1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
package mobvista.dmp.test
import java.util.regex.Pattern
import org.apache.hadoop.io.compress.GzipCodec
import org.apache.hadoop.io.{LongWritable, Text}
import org.apache.hadoop.mapreduce.lib.input.{FileSplit, TextInputFormat}
import org.apache.spark.rdd.NewHadoopRDD
import org.apache.spark.sql.SparkSession
class ATest {
}
object ATest {
val dataSplit = "\t"
val part = "QUERY"
val encode = "UTF-8"
val httpPrefix = "http://test.com"
val regex = "^id\\d+$"
def main(args: Array[String]): Unit = {
val spark = SparkSession.builder()
.getOrCreate()
import spark.implicits._
val sc = spark.sparkContext
val rdd = sc.newAPIHadoopFile[LongWritable, Text, TextInputFormat]("hdfs://ip-172-31-33-205.ec2.internal:8020/user/root/ga_install_repair/*/*/*")
.asInstanceOf[NewHadoopRDD[LongWritable, Text]]
rdd.mapPartitionsWithInputSplit((split, itr) => {
val path = split.asInstanceOf[FileSplit].getPath
val pattern = Pattern.compile("\\d{4}/\\d{2}/\\d{2}")
val math = pattern.matcher(path.toString)
math.find()
val date = math.group(0).replace("/", "")
itr.map(_._2.toString)
.map(_.split("\\|", -1))
.filter(array => {
array.length >= 4 && (array(1).equalsIgnoreCase("ios") || array(1).equalsIgnoreCase("android"))
})
.map(array => {
val platform = array(1)
var idType = ""
var plt = ""
if (platform.equalsIgnoreCase("ios")) {
idType = "idfa"
plt = "ios"
} else if (platform.equalsIgnoreCase("android")) {
idType = "gaid"
plt = "adr"
}
val date1 = date.substring(0, 4) + "-" + date.substring(4, 6) + "-" + date.substring(6, 8)
(array(0), idType, plt, array(2), array(3), date1)
})
})
.toDF("device_id", "device_type", "platform", "package_name", "store_link", "date")
.createOrReplaceTempView("tmp_ga_data")
val sql =
"""
|select t.device_id, t.device_type, t.platform, t.package_name, t.store_link, t.date
|from (
| select t.device_id, t.device_type, t.platform, t.package_name, t.store_link, t.date,
| row_number() over(partition by t.device_id, t.device_type order by t.date desc) as rk
| from tmp_ga_data t
|) t
|where t.rk='1'
""".stripMargin
spark.sql(sql)
.rdd
.map(_.mkString("\t"))
.saveAsTextFile("s3://mob-emr-test/feng.liang/ga_install_list_all/", classOf[GzipCodec])
}
}