创建dataframe的几种方式:
DataFrame也是一个分布式数据容器。与RDD类似,然而DataFrame更像传统数据库的二维表格,除了数据以外,还掌握数据的结构信息,即schema。同时,与Hive类似,DataFrame也支持嵌套数据类型(struct、array和map)。从API易用性的角度上 看, DataFrame API提供的是一套高层的关系操作,比函数式的RDD API要更加友好,门槛更低。
DataFrame的底层封装的是RDD,只不过RDD的泛型是Row类型。
1.读取json格式的文件创建dataframe
SparkConf conf = new SparkConf();
conf.setMaster("local").setAppName("jsonfile");
SparkContext sc = new SparkContext(conf);
//创建sqlContext
SQLContext sqlContext = new SQLContext(sc);
/**
* DataFrame的底层是一个一个的RDD RDD的泛型是Row类型。
* 以下两种方式都可以读取json格式的文件
*/
DataFrame df = sqlContext.read().format("json").load("sparksql/json");
// DataFrame df2 = sqlContext.read().json("sparksql/json.txt");
// df2.show();
/**
* DataFrame转换成RDD
*/
RDD<Row> rdd = df.rdd();
/**
* 显示 DataFrame中的内容,默认显示前20行。如果现实多行要指定多少行show(行数)
* 注意:当有多个列时,显示的列先后顺序是按列的ascii码先后显示。
*/
// df.show();
/**
* 树形的形式显示schema信息
*/
df.printSchema();
/**
* dataFram自带的API 操作DataFrame
*/
//select name from table
// df.select("name").show();
//select name age+10 as addage from table
df.select(df.col("name"),df.col("age").plus(10).alias("addage")).show();
//select name ,age from table where age>19
df.select(df.col("name"),df.col("age")).where(df.col("age").gt(19)).show();
//select count(*) from table group by age
df.groupBy(df.col("age")).count().show();
/**
* 将DataFrame注册成临时的一张表,这张表临时注册到内存中,是逻辑上的表,不会雾化到磁盘
*/
df.registerTempTable("jtable");
DataFrame sql = sqlContext.sql("select age,count(1) from jtable group by age");
DataFrame sql2 = sqlContext.sql("select * from jtable");
sc.stop();
2.通过json格式的rdd创建dataframe
SparkConf conf = new SparkConf();
conf.setMaster("local").setAppName("jsonRDD");
JavaSparkContext sc = new JavaSparkContext(conf);
SQLContext sqlContext = new SQLContext(sc);
JavaRDD<String> nameRDD = sc.parallelize(Arrays.asList(
"{\"name\":\"zhangsan\",\"age\":\"18\"}",
"{\"name\":\"lisi\",\"age\":\"19\"}",
"{\"name\":\"wangwu\",\"age\":\"20\"}"
));
JavaRDD<String> scoreRDD = sc.parallelize(Arrays.asList(
"{\"name\":\"zhangsan\",\"score\":\"100\"}",
"{\"name\":\"lisi\",\"score\":\"200\"}",
"{\"name\":\"wangwu\",\"score\":\"300\"}"
));
DataFrame namedf = sqlContext.read().json(nameRDD);
DataFrame scoredf = sqlContext.read().json(scoreRDD);
namedf.registerTempTable("name");
scoredf.registerTempTable("score");
DataFrame result = sqlContext.sql("select name.name,name.age,score.score from name,score where name.name = score.name");
result.show();
sc.stop();
3.非json格式的rdd创建dataframe
1)通过反射的方式将非json格式的rdd转换成dataframe
/**
* 注意:
* 1.自定义类必须是可序列化的
* 2.自定义类访问级别必须是Public
* 3.RDD转成DataFrame会把自定义类中字段的名称按assci码排序
*/
SparkConf conf = new SparkConf();
conf.setMaster("local").setAppName("RDD");
JavaSparkContext sc = new JavaSparkContext(conf);
SQLContext sqlContext = new SQLContext(sc);
JavaRDD<String> lineRDD = sc.textFile("sparksql/person.txt");
JavaRDD<Person> personRDD = lineRDD.map(new Function<String, Person>() {
/**
*
*/
private static final long serialVersionUID = 1L;
@Override
public Person call(String s) throws Exception {
Person p = new Person();
p.setId(s.split(",")[0]);
p.setName(s.split(",")[1]);
p.setAge(Integer.valueOf(s.split(",")[2]));
return p;
}
});
/**
* 传入进去Person.class的时候,sqlContext是通过反射的方式创建DataFrame
* 在底层通过反射的方式获得Person的所有field,结合RDD本身,就生成了DataFrame
*/
DataFrame df = sqlContext.createDataFrame(personRDD, Person.class);
df.show();
df.registerTempTable("person");
sqlContext.sql("select name from person where id = 2").show();
/**
* 将DataFrame转成JavaRDD
* 注意:
* 1.可以使用row.getInt(0),row.getString(1)...通过下标获取返回Row类型的数据,但是要注意列顺序问题---不常用
* 2.可以使用row.getAs("列名")来获取对应的列值。
*
*/
JavaRDD<Row> javaRDD = df.javaRDD();
JavaRDD<Person> map = javaRDD.map(new Function<Row, Person>() {
/**
*
*/
private static final long serialVersionUID = 1L;
@Override
public Person call(Row row) throws Exception {
Person p = new Person();
//p.setId(row.getString(1));
//p.setName(row.getString(2));
//p.setAge(row.getInt(0));
p.setId((String)row.getAs("id"));
p.setName((String)row.getAs("name"));
p.setAge((Integer)row.getAs("age"));
return p;
}
});
map.foreach(new VoidFunction<Person>() {
/**
*
*/
private static final long serialVersionUID = 1L;
@Override
public void call(Person t) throws Exception {
System.out.println(t);
}
});
sc.stop();
2)通过动态创建schema的方式创建dataframe
SparkConf conf = new SparkConf();
conf.setMaster("local").setAppName("rddStruct");
JavaSparkContext sc = new JavaSparkContext(conf);
SQLContext sqlContext = new SQLContext(sc);
JavaRDD<String> lineRDD = sc.textFile("./sparksql/person.txt");
/**
* 转换成Row类型的RDD
*/
JavaRDD<Row> rowRDD = lineRDD.map(new Function<String, Row>() {
/**
*
*/
private static final long serialVersionUID = 1L;
@Override
public Row call(String s) throws Exception {
return RowFactory.create(
String.valueOf(s.split(",")[0]),
String.valueOf(s.split(",")[1]),
Integer.valueOf(s.split(",")[2])
);
}
});
/**
* 动态构建DataFrame中的元数据,一般来说这里的字段可以来源自字符串,也可以来源于外部数据库
*/
List<StructField> asList =Arrays.asList(
DataTypes.createStructField("id", DataTypes.StringType, true),
DataTypes.createStructField("name", DataTypes.StringType, true),
DataTypes.createStructField("age", DataTypes.IntegerType, true)
);
StructType schema = DataTypes.createStructType(asList);
DataFrame df = sqlContext.createDataFrame(rowRDD, schema);
df.show();
sc.stop();
4.读取parquet文件创建dataframe
SparkConf conf = new SparkConf();
conf.setMaster("local").setAppName("parquet");
JavaSparkContext sc = new JavaSparkContext(conf);
SQLContext sqlContext = new SQLContext(sc);
JavaRDD<String> jsonRDD = sc.textFile("sparksql/json");
DataFrame df = sqlContext.read().json(jsonRDD);
/**
* 将DataFrame保存成parquet文件,SaveMode指定存储文件时的保存模式
* 保存成parquet文件有以下两种方式:
*/
df.write().mode(SaveMode.Overwrite).format("parquet").save("./sparksql/parquet");
df.write().mode(SaveMode.Overwrite).parquet("./sparksql/parquet");
df.show();
/**
* 加载parquet文件成DataFrame
* 加载parquet文件有以下两种方式:
*/
DataFrame load = sqlContext.read().format("parquet").load("./sparksql/parquet");
load = sqlContext.read().parquet("./sparksql/parquet");
load.show();
sc.stop();
5.读取jdbc中的数据创建dataframe
SparkConf conf = new SparkConf();
conf.setMaster("local").setAppName("mysql");
JavaSparkContext sc = new JavaSparkContext(conf);
SQLContext sqlContext = new SQLContext(sc);
/**
* 第一种方式读取MySql数据库表,加载为DataFrame
*/
Map<String, String> options = new HashMap<String,String>();
options.put("url", "jdbc:mysql://192.168.179.4:3306/spark");
options.put("driver", "com.mysql.jdbc.Driver");
options.put("user", "root");
options.put("password", "123456");
options.put("dbtable", "person");
DataFrame person = sqlContext.read().format("jdbc").options(options).load();
person.show();
person.registerTempTable("person");
/**
* 第二种方式读取MySql数据表加载为DataFrame
*/
DataFrameReader reader = sqlContext.read().format("jdbc");
reader.option("url", "jdbc:mysql://192.168.179.4:3306/spark");
reader.option("driver", "com.mysql.jdbc.Driver");
reader.option("user", "root");
reader.option("password", "123456");
reader.option("dbtable", "score");
DataFrame score = reader.load();
score.show();
score.registerTempTable("score");
DataFrame result =
sqlContext.sql("select person.id,person.name,score.score from person,score where person.name = score.name");
result.show();
/**
* 将DataFrame结果保存到Mysql中
*/
Properties properties = new Properties();
properties.setProperty("user", "root");
properties.setProperty("password", "123456");
result.write().mode(SaveMode.Overwrite).jdbc("jdbc:mysql://192.168.179.4:3306/spark", "result", properties);
sc.stop();
6.读取hive中的数据
./spark-submit
--master spark://node1:7077,node2:7077
--executor-cores 1
--executor-memory 2G
--total-executor-cores 1
--class com.bjsxt.sparksql.dataframe.CreateDFFromHive
/root/test/HiveTest.jar
SparkConf conf = new SparkConf();
conf.setAppName("hive");
JavaSparkContext sc = new JavaSparkContext(conf);
//HiveContext是SQLContext的子类。
HiveContext hiveContext = new HiveContext(sc);
hiveContext.sql("USE spark");
hiveContext.sql("DROP TABLE IF EXISTS student_infos");
//在hive中创建student_infos表
hiveContext.sql("CREATE TABLE IF NOT EXISTS student_infos (name STRING,age INT) row format delimited fields terminated by '\t' ");
hiveContext.sql("load data local inpath '/root/test/student_infos' into table student_infos");
hiveContext.sql("DROP TABLE IF EXISTS student_scores");
hiveContext.sql("CREATE TABLE IF NOT EXISTS student_scores (name STRING, score INT) row format delimited fields terminated by '\t'");
hiveContext.sql("LOAD DATA "
+ "LOCAL INPATH '/root/test/student_scores'"
+ "INTO TABLE student_scores");
/**
* 查询表生成DataFrame
*/
DataFrame goodStudentsDF = hiveContext.sql("SELECT si.name, si.age, ss.score "
+ "FROM student_infos si "
+ "JOIN student_scores ss "
+ "ON si.name=ss.name "
+ "WHERE ss.score>=80");
hiveContext.sql("DROP TABLE IF EXISTS good_student_infos");
goodStudentsDF.registerTempTable("goodstudent");
DataFrame result = hiveContext.sql("select * from goodstudent");
result.show();
/**
* 将结果保存到hive表 good_student_infos
*/
goodStudentsDF.write().mode(SaveMode.Overwrite).saveAsTable("good_student_infos");
Row[] goodStudentRows = hiveContext.table("good_student_infos").collect();
for(Row goodStudentRow : goodStudentRows) {
System.out.println(goodStudentRow);
}
sc.stop();
开创函数:
SparkConf conf = new SparkConf();
conf.setAppName("windowfun");
JavaSparkContext sc = new JavaSparkContext(conf);
HiveContext hiveContext = new HiveContext(sc);
hiveContext.sql("use spark");
hiveContext.sql("drop table if exists sales");
hiveContext.sql("create table if not exists sales (riqi string,leibie string,jine Int) "
+ "row format delimited fields terminated by '\t'");
hiveContext.sql("load data local inpath '/root/test/sales' into table sales");
/**
* 开窗函数格式:
* 【 rou_number() over (partitin by XXX order by XXX) 】
*/
DataFrame result = hiveContext.sql("select riqi,leibie,jine "
+ "from ("
+ "select riqi,leibie,jine,"
+ "row_number() over (partition by leibie order by jine desc) rank "
+ "from sales) t "
+ "where t.rank<=3");
result.show();
sc.stop();
UDF:
SparkConf conf = new SparkConf();
conf.setMaster("local");
conf.setAppName("udf");
JavaSparkContext sc = new JavaSparkContext(conf);
SQLContext sqlContext = new SQLContext(sc);
JavaRDD<String> parallelize = sc.parallelize(Arrays.asList("zhansan","lisi","wangwu"));
JavaRDD<Row> rowRDD = parallelize.map(new Function<String, Row>() {
/**
*
*/
private static final long serialVersionUID = 1L;
@Override
public Row call(String s) throws Exception {
return RowFactory.create(s);
}
});
List<StructField> fields = new ArrayList<StructField>();
fields.add(DataTypes.createStructField("name", DataTypes.StringType,true));
StructType schema = DataTypes.createStructType(fields);
DataFrame df = sqlContext.createDataFrame(rowRDD,schema);
df.registerTempTable("user");
/**
* 根据UDF函数参数的个数来决定是实现哪一个UDF UDF1,UDF2。。。。UDF1xxx
*/
sqlContext.udf().register("StrLen", new UDF1<String,Integer>() {
/**
*
*/
private static final long serialVersionUID = 1L;
@Override
public Integer call(String t1) throws Exception {
return t1.length();
}
}, DataTypes.IntegerType);
sqlContext.sql("select name ,StrLen(name) as length from user").show();
//sqlContext.udf().register("StrLen",new UDF2<String, Integer, Integer>() {
//
// /**
// *
// */
// private static final long serialVersionUID = 1L;
//
// @Override
// public Integer call(String t1, Integer t2) throws Exception {
//return t1.length()+t2;
// }
//} ,DataTypes.IntegerType );
//sqlContext.sql("select name ,StrLen(name,10) as length from user").show();
sc.stop();
UDAF:
SparkConf conf = new SparkConf();
conf.setMaster("local").setAppName("udaf");
JavaSparkContext sc = new JavaSparkContext(conf);
SQLContext sqlContext = new SQLContext(sc);
JavaRDD<String> parallelize = sc.parallelize(Arrays.asList("zhansan","lisi","wangwu","zhangsan","zhangsan","lisi"));
JavaRDD<Row> rowRDD = parallelize.map(new Function<String, Row>() {
/**
*
*/
private static final long serialVersionUID = 1L;
@Override
public Row call(String s) throws Exception {
return RowFactory.create(s);
}
});
List<StructField> fields = new ArrayList<StructField>();
fields.add(DataTypes.createStructField("name", DataTypes.StringType, true));
StructType schema = DataTypes.createStructType(fields);
DataFrame df = sqlContext.createDataFrame(rowRDD, schema);
df.registerTempTable("user");
/**
* 注册一个UDAF函数,实现统计相同值得个数
* 注意:这里可以自定义一个类继承UserDefinedAggregateFunction类也是可以的
*/
sqlContext.udf().register("StringCount", new UserDefinedAggregateFunction() {
/**
*
*/
private static final long serialVersionUID = 1L;
/**
* 更新 可以认为一个一个地将组内的字段值传递进来 实现拼接的逻辑
* buffer.getInt(0)获取的是上一次聚合后的值
* 相当于map端的combiner,combiner就是对每一个map task的处理结果进行一次小聚合
* 大聚和发生在reduce端.
* 这里即是:在进行聚合的时候,每当有新的值进来,对分组后的聚合如何进行计算
*/
@Override
public void update(MutableAggregationBuffer buffer, Row arg1) {
buffer.update(0, buffer.getInt(0)+1);
}
/**
* 合并 update操作,可能是针对一个分组内的部分数据,在某个节点上发生的 但是可能一个分组内的数据,会分布在多个节点上处理
* 此时就要用merge操作,将各个节点上分布式拼接好的串,合并起来
* buffer1.getInt(0) : 大聚和的时候 上一次聚合后的值
* buffer2.getInt(0) : 这次计算传入进来的update的结果
* 这里即是:最后在分布式节点完成后需要进行全局级别的Merge操作
*/
@Override
public void merge(MutableAggregationBuffer buffer1, Row buffer2) {
buffer1.update(0, buffer1.getInt(0) + buffer2.getInt(0));
}
/**
* 指定输入字段的字段及类型
*/
@Override
public StructType inputSchema() {
return DataTypes.createStructType(
Arrays.asList(DataTypes.createStructField("name",
DataTypes.StringType, true)));
}
/**
* 初始化一个内部的自己定义的值,在Aggregate之前每组数据的初始化结果
*/
@Override
public void initialize(MutableAggregationBuffer buffer) {
buffer.update(0, 0);
}
/**
* 最后返回一个和DataType的类型要一致的类型,返回UDAF最后的计算结果
*/
@Override
public Object evaluate(Row row) {
return row.getInt(0);
}
@Override
public boolean deterministic() {
//设置为true
return true;
}
/**
* 指定UDAF函数计算后返回的结果类型
*/
@Override
public DataType dataType() {
return DataTypes.IntegerType;
}
/**
* 在进行聚合操作的时候所要处理的数据的结果的类型
*/
@Override
public StructType bufferSchema() {
return
DataTypes.createStructType(
Arrays.asList(DataTypes.createStructField("bf", DataTypes.IntegerType,
true)));
}
});
sqlContext.sql("select name ,StringCount(name) from user group by name").show();
sc.stop();