[ 
https://jira.nuxeo.org/browse/NXP-5506?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=80159#action_80159
 ] 

Raimund Sacherer commented on NXP-5506:
---------------------------------------

I understand. Yes you are right, from a performance and database optimization 
viewpoint ecm:fulltext = "Auto Haus" is more efficient then 'AUTO' and 'HAUS'.

The point is that my client has scanned documents, where information in these 
documents was additional entered in metadata so it can be searched. Now this 
data is address information and there are a lot of them, so my client want to 
search for example Name:Auto, Surname:Haus, Street: Autohaus and the results 
should not include datasets where Auto is also in Surname or Street or vice 
versa. At least here in spain names can be surname and name and it's not so 
infrequent like it's in german for example, where the distinction of first and 
lastname is more clear.

The point of doing fulltext search instead of LIKE:
case insensitivity
accents get stripped automatically through the fulltext conversation
two names are easier found in surnames (everyone in spain has two surnames), 
streets are easier to find, etc


Of course the best solution would be a soundEX or levenstein or double 
metaphone, but I have no idea how to integrate this into searches in nuxeo ... 

> NXQL query with several ecm:fulltext references fails
> -----------------------------------------------------
>
>                 Key: NXP-5506
>                 URL: https://jira.nuxeo.org/browse/NXP-5506
>             Project: Nuxeo Enterprise Platform
>          Issue Type: Bug
>          Components: Core SQL Storage, Search / Query
>    Affects Versions: 5.3.2
>         Environment: PostgreSQL, MySQL, Oracle or SQL Server
>            Reporter: Florent Guillaume
>            Assignee: Florent Guillaume
>            Priority: Major
>             Fix For: 5.4
>
>
> When doing a NXQL query involving several ecm:fulltext, the generated SQL 
> query is incorrect and fails, because it mentions the fulltext table twice 
> without aliasing:
> SELECT "hierarchy"."id" AS "_C1"
>   FROM "hierarchy"
>   JOIN "fulltext" ON "fulltext"."id" = "hierarchy"."id"
>   JOIN "fulltext" ON "fulltext"."id" = "hierarchy"."id"
>    ...
> This does not happen with H2.

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: 
https://jira.nuxeo.org/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira

        
_______________________________________________
ECM-tickets mailing list
[email protected]
http://lists.nuxeo.com/mailman/listinfo/ecm-tickets

Reply via email to