Aller au contenu

Logpickr UDF Case Event

Présentation

UDF permettant de récupérer les informations associées à un CaseID existant

UDF Case Events requesting Druid

To get information about this UDF direclty in ksqlDB, use the command :

DESCRIBE FUNCTION LOGPICKR_CASE_EVENTS;

This UDF allows retrieving from the _vertex Druid DataSource information related to a given caseId. The information corresponds to the following columns :

  • __time
  • enddate
  • vertex_name

UDF Signature :

def logpickrCaseEvents(caseId: String, projectId: String, workgroupId: String, workgroupKey: String, host: String, port: String): util.List[Struct]

The output Struct format :

STRUCT<START_DATE VARCHAR(STRING), END_DATE VARCHAR(STRING), VERTEX_NAME VARCHAR(STRING)>

Here, START_DATE corresponds to the __time column, END_DATE to the enddate one, and finally VERTEX_NAME which corresponds to the vertex_name column

Nevertheless, it is an Array with this Structure that is returned, in order to retrieve those information for each line associated to the caseId

The SQL request used by the UDF is the following :

SELECT __time AS startdate, enddate, vertex_name AS vertexName
FROM "projectId_vertex"
WHERE vertex_name is not NULL AND caseid = 'caseIdParam'

where projectId corresponds to the Logpickr project ID and caseIdParam corresponds to UDF parameter related to the caseId

Parameters :

  • caseId : The caseId for which we want to get information
  • projectId : The id of the Logpickr project containing the information
  • workgroupId : The id of the Logpickr workgroup related to the project containing the information
  • workgroupKey : The key of the Logpickr workgroup related to the project containing the information
  • host : Corresponds to the Druid host
  • port : Corresponds to the Druid connexion port

Examples in ksqlDB

To follow the examples, start by using the following command to apply modifications of a STREAM on data inserted before the creation/display of the STREAM :

SET 'auto.offset.reset'='earliest';
  • Creation of the initial STREAM storing the caseId for which we want to get information :
CREATE STREAM s1 (
  caseId VARCHAR
  ) WITH (
  kafka_topic = 's1',
  partitions = 1,
  value_format = 'avro'
  );
  • Creation of the STREAM doing the call to the UDF (here the UDF parameters are given as an example and can be modified, except for the caseId, as it must correspond to the caseId column of the previous STREAM)
CREATE STREAM s2 AS SELECT 
    caseId, 
    logpickr_case_events(caseId, '8968a61c-1fb5-4721-9a8b-d09198eef06b', 'druid_system', 'lpkAuth', 'dev-api.logpickr.com', '8082') AS informations 
    FROM s1 EMIT CHANGES;
  • Insertion of a caseID for which we want to get information (in order to retrieve information for a given caseId, the caseId needs to exist in the Logpickr project)
INSERT INTO s1 (caseId) VALUES ('3');
  • Display of the final result
SELECT caseId, informations FROM s2 EMIT CHANGES;