案件初始化

src/main/scala/com/winhc/bigdata/spark/ng/judicial/JudicialCaseRelationAggsV3.scala

@@ -212,9 +212,9 @@ case class JudicialCaseRelationAggsV3(s: SparkSession, project: String, args_cas
   //替换id表
   val ads_judicial_case_relation_id = s" $project.ads_judicial_case_relation_id_v9"
   //主表
-  val ads_judicial_case_relation_r1 = s" $project.ads_judicial_case_relation_r1_v9"
+  val ads_judicial_case_relation_r1 = s" $project.ads_judicial_case_relation_r1_tmp"
   //明细表(增强)
-  val ads_judicial_case_relation_r3 = s" $project.ads_judicial_case_relation_r3_v9"
+  val ads_judicial_case_relation_r3 = s" $project.ads_judicial_case_relation_r3_tmp"
 
 
   val update = s"update"

src/main/scala/com/winhc/bigdata/spark/ng/judicial/JudicialCaseRelationRowkeyRelation_v3.scala

@@ -1,5 +1,6 @@
 package com.winhc.bigdata.spark.ng.judicial
 
+import com.alibaba.fastjson.JSON
 import com.winhc.bigdata.spark.implicits.BaseHelper._
 import com.winhc.bigdata.spark.implicits.RegexUtils.RichRegex
 import com.winhc.bigdata.spark.udf.BaseFunc
@@ -40,6 +41,7 @@ case class JudicialCaseRelationRowkeyRelation_v3(s: SparkSession,
     is_id_card_udf()
     json_parse_udf()
     spark.udf.register("case_equ", case_equ _)
+    spark.udf.register("case_no_split", case_no_split _)
     spark.udf.register("str_sort", (v1: String, v2: String) => Seq(v1, v2).filter(_ != null).sorted.mkString(""))
     spark.udf.register("match_case_no", (case_no: String) => pat matches case_no)
 
@@ -245,7 +247,7 @@ case class JudicialCaseRelationRowkeyRelation_v3(s: SparkSession,
     val inc_org_tab = s"winhc_ng.inc_ads_$tableName" + "_v9"
 
     val table_id = "rowkey"
-    val other_cols = Seq("plaintiff_info", "court_name", "case_no", "litigant_info", "defendant_info") ++ Seq(table_id, "ds", "connect_case_no")
+    val other_cols = Seq("plaintiff_info", "court_name", "case_no", "litigant_info", "defendant_info") ++ Seq(table_id, "ds", "connect_case_no","update_time")
 
     val ods_end_ds = getLastPartitionsOrElse(org_tab, "0")
     val tmp_tab = s"all_${tableName}_tmp_$ods_end_ds"
@@ -257,7 +259,7 @@ case class JudicialCaseRelationRowkeyRelation_v3(s: SparkSession,
            |SELECT  *
            |FROM    (
            |            SELECT  *
-           |                    ,ROW_NUMBER() OVER(PARTITION BY $table_id ORDER BY ds DESC ) AS num
+           |                    ,ROW_NUMBER() OVER(PARTITION BY $table_id ORDER BY ds DESC ,update_time DESC) AS num
            |            FROM    (
            |                        SELECT  ${other_cols.mkString(",")}
            |                        FROM    $inc_org_tab
@@ -273,7 +275,7 @@ case class JudicialCaseRelationRowkeyRelation_v3(s: SparkSession,
            |SELECT  *
            |FROM    (
            |            SELECT  *
-           |                    ,ROW_NUMBER() OVER(PARTITION BY $table_id ORDER BY ds DESC ) AS num
+           |                    ,ROW_NUMBER() OVER(PARTITION BY $table_id ORDER BY ds DESC ,update_time DESC) AS num
            |            FROM    (
            |                        SELECT  ${other_cols.mkString(",")}
            |                        FROM    $org_tab
@@ -292,7 +294,7 @@ case class JudicialCaseRelationRowkeyRelation_v3(s: SparkSession,
     sql(
       s"""
          |SELECT  *
-         |FROM    $tmp_tab lateral view OUTER explode(split(connect_case_no,'\\n')) t as single_connect_case_no
+         |FROM    $tmp_tab lateral view OUTER explode(case_no_split(connect_case_no)) t as single_connect_case_no
          |""".stripMargin)
       .createTempView(s"explode_$tmp_tab")
 
@@ -331,7 +333,7 @@ case class JudicialCaseRelationRowkeyRelation_v3(s: SparkSession,
            |SELECT $view
            |FROM    (
            |            SELECT  *
-           |                    ,ROW_NUMBER() OVER(PARTITION BY rowkey ORDER BY ds DESC ) AS num
+           |                    ,ROW_NUMBER() OVER(PARTITION BY rowkey ORDER BY ds DESC ,update_time DESC) AS num
            |            FROM    (
            |                    SELECT  *
            |                    FROM    $org_inc_ads_tab
@@ -349,7 +351,7 @@ case class JudicialCaseRelationRowkeyRelation_v3(s: SparkSession,
            |SELECT $view
            |FROM    (
            |            SELECT  *
-           |                    ,ROW_NUMBER() OVER(PARTITION BY rowkey ORDER BY ds DESC ) AS num
+           |                    ,ROW_NUMBER() OVER(PARTITION BY rowkey ORDER BY ds DESC ,update_time DESC) AS num
            |            FROM    (
            |                        SELECT  ${intersect_cols.mkString(",")}
            |                        FROM    $org_ads_tab
@@ -787,6 +789,19 @@ case class JudicialCaseRelationRowkeyRelation_v3(s: SparkSession,
     }
   }
 
+
+  private def case_no_split(no: String): List[String] = {
+    var list: mutable.Seq[String] = mutable.Seq.empty
+    if (StringUtils.isBlank(no)) return list.toList
+    val array = JSON.parseArray(no)
+
+    for (i <- 0 until array.size()) {
+      val case_no = array.getString(i)
+      list = list :+ case_no
+    }
+    list.toList
+  }
+
 }
 
 object JudicialCaseRelationRowkeyRelation_v3 {

src/main/scala/com/winhc/bigdata/spark/utils/case_connect_utils.scala

@@ -114,15 +114,15 @@ object case_connect_utils {
   }
 
   def main(args: Array[String]): Unit = {
-    val current_case_party_list: Seq[String] = Seq("李桂香")
-    val connect_case_party_list: Seq[String] = Seq("陈世赞")
-
-    val current_case_no = "（2018）桂0105执372号"
-    val connect_case_no = "（2018）桂0105执372号"
-    val current_court_name = "浙江省宁波市中级人民法院"
-    val connect_court_name = "浙江省宁波市中级人民法院"
-    val tn1 = "company_lawsuit"
-    val tn2 = "company_lawsuit"
+    val current_case_party_list: Seq[String] = Seq("顾晓智")
+    val connect_case_party_list: Seq[String] = Seq("罗彬","顾晓智")
+
+    val current_case_no = "(2022)川0112执3297号"
+    val connect_case_no = "(2022)川0112执3297号"
+    val current_court_name = null
+    val connect_court_name = "成都市龙泉驿区人民法院"
+    val tn1 = "company_zxr_restrict"
+    val tn2 = "company_zxr"
     //
     println(isConnect(current_case_party_list, connect_case_party_list, current_case_no, connect_case_no, current_court_name, connect_court_name, tn1, tn2))
   }

src/main/scala/com/winhc/bigdata/spark/utils/case_connect_utils_v2.scala

@@ -118,15 +118,15 @@ object case_connect_utils_v2 {
   }
 
   def main(args: Array[String]): Unit = {
-    val current_case_party_list: Seq[String] = Seq("徐靖良", "黄剑麟")
-    val connect_case_party_list: Seq[String] = Seq("张徐靖良", "黄剑麟", "云南长天商务有限公司", "云南滇沪房地产开发有限公司")
-
-    val current_case_no = "(2019)云民辖终279号"
-    val connect_case_no = "(2019)云民辖终279号"
-    val current_court_name = "云南省高级人民法院"
-    val connect_court_name = "云南省昆明市中级人民法院"
-    val tn1 = "wenshu_detail_v2"
-    val tn2 = "company_send_announcement"
+    val current_case_party_list: Seq[String] = Seq("顾晓智")
+    val connect_case_party_list: Seq[String] = Seq("罗彬,顾晓智")
+
+    val current_case_no = "(2022)川0112执3297号"
+    val connect_case_no = "(2022)川0112执3297号"
+    val current_court_name = null
+    val connect_court_name = "成都市龙泉驿区人民法院"
+    val tn1 = "company_zxr_restrict"
+    val tn2 = "company_zxr"
     //
     println(isConnect(current_case_party_list, connect_case_party_list, current_case_no, connect_case_no, current_court_name, connect_court_name, tn1, tn2))
   }