Issue with xsd:maxInclusive restriction for 'float' type

Hi Experts,
My first question in the forum.

I've an XML file and an XSD to validate that. I've an element 'Duration' for which I've given a restriction of range. For eg [32.25, 128000]. I've defined the element as a simply type. Pls see  below,

      <xsd:restriction base="xsd:float">
           <xsd:minInclusive value="32.25"/>
            <xsd:maxInclusive value="128000"/>

But the problem is with values. If I give values like 128000.001, 128000.002, 12800.003 and validate against the schema 'no errors are reported!!'. But if I give a value 128000.004 or more I'm getting 'cvc-maxInclusive-valid' failure, which is what I need.
1) I'm using Oxygen 11.1 version.
2) The problem starts with values with 3 or more fractional digits.
Could u pls tell me wr the problem is?
Who is Participating?
Geert BormansConnect With a Mentor Information ArchitectCommented:
The problem is in the precision of a xs:float
If this is important for you at that level of precision,
you should use xs:double
(I wonder actually why you would use xs:float after all)

I just had a look at the spec:
There I can read:
The value of maxInclusive ·must· be in the ·value space· of the ·base type·.
And from the definition, i get:
value = anySimpleType

My guess would be that in your xsd 128000 is an integer and therefore the check is done in the integer value-space.
It's just a thing worth trying. Change your rule to this and see if it works:
      <xsd:restriction base="xsd:float">
           <xsd:minInclusive value="32.25"/>
            <xsd:maxInclusive value="128000.0"/>

Open in new window

vipinreyarothAuthor Commented:
Hi Gertone, ChristoferDutz thanks for your inputs.
1) @Gertone: Changing to 'xsd:double' works with the precision of 11 digits. But I've to keep the type as 'float' ( that's a requirement). So I think restricting the number of fractions (upto 2 digits for floats) using a pattern would help.
2) @ChristoferDutz: I tried, but it doesn't work. :(
Cloud Class® Course: MCSA MCSE Windows Server 2012

This course teaches how to install and configure Windows Server 2012 R2.  It is the first step on your path to becoming a Microsoft Certified Solutions Expert (MCSE).

ChristoferDutzConnect With a Mentor Commented:
Ahh ... dammit ... gertone is correct.
The problem is the number is internally represented as 32bit floating point number which means that it can store about 8 digits + their offset ... if you would set the upper limit to 128 instead of 128000 then you would get precision for 3 more digits in the fractional part. The problem is that when parsing the number as float the fractional part is simply stripped off and therefore cannot be used for comparison.
Geert BormansInformation ArchitectCommented:
I was in the car for the past hour or so, so I missed your response.
You have conflicting requirements, you simply can't get the precision you want from an xs:float.

If you really want to use xs:float
and exclude some values at the high end,
use maxExclusive instead of maxInclusive (you will loose some valid values, but at least you get warned, being too strict is always better than not being strict enough)

Or use xs:double for the calculation in the schema and add a number type conversion after the validation in whatever technology you use to process the XML
At least your validation then happens on maximum precision
vipinreyarothAuthor Commented:
Thanks Gertone. It'll do.
Geert BormansInformation ArchitectCommented:
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

All Courses

From novice to tech pro — start learning today.