[
https://issues.apache.org/jira/browse/THRIFT-964?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12924791#action_12924791
]
Bryan Duxbury commented on THRIFT-964:
--------------------------------------
If it's just the memory size you're looking to improve on, then I might support
adding a type annotation that allowed a given double field to be represented in
memory as a float. However, you'd still need to decode it as a double then
truncate it into the float. This will probably take extra time, but it will
save you the bytes. That is, unless you're in a 64-bit JVM... I'm not sure how
much memory a float field takes in a class as compared to a double. It probably
depends greatly on the other fields.
> Support for single precision floating point numbers
> ---------------------------------------------------
>
> Key: THRIFT-964
> URL: https://issues.apache.org/jira/browse/THRIFT-964
> Project: Thrift
> Issue Type: New Feature
> Components: Compiler (General)
> Reporter: Arvind Jayaprakash
>
> We have an application that uses "float", the IEEE single precision numbers.
> The reasons are the two obvious ones: the lack of need for precision as high
> as "double" and data size given that we have millions of such numbers.
> I'd like to see support for this data format.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
