Solr Schema Design & Optimization

The schema for the core needs to be properly designed and optimized based on the following specifications.

The dataConfig

<dataConfig>

<dataSource type="JdbcDataSource"

driver="[login to view URL]"

url="jdbc:jtds:sqlserver:////………….." />

<document>

<entity name="parcels"

pk="PIN_18"

query="select PIN_18,[login to view URL](PIN_18, '[-.]', '') as StrippedPIN, [Owner] as OwnerName, RTRIM(ISNULL(Owner_Street, '') + ' ' + ISNULL(Owner_City_ST_ZIP, '')) as OwnerAddress, RTRIM(ISNULL(Property_Street, '') + ' ' + ISNULL(Property_City_ST_ZIP, '')) as PropertyAddress from [login to view URL]"

deltaImportQuery="select PIN_18, [login to view URL](PIN_18, '[-.]', '') as StrippedPIN, [Owner] as OwnerName, RTRIM(ISNULL(Owner_Street, '') + ' ' + ISNULL(Owner_City_ST_ZIP, '')) as OwnerAddress, RTRIM(ISNULL(Property_Street, '') + ' ' + ISNULL(Property_City_ST_ZIP, '')) as PropertyAddress from [login to view URL] WHERE id='${[login to view URL]}'"

deltaQuery="SELECT PIN_18 FROM [login to view URL] WHERE changed = 1"

>

</entity>

</document>

</dataConfig>

Current Schema File: Attached

Requirements :

A single search term is passed to the Solr Core like below

https://[[login to view URL]]/solr/parcelsmonroein/select?q=radhakrish*&wt=json&rows=10&indent=true&hl=true&hl.fl=*&hl.simple.pre=%3Cem%3E&hl.simple.post=%3C%2Fem%3E

In this case the search term is radhakrish*

The search is performed across all fields and returns results with highlights. This works well for wildcard searches even though the schema is not optimized.

We would like to

a. optimize the schema

b. improve the search for "phonetic" matches such as

Angela - Angie

Mike - Michael

Toni or Tony - Anthony

and if possible

7th - seventh

The name phonetic matches will apply for Owner names and the numerical for addresses.

Sample dataset is attached

Apache Solr

ID do Projeto: #10096196

Sobre o projeto