[OASIS Issue Tracker] (ODATA-554) Clarify representation of floating-point numbers

[OASIS Issues Tracker <workgroup_mailer@lists.oasis-open.org>]

     [ https://tools.oasis-open.org/issues/browse/ODATA-554?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Ralf Handl updated ODATA-554:
-----------------------------

    Component/s: Implementing OData

> Clarify representation of floating-point numbers
> ------------------------------------------------
>
>                 Key: ODATA-554
>                 URL: https://tools.oasis-open.org/issues/browse/ODATA-554
>             Project: OASIS Open Data Protocol (OData) TC
>          Issue Type: Improvement
>          Components: Implementing OData
>    Affects Versions: V4.0_CSD03
>         Environment: [Proposed]
>            Reporter: Ralf Handl
>             Fix For: V4.0_WD01
>
>
> We specify Edm.Double and Edm.Single to be IEEE754 binary floating-point numbers with 64 and 32 bits.
> Unfortunately the IEEE754 specifications are not freely available, and their publicly available excerpts in e.g. Wikipedia and XML Schema 1.1 Part 2 seem to diverge in the fine points of
> - normalized and denormalized values
> - smallest and largest positive and negative values
> Also we don't give guidance like "always produce normalized values, always accept all values and round appropriately".



--
This message was sent by Atlassian JIRA
(v6.2.2#6258)