SourceForge : artf6183: loop over decimal / float values

Project: JSDL-WG Trackers > [CLOSED] Parameter Sweep > View Artifact

Artifact artf6183 : loop over decimal / float values

Tracker:	[CLOSED] Parameter Sweep
Title:	loop over decimal / float values
Description:	Many variables for sweeping are decimal / float type values e.g. 0.1, 0.2, 0.3, 0.4 and so could appear in a looping function. Additional data such as rounding rules (up, down, truncate) and significant digits / figures may be specified.
Submitted By:	geoff williams
Submitted On:	04/03/2008 6:30 AM EDT
Last Modified:	06/25/2008 10:00 AM EDT
Closed:	06/25/2008 10:00 AM EDT

	Status / Comments		Change Log		Associations		Attachments

Status
Group: *
Status:*	Closed
Category: *
Customer: *
Priority: *	4
Assigned To: *	geoff williams
Reported in Release: *
Fixed in Release: *
Estimated Hours: *	0
Actual Hours: *	0

Comments

Andreas Savva: 06/25/2008 10:00 AM EDT
	Comment:	Same as artf6184
	Action:	Update Closed set to 06/25/2008 Status changed from Pending to Closed

geoff williams: 05/14/2008 12:14 PM EDT
	Comment:	see http://forge.ogf.org/sf/wiki/do/viewPage/projects.jsdl-wg/wiki/ParameterSweepExtensionLooping
	Action:	Update

Andreas Savva: 04/23/2008 11:03 AM EDT
	Comment:	Agreed in principle to do floats. details pending interoperability note required for implementors. we only care about lexical representation
	Action:	Update Status changed from Open to Pending

geoff williams: 04/17/2008 11:05 AM EDT

Comment:

Presumably we need to offer to users the ability to specify if they require float (32bit) or double (64bit).

xsd:decimal http://www.w3.org/TR/2001/REC-xmlschema-2-20010502/#decimal
xsd:float http://www.w3.org/TR/2001/REC-xmlschema-2-20010502/#float
xsd:double http://www.w3.org/TR/2001/REC-xmlschema-2-20010502/#double

xsd:float and xsd:double types allow "INF", "-INF", "NaN", 0 (positive 0), -0 (negative 0).
xsd:float and xsd:double allow scientific notation, xsd:decimal does not.

I'd prefer INF, -INF and NaN to be disallowed for the start and end attributes, and in addition 0 and -0 to also be disallowed for the step attribute.
 Trouble is that I can't find an exclude instruction in XML schema to define this.

I also can't find a way of avoiding the duplicated structure in the examples below, in that it would be nice to specify a generic element, e.g. 
LoopGeneric, with implementations differing only in their xsd type.

Float:

<xsd:element name="LoopFloat" substitutionGroup="sweep:Function">
  <xsd:complexType>
    <xsd:sequence>
      <xsd:element name="Exception" type="xsd:float"
                   maxOccurs="unbounded" minOccurs="0" />
    </xsd:sequence>

    <xsd:attribute name="start" type="xsd:float" use="required" />
    <xsd:attribute name="end" type="xsd:float" use="required" />
    <xsd:attribute name="step" type="xsd:float" use="required" />
  </xsd:complexType>
</xsd:element>

example:

<LoopFloat start="12.45e-2" end="12.89e-2" step="1e-3" />

12.45e-2 12.55e-2 12.65e-2 12.75e-2 12.85e-2

Double:

<xsd:element name="LoopDouble" substitutionGroup="sweep:Function">
  <xsd:complexType>
    <xsd:sequence>
      <xsd:element name="Exception" type="xsd:double"
                   maxOccurs="unbounded" minOccurs="0" />
    </xsd:sequence>

    <xsd:attributeGroup>
      <xsd:attribute name="start" type="xsd:double" use="required" />
      <xsd:attribute name="end" type="xsd:double" use="required" />
      <xsd:attribute name="step" type="xsd:double" use="required" />
    </xsd:attributeGroup>
  </xsd:complexType>
</xsd:element>

example:

<LoopDouble start="0.001" end="0.010" step="0.001">
  <Exception>0.003</Exception>
</LoopDouble>

0.001, 0.002, 0.004, 0.005, 0.006, 0.007, 0.008, 0.009, 0.010

Action:

Update

geoff williams: 04/03/2008 1:49 PM EDT
	Action:	Update Assigned To set to geoff williams

geoff williams: 04/03/2008 6:30 AM EDT
	Action:	Create

Return

< Previous

Next >