在我的测试方法中运行。
|
||||||
|
|
||||||
|
```json
|
||||||
|
"writer": {
|
||||||
|
"name": "neo4jWriter",
|
||||||
|
"parameter": {
|
||||||
|
"uri": "neo4j://localhost:7687",
|
||||||
|
"username": "neo4j",
|
||||||
|
"password": "Test@12343",
|
||||||
|
"database": "neo4j",
|
||||||
|
"cypher": "unwind $batch as row create(p:Person) set p.pbool = row.pbool,p.pstring = row.pstring,p.plong = row.plong,p.pshort = row.pshort,p.pdouble=row.pdouble,p.pstringarr=row.pstringarr,p.plocaldate=row.plocaldate",
|
||||||
|
"batchDataVariableName": "batch",
|
||||||
|
"batchSize": "33",
|
||||||
|
"properties": [
|
||||||
|
{
|
||||||
|
"name": "pbool",
|
||||||
|
"type": "BOOLEAN"
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"name": "pstring",
|
||||||
|
"type": "STRING"
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"name": "plong",
|
||||||
|
"type": "LONG"
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"name": "pshort",
|
||||||
|
"type": "SHORT"
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"name": "pdouble",
|
||||||
|
"type": "DOUBLE"
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"name": "pstringarr",
|
||||||
|
"type": "STRING_ARRAY",
|
||||||
|
"split": ","
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"name": "plocaldate",
|
||||||
|
"type": "LOCAL_DATE",
|
||||||
|
"dateFormat": "yyyy-MM-dd"
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
|
}
|
||||||
|
```
|
||||||
|
|
||||||
|
### 写关系
|
||||||
|
|
||||||
|
```json
|
||||||
|
"writer": {
|
||||||
|
"name": "neo4jWriter",
|
||||||
|
"parameter": {
|
||||||
|
"uri": "neo4j://localhost:7687",
|
||||||
|
"username": "neo4j",
|
||||||
|
"password": "Test@12343",
|
||||||
|
"database": "neo4j",
|
||||||
|
"cypher": "unwind $batch as row match(p1:Person) where p1.id = row.startNodeId match(p2:Person) where p2.id = row.endNodeId create (p1)-[:LINK]->(p2)",
|
||||||
|
"batchDataVariableName": "batch",
|
||||||
|
"batch_size": "33",
|
||||||
|
"properties": [
|
||||||
|
{
|
||||||
|
"name": "startNodeId",
|
||||||
|
"type": "STRING"
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"name": "endNodeId",
|
||||||
|
"type": "STRING"
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
|
}
|
||||||
|
```
|
||||||
|
|
||||||
|
### 节点/关系类型动态写
|
||||||
|
|
||||||
|
> 需要使用AOPC函数拓展，如果你的数据库没有，请安装APOC函数拓展
|
||||||
|
|
||||||
|
```json
|
||||||
|
"writer": {
|
||||||
|
"name": "neo4jWriter",
|
||||||
|
"parameter": {
|
||||||
|
"uri": "bolt://localhost:7687",
|
||||||
|
"username": "yourUserName",
|
||||||
|
"password": "yourPassword",
|
||||||
|
"database": "yourDataBase",
|
||||||
|
"cypher": "unwind $batch as row CALL apoc.cypher.doIt( 'create (n:`' + row.Label + '`{id:$id})' ,{id: row.id} ) YIELD value RETURN 1 ",
|
||||||
|
"batchDataVariableName": "batch",
|
||||||
|
"batch_size": "1",
|
||||||
|
"properties": [
|
||||||
|
{
|
||||||
|
"name": "Label",
|
||||||
|
"type": "STRING"
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"name": "id",
|
||||||
|
"type": "STRING"
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
|
}
|
||||||
|
```
|
||||||
|
|
||||||
|
## 注意事项
|
||||||
|
|
||||||
|
* properties定义的顺序需要与reader端顺序一一对应。
|
||||||
|
* 灵活使用map类型，可以免去很多数据加工的烦恼。在cypher中，可以根据 . 属性访问符号一直取值。比如 unwind $batch as row create (p) set p.name = row.prop.name,set p.age = row.prop.age，在这个例子中，prop是map类型，包含name和age两个属性。
|
||||||
|
* 如果提示事务超时，建议调大事务运行时间或者调小batchSize
|
||||||
|
* 如果用于更新场景，遇到死锁问题影响写入，建议二开源码加入死锁异常检测，并进行重试。
|
||||||
|
|
||||||
|
## 性能报告
|
||||||
|
|
||||||
|
**JVM参数**
|
||||||
|
|
||||||
|
16G G1垃圾收集器 8核心
|
||||||
|
|
||||||
|
**Neo4j数据库配置**
|
||||||
|
|
||||||
|
32核心，256G
|
||||||
|
|
||||||
|
**datax 配置**
|
||||||
|
|
||||||
|
* Channel 20 batchsize = 1000
|
||||||
|
* 任务平均流量：15.23MB/s
|
||||||
|
* 记录写入速度：44440 rec/s
|
||||||
|
* 读出记录总数：2222013
|
||||||
100
neo4jwriter/pom.xml
Normal file
100
neo4jwriter/pom.xml
Normal file
@ -0,0 +1,100 @@
|
|||||||
|
<?xml version="1.0" encoding="UTF-8"?>
|
||||||
|
<project xmlns="http://maven.apache.org/POM/4.0.0"
|
||||||
|
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
|
||||||
|
xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
|
||||||
|
<parent>
|
||||||
|
<groupId>com.alibaba.datax</groupId>
|
||||||
|
<artifactId>datax-all</artifactId>
|
||||||
|
<version>0.0.1-SNAPSHOT</version>
|
||||||
|
</parent>
|
||||||
|
<modelVersion>4.0.0</modelVersion>
|
||||||
|
|
||||||
|
<artifactId>neo4jwriter</artifactId>
|
||||||
|
<name>neo4jwriter</name>
|
||||||
|
<packaging>jar</packaging>
|
||||||
|
|
||||||
|
<properties>
|
||||||
|
<maven.compiler.source>8</maven.compiler.source>
|
||||||
|
<maven.compiler.target>8</maven.compiler.target>
|
||||||
|
<project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
|
||||||
|
<neo4j-java-driver.version>4.4.9</neo4j-java-driver.version>
|
||||||
|
<junit4.version>4.13.2</junit4.version>
|
||||||
|
<test.container.version>1.17.6</test.container.version>
|
||||||
|
</properties>
|
||||||
|
<dependencies>
|
||||||
|
<dependency>
|
||||||
|
<groupId>org.slf4j</groupId>
|
||||||
|
<artifactId>slf4j-api</artifactId>
|
||||||
|
</dependency>
|
||||||
|
<dependency>
|
||||||
|
<groupId>ch.qos.logback</groupId>
|
||||||
|
<artifactId>logback-classic</artifactId>
|
||||||
|
</dependency>
|
||||||
|
<dependency>
|
||||||
|
<groupId>org.neo4j.driver</groupId>
|
||||||
|
<artifactId>neo4j-java-driver</artifactId>
|
||||||
|
<version>${neo4j-java-driver.version}</version>
|
||||||
|
</dependency>
|
||||||
|
<dependency>
|
||||||
|
<groupId>com.alibaba.datax</groupId>
|
||||||
|
<artifactId>datax-common</artifactId>
|
||||||
|
<version>${datax-project-version}</version>
|
||||||
|
</dependency>
|
||||||
|
|
||||||
|
<dependency>
|
||||||
|
<groupId>org.testcontainers</groupId>
|
||||||
|
<artifactId>testcontainers</artifactId>
|
||||||
|
<version>${test.container.version}</version>
|
||||||
|
</dependency>
|
||||||
|
<!-- Testcontainers 1.x is tightly coupled with the JUnit 4.x rule API-->
|
||||||
|
<dependency>
|
||||||
|
<groupId>junit</groupId>
|
||||||
|
<artifactId>junit</artifactId>
|
||||||
|
<version>${junit4.version}</version>
|
||||||
|
<scope>test</scope>
|
||||||
|
</dependency>
|
||||||
|
</dependencies>
|
||||||
|
|
||||||
|
<build>
|
||||||
|
<resources>
|
||||||
|
<resource>
|
||||||
|
<directory>src/main/resources</directory>
|
||||||
|
<includes>
|
||||||
|
<include>**/*.*</include>
|
||||||
|
</includes>
|
||||||
|
<filtering>true</filtering>
|
||||||
|
</resource>
|
||||||
|
</resources>
|
||||||
|
|
||||||
|
<plugins>
|
||||||
|
<!-- compiler plugin -->
|
||||||
|
<plugin>
|
||||||
|
<artifactId>maven-compiler-plugin</artifactId>
|
||||||
|
<configuration>
|
||||||
|
<source>${jdk-version}</source>
|
||||||
|
<target>${jdk-version}</target>
|
||||||
|
<encoding>${project-sourceEncoding}</encoding>
|
||||||
|
</configuration>
|
||||||
|
</plugin>
|
||||||
|
<!-- assembly plugin -->
|
||||||
|
<plugin>
|
||||||
|
<artifactId>maven-assembly-plugin</artifactId>
|
||||||
|
<configuration>
|
||||||
|
<descriptors>
|
||||||
|
<descriptor>src/main/assembly/package.xml</descriptor>
|
||||||
|
</descriptors>
|
||||||
|
<finalName>datax</finalName>
|
||||||
|
</configuration>
|
||||||
|
<executions>
|
||||||
|
<execution>
|
||||||
|
<id>dwzip</id>
|
||||||
|
<phase>package</phase>
|
||||||
|
<goals>
|
||||||
|
<goal>single</goal>
|
||||||
|
</goals>
|
||||||
|
</execution>
|
||||||
|
</executions>
|
||||||
|
</plugin>
|
||||||
|
</plugins>
|
||||||
|
</build>
|
||||||
|
</project>
|
||||||
35
neo4jwriter/src/main/assembly/package.xml
Normal file
35
neo4jwriter/src/main/assembly/package.xml
Normal file
@ -0,0 +1,35 @@
|
|||||||
|
<assembly
|
||||||
|
xmlns="http://maven.apache.org/plugins/maven-assembly-plugin/assembly/1.1.0"
|
||||||
|
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
|
||||||
|
xsi:schemaLocation="http://maven.apache.org/plugins/maven-assembly-plugin/assembly/1.1.0 http://maven.apache.org/xsd/assembly-1.1.0.xsd">
|
||||||
|
<id></id>
|
||||||
|
<formats>
|
||||||
|
<format>dir</format>
|
||||||
|
</formats>
|
||||||
|
<includeBaseDirectory>false</includeBaseDirectory>
|
||||||
|
<fileSets>
|
||||||
|
<fileSet>
|
||||||
|
<directory>src/main/resources</directory>
|
||||||
|
<includes>
|
||||||
|
<include>plugin.json</include>
|
||||||
|
<include>plugin_job_template.json</include>
|
||||||
|
</includes>
|
||||||
|
<outputDirectory>plugin/writer/neo4jwriter</outputDirectory>
|
||||||
|
</fileSet>
|
||||||
|
<fileSet>
|
||||||
|
<directory>target/</directory>
|
||||||
|
<includes>
|
||||||
|
<include>neo4jwriter-0.0.1-SNAPSHOT.jar</include>
|
||||||
|
</includes>
|
||||||
|
<outputDirectory>plugin/writer/neo4jwriter</outputDirectory>
|
||||||
|
</fileSet>
|
||||||
|
</fileSets>
|
||||||
|
|
||||||
|
<dependencySets>
|
||||||
|
<dependencySet>
|
||||||
|
<useProjectArtifact>false</useProjectArtifact>
|
||||||
|
<outputDirectory>plugin/writer/neo4jwriter/libs</outputDirectory>
|
||||||
|
<scope>runtime</scope>
|
||||||
|
</dependencySet>
|
||||||
|
</dependencySets>
|
||||||
|
</assembly>
|
||||||
@ -0,0 +1,256 @@
|
|||||||
|
package com.alibaba.datax.plugin.writer.neo4jwriter;
|
||||||
|
|
||||||
|
|
||||||
|
import com.alibaba.datax.common.element.Column;
|
||||||
|
import com.alibaba.datax.common.element.Record;
|
||||||
|
import com.alibaba.datax.common.exception.DataXException;
|
||||||
|
import com.alibaba.datax.common.plugin.TaskPluginCollector;
|
||||||
|
import com.alibaba.datax.common.util.Configuration;
|
||||||
|
import com.alibaba.datax.common.util.RetryUtil;
|
||||||
|
import com.alibaba.datax.plugin.writer.neo4jwriter.adapter.DateAdapter;
|
||||||
|
import com.alibaba.datax.plugin.writer.neo4jwriter.adapter.ValueAdapter;
|
||||||
|
import com.alibaba.datax.plugin.writer.neo4jwriter.config.Neo4jProperty;
|
||||||
|
import com.alibaba.datax.plugin.writer.neo4jwriter.exception.Neo4jErrorCode;
|
||||||
|
import com.alibaba.fastjson2.JSON;
|
||||||
|
import org.apache.commons.lang3.StringUtils;
|
||||||
|
import org.neo4j.driver.*;
|
||||||
|
import org.neo4j.driver.exceptions.Neo4jException;
|
||||||
|
import org.neo4j.driver.internal.value.MapValue;
|
||||||
|
import org.slf4j.Logger;
|
||||||
|
import org.slf4j.LoggerFactory;
|
||||||
|
|
||||||
|
import java.util.*;
|
||||||
|
import java.util.concurrent.TimeUnit;
|
||||||
|
|
||||||
|
import static com.alibaba.datax.plugin.writer.neo4jwriter.config.ConfigConstants.*;
|
||||||
|
import static com.alibaba.datax.plugin.writer.neo4jwriter.exception.Neo4jErrorCode.DATABASE_ERROR;
|
||||||
|
|
||||||
|
public class Neo4jClient {
|
||||||
|
private static final Logger LOGGER = LoggerFactory.getLogger(Neo4jClient.class);
|
||||||
|
private Driver driver;
|
||||||
|
|
||||||
|
private WriteConfig writeConfig;
|
||||||
|
private RetryConfig retryConfig;
|
||||||
|
private TaskPluginCollector taskPluginCollector;
|
||||||
|
|
||||||
|
private Session session;
|
||||||
|
|
||||||
|
private List<MapValue> writerBuffer;
|
||||||
|
|
||||||
|
|
||||||
|
public Neo4jClient(Driver driver,
|
||||||
|
WriteConfig writeConfig,
|
||||||
|
RetryConfig retryConfig,
|
||||||
|
TaskPluginCollector taskPluginCollector) {
|
||||||
|
this.driver = driver;
|
||||||
|
this.writeConfig = writeConfig;
|
||||||
|
this.retryConfig = retryConfig;
|
||||||
|
this.taskPluginCollector = taskPluginCollector;
|
||||||
|
this.writerBuffer = new ArrayList<>(writeConfig.batchSize);
|
||||||
|
}
|
||||||
|
|
||||||
|
public void init() {
|
||||||
|
String database = writeConfig.database;
|
||||||
|
//neo4j 3.x 没有数据库
|
||||||
|
if (null != database && !"".equals(database)) {
|
||||||
|
this.session = driver.session(SessionConfig.forDatabase(database));
|
||||||
|
} else {
|
||||||
|
this.session = driver.session();
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
public static Neo4jClient build(Configuration config, TaskPluginCollector taskPluginCollector) {
|
||||||
|
|
||||||
|
Driver driver = buildNeo4jDriver(config);
|
||||||
|
String cypher = checkCypher(config);
|
||||||
|
String database = config.getString(DATABASE.getKey());
|
||||||
|
String batchVariableName = config.getString(BATCH_DATA_VARIABLE_NAME.getKey(),
|
||||||
|
BATCH_DATA_VARIABLE_NAME.getDefaultValue());
|
||||||
|
List<Neo4jProperty> neo4jProperties = JSON.parseArray(config.getString(NEO4J_PROPERTIES.getKey()), Neo4jProperty.class);
|
||||||
|
int batchSize = config.getInt(BATCH_SIZE.getKey(), BATCH_SIZE.getDefaultValue());
|
||||||
|
int retryTimes = config.getInt(RETRY_TIMES.getKey(), RETRY_TIMES.getDefaultValue());
|
||||||
|
|
||||||
|
return new Neo4jClient(driver,
|
||||||
|
new WriteConfig(cypher, database, batchVariableName, neo4jProperties, batchSize),
|
||||||
|
new RetryConfig(retryTimes, config.getLong(RETRY_SLEEP_MILLS.getKey(), RETRY_SLEEP_MILLS.getDefaultValue())),
|
||||||
|
taskPluginCollector
|
||||||
|
);
|
||||||
|
}
|
||||||
|
|
||||||
|
private static String checkCypher(Configuration config) {
|
||||||
|
String cypher = config.getString(CYPHER.getKey());
|
||||||
|
if (StringUtils.isBlank(cypher)) {
|
||||||
|
throw DataXException.asDataXException(Neo4jErrorCode.CONFIG_INVALID, "cypher must not null or empty");
|
||||||
|
}
|
||||||
|
return cypher;
|
||||||
|
}
|
||||||
|
|
||||||
|
private static Driver buildNeo4jDriver(Configuration config) {
|
||||||
|
|
||||||
|
Config.ConfigBuilder configBuilder = Config.builder().withMaxConnectionPoolSize(1);
|
||||||
|
String uri = checkUriConfig(config);
|
||||||
|
|
||||||
|
//connection timeout
|
||||||
|
//连接超时时间
|
||||||
|
Long maxConnTime = config.getLong(MAX_CONNECTION_TIMEOUT_SECONDS.getKey(), MAX_TRANSACTION_RETRY_TIME.getDefaultValue());
|
||||||
|
configBuilder
|
||||||
|
.withConnectionAcquisitionTimeout(
|
||||||
|
maxConnTime * 2, TimeUnit.SECONDS)
|
||||||
|
.withConnectionTimeout(maxConnTime, TimeUnit.SECONDS);
|
||||||
|
|
||||||
|
|
||||||
|
//transaction timeout
|
||||||
|
//事务运行超时时间
|
||||||
|
Long txRetryTime = config.getLong(MAX_TRANSACTION_RETRY_TIME.getKey(), MAX_TRANSACTION_RETRY_TIME.getDefaultValue());
|
||||||
|
configBuilder.withMaxTransactionRetryTime(txRetryTime, TimeUnit.SECONDS);
|
||||||
|
String username = config.getString(USERNAME.getKey());
|
||||||
|
String password = config.getString(PASSWORD.getKey());
|
||||||
|
String bearerToken = config.getString(BEARER_TOKEN.getKey());
|
||||||
|
String kerberosTicket = config.getString(KERBEROS_TICKET.getKey());
|
||||||
|
|
||||||
|
if (StringUtils.isNotBlank(username) && StringUtils.isNotBlank(password)) {
|
||||||
|
|
||||||
|
return GraphDatabase.driver(uri, AuthTokens.basic(username, password), configBuilder.build());
|
||||||
|
|
||||||
|
} else if (StringUtils.isNotBlank(bearerToken)) {
|
||||||
|
|
||||||
|
return GraphDatabase.driver(uri, AuthTokens.bearer(bearerToken), configBuilder.build());
|
||||||
|
|
||||||
|
} else if (StringUtils.isNotBlank(kerberosTicket)) {
|
||||||
|
|
||||||
|
return GraphDatabase.driver(uri, AuthTokens.kerberos(kerberosTicket), configBuilder.build());
|
||||||
|
|
||||||
|
}
|
||||||
|
|
||||||
|
throw DataXException.asDataXException(Neo4jErrorCode.CONFIG_INVALID, "Invalid Auth config.");
|
||||||
|
}
|
||||||
|
|
||||||
|
private static String checkUriConfig(Configuration config) {
|
||||||
|
String uri = config.getString(URI.getKey());
|
||||||
|
if (null == uri || uri.length() == 0) {
|
||||||
|
throw DataXException.asDataXException(Neo4jErrorCode.CONFIG_INVALID, "Invalid uri configuration");
|
||||||
|
}
|
||||||
|
return uri;
|
||||||
|
}
|
||||||
|
|
||||||
|
public void destroy() {
|
||||||
|
tryFlushBuffer();
|
||||||
|
if (driver != null) {
|
||||||
|
driver.close();
|
||||||
|
}
|
||||||
|
if (session != null) {
|
||||||
|
session.close();
|
||||||
|
}
|
||||||
|
DateAdapter.destroy();
|
||||||
|
}
|
||||||
|
|
||||||
|
private void tryFlushBuffer() {
|
||||||
|
if (!writerBuffer.isEmpty()) {
|
||||||
|
doWrite(writerBuffer);
|
||||||
|
writerBuffer.clear();
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
private void tryBatchWrite() {
|
||||||
|
if (!writerBuffer.isEmpty() && writerBuffer.size() >= writeConfig.batchSize) {
|
||||||
|
doWrite(writerBuffer);
|
||||||
|
writerBuffer.clear();
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
private void doWrite(List<MapValue> values) {
|
||||||
|
Value batchValues = Values.parameters(this.writeConfig.batchVariableName, values);
|
||||||
|
Query query = new Query(this.writeConfig.cypher, batchValues);
|
||||||
|
// LOGGER.debug("query:{}", query.text());
|
||||||
|
// LOGGER.debug("batch:{}", toUnwindStr(values));
|
||||||
|
try {
|
||||||
|
RetryUtil.executeWithRetry(() -> {
|
||||||
|
session.writeTransaction(tx -> tx.run(query));
|
||||||
|
return null;
|
||||||
|
}, this.retryConfig.retryTimes, retryConfig.retrySleepMills, true,
|
||||||
|
Collections.singletonList(Neo4jException.class));
|
||||||
|
} catch (Exception e) {
|
||||||
|
LOGGER.error("an exception occurred while writing to the database,message:{}", e.getMessage());
|
||||||
|
throw DataXException.asDataXException(DATABASE_ERROR, e.getMessage());
|
||||||
|
}
|
||||||
|
|
||||||
|
|
||||||
|
}
|
||||||
|
|
||||||
|
private String toUnwindStr(List<MapValue> values) {
|
||||||
|
StringJoiner joiner = new StringJoiner(",");
|
||||||
|
for (MapValue value : values) {
|
||||||
|
joiner.add(value.toString());
|
||||||
|
}
|
||||||
|
return "[" + joiner + "]";
|
||||||
|
}
|
||||||
|
|
||||||
|
public void tryWrite(Record record) {
|
||||||
|
MapValue neo4jValue = checkAndConvert(record);
|
||||||
|
writerBuffer.add(neo4jValue);
|
||||||
|
tryBatchWrite();
|
||||||
|
}
|
||||||
|
|
||||||
|
private MapValue checkAndConvert(Record record) {
|
||||||
|
int sourceColNum = record.getColumnNumber();
|
||||||
|
List<Neo4jProperty> neo4jProperties = writeConfig.neo4jProperties;
|
||||||
|
if (neo4jProperties == null || neo4jProperties.size() != sourceColNum) {
|
||||||
|
throw new DataXException(Neo4jErrorCode.CONFIG_INVALID, "the read and write columns do not match!");
|
||||||
|
}
|
||||||
|
Map<String, Value> data = new HashMap<>(sourceColNum * 4 / 3);
|
||||||
|
for (int i = 0; i < sourceColNum; i++) {
|
||||||
|
Column column = record.getColumn(i);
|
||||||
|
Neo4jProperty neo4jProperty = neo4jProperties.get(i);
|
||||||
|
try {
|
||||||
|
|
||||||
|
Value value = ValueAdapter.column2Value(column, neo4jProperty);
|
||||||
|
data.put(neo4jProperty.getName(), value);
|
||||||
|
} catch (Exception e) {
|
||||||
|
LOGGER.info("dirty record：{},message :{}", column, e.getMessage());
|
||||||
|
this.taskPluginCollector.collectDirtyRecord(record, e.getMessage());
|
||||||
|
}
|
||||||
|
}
|
||||||
|
return new MapValue(data);
|
||||||
|
}
|
||||||
|
|
||||||
|
public List<Neo4jProperty> getNeo4jFields() {
|
||||||
|
return this.writeConfig.neo4jProperties;
|
||||||
|
}
|
||||||
|
|
||||||
|
|
||||||
|
static class RetryConfig {
|
||||||
|
int retryTimes;
|
||||||
|
long retrySleepMills;
|
||||||
|
|
||||||
|
RetryConfig(int retryTimes, long retrySleepMills) {
|
||||||
|
this.retryTimes = retryTimes;
|
||||||
|
this.retrySleepMills = retrySleepMills;
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
static class WriteConfig {
|
||||||
|
String cypher;
|
||||||
|
|
||||||
|
String database;
|
||||||
|
|
||||||
|
String batchVariableName;
|
||||||
|
|
||||||
|
List<Neo4jProperty> neo4jProperties;
|
||||||
|
|
||||||
|
int batchSize;
|
||||||
|
|
||||||
|
public WriteConfig(String cypher,
|
||||||
|
String database,
|
||||||
|
String batchVariableName,
|
||||||
|
List<Neo4jProperty> neo4jProperties,
|
||||||
|
int batchSize) {
|
||||||
|
this.cypher = cypher;
|
||||||
|
this.database = database;
|
||||||
|
this.batchVariableName = batchVariableName;
|
||||||
|
this.neo4jProperties = neo4jProperties;
|
||||||
|
this.batchSize = batchSize;
|
||||||
|
}
|
||||||
|
|
||||||
|
|
||||||
|
}
|
||||||
|
}
|
||||||
@ -0,0 +1,64 @@
|
|||||||
|
package com.alibaba.datax.plugin.writer.neo4jwriter;
|
||||||
|
|
||||||
|
import com.alibaba.datax.common.plugin.RecordReceiver;
|
||||||
|
import com.alibaba.datax.common.spi.Writer;
|
||||||
|
import com.alibaba.datax.common.util.Configuration;
|
||||||
|
import com.alibaba.datax.common.element.Record;
|
||||||
|
import org.slf4j.Logger;
|
||||||
|
import org.slf4j.LoggerFactory;
|
||||||
|
|
||||||
|
import java.util.ArrayList;
|
||||||
|
import java.util.List;
|
||||||
|
|
||||||
|
public class Neo4jWriter extends Writer {
|
||||||
|
public static class Job extends Writer.Job {
|
||||||
|
private static final Logger LOGGER = LoggerFactory.getLogger(Job.class);
|
||||||
|
|
||||||
|
private Configuration jobConf = null;
|
||||||
|
@Override
|
||||||
|
public void init() {
|
||||||
|
LOGGER.info("Neo4jWriter Job init success");
|
||||||
|
this.jobConf = getPluginJobConf();
|
||||||
|
}
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public void destroy() {
|
||||||
|
LOGGER.info("Neo4jWriter Job destroyed");
|
||||||
|
}
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public List<Configuration> split(int mandatoryNumber) {
|
||||||
|
List<Configuration> configurations = new ArrayList<Configuration>(mandatoryNumber);
|
||||||
|
for (int i = 0; i < mandatoryNumber; i++) {
|
||||||
|
configurations.add(this.jobConf.clone());
|
||||||
|
}
|
||||||
|
return configurations;
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
public static class Task extends Writer.Task {
|
||||||
|
private static final Logger TASK_LOGGER = LoggerFactory.getLogger(Task.class);
|
||||||
|
private Neo4jClient neo4jClient;
|
||||||
|
@Override
|
||||||
|
public void init() {
|
||||||
|
Configuration taskConf = super.getPluginJobConf();
|
||||||
|
this.neo4jClient = Neo4jClient.build(taskConf,getTaskPluginCollector());
|
||||||
|
this.neo4jClient.init();
|
||||||
|
TASK_LOGGER.info("neo4j writer task init success.");
|
||||||
|
}
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public void destroy() {
|
||||||
|
this.neo4jClient.destroy();
|
||||||
|
TASK_LOGGER.info("neo4j writer task destroyed.");
|
||||||
|
}
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public void startWrite(RecordReceiver receiver) {
|
||||||
|
Record record;
|
||||||
|
while ((record = receiver.getFromReader()) != null){
|
||||||
|
this.neo4jClient.tryWrite(record);
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
@ -0,0 +1,70 @@
|
|||||||
|
package com.alibaba.datax.plugin.writer.neo4jwriter.adapter;
|
||||||
|
|
||||||
|
|
||||||
|
import com.alibaba.datax.plugin.writer.neo4jwriter.config.Neo4jProperty;
|
||||||
|
import org.testcontainers.shaded.com.google.common.base.Supplier;
|
||||||
|
|
||||||
|
import java.time.LocalDate;
|
||||||
|
import java.time.LocalDateTime;
|
||||||
|
import java.time.LocalTime;
|
||||||
|
import java.time.format.DateTimeFormatter;
|
||||||
|
|
||||||
|
/**
|
||||||
|
* @author fuyouj
|
||||||
|
*/
|
||||||
|
public class DateAdapter {
|
||||||
|
private static final ThreadLocal<DateTimeFormatter> LOCAL_DATE_FORMATTER_MAP = new ThreadLocal<>();
|
||||||
|
private static final ThreadLocal<DateTimeFormatter> LOCAL_TIME_FORMATTER_MAP = new ThreadLocal<>();
|
||||||
|
private static final ThreadLocal<DateTimeFormatter> LOCAL_DATE_TIME_FORMATTER_MAP = new ThreadLocal<>();
|
||||||
|
private static final String DEFAULT_LOCAL_DATE_FORMATTER = "yyyy-MM-dd";
|
||||||
|
private static final String DEFAULT_LOCAL_TIME_FORMATTER = "HH:mm:ss";
|
||||||
|
private static final String DEFAULT_LOCAL_DATE_TIME_FORMATTER = "yyyy-MM-dd HH:mm:ss";
|
||||||
|
|
||||||
|
|
||||||
|
public static LocalDate localDate(String text, Neo4jProperty neo4jProperty) {
|
||||||
|
if (LOCAL_DATE_FORMATTER_MAP.get() != null) {
|
||||||
|
return LocalDate.parse(text, LOCAL_DATE_FORMATTER_MAP.get());
|
||||||
|
}
|
||||||
|
|
||||||
|
String format = getOrDefault(neo4jProperty::getDateFormat, DEFAULT_LOCAL_DATE_FORMATTER);
|
||||||
|
DateTimeFormatter dateTimeFormatter = DateTimeFormatter.ofPattern(format);
|
||||||
|
LOCAL_DATE_FORMATTER_MAP.set(dateTimeFormatter);
|
||||||
|
return LocalDate.parse(text, dateTimeFormatter);
|
||||||
|
}
|
||||||
|
|
||||||
|
public static String getOrDefault(Supplier<String> dateFormat, String defaultFormat) {
|
||||||
|
String format = dateFormat.get();
|
||||||
|
if (null == format || "".equals(format)) {
|
||||||
|
return defaultFormat;
|
||||||
|
} else {
|
||||||
|
return format;
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
public static void destroy() {
|
||||||
|
LOCAL_DATE_FORMATTER_MAP.remove();
|
||||||
|
LOCAL_TIME_FORMATTER_MAP.remove();
|
||||||
|
LOCAL_DATE_TIME_FORMATTER_MAP.remove();
|
||||||
|
}
|
||||||
|
|
||||||
|
public static LocalTime localTime(String text, Neo4jProperty neo4JProperty) {
|
||||||
|
if (LOCAL_TIME_FORMATTER_MAP.get() != null) {
|
||||||
|
return LocalTime.parse(text, LOCAL_TIME_FORMATTER_MAP.get());
|
||||||
|
}
|
||||||
|
|
||||||
|
String format = getOrDefault(neo4JProperty::getDateFormat, DEFAULT_LOCAL_TIME_FORMATTER);
|
||||||
|
DateTimeFormatter dateTimeFormatter = DateTimeFormatter.ofPattern(format);
|
||||||
|
LOCAL_TIME_FORMATTER_MAP.set(dateTimeFormatter);
|
||||||
|
return LocalTime.parse(text, dateTimeFormatter);
|
||||||
|
}
|
||||||
|
|
||||||
|
public static LocalDateTime localDateTime(String text, Neo4jProperty neo4JProperty) {
|
||||||
|
if (LOCAL_DATE_TIME_FORMATTER_MAP.get() != null){
|
||||||
|
return LocalDateTime.parse(text,LOCAL_DATE_TIME_FORMATTER_MAP.get());
|
||||||
|
}
|
||||||
|
String format = getOrDefault(neo4JProperty::getDateFormat, DEFAULT_LOCAL_DATE_TIME_FORMATTER);
|
||||||
|
DateTimeFormatter dateTimeFormatter = DateTimeFormatter.ofPattern(format);
|
||||||
|
LOCAL_DATE_TIME_FORMATTER_MAP.set(dateTimeFormatter);
|
||||||
|
return LocalDateTime.parse(text, dateTimeFormatter);
|
||||||
|
}
|
||||||
|
}
|
||||||
@ -0,0 +1,95 @@
|
|||||||
|
package com.alibaba.datax.plugin.writer.neo4jwriter.adapter;
|
||||||
|
|
||||||
|
|
||||||
|
import com.alibaba.datax.common.element.Column;
|
||||||
|
import com.alibaba.datax.plugin.writer.neo4jwriter.config.Neo4jProperty;
|
||||||
|
import com.alibaba.datax.plugin.writer.neo4jwriter.element.PropertyType;
|
||||||
|
import com.alibaba.fastjson2.JSON;
|
||||||
|
import org.neo4j.driver.Value;
|
||||||
|
import org.neo4j.driver.Values;
|
||||||
|
import org.neo4j.driver.internal.value.NullValue;
|
||||||
|
|
||||||
|
import java.util.ArrayList;
|
||||||
|
import java.util.Collections;
|
||||||
|
import java.util.List;
|
||||||
|
import java.util.Map;
|
||||||
|
import java.util.function.Function;
|
||||||
|
|
||||||
|
/**
|
||||||
|
* @author fuyouj
|
||||||
|
*/
|
||||||
|
public class ValueAdapter {
|
||||||
|
|
||||||
|
|
||||||
|
public static Value column2Value(final Column column, final Neo4jProperty neo4JProperty) {
|
||||||
|
|
||||||
|
String typeStr = neo4JProperty.getType();
|
||||||
|
PropertyType type = PropertyType.fromStrIgnoreCase(typeStr);
|
||||||
|
if (column.asString() == null) {
|
||||||
|
return NullValue.NULL;
|
||||||
|
}
|
||||||
|
|
||||||
|
switch (type) {
|
||||||
|
case NULL:
|
||||||
|
return NullValue.NULL;
|
||||||
|
case MAP:
|
||||||
|
return Values.value(JSON.parseObject(column.asString(), Map.class));
|
||||||
|
case BOOLEAN:
|
||||||
|
return Values.value(column.asBoolean());
|
||||||
|
case STRING:
|
||||||
|
return Values.value(column.asString());
|
||||||
|
case INTEGER:
|
||||||
|
case LONG:
|
||||||
|
return Values.value(column.asLong());
|
||||||
|
case SHORT:
|
||||||
|
return Values.value(Short.valueOf(column.asString()));
|
||||||
|
case FLOAT:
|
||||||
|
case DOUBLE:
|
||||||
|
return Values.value(column.asDouble());
|
||||||
|
case BYTE_ARRAY:
|
||||||
|
return Values.value(parseArrayType(neo4JProperty, column.asString(), Byte::valueOf));
|
||||||
|
case CHAR_ARRAY:
|
||||||
|
return Values.value(parseArrayType(neo4JProperty, column.asString(), (s) -> s.charAt(0)));
|
||||||
|
case BOOLEAN_ARRAY:
|
||||||
|
return Values.value(parseArrayType(neo4JProperty, column.asString(), Boolean::valueOf));
|
||||||
|
case STRING_ARRAY:
|
||||||
|
case Object_ARRAY:
|
||||||
|
case LIST:
|
||||||
|
return Values.value(parseArrayType(neo4JProperty, column.asString(), Function.identity()));
|
||||||
|
case LONG_ARRAY:
|
||||||
|
return Values.value(parseArrayType(neo4JProperty, column.asString(), Long::valueOf));
|
||||||
|
case INT_ARRAY:
|
||||||
|
return Values.value(parseArrayType(neo4JProperty, column.asString(), Integer::valueOf));
|
||||||
|
case SHORT_ARRAY:
|
||||||
|
return Values.value(parseArrayType(neo4JProperty, column.asString(), Short::valueOf));
|
||||||
|
case DOUBLE_ARRAY:
|
||||||
|
case FLOAT_ARRAY:
|
||||||
|
return Values.value(parseArrayType(neo4JProperty, column.asString(), Double::valueOf));
|
||||||
|
case LOCAL_DATE:
|
||||||
|
return Values.value(DateAdapter.localDate(column.asString(), neo4JProperty));
|
||||||
|
case LOCAL_TIME:
|
||||||
|
return Values.value(DateAdapter.localTime(column.asString(), neo4JProperty));
|
||||||
|
case LOCAL_DATE_TIME:
|
||||||
|
return Values.value(DateAdapter.localDateTime(column.asString(), neo4JProperty));
|
||||||
|
default:
|
||||||
|
return Values.value(column.getRawData());
|
||||||
|
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
|
||||||
|
private static <R> List<R> parseArrayType(final Neo4jProperty neo4JProperty,
|
||||||
|
final String strValue,
|
||||||
|
final Function<String, R> convertFunc) {
|
||||||
|
if (null == strValue || "".equals(strValue)) {
|
||||||
|
return Collections.emptyList();
|
||||||
|
}
|
||||||
|
String split = neo4JProperty.getSplitOrDefault();
|
||||||
|
String[] strArr = strValue.split(split);
|
||||||
|
List<R> ans = new ArrayList<>();
|
||||||
|
for (String s : strArr) {
|
||||||
|
ans.add(convertFunc.apply(s));
|
||||||
|
}
|
||||||
|
return ans;
|
||||||
|
}
|
||||||
|
}
|
||||||
@ -0,0 +1,116 @@
|
|||||||
|
package com.alibaba.datax.plugin.writer.neo4jwriter.config;
|
||||||
|
|
||||||
|
|
||||||
|
import java.util.List;
|
||||||
|
|
||||||
|
/**
|
||||||
|
* @author fuyouj
|
||||||
|
*/
|
||||||
|
public final class ConfigConstants {
|
||||||
|
|
||||||
|
public static final Long DEFAULT_MAX_TRANSACTION_RETRY_SECONDS = 30L;
|
||||||
|
|
||||||
|
public static final Long DEFAULT_MAX_CONNECTION_SECONDS = 30L;
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
public static final Option<Integer> RETRY_TIMES =
|
||||||
|
Option.<Integer>builder()
|
||||||
|
.key("retryTimes")
|
||||||
|
.defaultValue(3)
|
||||||
|
.desc("The number of overwrites when an error occurs")
|
||||||
|
.build();
|
||||||
|
|
||||||
|
public static final Option<Long> RETRY_SLEEP_MILLS =
|
||||||
|
Option.<Long>builder()
|
||||||
|
.key("retrySleepMills")
|
||||||
|
.defaultValue(3000L)
|
||||||
|
.build();
|
||||||
|
|
||||||
|
/**
|
||||||
|
* cluster mode please reference
|
||||||
|
* <a href="https://neo4j.com/docs/java-manual/current/client-applications/">how to connect cluster mode</a>
|
||||||
|
*/
|
||||||
|
public static final Option<String> URI =
|
||||||
|
Option.<String>builder()
|
||||||
|
.key("uri")
|
||||||
|
.noDefaultValue()
|
||||||
|
.desc("uir of neo4j database")
|
||||||
|
.build();
|
||||||
|
|
||||||
|
public static final Option<String> USERNAME =
|
||||||
|
Option.<String>builder()
|
||||||
|
.key("username")
|
||||||
|
.noDefaultValue()
|
||||||
|
.desc("username for accessing the neo4j database")
|
||||||
|
.build();
|
||||||
|
|
||||||
|
public static final Option<String> PASSWORD =
|
||||||
|
Option.<String>builder()
|
||||||
|
.key("password")
|
||||||
|
.noDefaultValue()
|
||||||
|
.desc("password for accessing the neo4j database")
|
||||||
|
.build();
|
||||||
|
|
||||||
|
public static final Option<String> BEARER_TOKEN =
|
||||||
|
Option.<String>builder()
|
||||||
|
.key("bearerToken")
|
||||||
|
.noDefaultValue()
|
||||||
|
.desc("base64 encoded bearer token of the Neo4j. for Auth.")
|
||||||
|
.build();
|
||||||
|
|
||||||
|
public static final Option<String> KERBEROS_TICKET =
|
||||||
|
Option.<String>builder()
|
||||||
|
.key("kerberosTicket")
|
||||||
|
.noDefaultValue()
|
||||||
|
.desc("base64 encoded kerberos ticket of the Neo4j. for Auth.")
|
||||||
|
.build();
|
||||||
|
|
||||||
|
public static final Option<String> DATABASE =
|
||||||
|
Option.<String>builder()
|
||||||
|
.key("database")
|
||||||
|
.noDefaultValue()
|
||||||
|
.desc("database name.")
|
||||||
|
.build();
|
||||||
|
|
||||||
|
public static final Option<String> CYPHER =
|
||||||
|
Option.<String>builder()
|
||||||
|
.key("cypher")
|
||||||
|
.noDefaultValue()
|
||||||
|
.desc("cypher query.")
|
||||||
|
.build();
|
||||||
|
|
||||||
|
public static final Option<Long> MAX_TRANSACTION_RETRY_TIME =
|
||||||
|
Option.<Long>builder()
|
||||||
|
.key("maxTransactionRetryTimeSeconds")
|
||||||
|
.defaultValue(DEFAULT_MAX_TRANSACTION_RETRY_SECONDS)
|
||||||
|
.desc("maximum transaction retry time(seconds). transaction fail if exceeded.")
|
||||||
|
.build();
|
||||||
|
public static final Option<Long> MAX_CONNECTION_TIMEOUT_SECONDS =
|
||||||
|
Option.<Long>builder()
|
||||||
|
.key("maxConnectionTimeoutSeconds")
|
||||||
|
.defaultValue(DEFAULT_MAX_CONNECTION_SECONDS)
|
||||||
|
.desc("The maximum amount of time to wait for a TCP connection to be established (seconds).")
|
||||||
|
.build();
|
||||||
|
|
||||||
|
public static final Option<String> BATCH_DATA_VARIABLE_NAME =
|
||||||
|
Option.<String>builder()
|
||||||
|
.key("batchDataVariableName")
|
||||||
|
.defaultValue("batch")
|
||||||
|
.desc("in a cypher statement, a variable name that represents a batch of data")
|
||||||
|
.build();
|
||||||
|
|
||||||
|
public static final Option<List<Neo4jProperty>> NEO4J_PROPERTIES =
|
||||||
|
Option.<List<Neo4jProperty>>builder()
|
||||||
|
.key("properties")
|
||||||
|
.noDefaultValue()
|
||||||
|
.desc("neo4j node or relation`s props")
|
||||||
|
.build();
|
||||||
|
|
||||||
|
public static final Option<Integer> BATCH_SIZE =
|
||||||
|
Option.<Integer>builder().
|
||||||
|
key("batchSize")
|
||||||
|
.defaultValue(1000)
|
||||||
|
.desc("max batch size")
|
||||||
|
.build();
|
||||||
|
}
|
||||||
@ -0,0 +1,82 @@
|
|||||||
|
package com.alibaba.datax.plugin.writer.neo4jwriter.config;
|
||||||
|
|
||||||
|
/**
|
||||||
|
* 由于dataX并不能传输数据的元数据，所以只能在writer端定义每列数据的名字
|
||||||
|
* datax does not support data metadata,
|
||||||
|
* only the name of each column of data can be defined on neo4j writer
|
||||||
|
*
|
||||||
|
* @author fuyouj
|
||||||
|
*/
|
||||||
|
public class Neo4jProperty {
|
||||||
|
public static final String DEFAULT_SPLIT = ",";
|
||||||
|
|
||||||
|
/**
|
||||||
|
* name of neo4j field
|
||||||
|
*/
|
||||||
|
private String name;
|
||||||
|
|
||||||
|
/**
|
||||||
|
* neo4j type
|
||||||
|
* reference by org.neo4j.driver.Values
|
||||||
|
*/
|
||||||
|
private String type;
|
||||||
|
|
||||||
|
/**
|
||||||
|
* for date
|
||||||
|
*/
|
||||||
|
private String dateFormat;
|
||||||
|
|
||||||
|
/**
|
||||||
|
* for array type
|
||||||
|
*/
|
||||||
|
private String split;
|
||||||
|
|
||||||
|
public Neo4jProperty() {
|
||||||
|
}
|
||||||
|
|
||||||
|
public Neo4jProperty(String name, String type, String format, String split) {
|
||||||
|
this.name = name;
|
||||||
|
this.type = type;
|
||||||
|
this.dateFormat = format;
|
||||||
|
this.split = split;
|
||||||
|
}
|
||||||
|
|
||||||
|
public String getName() {
|
||||||
|
return name;
|
||||||
|
}
|
||||||
|
|
||||||
|
public void setName(String name) {
|
||||||
|
this.name = name;
|
||||||
|
}
|
||||||
|
|
||||||
|
public String getType() {
|
||||||
|
return type;
|
||||||
|
}
|
||||||
|
|
||||||
|
public void setType(String type) {
|
||||||
|
this.type = type;
|
||||||
|
}
|
||||||
|
|
||||||
|
public String getDateFormat() {
|
||||||
|
return dateFormat;
|
||||||
|
}
|
||||||
|
|
||||||
|
public void setDateFormat(String dateFormat) {
|
||||||
|
this.dateFormat = dateFormat;
|
||||||
|
}
|
||||||
|
|
||||||
|
public String getSplit() {
|
||||||
|
return getSplitOrDefault();
|
||||||
|
}
|
||||||
|
|
||||||
|
public String getSplitOrDefault() {
|
||||||
|
if (split == null || "".equals(split)) {
|
||||||
|
return DEFAULT_SPLIT;
|
||||||
|
}
|
||||||
|
return split;
|
||||||
|
}
|
||||||
|
|
||||||
|
public void setSplit(String split) {
|
||||||
|
this.split = split;
|
||||||
|
}
|
||||||
|
}
|
||||||
@ -0,0 +1,65 @@
|
|||||||
|
package com.alibaba.datax.plugin.writer.neo4jwriter.config;
|
||||||
|
|
||||||
|
|
||||||
|
public class Option<T> {
|
||||||
|
|
||||||
|
public static class Builder<T> {
|
||||||
|
private String key;
|
||||||
|
private String desc;
|
||||||
|
|
||||||
|
private T defaultValue;
|
||||||
|
|
||||||
|
public Builder<T> key(String key) {
|
||||||
|
this.key = key;
|
||||||
|
return this;
|
||||||
|
}
|
||||||
|
|
||||||
|
public Builder<T> desc(String desc) {
|
||||||
|
this.desc = desc;
|
||||||
|
return this;
|
||||||
|
}
|
||||||
|
|
||||||
|
public Builder<T> defaultValue(T defaultValue) {
|
||||||
|
this.defaultValue = defaultValue;
|
||||||
|
return this;
|
||||||
|
}
|
||||||
|
|
||||||
|
public Builder<T> noDefaultValue() {
|
||||||
|
return this;
|
||||||
|
}
|
||||||
|
|
||||||
|
public Option<T> build() {
|
||||||
|
return new Option<>(this.key, this.desc, this.defaultValue);
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
private final String key;
|
||||||
|
private final String desc;
|
||||||
|
|
||||||
|
private final T defaultValue;
|
||||||
|
|
||||||
|
public Option(String key, String desc, T defaultValue) {
|
||||||
|
this.key = key;
|
||||||
|
this.desc = desc;
|
||||||
|
this.defaultValue = defaultValue;
|
||||||
|
}
|
||||||
|
|
||||||
|
public static <T> Builder<T> builder(){
|
||||||
|
return new Builder<>();
|
||||||
|
}
|
||||||
|
|
||||||
|
public String getKey() {
|
||||||
|
return key;
|
||||||
|
}
|
||||||
|
|
||||||
|
public String getDesc() {
|
||||||
|
return desc;
|
||||||
|
}
|
||||||
|
|
||||||
|
public T getDefaultValue() {
|
||||||
|
if (defaultValue == null){
|
||||||
|
throw new IllegalStateException(key + ":defaultValue is null");
|
||||||
|
}
|
||||||
|
return defaultValue;
|
||||||
|
}
|
||||||
|
}
|
||||||
@ -0,0 +1,40 @@
|
|||||||
|
package com.alibaba.datax.plugin.writer.neo4jwriter.element;
|
||||||
|
|
||||||
|
import java.util.Arrays;
|
||||||
|
|
||||||
|
/**
|
||||||
|
* @see org.neo4j.driver.Values
|
||||||
|
* @author fuyouj
|
||||||
|
*/
|
||||||
|
public enum PropertyType {
|
||||||
|
NULL,
|
||||||
|
BOOLEAN,
|
||||||
|
STRING,
|
||||||
|
LONG,
|
||||||
|
SHORT,
|
||||||
|
INTEGER,
|
||||||
|
DOUBLE,
|
||||||
|
FLOAT,
|
||||||
|
LOCAL_DATE,
|
||||||
|
LOCAL_TIME,
|
||||||
|
LOCAL_DATE_TIME,
|
||||||
|
LIST,
|
||||||
|
MAP,
|
||||||
|
CHAR_ARRAY,
|
||||||
|
BYTE_ARRAY,
|
||||||
|
BOOLEAN_ARRAY,
|
||||||
|
STRING_ARRAY,
|
||||||
|
LONG_ARRAY,
|
||||||
|
INT_ARRAY,
|
||||||
|
SHORT_ARRAY,
|
||||||
|
DOUBLE_ARRAY,
|
||||||
|
FLOAT_ARRAY,
|
||||||
|
Object_ARRAY;
|
||||||
|
|
||||||
|
public static PropertyType fromStrIgnoreCase(String typeStr) {
|
||||||
|
return Arrays.stream(PropertyType.values())
|
||||||
|
.filter(e -> e.name().equalsIgnoreCase(typeStr))
|
||||||
|
.findFirst()
|
||||||
|
.orElse(PropertyType.STRING);
|
||||||
|
}
|
||||||
|
}
|
||||||
@ -0,0 +1,37 @@
|
|||||||
|
package com.alibaba.datax.plugin.writer.neo4jwriter.exception;
|
||||||
|
|
||||||
|
import com.alibaba.datax.common.spi.ErrorCode;
|
||||||
|
|
||||||
|
|
||||||
|
public enum Neo4jErrorCode implements ErrorCode {
|
||||||
|
|
||||||
|
/**
|
||||||
|
* Invalid configuration
|
||||||
|
* 配置校验异常
|
||||||
|
*/
|
||||||
|
CONFIG_INVALID("NEO4J_ERROR_01","invalid configuration"),
|
||||||
|
/**
|
||||||
|
* database error
|
||||||
|
* 在执行写入到数据库时抛出的异常，可能是权限异常，也可能是连接超时，或者是配置到了从节点。
|
||||||
|
* 如果是更新操作，还会有死锁异常。具体原因根据报错信息确定，但是这与dataX无关。
|
||||||
|
*/
|
||||||
|
DATABASE_ERROR("NEO4J_ERROR_02","database error");
|
||||||
|
|
||||||
|
private final String code;
|
||||||
|
private final String description;
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public String getCode() {
|
||||||
|
return code;
|
||||||
|
}
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public String getDescription() {
|
||||||
|
return description;
|
||||||
|
}
|
||||||
|
|
||||||
|
Neo4jErrorCode(String code, String description) {
|
||||||
|
this.code = code;
|
||||||
|
this.description = description;
|
||||||
|
}
|
||||||
|
}
|
||||||
6
neo4jwriter/src/main/resources/plugin.json
Normal file
6
neo4jwriter/src/main/resources/plugin.json
Normal file
@ -0,0 +1,6 @@
|
|||||||
|
{
|
||||||
|
"name": "neo4jWriter",
|
||||||
|
"class": "com.alibaba.datax.plugin.writer.neo4jwriter.Neo4jWriter",
|
||||||
|
"description": "dataX neo4j 写插件",
|
||||||
|
"developer": "付有杰"
|
||||||
|
}
|
||||||
42
neo4jwriter/src/main/resources/plugin_job_template.json
Normal file
42
neo4jwriter/src/main/resources/plugin_job_template.json
Normal file
@ -0,0 +1,42 @@
|
|||||||
|
{
|
||||||
|
"uri": "neo4j://localhost:7687",
|
||||||
|
"username": "neo4j",
|
||||||
|
"password": "Test@12343",
|
||||||
|
"database": "neo4j",
|
||||||
|
"cypher": "unwind $batch as row create(p:Person) set p.pbool = row.pbool,p.pstring = row.pstring,p.plong = row.plong,p.pshort = row.pshort,p.pdouble=row.pdouble,p.pstringarr=row.pstringarr,p.plocaldate=row.plocaldate",
|
||||||
|
"batchDataVariableName": "batch",
|
||||||
|
"batchSize": "33",
|
||||||
|
"properties": [
|
||||||
|
{
|
||||||
|
"name": "pbool",
|
||||||
|
//type 忽略大小写
|
||||||
|
"type": "BOOLEAN"
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"name": "pstring",
|
||||||
|
"type": "STRING"
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"name": "plong",
|
||||||
|
"type": "LONG"
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"name": "pshort",
|
||||||
|
"type": "SHORT"
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"name": "pdouble",
|
||||||
|
"type": "DOUBLE"
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"name": "pstringarr",
|
||||||
|
"type": "STRING_ARRAY",
|
||||||
|
"split": ","
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"name": "plocaldate",
|
||||||
|
"type": "LOCAL_DATE",
|
||||||
|
"dateFormat": "yyyy-MM-dd"
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
@ -0,0 +1,257 @@
|
|||||||
|
package com.alibaba.datax.plugin.writer;
|
||||||
|
|
||||||
|
|
||||||
|
import com.alibaba.datax.common.element.Record;
|
||||||
|
import com.alibaba.datax.common.element.StringColumn;
|
||||||
|
import com.alibaba.datax.common.util.Configuration;
|
||||||
|
import com.alibaba.datax.plugin.writer.mock.MockRecord;
|
||||||
|
import com.alibaba.datax.plugin.writer.mock.MockUtil;
|
||||||
|
import com.alibaba.datax.plugin.writer.neo4jwriter.Neo4jClient;
|
||||||
|
import com.alibaba.datax.plugin.writer.neo4jwriter.config.Neo4jProperty;
|
||||||
|
import com.alibaba.datax.plugin.writer.neo4jwriter.element.PropertyType;
|
||||||
|
import org.junit.After;
|
||||||
|
import org.junit.Before;
|
||||||
|
import org.junit.Test;
|
||||||
|
import org.neo4j.driver.*;
|
||||||
|
import org.neo4j.driver.types.Node;
|
||||||
|
import org.slf4j.Logger;
|
||||||
|
import org.slf4j.LoggerFactory;
|
||||||
|
import org.testcontainers.containers.GenericContainer;
|
||||||
|
import org.testcontainers.containers.Network;
|
||||||
|
import org.testcontainers.containers.output.Slf4jLogConsumer;
|
||||||
|
import org.testcontainers.lifecycle.Startables;
|
||||||
|
import org.testcontainers.shaded.org.awaitility.Awaitility;
|
||||||
|
import org.testcontainers.utility.DockerImageName;
|
||||||
|
import org.testcontainers.utility.DockerLoggerFactory;
|
||||||
|
|
||||||
|
import java.io.File;
|
||||||
|
import java.net.URI;
|
||||||
|
import java.util.ArrayList;
|
||||||
|
import java.util.Arrays;
|
||||||
|
import java.util.List;
|
||||||
|
import java.util.concurrent.TimeUnit;
|
||||||
|
import java.util.stream.Stream;
|
||||||
|
|
||||||
|
import static org.junit.Assert.assertEquals;
|
||||||
|
import static org.junit.Assert.assertTrue;
|
||||||
|
|
||||||
|
|
||||||
|
public class Neo4jWriterTest {
|
||||||
|
|
||||||
|
private static final Logger LOGGER = LoggerFactory.getLogger(Neo4jWriterTest.class);
|
||||||
|
private static final int MOCK_NUM = 100;
|
||||||
|
private static final String CONTAINER_IMAGE = "neo4j:5.9.0";
|
||||||
|
|
||||||
|
private static final String CONTAINER_HOST = "neo4j-host";
|
||||||
|
private static final int HTTP_PORT = 7474;
|
||||||
|
private static final int BOLT_PORT = 7687;
|
||||||
|
private static final String CONTAINER_NEO4J_USERNAME = "neo4j";
|
||||||
|
private static final String CONTAINER_NEO4J_PASSWORD = "Test@12343";
|
||||||
|
private static final URI CONTAINER_URI = URI.create("neo4j://localhost:" + BOLT_PORT);
|
||||||
|
|
||||||
|
protected static final Network NETWORK = Network.newNetwork();
|
||||||
|
|
||||||
|
private GenericContainer<?> container;
|
||||||
|
private Driver neo4jDriver;
|
||||||
|
private Session neo4jSession;
|
||||||
|
|
||||||
|
@Before
|
||||||
|
public void init() {
|
||||||
|
DockerImageName imageName = DockerImageName.parse(CONTAINER_IMAGE);
|
||||||
|
container =
|
||||||
|
new GenericContainer<>(imageName)
|
||||||
|
.withNetwork(NETWORK)
|
||||||
|
.withNetworkAliases(CONTAINER_HOST)
|
||||||
|
.withExposedPorts(HTTP_PORT, BOLT_PORT)
|
||||||
|
.withEnv(
|
||||||
|
"NEO4J_AUTH",
|
||||||
|
CONTAINER_NEO4J_USERNAME + "/" + CONTAINER_NEO4J_PASSWORD)
|
||||||
|
.withEnv("apoc.export.file.enabled", "true")
|
||||||
|
.withEnv("apoc.import.file.enabled", "true")
|
||||||
|
.withEnv("apoc.import.file.use_neo4j_config", "true")
|
||||||
|
.withEnv("NEO4J_PLUGINS", "[\"apoc\"]")
|
||||||
|
.withLogConsumer(
|
||||||
|
new Slf4jLogConsumer(
|
||||||
|
DockerLoggerFactory.getLogger(CONTAINER_IMAGE)));
|
||||||
|
container.setPortBindings(
|
||||||
|
Arrays.asList(
|
||||||
|
String.format("%s:%s", HTTP_PORT, HTTP_PORT),
|
||||||
|
String.format("%s:%s", BOLT_PORT, BOLT_PORT)));
|
||||||
|
Startables.deepStart(Stream.of(container)).join();
|
||||||
|
LOGGER.info("container started");
|
||||||
|
Awaitility.given()
|
||||||
|
.ignoreExceptions()
|
||||||
|
.await()
|
||||||
|
.atMost(30, TimeUnit.SECONDS)
|
||||||
|
.untilAsserted(this::initConnection);
|
||||||
|
}
|
||||||
|
|
||||||
|
@Test
|
||||||
|
public void testCreateNodeAllTypeField() {
|
||||||
|
final Result checkExists = neo4jSession.run("MATCH (p:Person) RETURN p limit 1");
|
||||||
|
if (checkExists.hasNext()) {
|
||||||
|
neo4jSession.run("MATCH (p:Person) delete p");
|
||||||
|
}
|
||||||
|
|
||||||
|
Configuration configuration = Configuration.from(new File("src/test/resources/allTypeFieldNode.json"));
|
||||||
|
Neo4jClient neo4jClient = Neo4jClient.build(configuration, null);
|
||||||
|
|
||||||
|
neo4jClient.init();
|
||||||
|
for (int i = 0; i < MOCK_NUM; i++) {
|
||||||
|
neo4jClient.tryWrite(mockAllTypeFieldTestNode(neo4jClient.getNeo4jFields()));
|
||||||
|
}
|
||||||
|
neo4jClient.destroy();
|
||||||
|
|
||||||
|
|
||||||
|
Result result = neo4jSession.run("MATCH (p:Person) return p");
|
||||||
|
// nodes
|
||||||
|
assertTrue(result.hasNext());
|
||||||
|
int cnt = 0;
|
||||||
|
while (result.hasNext()) {
|
||||||
|
org.neo4j.driver.Record record = result.next();
|
||||||
|
record.get("p").get("pbool").asBoolean();
|
||||||
|
record.get("p").get("pstring").asString();
|
||||||
|
record.get("p").get("plong").asLong();
|
||||||
|
record.get("p").get("pshort").asInt();
|
||||||
|
record.get("p").get("pdouble").asDouble();
|
||||||
|
List list = (List) record.get("p").get("pstringarr").asObject();
|
||||||
|
record.get("p").get("plocaldate").asLocalDate();
|
||||||
|
cnt++;
|
||||||
|
|
||||||
|
}
|
||||||
|
assertEquals(cnt, MOCK_NUM);
|
||||||
|
}
|
||||||
|
|
||||||
|
|
||||||
|
/**
|
||||||
|
* 创建关系 必须先有节点
|
||||||
|
* 所以先创建节点再模拟关系
|
||||||
|
*/
|
||||||
|
@Test
|
||||||
|
public void testCreateRelation() {
|
||||||
|
final Result checkExists = neo4jSession.run("MATCH (p1:Person)-[r:LINK]->(p1:Person) return r limit 1");
|
||||||
|
if (checkExists.hasNext()) {
|
||||||
|
neo4jSession.run("MATCH (p1:Person)-[r:LINK]->(p1:Person) delete r,p1,p2");
|
||||||
|
}
|
||||||
|
|
||||||
|
String createNodeCql = "create (p:Person) set p.id = '%s'";
|
||||||
|
Configuration configuration = Configuration.from(new File("src/test/resources/relationship.json"));
|
||||||
|
|
||||||
|
Neo4jClient neo4jClient = Neo4jClient.build(configuration, null);
|
||||||
|
neo4jClient.init();
|
||||||
|
//创建节点为后续写关系做准备
|
||||||
|
//Create nodes to prepare for subsequent write relationships
|
||||||
|
for (int i = 0; i < MOCK_NUM; i++) {
|
||||||
|
neo4jSession.run(String.format(createNodeCql, i + "start"));
|
||||||
|
neo4jSession.run(String.format(createNodeCql, i + "end"));
|
||||||
|
Record record = new MockRecord();
|
||||||
|
record.addColumn(new StringColumn(i + "start"));
|
||||||
|
record.addColumn(new StringColumn(i + "end"));
|
||||||
|
neo4jClient.tryWrite(record);
|
||||||
|
|
||||||
|
}
|
||||||
|
neo4jClient.destroy();
|
||||||
|
|
||||||
|
Result result = neo4jSession.run("MATCH (start:Person)-[r:LINK]->(end:Person) return r,start,end");
|
||||||
|
// relationships
|
||||||
|
assertTrue(result.hasNext());
|
||||||
|
int cnt = 0;
|
||||||
|
while (result.hasNext()) {
|
||||||
|
org.neo4j.driver.Record record = result.next();
|
||||||
|
|
||||||
|
Node startNode = record.get("start").asNode();
|
||||||
|
assertTrue(startNode.hasLabel("Person"));
|
||||||
|
assertTrue(startNode.asMap().containsKey("id"));
|
||||||
|
|
||||||
|
Node endNode = record.get("end").asNode();
|
||||||
|
assertTrue(startNode.hasLabel("Person"));
|
||||||
|
assertTrue(endNode.asMap().containsKey("id"));
|
||||||
|
|
||||||
|
|
||||||
|
String name = record.get("r").type().name();
|
||||||
|
assertEquals("RELATIONSHIP", name);
|
||||||
|
cnt++;
|
||||||
|
}
|
||||||
|
assertEquals(cnt, MOCK_NUM);
|
||||||
|
}
|
||||||
|
|
||||||
|
/**
|
||||||
|
* neo4j中,Label和关系类型,想动态的写，需要借助于apoc函数
|
||||||
|
*/
|
||||||
|
@Test
|
||||||
|
public void testUseApocCreateDynamicLabel() {
|
||||||
|
List<String> dynamicLabel = new ArrayList<>();
|
||||||
|
for (int i = 0; i < MOCK_NUM; i++) {
|
||||||
|
dynamicLabel.add("Label" + i);
|
||||||
|
}
|
||||||
|
//删除原有数据
|
||||||
|
//remove test data if exist
|
||||||
|
//这种占位符的方式不支持批量动态写,当然可以使用union拼接，但是性能不好
|
||||||
|
String query = "match (p:%s) return p";
|
||||||
|
String delete = "match (p:%s) delete p";
|
||||||
|
for (String label : dynamicLabel) {
|
||||||
|
Result result = neo4jSession.run(String.format(query, label));
|
||||||
|
if (result.hasNext()) {
|
||||||
|
neo4jSession.run(String.format(delete, label));
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
Configuration configuration = Configuration.from(new File("src/test/resources/dynamicLabel.json"));
|
||||||
|
Neo4jClient neo4jClient = Neo4jClient.build(configuration, null);
|
||||||
|
|
||||||
|
neo4jClient.init();
|
||||||
|
for (int i = 0; i < dynamicLabel.size(); i++) {
|
||||||
|
Record record = new MockRecord();
|
||||||
|
record.addColumn(new StringColumn(dynamicLabel.get(i)));
|
||||||
|
record.addColumn(new StringColumn(String.valueOf(i)));
|
||||||
|
neo4jClient.tryWrite(record);
|
||||||
|
}
|
||||||
|
neo4jClient.destroy();
|
||||||
|
|
||||||
|
//校验脚本的批量写入是否正确
|
||||||
|
int cnt = 0;
|
||||||
|
for (int i = 0; i < dynamicLabel.size(); i++) {
|
||||||
|
String label = dynamicLabel.get(i);
|
||||||
|
Result result = neo4jSession.run(String.format(query, label));
|
||||||
|
while (result.hasNext()) {
|
||||||
|
org.neo4j.driver.Record record = result.next();
|
||||||
|
Node node = record.get("p").asNode();
|
||||||
|
assertTrue(node.hasLabel(label));
|
||||||
|
assertEquals(node.asMap().get("id"), i + "");
|
||||||
|
cnt++;
|
||||||
|
}
|
||||||
|
}
|
||||||
|
assertEquals(cnt, MOCK_NUM);
|
||||||
|
|
||||||
|
}
|
||||||
|
|
||||||
|
|
||||||
|
private Record mockAllTypeFieldTestNode(List<Neo4jProperty> neo4JProperties) {
|
||||||
|
Record mock = new MockRecord();
|
||||||
|
for (Neo4jProperty field : neo4JProperties) {
|
||||||
|
mock.addColumn(MockUtil.mockColumnByType(PropertyType.fromStrIgnoreCase(field.getType())));
|
||||||
|
}
|
||||||
|
return mock;
|
||||||
|
}
|
||||||
|
|
||||||
|
@After
|
||||||
|
public void destroy() {
|
||||||
|
if (neo4jSession != null) {
|
||||||
|
neo4jSession.close();
|
||||||
|
}
|
||||||
|
if (neo4jDriver != null) {
|
||||||
|
neo4jDriver.close();
|
||||||
|
}
|
||||||
|
if (container != null) {
|
||||||
|
container.close();
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
private void initConnection() {
|
||||||
|
neo4jDriver =
|
||||||
|
GraphDatabase.driver(
|
||||||
|
CONTAINER_URI,
|
||||||
|
AuthTokens.basic(CONTAINER_NEO4J_USERNAME, CONTAINER_NEO4J_PASSWORD));
|
||||||
|
neo4jSession = neo4jDriver.session(SessionConfig.forDatabase("neo4j"));
|
||||||
|
}
|
||||||
|
}
|
||||||
@ -0,0 +1,104 @@
|
|||||||
|
package com.alibaba.datax.plugin.writer.mock;
|
||||||
|
|
||||||
|
|
||||||
|
import com.alibaba.datax.common.element.Column;
|
||||||
|
import com.alibaba.datax.common.element.Record;
|
||||||
|
import com.alibaba.fastjson2.JSON;
|
||||||
|
|
||||||
|
import java.util.ArrayList;
|
||||||
|
import java.util.HashMap;
|
||||||
|
import java.util.List;
|
||||||
|
import java.util.Map;
|
||||||
|
|
||||||
|
public class MockRecord implements Record {
|
||||||
|
private static final int RECORD_AVERGAE_COLUMN_NUMBER = 16;
|
||||||
|
|
||||||
|
private List<Column> columns;
|
||||||
|
|
||||||
|
private int byteSize;
|
||||||
|
|
||||||
|
|
||||||
|
private Map<String, Object> meta;
|
||||||
|
|
||||||
|
public MockRecord() {
|
||||||
|
this.columns = new ArrayList<>(RECORD_AVERGAE_COLUMN_NUMBER);
|
||||||
|
}
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public void addColumn(Column column) {
|
||||||
|
columns.add(column);
|
||||||
|
incrByteSize(column);
|
||||||
|
}
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public Column getColumn(int i) {
|
||||||
|
if (i < 0 || i >= columns.size()) {
|
||||||
|
return null;
|
||||||
|
}
|
||||||
|
return columns.get(i);
|
||||||
|
}
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public void setColumn(int i, final Column column) {
|
||||||
|
if (i < 0) {
|
||||||
|
throw new IllegalArgumentException("不能给index小于0的column设置值");
|
||||||
|
}
|
||||||
|
|
||||||
|
if (i >= columns.size()) {
|
||||||
|
expandCapacity(i + 1);
|
||||||
|
}
|
||||||
|
|
||||||
|
decrByteSize(getColumn(i));
|
||||||
|
this.columns.set(i, column);
|
||||||
|
incrByteSize(getColumn(i));
|
||||||
|
}
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public String toString() {
|
||||||
|
Map<String, Object> json = new HashMap<String, Object>();
|
||||||
|
json.put("size", this.getColumnNumber());
|
||||||
|
json.put("data", this.columns);
|
||||||
|
return JSON.toJSONString(json);
|
||||||
|
}
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public int getColumnNumber() {
|
||||||
|
return this.columns.size();
|
||||||
|
}
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public int getByteSize() {
|
||||||
|
return byteSize;
|
||||||
|
}
|
||||||
|
|
||||||
|
public int getMemorySize() {
|
||||||
|
throw new UnsupportedOperationException();
|
||||||
|
}
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public void setMeta(Map<String, String> meta) {
|
||||||
|
|
||||||
|
}
|
||||||
|
|
||||||
|
@Override
|
||||||
|
public Map<String, String> getMeta() {
|
||||||
|
return null;
|
||||||
|
}
|
||||||
|
|
||||||
|
private void decrByteSize(final Column column) {
|
||||||
|
}
|
||||||
|
|
||||||
|
private void incrByteSize(final Column column) {
|
||||||
|
}
|
||||||
|
|
||||||
|
private void expandCapacity(int totalSize) {
|
||||||
|
if (totalSize <= 0) {
|
||||||
|
return;
|
||||||
|
}
|
||||||
|
|
||||||
|
int needToExpand = totalSize - columns.size();
|
||||||
|
while (needToExpand-- > 0) {
|
||||||
|
this.columns.add(null);
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
@ -0,0 +1,50 @@
|
|||||||
|
package com.alibaba.datax.plugin.writer.mock;
|
||||||
|
|
||||||
|
|
||||||
|
import com.alibaba.datax.common.element.*;
|
||||||
|
import com.alibaba.datax.plugin.writer.neo4jwriter.element.PropertyType;
|
||||||
|
import com.alibaba.fastjson2.JSON;
|
||||||
|
|
||||||
|
import java.time.LocalDate;
|
||||||
|
import java.time.format.DateTimeFormatter;
|
||||||
|
import java.util.HashMap;
|
||||||
|
import java.util.Map;
|
||||||
|
import java.util.Random;
|
||||||
|
|
||||||
|
public class MockUtil {
|
||||||
|
|
||||||
|
public static Column mockColumnByType(PropertyType type) {
|
||||||
|
Random random = new Random();
|
||||||
|
switch (type) {
|
||||||
|
case SHORT:
|
||||||
|
return new StringColumn("1");
|
||||||
|
case BOOLEAN:
|
||||||
|
return new BoolColumn(random.nextInt() % 2 == 0);
|
||||||
|
case INTEGER:
|
||||||
|
case LONG:
|
||||||
|
return new LongColumn(random.nextInt(Integer.MAX_VALUE));
|
||||||
|
case FLOAT:
|
||||||
|
case DOUBLE:
|
||||||
|
return new DoubleColumn(random.nextDouble());
|
||||||
|
case NULL:
|
||||||
|
return null;
|
||||||
|
case BYTE_ARRAY:
|
||||||
|
return new BytesColumn(new byte[]{(byte) (random.nextInt() % 2)});
|
||||||
|
case LOCAL_DATE:
|
||||||
|
return new StringColumn(LocalDate.now().format(DateTimeFormatter.ofPattern("yyyy-MM-dd")));
|
||||||
|
case MAP:
|
||||||
|
return new StringColumn(JSON.toJSONString(propmap()));
|
||||||
|
case STRING_ARRAY:
|
||||||
|
return new StringColumn("[1,1,1,1,1,1,1]");
|
||||||
|
default:
|
||||||
|
return new StringColumn("randomStr" + random.nextInt(Integer.MAX_VALUE));
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
public static Map<String, Object> propmap() {
|
||||||
|
Map<String, Object> prop = new HashMap<>();
|
||||||
|
prop.put("name", "neo4jWriter");
|
||||||
|
prop.put("age", "1");
|
||||||
|
return prop;
|
||||||
|
}
|
||||||
|
}
|
||||||
41
neo4jwriter/src/test/resources/allTypeFieldNode.json
Normal file
41
neo4jwriter/src/test/resources/allTypeFieldNode.json
Normal file
@ -0,0 +1,41 @@
|
|||||||
|
{
|
||||||
|
"uri": "neo4j://localhost:7687",
|
||||||
|
"username":"neo4j",
|
||||||
|
"password":"Test@12343",
|
||||||
|
"database":"neo4j",
|
||||||
|
"cypher": "unwind $batch as row create(p:Person) set p.pbool = row.pbool,p.pstring = row.pstring,p.plong = row.plong,p.pshort = row.pshort,p.pdouble=row.pdouble,p.pstringarr=row.pstringarr,p.plocaldate=row.plocaldate",
|
||||||
|
"batchDataVariableName": "batch",
|
||||||
|
"batchSize": "33",
|
||||||
|
"properties": [
|
||||||
|
{
|
||||||
|
"name": "pbool",
|
||||||
|
"type": "BOOLEAN"
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"name": "pstring",
|
||||||
|
"type": "STRING"
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"name": "plong",
|
||||||
|
"type": "LONG"
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"name": "pshort",
|
||||||
|
"type": "SHORT"
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"name": "pdouble",
|
||||||
|
"type": "DOUBLE"
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"name": "pstringarr",
|
||||||
|
"type": "STRING_ARRAY",
|
||||||
|
"split": ","
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"name": "plocaldate",
|
||||||
|
"type": "LOCAL_DATE",
|
||||||
|
"dateFormat": "yyyy-MM-dd"
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
19
neo4jwriter/src/test/resources/dynamicLabel.json
Normal file
19
neo4jwriter/src/test/resources/dynamicLabel.json
Normal file
@ -0,0 +1,19 @@
|
|||||||
|
{
|
||||||
|
"uri": "bolt://localhost:7687",
|
||||||
|
"username":"neo4j",
|
||||||
|
"password":"Test@12343",
|
||||||
|
"database":"neo4j",
|
||||||
|
"cypher": "unwind $batch as row CALL apoc.cypher.doIt( 'create (n:`' + row.Label + '`{id:$id})' ,{id: row.id} ) YIELD value RETURN 1 ",
|
||||||
|
"batchDataVariableName": "batch",
|
||||||
|
"batchSize": "33",
|
||||||
|
"properties": [
|
||||||
|
{
|
||||||
|
"name": "Label",
|
||||||
|
"type": "string"
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"name": "id",
|
||||||
|
"type": "STRING"
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
19
neo4jwriter/src/test/resources/relationship.json
Normal file
19
neo4jwriter/src/test/resources/relationship.json
Normal file
@ -0,0 +1,19 @@
|
|||||||
|
{
|
||||||
|
"uri": "neo4j://localhost:7687",
|
||||||
|
"username":"neo4j",
|
||||||
|
"password":"Test@12343",
|
||||||
|
"database":"neo4j",
|
||||||
|
"cypher": "unwind $batch as row match(p1:Person) where p1.id = row.startNodeId match(p2:Person) where p2.id = row.endNodeId create (p1)-[:LINK]->(p2)",
|
||||||
|
"batchDataVariableName": "batch",
|
||||||
|
"batchSize": "33",
|
||||||
|
"properties": [
|
||||||
|
{
|
||||||
|
"name": "startNodeId",
|
||||||
|
"type": "STRING"
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"name": "endNodeId",
|
||||||
|
"type": "STRING"
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
Some files were not shown because too many files have changed in this diff Show More
Loading…
Reference in New Issue
Block a user