You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@nifi.apache.org by "qingbo li (JIRA)" <ji...@apache.org> on 2019/03/30 07:58:00 UTC
[jira] [Updated] (NIFI-6165) PutSolrRecord Improvement

     [ https://issues.apache.org/jira/browse/NIFI-6165?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

qingbo li updated NIFI-6165:
----------------------------
    Affects Version/s: 1.9.1
          Description: 
{code:java}
while ((record = reader.nextRecord()) != null) {
SolrInputDocument inputDoc = new SolrInputDocument();
writeRecord(record, inputDoc,fieldList,EMPTY_STRING);
inputDocumentList.add(inputDoc);
if(inputDocumentList.size()==batchSize) {
index(isSolrCloud, collection, commitWithin, contentStreamPath, requestParams, inputDocumentList);
inputDocumentList = new ArrayList<>();
}
index(isSolrCloud, collection, commitWithin, contentStreamPath, requestParams, inputDocumentList);
}


I think this second index method shoud outside of while ,and I don't catch you why must new ArrayList after first index, index method had cleared the SolrInputDocument List.{code}

  was:
{code:java}
while ((record = reader.nextRecord()) != null) {
SolrInputDocument inputDoc = new SolrInputDocument();
writeRecord(record, inputDoc,fieldList,EMPTY_STRING);
inputDocumentList.add(inputDoc);
if(inputDocumentList.size()==batchSize) {
index(isSolrCloud, collection, commitWithin, contentStreamPath, requestParams, inputDocumentList);
inputDocumentList = new ArrayList<>();
}
index(isSolrCloud, collection, commitWithin, contentStreamPath, requestParams, inputDocumentList);
}

{code}


> PutSolrRecord Improvement
> -------------------------
>
>                 Key: NIFI-6165
>                 URL: https://issues.apache.org/jira/browse/NIFI-6165
>             Project: Apache NiFi
>          Issue Type: Improvement
>    Affects Versions: 1.9.1
>         Environment: nifi1.9
>            Reporter: qingbo li
>            Priority: Major
>
> {code:java}
> while ((record = reader.nextRecord()) != null) {
> SolrInputDocument inputDoc = new SolrInputDocument();
> writeRecord(record, inputDoc,fieldList,EMPTY_STRING);
> inputDocumentList.add(inputDoc);
> if(inputDocumentList.size()==batchSize) {
> index(isSolrCloud, collection, commitWithin, contentStreamPath, requestParams, inputDocumentList);
> inputDocumentList = new ArrayList<>();
> }
> index(isSolrCloud, collection, commitWithin, contentStreamPath, requestParams, inputDocumentList);
> }
> I think this second index method shoud outside of while ,and I don't catch you why must new ArrayList after first index, index method had cleared the SolrInputDocument List.{code}



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)