[ 
https://issues.apache.org/jira/browse/NIFI-7403?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

ZhangCheng updated NIFI-7403:
-----------------------------
    Description: 
PutSQL processor support <Support Fragmented Transactions>,if we set this 
property true, I think it means The PutSQL processor will excute these sqls of 
one transaction Transactionally!!

But we find that when we set the <Rollback On Failure> false, those sqls of one 
transaction do not excute transactionally,some sucess and some failure. I think 
it's wrong.

I think, if we set <Support Fragmented Transactions> true, it should be 
executed Transactionally, no matter <Rollback On Failure> is true or false.



  was:
PutSQL processor support <Support Fragmented Transactions>,if we set this 
property true, I think it means The PutSQL processor will excute these sqls of 
one transaction Transactionally!!

But we find that when we set the <Rollback On Failure> false, those sqls of one 
transaction do not excute transactionally,some sucess and some failure. I think 
it's wrong.

I think, if we set <Support Fragmented Transactions> true, it should be 
executed Transactionally, no matter <Rollback On Failure> is true or false.

I see the code, only PutSQL has the <Support Fragmented Transactions>,  it 
maybe improve this feature at a small cost.

modify code design:

step1:  Maybe other Processors would support the  <Support Fragmented 
Transactions>(such as PutDatabaseRecord), we should move the  <Support 
Fragmented Transactions> from PutSQL.java to Put.java( I think it's a rational 
design that `Put.java`  define the <Support Fragmented Transactions> property )


{code:java}
public static final PropertyDescriptor SUPPORT_TRANSACTIONS = new 
PropertyDescriptor.Builder()
            .name("Support Fragmented Transactions")
           ...
{code}

step2: Additionally, I think the Put.java can extract the RelationShips of the 
processors those use the Put.java(PutSQL PutDatabaseRecord, PutHiveQL...We can 
see that these processors who use the Put.java have the same Relationships, I 
this this is the `Put`'s common feature) 


{code:java}
static final Relationship REL_SUCCESS = new Relationship.Builder()
            .name("success")
            .description("A FlowFile is routed to this relationship after the 
database is successfully updated")
            .build();
    static final Relationship REL_RETRY = new Relationship.Builder()
            .name("retry")
            .description("A FlowFile is routed to this relationship if the 
database cannot be updated but attempting the operation again may succeed")
            .build();
    static final Relationship REL_FAILURE = new Relationship.Builder()
            .name("failure")
            .description("A FlowFile is routed to this relationship if the 
database cannot be updated and retrying the operation will also fail, "
                    + "such as an invalid query or an integrity constraint 
violation")
            .build();
{code}


step3: in Put.java `onTrigger` method,  after the `putFlowFiles` and before the 
`onCompleted.apply`, we try to get the value of <Rollback On Failure>, if true 
, check the `transferredFlowFiles` , if there are flowfiles don't route to 
`Success`, we should reroute these `transferredFlowFiles`(retry > failure),and 
do `onFailed`(if it's not null)

{code:java}
 try {
                    putFlowFiles(context, session, functionContext, connection, 
flowFiles, result);
                } catch (DiscontinuedException e) {
                    // Whether it was an error or semi normal is depends on the 
implementation and reason why it wanted to discontinue.
                    // So, no logging is needed here.
                }
...
                
                if(context.getProperty(SUPPORT_TRANSACTIONS).asBoolean()){
                    //TODO   do sth
                }

                // OnCompleted processing.
                if (onCompleted != null) {
                    onCompleted.apply(context, session, functionContext, 
connection);
                }

                // Transfer FlowFiles.
                transferFlowFiles.apply(context, session, functionContext, 
result);
{code}




> Put.java improvement(PutSQL's transactions support)
> ---------------------------------------------------
>
>                 Key: NIFI-7403
>                 URL: https://issues.apache.org/jira/browse/NIFI-7403
>             Project: Apache NiFi
>          Issue Type: Improvement
>          Components: Extensions
>    Affects Versions: 1.11.4
>            Reporter: ZhangCheng
>            Priority: Major
>              Labels: pull-request-available
>             Fix For: 1.12.0
>
>          Time Spent: 20m
>  Remaining Estimate: 0h
>
> PutSQL processor support <Support Fragmented Transactions>,if we set this 
> property true, I think it means The PutSQL processor will excute these sqls 
> of one transaction Transactionally!!
> But we find that when we set the <Rollback On Failure> false, those sqls of 
> one transaction do not excute transactionally,some sucess and some failure. I 
> think it's wrong.
> I think, if we set <Support Fragmented Transactions> true, it should be 
> executed Transactionally, no matter <Rollback On Failure> is true or false.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)

Reply via email to