[
https://issues.apache.org/jira/browse/BLUR-1?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13437791#comment-13437791
]
Aaron McCurry edited comment on BLUR-1 at 8/20/12 10:45 PM:
------------------------------------------------------------
To allow for types now before any API changes are made, the analyzer class name
could allow for defining types to be indexed. Granted the data will still have
to be passed as a string and stored as a string but would be indexed the given
type. This should allow the support of all the given in this task.
These are the following fields in the API where classnames for analyzers can be
defined.
ColumnDefinition.analyzerClassName
AlternateColumnDefinition.analyzerClassName
AnalyzerDefinition.fullTextAnalyzerClassName
For example:
ColumnDefinition.analyzerClassName = "string"
instead of:
ColumnDefinition.analyzerClassName =
"org.apache.lucene.analysis.standard.StandardAnalyzer"
And for Long:
ColumnDefinition.analyzerClassName = "long"
or
ColumnDefinition.analyzerClassName = "long,4"
- The number 4 there would indicate the precision step.
was (Author: amccurry):
To allow for types now before any API changes are made, the analyzer class
name could allow for defining types to be indexed. Granted the data will still
have to be passed as a string and stored as a string but would be indexed the
given type. This should allow the support of all the given in this task.
These are the following fields in the API where classnames for analyzers can be
defined.
ColumnDefinition.analyzerClassName
AlternateColumnDefinition.analyzerClassName
AnalyzerDefinition.fullTextAnalyzerClassName
> Create Formal Types
> -------------------
>
> Key: BLUR-1
> URL: https://issues.apache.org/jira/browse/BLUR-1
> Project: Apache Blur
> Issue Type: New Feature
> Reporter: Aaron McCurry
>
> Currently all data in Blur has to be represented as a String, though
> different analyzers can be used to provide the effect of types. The Thrift
> RPC will have to changed to accommodate this new feature. The initial list
> of types that should be supported are:
> -String
> -Long
> -Date
> -Int
> -Double
> -Float
> -GIS (Lat + Long)
> -Binary
> -Security Label
> -Custom
> The security label would allow for a formal implementation of access control
> per document (Record in current RPC API terminology).
> The custom type, would allow for creating your own named type that can be
> installed into Blur and then referenced by name in the RPC.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
