why is hibernating substituting primary key projectReleasePK with "id"?
----------------------------------------------------------------------------------
DEBUG [15:26:19,023] [] [net.sf.hibernate.impl.SessionImpl:1328] - find: from prjtsk in class ProjectReleaseTaskSubTask where prjtsk.projectReleaseTaskSubTaskPK.projectRelease.projectReleasePK.project.prjCode = '1130'
DEBUG [15:26:19,083] [] [net.sf.hibernate.hql.QueryTranslator:144] - compiling query
net.sf.hibernate.QueryException: could not resolve property type: projectReleaseTaskSubTaskPK.projectRelease.id.project.id [from prjtsk in class ProjectReleaseTaskSubTask where prjtsk.projectReleaseTaskSubTaskPK.projectRelease.projectReleasePK.project.prjCode = '1130']

net.sf.hibernate.QueryException: could not resolve property type: projectReleaseTaskSubTaskPK.projectRelease.id.project.id [from prjtsk in class ProjectReleaseTaskSubTask where prjtsk.projectReleaseTaskSubTaskPK.projectRelease.projectReleasePK.project.prjCode = '1130']
at net.sf.hibernate.hql.PathExpressionParser.getPropertyType(PathExpressionParser.java:235)
at net.sf.hibernate.hql.PathExpressionParser.end(PathExpressionParser.java:281)
at net.sf.hibernate.hql.WhereParser.doPathExpression(WhereParser.java:366)
at net.sf.hibernate.hql.WhereParser.doToken(WhereParser.java:393)
at net.sf.hibernate.hql.WhereParser.token(WhereParser.java:279)
at net.sf.hibernate.hql.ClauseParser.token(ClauseParser.java:87)
at net.sf.hibernate.hql.PreprocessingParser.token(PreprocessingParser.java:120)
at net.sf.hibernate.hql.ParserHelper.parse(ParserHelper.java:29)
at net.sf.hibernate.hql.QueryTranslator.compile(QueryTranslator.java:146)
at net.sf.hibernate.hql.QueryTranslator.compile(QueryTranslator.java:133)
at net.sf.hibernate.impl.SessionFactoryImpl.getQuery(SessionFactoryImpl.java:352)
at net.sf.hibernate.impl.SessionFactoryImpl.getQuery(SessionFactoryImpl.java:330)
at net.sf.hibernate.impl.SessionImpl.getQueries(SessionImpl.java:1368)
at net.sf.hibernate.impl.SessionImpl.find(SessionImpl.java:1332)
at net.sf.hibernate.impl.SessionImpl.find(SessionImpl.java:1322)
at net.sf.hibernate.impl.SessionImpl.find(SessionImpl.java:1314)

Looks like hibernate doesn't like long names and replaces randomly the primary key class names with "id".
Is there a limit defined somewhere to the length of the query name?

Now this is strange. When i change the query to clip off project.prjCode and use only till prjtsk.projectReleaseTaskSubTaskPK.projectRelease.projectReleasePK and fetch for one of its attributes like
prjtsk.projectReleaseTaskSubTaskPK.projectRelease.projectReleasePK.title,
it works fine but as soon as project.prjCode is added, it starts to substitute the **PK's with id's.

Any suggestions?
I've tried shortenening the names of my classes as well still the same.

I guess this is a mapping bug.
If you post a minimzed example with the information asked by the red box, maybe someone can spot the problem. Often I find bugs while preparing this information for a problem!

HTH
Ernst

The mapping is fine as you can see from the "from" clause. Its only when its compiling the query, hibernate is somehow replacing the primary key attributes with just "id".

With the same mapping if i just query the "Project" table and traverse through release and task sub task classes, everything is fine. Its doing this only when i try to query from the task sub task table that has a matching prjoect code.

It looks like a wrong query, you have provided no information, but it looks like you want to select this :
from prjtsk in class ProjectReleaseTaskSubTask where prjtsk.projectRelease.project.prjCode = '1130'

Here are the mappings :
--------------------------------------------------

<class
name="Project"
table="PRJ"
>

<id
name="prjCode"
type="java.lang.String"
column="PRJ_Code"
>
<generator class="assigned" />
</id>

............

<set
name="projectReleases"
lazy="true"
inverse="true"
>
<key>
<column name="PRJ_Code" />
</key>
<one-to-many
class="ProjectRelease"
/>
</set>

--------------------------------------------------------------------------



<class
name="ProjectRelease"
table="REL"
>

<composite-id name="projectReleasePK" class="ProjectReleasePK">
<key-property
name="prlseNum"
column="PRLSE_NUM"
type="java.lang.String"
length="4"
/>

<key-many-to-one
name="project"
class="Project"
>
<column name="PRJ_CD" />
</key-many-to-one>
</composite-id>

......................................

<set
name="projectRelTaskSubTasks"
lazy="true"
inverse="true"
>
<key>
<column name="PRJ_CD" />
<column name="PRLSE_NUM" />
</key>
<one-to-many
class="ProjectRelTaskSubTask"
/>
</set>


---------------------------------------------------------------------------------

<class
name="ProjectReleaseTaskSubTask"
table="RTS"
>

<composite-id name="projectReleaseTaskSubTaskPK" class="ProjectReleaseTaskSubTaskPK">
<key-property
name="prlseTaskCode"
column="PRLSE_TASK_CD"
type="java.lang.String"
length="4"
/>
<key-property
name="prlseStskCode"
column="PRLSE_STSK_CD"
type="java.lang.String"
length="2"
/>

<key-many-to-one
name="projectRelease"
class="ProjectRelease"
>
<column name="PRJ_CD" />
<column name="PRLSE_NUM" />
</key-many-to-one>
</composite-id>

Has anyone come across this problem? Where you have three table related with composite primary keys?
Does anyone see any problem with the mapping or the query?