Member since
06-20-2017
19
Posts
4
Kudos Received
3
Solutions
My Accepted Solutions
Title | Views | Posted |
---|---|---|
3029 | 08-22-2017 01:24 PM | |
3467 | 08-16-2017 01:59 PM | |
1182 | 08-10-2017 08:57 AM |
08-22-2017
04:46 PM
You can use the ReplaceText processor. Use ${attribute_name} to write an attribute in your text.
... View more
08-22-2017
01:24 PM
With the UpdateAttribute, you can add your own property. (With the '+' button in the 'Properties' tab) Add a property with the name : Sis_Grp.Dom and the value : ${ZComponent:substringAfter('"'):substringBefore('"')} (the argument for the substringBefore is double quote between simple quotes.) and a second with the name : Name and the value : ${ZComponent:substringAfter('Name="'):substringBefore('"')} You can find all the Nifi Expression language here
... View more
08-21-2017
07:20 AM
Do you have an error ?
... View more
08-18-2017
02:21 PM
1 - ExecuteSQL -> 2 - ConvertAvroToJSON -> 3 - splitJSon -> 4 - EvaluateJsonPath -> 5 - UpdateAttribute -> 6 ExecuteSQL -> 7 - replaceText -> 8- putSQL. 1 - 2 - 3 : query your table and split by row. You will have one flowfile by row. Use $ for the JsonPathExpression in the split processor. 4 - Retrieve the query_id and the query_execute in two attributes of each flowfile (values of your properties are $.query_id and $.query_execute) 5 - Create a new attribute query_startTime with the value ${now():format('HH:mm:ss') 6 - Execute the query ${query_execute} 7 - Write your insert query : INSERT INTO queryExecutionInfo VALUES (${query_id}, ${query_startTime},{${now():format('HH:mm:ss')); 8 - Execute your insert I think this flow works, maybe someone will have something efficient.
... View more
08-18-2017
09:10 AM
1 Kudo
Hi A common flow to modify your data is : ConvertAvroToJson -> splitJson -> EvaluateJsonPath -> UpdateAttribute So you will have one flowfile by row and with the EvaluateJsonPath you convert your fields as attributes of this flowfile. After that you can use UpdateAttribute to modify an attribute or add new one. You can use Nifi Expression Language.
... View more
08-16-2017
01:59 PM
1 Kudo
Hi, I think you can do that with the QueryRecord processor. Add a new property, whatever the name and the content may be something like "select Subject, count(Name) as nb from flowfile group by Subject". Set correctly the schemas for the reader and the writer and you will find the result of the query as a new relationship of your processor. (A new relationship with the name of your property will appear)
... View more
08-10-2017
08:57 AM
Hi, 1. You can find logs for Schema Registry under this path : /usr/hdf/current/registry/logs 2. Remove all the spaces you have in the name of your properties and it would be fine {
"type" : "record",
"namespace" : "poc",
"name" : "Employee",
"fields" : [
{ "name" : "Name", "type" : "string"},
{ "name" : "Age", "type" : "int"}
]
}
... View more
07-31-2017
08:00 AM
Thanks a lot for this detailed response ! @Greg Keys
... View more
07-28-2017
01:15 PM
1 Kudo
With the new version of HDF (3.0), we find Nifi and the new application : Streaming analystics manager. What is the main difference between Nifi and the stream builder module of SAM ? How can/should we use these two applications together ? After a quick time to read the description of this module, it seems it's really close to Nifi. Thank you !
... View more
Labels:
- Labels:
-
Apache NiFi
-
Cloudera DataFlow (CDF)