On 5/11/05, David Easley <[EMAIL PROTECTED]> wrote:
> I expected my case to go south of the else clause but instead it went north
> and used the 3 param version of _infoFactory.createCollection which always
> uses the default collection type.
I'm experiencing the same behavior with my test case and it's due to
the fact that the xmlType in this snippet (line 244):
XMLType xmlType = component.getXMLType();
is not able to be retreived. Therefore the code block wrapped in this
condition:
if (xmlType != null) {
is never entered causing the boolean named simpleTypeCollection is
never set to true hence the lack of recognition of the collection
type.
> Here's a very suspicious code snippet I noticed along the way:
>
> org.exolab.castor.builder.FieldInfoFactory:
> 117 public CollectionInfo createCollection
> 118 (XSType contentType, String name, String elementName, String
> collectionName)
> 119 {
> 120
> 121 String temp = collectionName;
> 122 if (temp == null || temp.length() >0)
> 123 temp = _default;
> 124
> 125 CollectionInfo cInfo = null;
> 126 if (temp.equals(VECTOR)) {
> 127 cInfo = new CollectionInfo(contentType,name,elementName);
> 128 } else if (temp.equals(ARRAY_LIST)) {
> 129 cInfo = new CollectionInfoJ2(contentType,name,elementName);
> 130 } else if (temp.equals(ODMG)) {
> 131 cInfo = new
> CollectionInfoODMG30(contentType,name,elementName);
> 132 }
> ...
>
> Surely line 122 should be:
>
> 122 if (temp == null || temp.length() == 0)
>
> i.e. the comment for the 121-123 block would be
> // If no collection name has been specified, use the default
Yes, this is definitely incorrect and should be patched. I can do
this, but I'd like to commit both fixes at the same time.
> I tentatively fixed this, but SourceGenerator still took no notice of the
> member element's collection attribute (collectionName and _default and both
> "vector", despite me specifying some other type in the bindings file).
>
> So, two questions:
> 1. Does anyone know how to make SourceGenerator take any notice of the
> member element's collection attribute?
It appears that this is either a bug or a missing feature. I need
either Andrew of Keith to chime in here simply because they're more
experienced in this are of the code.
> 2. Do any Castor commiters think line 122 looks wrong?
Yes, see my comment above on this one.
Bruce
--
perl -e 'print unpack("u30","D0G)[EMAIL
PROTECTED]&5R\"F)R=6-E+G-N>61E<D\!G;6%I;\"YC;VT*"
);'
The Castor Project
http://www.castor.org/
Apache Geronimo
http://geronimo.apache.org/
-------------------------------------------------
If you wish to unsubscribe from this list, please
send an empty message to the following address:
[EMAIL PROTECTED]
-------------------------------------------------
