it is safe to use different column names as Shalin suggested. After
all a row is a map with the column value as the key. If you map
multiple values to the same column it may overwrite each other. use
explicit 'name' attributes
On Wed, Feb 4, 2009 at 2:17 AM, Shalin Shekhar Mangar
<shalinman...@gmail.com> wrote:
> On Wed, Feb 4, 2009 at 1:35 AM, Fergus McMenemie <fer...@twig.me.uk> wrote:
>
>> <entity name="x"
>> dataSource="myfilereader"
>> processor="XPathEntityProcessor"
>> url="${jc.fileAbsolutePath}"
>> stream="false"
>> forEach="/record">
>> <field column="para" xpath="/record/sect1/para" />
>> <field column="para" xpath="/record/list/listitem/para" />
>> <field column="para" xpath="/a/b/c/para" />
>> <field column="para" xpath="/d/e/f/g/para" />
>>
>> Below is the line from my schema.xml
>>
>> <field name="para" type="text" indexed="true" stored="true"
>> multiValued="true"/>
>>
>> Now a given document will only have one style of layout, and of course
>> the /a/b/c /d/e/f/g stuff is made up. For a document that has a single
>> <para>Hello world</para> element I see search results as follows, the
>> one <para> string seems to have been entered into the index four times.
>> I only saw duplicate results before adding the extra made-up stuff.
>>
>>
> I think there is something fishy with the XPathEntityProcessor. For now, I
> think you can work around by giving each field a different 'column' and
> attribute 'name=para' on each of them.
>
> --
> Regards,
> Shalin Shekhar Mangar.
>
--
--Noble Paul
