forked from jerryuncle/TJFH
-
Notifications
You must be signed in to change notification settings - Fork 0
/
st_river_r_zt.scala
235 lines (201 loc) · 9.47 KB
/
st_river_r_zt.scala
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
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
package cn.gistack.gdmp.datawrangling.script
import cn.gistack.gdmp.datawrangling.script.service.IDataWranglingScript
import java.sql.DriverManager
import java.time.{LocalDate, LocalDateTime}
import java.time.format.DateTimeFormatter
import java.util.Properties
import org.apache.spark.broadcast.Broadcast
import org.apache.spark.sql.functions._
import org.apache.spark.sql.{DataFrame, SparkSession}
import scala.collection.mutable.ListBuffer
/**
* @author SuperHuang
* @date 2021/12/06 10:32
* */
class DataWranglingScript extends BaseDataWranglingScript with IDataWranglingScript {
override def executeScript(): Unit = {
val spark = SparkSession
.builder()
.master("local[*]")
.appName("test")
.config("spark.sql.extensions", "io.delta.sql.DeltaSparkSessionExtension")
.config("spark.sql.catalog.spark_catalog", "org.apache.spark.sql.delta.catalog.DeltaCatalog")
.getOrCreate()
var start_tm = "2024-06-01"
var end_tm = "2024-07-01"
var is_current = "true"
var sd = 2
//获取历史记录最新时间
var collect_tm = "2023-04-24"
if(is_current.equals("true")){
val his_max_tm_sql_tb = """ select max(tm)-0.1 as tm from md.st_river_r where data_source is null or data_source in('dzp_sq','lk') """
val lastest_tm_df = readMD(spark,his_max_tm_sql_tb)
val lastest_tm_cache = lastest_tm_df
if(lastest_tm_cache.take(1)(0)(0) !=null ){
println("st_river_r 前一次历史最新时间是:"+lastest_tm_cache.take(1)(0).getTimestamp(0))
val timestamp = lastest_tm_cache.select("tm").take(1)(0).getTimestamp(0)
val formatter = DateTimeFormatter.ofPattern("yyyy-MM-dd HH:mm:ss")
val data_last_tm = formatter.format(timestamp.toLocalDateTime)
println(data_last_tm)
collect_tm = data_last_tm
}else{
println("st_river_r 前一次历史最新时间是Null,因此数据将从最早的时间采集")
collect_tm= "2024-06-01"
}
}
var tm = new ListBuffer[String]
var sqktm = new ListBuffer[String]
if(is_current == "true"){
println(s"delete from md.st_river_r where tm>='$collect_tm' and (data_source is null or data_source in('dzp_sq','lk'))")
deleteData(s"delete from md.st_river_r where tm>='$collect_tm' and (data_source is null or data_source in('dzp_sq','lk'))")
println(s"==============tm>='$collect_tm' and (data_source is null or data_source in('dzp_sq','lk')) delete finish")
val strings = dateList(collect_tm, LocalDate.now().plusDays(1).toString+" 00:00:00", sd)
val sqkstr = sqk_list(collect_tm, LocalDate.now().plusDays(1).toString+" 00:00:00", sd)
tm = strings
sqktm=sqkstr
}else{
println(s"delete from md.st_river_r where tm>='$start_tm' and tm<'$end_tm' and (data_source is null or data_source in('dzp_sq','lk'))")
deleteData(s"delete from md.st_river_r where tm>='$start_tm' and tm<'$end_tm' and (data_source is null or data_source in('dzp_sq','lk'))")
println(s"==============tm>='$start_tm' and tm<'$end_tm' and (data_source is null or data_source in('dzp_sq','lk')) delete finish")
val strings = dateList(start_tm, end_tm, sd)
tm = strings
val sqkstr = sqk_list(start_tm, end_tm, sd)
sqktm=sqkstr
}
import spark.implicits._
val st_list = readMD(spark,"select * from md.att_st_base where st_extsttp like '%ZZ%' or st_extsttp like '%ZQ%'")
.select(col("st_code"))
println(st_list.count())
val broadcastCodes: Broadcast[Array[String]] = spark.sparkContext.broadcast(st_list.as[String].collect())
val prop = new Properties()
prop.put("user", "wqualitykz") //表示用户名
prop.put("password", "kz2018") //表示密码
prop.put("driver", "oracle.jdbc.driver.OracleDriver")
val sqk = spark.read.jdbc(url="jdbc:oracle:thin:@10.12.4.29:1521/meetHydro",table="HYDROKZ.DZP_SQ",sqktm.toArray,prop)
.filter(row => broadcastCodes.value.contains(row.getAs[String]("ZH")))
.select(
col("ZH").alias("guid"),
col("ZH").alias("st_code"),
col("YMDHM").alias("tm").cast("timestamp"),
(col("UP_SW")+col("DJSZGC")).alias("z").cast("decimal(7,3)"),
col("LL").alias("q").cast("decimal(9,3)"),
lit(null).alias("xsa").cast("decimal(9,3)"),
lit(null).alias("xsavv").cast("decimal(5,3)"),
lit(null).alias("xsmxv").cast("decimal(5,3)"),
lit(null).alias("flwchrcd").cast("string"),
lit(null).alias("wptn").cast("string"),
lit(null).alias("msqmt").cast("string"),
lit(null).alias("msamt").cast("string"),
lit(null).alias("msvmt").cast("string"),
lit(current_timestamp()).alias("eff_time"),
col("LEIXING").alias("data_type").cast("int"),
col("RJLL").alias("q_avg").cast("decimal(7,3)"),
col("XJLL").alias("xjll").cast("decimal(7,3)"),
col("YJLL").alias("yjll").cast("decimal(7,3)"),
lit("dzp_sq").alias("data_source").cast("string"),
col("JJSW").alias("jjsw").cast("decimal(7,3)"),
col("DJSZGC").alias("djszgc").cast("decimal(7,3)"),
col("UP_SW").alias("z_gc").cast("decimal(7,3)"),
col("EXKEY").alias("exkey").cast("string")
)
val res_rz = spark.read.jdbc(url="jdbc:oracle:thin:@10.12.4.29:1521/meetHydro",table="HYDROKZ.ST_RIVER_R",tm.toArray,prop)
.select(
col("STCD").alias("guid"),
col("STCD").alias("st_code"),
col("TM").alias("tm").cast("timestamp"),
col("Z").alias("z").cast("decimal(7,3)"),
col("Q").alias("q").cast("decimal(9,3)"),
col("XSA").alias("xsa"),
col("XSAVV").alias("xsavv").cast("decimal(5,3)"),
col("XSMXV").alias("xsmxv").cast("decimal(5,3)"),
col("FLWCHRCD").alias("flwchrcd"),
col("WPTN").alias("wptn"),
col("MSQMT").alias("msqmt"),
col("MSAMT").alias("msamt"),
col("MSVMT").alias("msvmt"),
lit(current_timestamp()).alias("eff_time"),
lit(1).alias("data_type").cast("int"),
lit(null).alias("q_avg").cast("decimal(7,3)"),
lit(null).alias("xjll").cast("decimal(7,3)"),
lit(null).alias("yjll").cast("decimal(7,3)"),
lit("lk").alias("data_source").cast("string"),
lit(null).alias("jjsw").cast("decimal(7,3)"),
lit(null).alias("djszgc").cast("decimal(7,3)"),
lit(null).alias("z_gc").cast("decimal(7,3)"),
lit(null).alias("exkey").cast("string")
).join(sqk, Seq("guid", "st_code","tm"), "left_anti")
val result = sqk.unionAll(res_rz)
println(result.count())
result.printSchema()
result.show()
writeRdjc(result,"st_river_r")
spark.stop()
}
def readMD(spark: SparkSession,table:String): DataFrame ={
val reader = spark.read.format("jdbc")
.option("dirver", "com.kingbase8.Driver")
.option("url", "jdbc:kingbase8://10.12.40.26:54321/tjfhdd")
.option("dbtable", s"($table) AS t")
.option("user", "kingbase")
.option("password", "L#zst&r&j5vZ")
.load()
reader
}
def dateList(start:String,end:String,slid:Int): ListBuffer[String] ={
val tm_list = new ListBuffer[String]()
val dateFormatter = DateTimeFormatter.ofPattern("yyyy-MM-dd HH:mm:ss")
val startDate = LocalDateTime.parse(start, dateFormatter)
val endDate = LocalDateTime.parse(end, dateFormatter)
//.minusDays(1)
// 构建日期序列
val dateRange = Iterator.iterate(startDate)(_ plusHours 1).takeWhile(!_.isAfter(endDate)).toList
// 遍历日期序列,将连续的两个日期作为参数传入函数
dateRange.sliding(slid).foreach(pair => {
val tm_filter = "tm>=to_date('"+pair.head.format(dateFormatter)+ "','yyyy-mm-dd hh24:mi:ss') and tm<to_date('"+pair.last.format(dateFormatter)+"','yyyy-mm-dd hh24:mi:ss')"
tm_list.append(tm_filter)
})
tm_list.foreach(r=>println("加载时间段:"+r))
tm_list
}
def sqk_list(start:String,end:String,slid:Int): ListBuffer[String] ={
val dzp_sq_list = new ListBuffer[String]()
val dateFormatter = DateTimeFormatter.ofPattern("yyyy-MM-dd HH:mm:ss")
val startDate = LocalDateTime.parse(start, dateFormatter)
val endDate = LocalDateTime.parse(end, dateFormatter)
//.minusDays(1)
// 构建日期序列
val dateRange = Iterator.iterate(startDate)(_ plusHours 1).takeWhile(!_.isAfter(endDate)).toList
// 遍历日期序列,将连续的两个日期作为参数传入函数
dateRange.sliding(slid).foreach(pair => {
val tm_filter = "YMDHM>=to_date('"+pair.head.format(dateFormatter)+ "','yyyy-mm-dd hh24:mi:ss') and YMDHM<to_date('"+pair.last.format(dateFormatter)+"','yyyy-mm-dd hh24:mi:ss')"
dzp_sq_list.append(tm_filter)
})
dzp_sq_list.foreach(r=>println("dzp_sq加载时间段:"+r))
dzp_sq_list
}
def deleteData(sql:String): Unit = {
// connect to Dm database
val dmUrl = "jdbc:kingbase8://10.12.40.26:54321/tjfhdd"
val dmUser = "kingbase"
val dmPassword = "L#zst&r&j5vZ"
val dmConn = DriverManager.getConnection(dmUrl, dmUser, dmPassword)
dmConn.setAutoCommit(false)
val insertStatement = dmConn.createStatement()
val insertQuery = s"$sql"
insertStatement.executeUpdate(insertQuery)
dmConn.commit()
dmConn.setAutoCommit(true)
insertStatement.close()
dmConn.close()
}
def writeRdjc(frame:DataFrame,table:String): Unit ={
frame.write
.mode("append")
.format("jdbc")
.option("url", "jdbc:kingbase8://10.12.40.26:54321/tjfhdd")
.option("dbtable", s"md.$table")
.option("user", "kingbase")
.option("password", "L#zst&r&j5vZ")
.save()
}
}