一键同步mysql到数仓（airflow调度）
作者：互联网
经常会接到产品的需求：同步***表到hive，做分析。（做多了感觉很烦，就写一个工具）
一：背景、功能、流程介绍
1.背景：
1.数仓使用hive存储，datax导数据、airflow调度
2.虽然数据产品同学对datax进行了封装，可以点点点完成mysql表的同步，但是过程太复杂了
还需要自己手动建表，还不支持修改。就萌生了自己写一个工具的想法
2.功能
就是通过mysql配置完成hive的一般建表，airflow调度任务的生成
3.流程
1.配置mysql链接
2.根据mysql数据类型，生成对应的hive表结构，建表
3.生成airflow调度任务（读取mysql表，调用datax，修复分区）
二：代码
1.配置文件介绍：
MysqlToHive.properties
jdbcalias:ptx_read #mysql别名要和同步的数据库的别名保持一致
table:be_product #要同步的表名
hive_prefix:ods.ods_product_ ##生成hive表的前缀
hive_suffix:_dd ##增量表还是全量表
owner=xiaolong.wu ##airflow任务的owner
lifecycle=180 ##hive表的生命周期，数据数据产品删除数据
airflowpath=/airflow/dags/ods/ ##生成airflow任务文件的路径
s3path=s3://path ##datax写hive表需要的基本路径
jdbc1alias : hive ##可以写多个mysql链接，不用一个来回改
jdbc1host : 127.0.0.1
jdbc1port : 3306
jdbc1user : root
jdbc1passwd : **
jdbc1db_name : test
jdbc2alias:read
jdbc2host : 127.0.0.1
jdbc2port : 3306
jdbc2user : root
jdbc2passwd :**
jdbc2db_name :test
2.基本代码：
MysqlToHive.java
import java.io.*;
import java.sql.Connection;
import java.sql.DriverManager;
import java.sql.ResultSet;
import java.sql.Statement;
import java.util.ArrayList;
import java.util.List;
import java.util.Properties;
class Database {//mysql配置工具类，非重点
private String alias;
private String host;
private int port;
private String user;
private String passwd;
private String db_name;
public String getAlias() {return alias;}
public void setAlias(String alias) {this.alias = alias;}
public String getHost() {return host;}
public void setHost(String host) {this.host = host;}
public int getPort() {return port;}
public void setPort(int port) {this.port = port;}
public String getUser() {return user;}
public void setUser(String user) {this.user = user;}
public String getPasswd() {return passwd;}
public void setPasswd(String passwd) {this.passwd = passwd;}
public String getDb_name() {return db_name;}
public void setDb_name(String db_name) {this.db_name = db_name;}
@Override
public String toString() {
return "Database{" +"alias='" + alias + '\'' +", host='" + host + '\'' +", port=" + port +
", user='" + user + '\'' +", passwd='" + passwd + '\'' +", db_name='" + db_name + '\'' +'}';
}
}
public class MysqlToHive {
public static String jdbcalias;
public static String table;
public static String hive_prefix;
public static String hive_suffix;
public static String owner;
public static String lifecycle;
public static String airflowpath;
public static String s3path;
public static Database database = new Database();
public static List<Database> databasesList = new ArrayList<Database>();
public static List<String> mysqlTableColumn = new ArrayList<String>();
public static void parseProperties(Properties pp){
jdbcalias = pp.getProperty("jdbcalias");
table = pp.getProperty("table");
hive_prefix = pp.getProperty("hive_prefix");
owner = pp.getProperty("owner");
lifecycle = pp.getProperty("lifecycle");
airflowpath = pp.getProperty("airflowpath");
s3path = pp.getProperty("s3path");
hive_suffix = pp.getProperty("hive_suffix");
int dbindex = 1;//根据mysql链接的别名，找到对应的mysql配置
while(pp.getProperty("jdbc"+dbindex+"alias")!= null){
Database databaseItem = new Database();
databaseItem.setDb_name(pp.getProperty("jdbc"+dbindex+"db_name"));
databaseItem.setHost(pp.getProperty("jdbc"+dbindex+"host"));
databaseItem.setAlias(pp.getProperty("jdbc"+dbindex+"alias"));
databaseItem.setPasswd(pp.getProperty("jdbc"+dbindex+"passwd"));
databaseItem.setPort(Integer.parseInt(pp.getProperty("jdbc"+dbindex+"port")));
databaseItem.setUser(pp.getProperty("jdbc"+dbindex+"user"));
System.out.println(databaseItem.toString());
databasesList.add(databaseItem);
if(databaseItem.getAlias().equals(jdbcalias)){
database =databasesList.get(dbindex-1);
}
dbindex++;
}
}
//读取配置文件
public static void readDbPropertiesFile (String fileName){
Properties pp = new Properties();
try {
InputStream fps = Thread.currentThread().getContextClassLoader().getResourceAsStream(fileName);
pp.load(fps);
parseProperties(pp);
fps.close();
} catch (Exception e) {
e.printStackTrace();
}
}
//链接mysql，拿到对应的表结构，为后续生成hive表做准备
public static void readTableFormatted () throws Exception {
String sql="show full fields from " + table;
Connection con=null;
Statement st=null;
ResultSet rs=null;
Class.forName("com.mysql.cj.jdbc.Driver");
System.out.println("jdbc:mysql://"+database.getHost()+":"+database.getPort()+"/"+database.getDb_name()+"?serverTimezone=UTC");
con= DriverManager.getConnection("jdbc:mysql://"+database.getHost()+":"+database.getPort()+"/"+database.getDb_name()+"?serverTimezone=UTC", database.getUser(),database.getPasswd());
st=con.createStatement();
rs=st.executeQuery(sql);
while(rs.next())
{
// System.out.println(rs.getString("Field").toLowerCase()+"|"+rs.getString("Type").toLowerCase() +"|"+ rs.getString("Comment"));
mysqlTableColumn.add(rs.getString("Field").toLowerCase()+"|"+rs.getString("Type").toLowerCase() +"|"+ rs.getString("Comment"));
}
}
public static int getmysqlLength(String type) {
return Integer.parseInt(type.substring(type.indexOf("(")+1,type.indexOf(")")));
}
//根据mysql类型，生成对应的hive类型，并建hive表
public static void buildExecuteHiveSql () throws IOException, InterruptedException {
StringBuilder hiveSqlBuilder = new StringBuilder();
hiveSqlBuilder.append("CREATE TABLE "+hive_prefix+table+hive_suffix+" ( \n");
for (int i = 0; i < mysqlTableColumn.size(); i++) {
String []fieldAndType= mysqlTableColumn.get(i).split("\\|");
hiveSqlBuilder.append(fieldAndType[0] + " ");
if(fieldAndType[1].contains("bigint") || fieldAndType[1].contains("int") || fieldAndType[1].contains("smallint") || fieldAndType[1].contains("tinyint")){
hiveSqlBuilder.append("bigint");
}
else if(fieldAndType[1].contains("binary") || fieldAndType[1].contains("varbinary") ){
hiveSqlBuilder.append("binary");
}
else if(fieldAndType[1].contains("date") ){
hiveSqlBuilder.append("date");
}
else if(fieldAndType[1].contains("double") || fieldAndType[1].contains("float") || fieldAndType[1].contains("decimal")){
hiveSqlBuilder.append("double");
}
else if(fieldAndType[1].contains("char") || fieldAndType[1].contains("varchar") || fieldAndType[1].contains("mediumtext")
|| fieldAndType[1].contains("datetime") || fieldAndType[1].contains("time") || fieldAndType[1].contains("timestamp")){
hiveSqlBuilder.append("string");
}
String comment = "";
if(fieldAndType.length==3){comment = fieldAndType[2];};
hiveSqlBuilder.append(" comment '"+comment+ "' ,");
hiveSqlBuilder.append("\n");
}
hiveSqlBuilder.deleteCharAt(hiveSqlBuilder.length()-2); //去除最后的回车和，
hiveSqlBuilder.append(") PARTITIONED BY ( dt string COMMENT '(一级分区)' ) \n");
hiveSqlBuilder.append("ROW FORMAT DELIMITED STORED AS PARQUET \n");
hiveSqlBuilder.append("TBLPROPERTIES ('lifecycle'='"+lifecycle+"','owner'='"+owner+"','parquet.compression'='snappy');");
System.out.println(hiveSqlBuilder.toString());
Process process = new ProcessBuilder("hive","-e","\""+hiveSqlBuilder.toString()+"\"").redirectErrorStream(true).start();;
BufferedReader br = new BufferedReader(new InputStreamReader(process.getInputStream()));
String line;
while ((line = br.readLine()) != null) {
System.out.println(line);
}
process.waitFor();
}
//拼接mysql查询语句，用于airflow调度中查询mysql的sql语句
public static void printAirflowJobGetSelectSql (StringBuilder mysqlSelectBuilder,StringBuilder hiveTypeBuilder){
mysqlSelectBuilder.append("select ");
for (int i = 0; i < mysqlTableColumn.size(); i++) {
String []fieldAndType= mysqlTableColumn.get(i).split("\\|");
mysqlSelectBuilder.append(fieldAndType[0] + ",");
hiveTypeBuilder.append("{\"name\":\""+fieldAndType[0]+"\",\"type\":\"");
if(fieldAndType[1].contains("bigint") || fieldAndType[1].contains("int") || fieldAndType[1].contains("smallint") || fieldAndType[1].contains("tinyint")){
hiveTypeBuilder.append("bigint");
}
else if(fieldAndType[1].contains("binary") || fieldAndType[1].contains("varbinary") ){
hiveTypeBuilder.append("binary");
}
else if(fieldAndType[1].contains("date") ){
hiveTypeBuilder.append("date");
}
else if(fieldAndType[1].contains("double") || fieldAndType[1].contains("float") || fieldAndType[1].contains("decimal")){
hiveTypeBuilder.append("double");
}
else if(fieldAndType[1].contains("char") || fieldAndType[1].contains("varchar") || fieldAndType[1].contains("mediumtext")
|| fieldAndType[1].contains("datetime") || fieldAndType[1].contains("time") || fieldAndType[1].contains("timestamp")){
hiveTypeBuilder.append("string");
}
hiveTypeBuilder.append("\"},");
}
hiveTypeBuilder.deleteCharAt(hiveTypeBuilder.length()-1);
mysqlSelectBuilder.deleteCharAt(mysqlSelectBuilder.length()-1);
mysqlSelectBuilder.append(" from " + table);
}
//在固定路径下生成airflow文件，就是生成调度
//该部分涉及到公司封装的代码太多了，就不保留了
public static void printAirflowJob () throws FileNotFoundException {
String db = hive_prefix.substring(0,hive_prefix.indexOf("."));
String odsTableName = hive_prefix.substring(hive_prefix.indexOf(".")+1);
if(new File(odsTableName).exists()){
System.out.println("folder exist,please delete the folder "+airflowpath+odsTableName+table+hive_suffix);
}
else{
StringBuilder mysqlSelectBuilder = new StringBuilder();
StringBuilder hiveTypeBuilder = new StringBuilder();
printAirflowJobGetSelectSql(mysqlSelectBuilder,hiveTypeBuilder);
File dir = new File(airflowpath+odsTableName+table+hive_suffix);
dir.mkdirs();
PrintWriter pw = new PrintWriter(airflowpath+odsTableName+table+hive_suffix+"/"+odsTableName+table+hive_suffix+"_dag.py");
pw.println("import airflow");
pw.println("job_name='"+hive_prefix+table+hive_suffix+"'");
pw.println("job_owner='"+owner+"'");
pw.println("default_job_retries=1");
pw.println("default_job_retry_delay=timedelta(minutes=5)");
pw.println("default_start_date=airflow.utils.dates.days_ago(1)");
pw.println("dag_schedule_interval='12 1 * * *'");
pw.println("");
pw.println("");
pw.println("hive_table_name = job_name");
pw.println("");
pw.println("default_args = {");
pw.println(" 'owner': job_owner,");
pw.println(" 'depends_on_past': False,");
pw.println(" 'start_date': default_start_date,");
pw.println(" 'email': default_email_to,");
pw.println(" 'email_on_failure': False,");
pw.println(" 'email_on_retry': False,");
pw.println(" 'retries': default_job_retries,");
pw.println(" 'retry_delay': default_job_retry_delay,");
pw.println(" 'pool': default_pool,");
pw.println(" 'priority_weight': 10");
pw.println("}");
pw.println("");
}
}
public static void main(String[] args) throws Exception {
readDbPropertiesFile("MysqlToHive.properties");
readTableFormatted();
buildExecuteHiveSql();
printAirflowJob();
}
}
三：执行样例
1.mysql样例：
CREATE TABLE mysql_column_test (
bigint_test bigint(10),
int_test int(10),
smallint_test smallint(10),
binary_test binary(20),
varbinary_test varbinary(20),
decimal_test decimal(30,5),
double_test double(30,5),
float_test float(30,5),
char_test char(40),
varchar_test varchar(40),
mediumtext_test mediumtext,
date_test date,
datetime_test datetime,
time_test time,
timestamp_test timestamp DEFAULT CURRENT_TIMESTAMP
) ENGINE=InnoDB DEFAULT CHARSET=utf8;
insert into mysql_column_test(bigint_test,int_test,smallint_test,binary_test,varbinary_test,decimal_test,double_test,float_test,char_test,varchar_test,mediumtext_test,
date_test,datetime_test,time_test) values(1,2,3,UNHEX('4'),UNHEX('5'),6.1,7.1,8.1,9.1,'10','11','2022-01-01','2020-09-14 23:18:17',CURRENT_TIMESTAMP);
2.代码执行：直接复制代码过去，新建文件，执行
javac -cp mysql-connector-java-8.0.18.jar MysqlToHive.java
java -classpath mysql-connector-java-8.0.18.jar: MysqlToHive
标签：数仓,airflow,String,contains,hive,mysql,test,fieldAndType,public 来源： https://www.cnblogs.com/wuxiaolong4/p/16462217.html