Hello,
We just published a few advices regarding indexing data for Solr (and
whatever).
http://blog.griddynamics.com/2015/07/how-to-import-structured-data-into-solr.html.
Indexing blocks by DIH is possible with specifying child=”true”
<https://issues.apache.org/jira/browse/SOLR-5147?focusedCommentId=14263373&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-14263373>
.
On Mon, Jul 6, 2015 at 9:18 PM, SHANKAR REDDY <sankara.teluku...@gmail.com>
wrote:
> Upayavira,
> Thanks for the update. I am using Data Import handler for MySQL as below.
>
>
> <dataConfig>
> <dataSource type="JdbcDataSource"
> driver="com.mysql.jdbc.Driver"
> url="jdbc:mysql://XXXXXXXXXX:3306/XXXXX"
> user="XXXXXX"
> password="XXXXXXX" />
> <document>
> <entity name="header" query="select * from
> l_report_header_data where status_master_id = 0">
> <field column="REPORT_HEADER_ID" name="id"/>
> <uniqueKey>id</uniqueKey>
> <!--- Header Data -->
> <field column="DISPLAY_VALUE"
> name="display_value_txt" />
> <field column="DESCRIPTION" name="description_txt"
> />
> <field column="REPORT_OWNER"
> name="report_owner_txt" />
> <field column="DATA_SOURCE" name="data_source_txt"
> />
> <field column="REPORT_TYPES_ID"
> name="report_types_id_txt" />
>
> <!--detail entity -->
> <entity name="detail" query="select * from
> l_report_detail_data where REPORT_HEADER_ID = '${header.REPORT_HEADER_ID}'"
> >
> <field column="IOBJNM" name="iobjnm_txt" />
> <field column="TXTLG" name="txtlg_txt" />
> <field column="IOBJTP" name="iobjtp_txt" />
> </entity>
> </entity>
> </document>
> </dataConfig>
>
>
> Please suggest.
>
>
> -Sankara
>
>
> On Sun, Jul 5, 2015 at 11:52 PM, Upayavira <u...@odoko.co.uk> wrote:
>
> > It looks like you are indexing the detail_txt fields as multivalued.
> > Perhaps you could index them as parent/child documents, and use block
> > joins and the expand component to retrieve them.
> >
> > You can index the doc as:
> >
> > <doc>
> > <field name="id">0BICA_PROBLEM_AUTHOR</field>
> > <field name="type">parent</field>
> > ...
> > <doc>
> > <field name="id">0RS_AUTHOR</field>
> > <field name="TXTLG">Author</field>
> > <field name="IOBJTP">CHA</field>
> > <field name="type">child</field>
> > </doc>
> > <doc>...</doc>
> > </doc>
> >
> > If you want to query the parent docs, just query as normal, and add
> > expand=true to see your child docs in your results (lower down). Or, use
> > the {!parent} or {!child} queries to select documents based upon
> > parent/child relationships.
> >
> > Upayavira
> >
> > On Mon, Jul 6, 2015, at 04:41 AM, SHANKAR REDDY wrote:
> > > Team,
> > > I have a requirement like getting the list of the child tables along
> > > with
> > > parent records as the below pattern.
> > >
> > > "id":"0BICA_PROBLEM_AUTHOR",
> > > "description_txt":[""],
> > > "display_value_txt":["Analyser : Problems/Category/Author"],
> > > "report_types_id_txt":["1"],
> > > "report_categories_id_txt":["30585"],
> > > "category_name_txt":["0RS_RPODS"],
> > > icon_url_txt":["images/icon_report_crystal.png"]
> > > "detail_txt":[{"IOBJNM": "0RS_AUTHOR","TXTLG":"Author","IOBJTP":"CHA"},
> > > {"IOBJNM": "0RS_CCAT","TXTLG":" CheckCategory (inte","IOBJTP":" CHA"},
> > > {"IOBJNM": "0RS_PRIORTY","TXTLG":"Priority CHA","IOBJTP":""},
> > > {"IOBJNM": "PRIOSEL","TXTLG":"Priority Selection","IOBJTP":"KFG""}]
> > > }
> > >
> > > From the above detail_txt child entity ( Child Table). With the current
> > > and
> > > default implementation, I am getting the response as below.
> > >
> > >
> > > {
> > > "data_source_txt":["0RS_RPOD"],
> > > "id":"0BICA_PROBLEM_AUTHOR",
> > > "description_txt":[""],
> > > "display_value_txt":["Analyser : Problems/Category/Author"],
> > > "report_types_id_txt":["1"],
> > > "report_categories_id_txt":["30585"],
> > > "category_name_txt":["0RS_RPODS"],
> > > "icon_url_txt":["images/icon_report_crystal.png"],
> > > "report_type_display_value_txt":["Bex"],
> > > "iobjtp_txt":["CHA",
> > > "CHA",
> > > "CHA",
> > > "KFG"],
> > > "txtlg_txt":["Author",
> > > "CheckCategory (inte",
> > > "Priority",
> > > "Priority Selection"],
> > > "iobjnm_txt":["0RS_AUTHOR",
> > > "0RS_CCAT",
> > > "0RS_PRIORTY",
> > > "PRIOSEL"],
> > > "_version_":1505820339930660864}
> > >
> > > As seen above I am getting the array of columns for each of fields
> > > instead
> > > of each row in the entity as a block.
> > > Please provide your suggestions on this. I am using latest version of
> > > solr
> > > APACHE SOLR™ 5.2.1
> > >
> > > Please provide your suggestion to meet the requirement.
> > >
> > > -Shankar
> >
>
--
Sincerely yours
Mikhail Khludnev
Principal Engineer,
Grid Dynamics
<http://www.griddynamics.com>
<mkhlud...@griddynamics.com>
