manifoldcf-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Prasad Perera (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (CONNECTORS-996) Handle Delete documents in CMIS/Alfresco/Google Driver Connectors
Date Wed, 16 Jul 2014 16:20:04 GMT

    [ https://issues.apache.org/jira/browse/CONNECTORS-996?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14063676#comment-14063676
] 

Prasad Perera commented on CONNECTORS-996:
------------------------------------------

Hi Karl,

I meant setting the rval[i] to null in exception catch, which picks up by the  WorkerThread
as a deleted document (null version) for a re-crawling job (if I'm not mistaken) in MCF version
1.6.

If this a proper way to do it, It will be nice to see this in 1.6 and future versions.
 

> Handle Delete documents in CMIS/Alfresco/Google Driver Connectors
> -----------------------------------------------------------------
>
>                 Key: CONNECTORS-996
>                 URL: https://issues.apache.org/jira/browse/CONNECTORS-996
>             Project: ManifoldCF
>          Issue Type: Bug
>          Components: Alfresco connector, CMIS connector, GoogleDrive connector
>    Affects Versions: ManifoldCF 1.6.1
>            Reporter: Prasad Perera
>            Assignee: Karl Wright
>            Priority: Minor
>             Fix For: ManifoldCF 1.7
>
>
> Steps to reproduce:
> Create a repository connector using CMIS connector.
> Create an incremental crawling job and let document get crawled one time.
> In the next re-crawl interval, delete one or more documents from the repository end.
You may start seeing below logs:
> FATAL 2014-07-16 11:10:58,376 (Worker thread '19') - Error tossed: Object not found:
workspace://SpacesStore/362c30aa-cf01-4f24-aa4a-8b17ef0375c5;1.0
> org.apache.chemistry.opencmis.commons.exceptions.CmisObjectNotFoundException: Object
not found: workspace://SpacesStore/362c30aa-cf01-4f24-aa4a-8b17ef0375c5;1.0
>         at org.apache.chemistry.opencmis.client.bindings.spi.atompub.AbstractAtomPubService.convertStatusCode(AbstractAtomPubService.java:462)
>         at org.apache.chemistry.opencmis.client.bindings.spi.atompub.AbstractAtomPubService.read(AbstractAtomPubService.java:601)
>         at org.apache.chemistry.opencmis.client.bindings.spi.atompub.AbstractAtomPubService.getObjectInternal(AbstractAtomPubService.java:824)
>         at org.apache.chemistry.opencmis.client.bindings.spi.atompub.ObjectServiceImpl.getObject(ObjectServiceImpl.java:608)
>         at org.apache.chemistry.opencmis.client.runtime.SessionImpl.getObject(SessionImpl.java:410)
>         at org.apache.chemistry.opencmis.client.runtime.SessionImpl.getObject(SessionImpl.java:388)
>         at org.apache.manifoldcf.crawler.connectors.cmis.CmisRepositoryConnector.getDocumentVersions(CmisRepositoryConnector.java:1261)
>         at org.apache.manifoldcf.crawler.connectors.BaseRepositoryConnector.getDocumentVersions(BaseRepositoryConnector.java:362)
>         at org.apache.manifoldcf.crawler.connectors.BaseRepositoryConnector.getDocumentVersions(BaseRepositoryConnector.java:343)
>         at org.apache.manifoldcf.crawler.connectors.BaseRepositoryConnector.getDocumentVersions(BaseRepositoryConnector.java:322)
>         at org.apache.manifoldcf.crawler.connectors.BaseRepositoryConnector.getDocumentVersions(BaseRepositoryConnector.java:299)
> org.apache.chemistry.opencmis.commons.exceptions.CmisObjectNotFoundException is thrown
for the seeded documents which got deleted. 
> Would it be proper to catch this exception at getDocumentVersions as :
> while (i < rval.length) {
>             try {
>                 CmisObject cmisObject = session.getObject(documentIdentifiers[i]);
> //
> // Rest of the code 
> //
>             } catch (CmisObjectNotFoundException e) { 
>                 rval[i] = null;
>             }
> }
> And forward document delete call for the output connectors ?



--
This message was sent by Atlassian JIRA
(v6.2#6252)

Mime
View raw message