I'm using solr 1.3 and am trying to get a delta-import with the DIH.
Recently the wiki, http://wiki.apache.org/solr/DataImportHandler, was
updated explaining that delta import is a 1.4 feature now but it was
still possible get a delta using the full import example here,
http://wiki.apache.org/solr/DataImportHandlerFaq#fullimportdelta. I
tried this but each time I run DIH, it reimports all rows and updates.
Below is my data-config.xml. I set rootEntity to false and issued
command=full-import&clean=false&optimize=false through DIH. Am I
doing something wrong here or is the DataImportHandlerFaq incorrect?
<dataConfig>
<dataSource driver="com.mysql.jdbc.Driver"
url="jdbc:mysql://pencil-somewhere.com:22222/SomeDB" user="someUser"
password="somePassword"/>
<document name="">
<entity name = "item" rootEntity="false"
query = "select DId from 2_Doc where
ModifiedDate > '${dataimporter.last_index_time}'
and DocType != 'Research Articles'">
<entity name="feature" pk="DId"
transformer="RegexTransformer"
query = "SELECT d.DId, d.SiteId,
d.DocTitle, d.DocURL, d.DocDesc,
d.DocType, d.Tags, d.Source,
d.Last90DaysRFIsPercent,
d.ModifiedDate, d.DocGuid, d.Author,
i.Industry FROM 2_Doc d LEFT
OUTER JOIN tmp_DocIndustry i
ON (d.DocId=i.DocId AND
d.SiteId=i.SiteId) where d.DocType != 'Research articles'
and d.DId = '${item.DId}' and
d.ModifiedDate > '${dataimporter.last_index_time}'">
<field column = "DId" name ="did"/>
<field column = "SiteId" name ="SiteId"/>
<field column = "DocId" name ="DocId"/>
<field column = "DocTitle" name ="DocTitle"/>
<field column = "DocURL" name ="DocURL"/>
<field column = "DocDesc" name ="DocDesc" />
<field column = "Snippet"
regex="^(.{0,800})\b.*$" sourceColName="DocDesc"/>
<field column = "DocType" name ="DocType"/>
<field column = "Tags" name ="Tags"
splitBy=";" sourceColName="Tags"/>
<field column = "Source" name ="Source"/>
<field column =
"Last90DaysRFIsPercent" name ="Last90DaysRFIsPercent"/>
<field column = "ModifiedDate" name
="ModifiedDate"/>
<field column = "DocGuid" name ="DocGuid"/>
<field column = "Author" name ="Author"/>
<field column = "Industry" name
="Industry" sourceColName="Industry"/>
</entity>
</entity>
</document>
</dataConfig>
Thanks,
-Tim
