Hi All
I don't know if this is a bug or expected behaviour, but regardless I
think it should be handled a little better as it is a nightmare to
debug...
I have a situation where I have two tables as follows:
<object name="Agency">
<hasMany name="Agency">
<relate from="AgencyUUID" to="ParentAgencyUUID"/>
</hasMany>
<hasMany name="SON">
<link name="SON_Agency"/>
</hasMany>
</object>
<object name="Contract">
<hasMany name="Contract">
<relate from="ContractUUID" to="parentContractUUID"/>
</hasMany>
<hasMany name="Agency">
<link name="Contract_Agency"/>
</hasMany>
</object>
<object name="Contract_Agency">
<hasOne name="Contract">
<relate from="ContractUUID" to="ContractUUID"/>
</haContracte>
<hasOne name="Agency">
<relate from="AgencyUUID" to="AgencyUUID"/>
</hasOne>
</object>
In plain english, I've got a recursive heirarchy of agencies, each
agency can have many contracts and contracts can be amended repeatedly
by posting a new contract with a parentContractUUID pointing back at
the original contract.
The problem is that in some conditions the iterator tries to create
the join from Agency to Contract using the Contract_Agency table, but
it gets mixed up between the <hasOne name="Contract"> on the
Contract_Agency table and the <hasMany name="Contract"> on the
Contract table.
The end result is an "Field Does Not Exist: The field
'parentContractUUID' does not exist for the 'Contract_Agency' object."
error in the abstractMetadata.cfc.
I can work around this easily by aliasing the contract amendment
relationship (<hasMany name="Contract" alias="ContractAmendment">).
As far as I can tell the code that makes the leap from one
relationship to the other is:
<!--- find the relationship to/from this object --->
<cfif
getObjectMetadata().hasRelationship(arguments.relationshipAlias)>
<cfset relationship =
getObjectMetadata().getRelationship(arguments.relationshipAlias) />
<cfelseif
ToObjectMetadata.hasRelationship(arguments.relationshipAlias)>
<cfset relationship =
ToObjectMetadata.getRelationship(arguments.relationshipAlias) />
...in object.cfc (around line 177).
My main concern is that it has take us a long time to work it all out
- others could easily fall into the same trap and when they do its a
long road to working out what went wrong.
What do you guys think? Should this be improved? If so how?
--
Mark Stanton
Gruden Pty Ltd
http://www.gruden.com
-- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
Reactor for ColdFusion Mailing List
[email protected]
Archives at: http://www.mail-archive.com/reactor%40doughughes.net/
-- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
