The behavior you describe is how optional is defined to work when
marshalling - if the value is the default ("false", in the case of a
boolean), it is not generated in the output. To get the behavior you
want, you'd need to have a separate flag to indicate the presence or
absence of the element. You can use the test-method and flag-method
attributes (part of the property group:
http://jibx.sourceforge.net/binding/binding-attributes.html#property) to
work with this added flag in the binding.
- Dennis
Dennis M. Sosnoski
XML and Web Services in Java
Training and Consulting
http://www.sosnoski.com - http://www.sosnoski.co.nz
Seattle, WA +1-425-939-0576 - Wellington, NZ +64-4-298-6117
[email protected] wrote:
> Hello:
>
> I have a document that requires serialization of boolean values
> in a 'yes' and 'no' format. I have defined a format in my
> binding.xml as follows to accomplish this trivial conversion:
>
>
> <format type="boolean" label="yesNoBoolean"
> serializer="ConversionUtils.serializeYesNoBoolean"
> deserializer="ConversionUtils.deserializeYesNoBoolean"/>
>
> I then reference the format in my mapping as follows:
>
> <value style="element" name="isGateway"
> get-method="isGateway" set-method="setGateway"
> format="yesNoBoolean" usage="optional"/>
>
> When I test a test-case input document via the JiBX roundtrip or
> my own JUnit test case this works just fine if (and only if) the
> underlying boolean is true. As soon as I test with a false
> value, the serialization process no longer creates the associated
> element. I presume that this may be some sort of optimization on
> part of the usage=?optional? attribute attempting to optimize the
> output or something? However, this causes the roundtrip to fail
> as well as my JUnit test case which are both expecting to see the
> attribute with a specified value.
>
> I have debuged the yesNoBoolean serializer and deserializers and
> have therefore verified that the serializers are never called
> when the underlying value is false (only when true) and that they
> are performing as would be expected otherwise and without failure.
>
> I also tried specifying a default=?no? on the format. While this
> did not produce any errors in and of itself, it also did not fix
> the roundtrip or JUnit test case failures.
>
> Finally, and most telling, if I remove the usage=?optional? from
> the value statement in the mapping, all works fine and as
> expected. However, this is not what I need, as this element
> value is truly optional.
>
> Is this a known feature/bug relative to the combination of format
> and optional values? Am I missing something else? Is there a
> way to achieve the truly optional behavior for the value, while
> using the yeNoBoolean format.
>
> TIA!
>
>
> ------------------------------------------------------------------------------
> Crystal Reports - New Free Runtime and 30 Day Trial
> Check out the new simplified licensing option that enables unlimited
> royalty-free distribution of the report engine for externally facing
> server and web deployment.
> http://p.sf.net/sfu/businessobjects
> _______________________________________________
> jibx-users mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/jibx-users
>
>
------------------------------------------------------------------------------
Are you an open source citizen? Join us for the Open Source Bridge conference!
Portland, OR, June 17-19. Two days of sessions, one day of unconference: $250.
Need another reason to go? 24-hour hacker lounge. Register today!
http://ad.doubleclick.net/clk;215844324;13503038;v?http://opensourcebridge.org
_______________________________________________
jibx-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/jibx-users
