Hi Steve,

thank you for your reply, it been quite long time to reply to you back.

i have tried what you suggested, and there were some improvements in terms
of searching and getting the results.

however, the team is facing some difficulty in searching using shortcut of
the indexed names which forced us to revert back the suggested changes..

below are the examples which we are facing:


---------------------------------
*Example 1:*

*Indexed Text*
بنك مسقط - مركز الاتصال

*Searched*
مسقط الات

*Remarks of Example 1*
unable to get the indexed result unless I typed the two words fully (مسقط
الاتصال)


{
  "responseHeader": {
    "status": 0,
    "QTime": 0,
    "params": {
      "indent": "true",
      "q": "businessNmBl:(مسقط الات)",
      "_": "1499758511717",
      "wt": "json"
    }
  },
  "response": {
    "numFound": 0,
    "start": 0,
    "docs": []
  }
}


---------------------------------

*Example 2:*

*Indexed Text
*الطيران العماني - مركز الاتصال

*Searched*
الطير الات

*Remarks*
unable to get the indexed result unless I typed the two words fully (الطيران
الاتصال)


{
  "responseHeader": {
    "status": 0,
    "QTime": 2,
    "params": {
      "indent": "true",
      "q": "businessNmBl:(طير الات)",
      "_": "1499758649600",
      "wt": "json"
    }
  },
  "response": {
    "numFound": 0,
    "start": 0,
    "docs": []
  }
}



Please be noted that the existing configuration (which we are facing
problems with Hamzzh (ء) and etc. )  on production is working with the above
examples. its not working only once we implement your suggested
configuration. 

Thanks in advance





--
View this message in context: 
http://lucene.472066.n3.nabble.com/Arabic-words-search-in-solr-tp4317733p4345392.html
Sent from the Solr - User mailing list archive at Nabble.com.

Reply via email to