I have been successfully using NHibernate for quite some time now and
have been able to solve a lot of pitfalls with an application that I
developed with it and that is running in production. The recent hurdle
really has me scratching my head, though.
Recently I had to expand the class library with some new classes that
are nested as children to some already existing classes. I just copied
the same model for aggregate mapping that I already was successfully
using, but this time it does not work.
Now when I use the following in the parent mapping file:
<bag name="SeaInfoItems" table="EDIImport_SeaInfo" lazy="false" cascade="save-update">
<key column="EDI_FK_OWNERID"/>
<one-to-many class="FargoGate.AppLib.EdiImportSeaInfo, FargoGate.AppLib"/>
</bag>
I can choose to, in the child class, either use:
<property name="EDI_FK_OWNERID" column="EDI_FK_OWNERID" />
...which gives me the infamous "Invalid Index n for this
SqlParameterCollection with Count=n" error.
OR I try with this solution I found after some Googling:
<property name="EDI_FK_OWNERID" column="EDI_FK_OWNERID" insert="false" update="false" />
...which gives me a "Cannot insert the value NULL into column
'EDI_FK_OWNERID'... column does not allow nulls." error.
So basically I have to choose between pest and cholera.
What I don't get is that it works flawlessly for the already existing
aggregate classes, and I really cannot spot the difference. The only
thing is that this foreign key (EDI_FK_OWNERID) could refer to two
different parent tables. Bad database design, I know, but I didn't
design it, and it is my task to develop up to it for better or worse. I cannot change the database design.
The other difference is that I totally removed the foreign key reference from the already existing child classes (the mappings as well as the class members). I tried to emulate that of course, but of no avail.
Also I discovered that one of the new classes (which is quite small) also works fine. But I cannot see what the difference is here either. I am stumped!
Anyone has a clue?
Aaargh! I was put so much on a wrong leg with this infamous “Invalid Index n for this SqlParameterCollection with Count=n” error that I overlooked the obvious: A duplicate mapping of a field for ONE of the classes.
In that particular mapping I left this error, where the primary key is also defined as a property:
<id name="ID" column="ID">
<generator class="guid" />
</id>
<property name="ID" column="ID" />
Now that was a waste of time trying to debug that!
Related
I am pretty new to NHibernate and I'm trying to create a mapping file to extend a data model project. The particular table I am mapping is called AttributeDef in the following image, the column ControlType actually relates to a lookup in the table called Code (yes, I know - there should be an FK constraint but this sort of thing is quite common in this project so please ignore the obvious howlers and focus on the question). In most cases tables which reference Code also have a column which contains the ID from the table CodeSet as the key in Code is, almost inevitably, a composite key, but not in this case presumably because the original author figured "Hey they're all from the same codeset so what's the point?".
Now if there was a column in AttributeDef which contained the CodeSet value then the mapping wouldn't be much of a problem. The mapping for the Code entity looks like this:
<hibernate-mapping xmlns="urn:nhibernate-mapping-2.2"
assembly="Activus.DataModel" namespace="Activus.DataModel">
<class name="Code" table="Code" mutable="false">
<composite-id name="CompositeCodeId" class="CompositeCodeId">
<key-property name="CodeId" column="CodeId"/>
<key-property name="CodeSet" column="CodeSet"/>
</composite-id>
<property name="Description" column="Description" type="string" length="100" not-null="true"/>
<property name="ExternalRef" column ="ExternalRef" type ="string" length ="256" not-null ="true"/>
<property name="InternalRef" column ="InternalRef" type ="string" not-null ="false"/>
<many-to-one name="CodeSet" class="CodeSet" column="CodeSet" not-null="true" insert="false" update="false"/>
</class>
</hibernate-mapping>
Therefore if there was a column in AttributeDef for the CodeSet value (notionally called FormControlCodeSet in this example) then in my AttributeDef mapping file I would include
<many-to-one name="ControlType" class="Code" not-null="false">
<column name="ControlType" />
<column name="FormControlCodeSet" />
</many-to-one>
And all should be well. The problem is that to add that column to AttributeDef would be very invasive as I would then have to make a LOT of other changes to accommodate this and that would increase the risk factor of the change I'm making to a point which might well be unacceptable (from the client's point of view given their time frame).
So, much as it's a horrible, horrible thing to contemplate, Is it possible to substitute the line
<column name="FormControlCodeSet" />
With a (whisper it) hard coded value? That value hasn't changed in a decade and isn't likely to anytime soon but it would get us past this change and would highlight the need to scope out and implement the inclusion of the extra column. I recognise how dreadful this is but unfortunately a lot of this database isn't really that well suited to ORM despite it's being shoe-horned in anyway.
You do not have to whisper requirements, when working with NHiberante. Because cooperation with legacy DB (i.e. fixed DB schema) is pretty standard, NHibernate does support many different settings.
One of these is a pair of (substitutional) settings: "column" vs "formula". The first takes the column as is, the second could do whatever we need. Take column, columns, pass a constant. So:
<many-to-one name="ControlType" class="Code" not-null="false">
<column name="ControlType" />
<!-- let's replace this -->
<!--<column name="FormControlCodeSet" />-->
<!-- 1) with a constant number 123-->
<formula>123</formula>
<!-- 2) with a constant string 'constantString' -->
<formula>'constantString'</formula>
</many-to-one>
Not sure if the FormControlCodeSet constant should be int or string, but as shown above, either option is possible.
I have this Instrument entity:
<class name="Instrument" table="Instruments" mutable="false">
<id name="ID" column="INSTRUMENT_ID" unsaved-value="0">
<generator class="assigned" />
</id>
<property name="....." />
<property name="....." />
</class>
This entity is used in many-to-one relationship in other entity (InstrumentSelection). This is many-to-one mapping info:
<many-to-one name="Instrument" access="field.camelcase" column="Instrument_ID" update="false" class="Instrument" not-null="true" fetch="join" lazy="false" />
The issue I've it that when I save InstrumentSelection entity with Save:
Transact(() => session.Save(entity));
I get error in logs:
2012-12-20 14:09:54,607 WARN 12 NHibernate.Engine.ForeignKeys - Unable
to determine if Instrument with assigned
identifier 11457 is transient or detached; querying the database. Use
explicit Save() or Update() in session to prevent this.
A few facts about Instrument entity:
It's just a reference entity
It's immutable entity
It can not be added / inserted via application. I get rows in database from external feed.
Question (version 1): A my question is: is there a way to instruct NHibernate to always consider Instrument entity as detached? I mean - if an instance of Instrument exists in application it means that it's present in database. So there is no too much sense in quering the database.
EDIT 1: Because Question (version 1) was not answered yet, let me change it slightly:
Question (version 2): What could be the behaviour that NHibernate is still trying to work out whether entity is detached/transient? I think I have my mapping configured correctly (unsaved-value, generator).
The problem is that when you save the InstrumentSelection, NHibernate is cascading the operation to save the child Instruments. My first suggestion is to set cascade to none on the InstrumentSelect side of the relationship.
My second suggestion is to use an interceptor as shown in this answer.
Still a n00b in NHibernate :(
I want to add a composite index to the hbm.xml of one of my POCOs, for performance purposes. It does not relate directly to a class, but rather two common values I will be querying against.
I think I need to do a <map></map> entry in the XML, but the XSD is asking for a class name on the composite-key element, and there is no direct relationship, per se... plus it's asking for more information than I think I would need to provide. How do I do this?
<map name="PropertyKeys">
<key>
<column name="StockID" />
<column name="PropertyName" />
</key>
<composite-index class="Something?">
<key-property name="What goes here?" />
</composite-index>
<what else goes here?>
</map>
Have you read this http://ayende.com/blog/4045/nhibernate-mapping-map It explains when you would use a map and further on it explains the composite-index
However without more background info it is not 100% certain that you need a map!
I'm doing some per table inheritance and all is working great- but I'm noticing that when I want the base entity (base table data) NHProf is showing a left outter join on the child entity / (related table)
How can I set the default behavior to only query the needed data - for example: When I want a list of parent elements (and only that data) the query only returns me that element.
right now my mapping is similar to the below:
<?xml version="1.0" encoding="utf-8" ?>
<hibernate-mapping xmlns="urn:nhibernate-mapping-2.2">
<class name="FormBase, ClassLibrary1" table="tbl_FormBase">
<id name="BaseID" column="ID" type="Int32" unsaved-value="0">
<generator class="native" />
</id>
<property name="ImportDate" column="ImportDate" type="datetime" not-null="false" />
<joined-subclass table="tbl_Form" name="Form, ClassLibrary1">
<key column="ID"/>
<property name="gendate" column="gendate" type="string" not-null="false" />
</joined-subclass>
</class>
</hibernate-mapping>
And the example where I want all the data back vs ONLY the parent entity is shown below:
Dim r As New FormRepository()
Dim forms As List(Of Form) = r.GetFormCollection().ToList()
Dim fbr As New FormBaseRepository()
Dim fb As List(Of FormBase) = fbr.GetFormBaseCollection().ToList()
You can't. It's called "implicit polymorphism" and it's a rather nice (albeit unwanted in your case :-) ) feature provided by Hibernate. When you query a list of base objects, the actual instances returned are of the actual concrete implementations. Hence the left join is needed for Hibernate to find out whether particular entity is a FormBase or a Form.
Update (too big to fit in comment):
The general issue here is that if you were to trick Hibernate into loading only the base entity you may end up with inconsistent session state. Consider the following:
Form instance (that is persisted to both form_base and form tables) was somehow loaded as FormBase.
You've deleted it.
During flush Hibernate (which thinks we're dealing with FormBase and thus is blissfully unaware that there are 2 tables involved) issues a DELETE FROM form statement which throws an exception as FK is violated.
Implicit polymorphism exists to prevent that from happening - Form is always a Form, never a FormBase. You could, of course, use "table-per-hierarchy" mapping where everything is in the same table and thus no joins are needed but you'll end up with (potentially) a lot of NULL columns and - ergo - inability to specify not-null on children's properties.
All that said, if this is REALLY a huge performance issue for you (which it normally shouldn't be - presumably it's an indexed join), you could try using a native query to just return FormBase instances.
Our application (sadly) uses an MDB back-end database (I.e. JET engine).
One of the items being persisted to the database is an "event" object. The object is persisted to a table with an ID (EventLogID) that is an Autonumber field. The NHibernate mapping is as follows:
<class name="EventLogEntry" table="tblEventLog" proxy="IEventLogEntry">
<id name="Id">
<column name="EventLogID" not-null="true" />
<generator class="native" />
</id>
<property name="Source" column="ErrorLogSource" />
<property name="Text" column="EventLogText" />
<property name="Time" column="EventLogTime" />
<property name="User" column="UserID" />
<property name="Device" column="EventDeviceID" />
</class>
According to the log file, on some occasions when NHibernate attempts to obtain the identity, it receives the value "0". Later, when Flush is called, NHibernate suffers from an assertion failure.
Can anyone suggest why this might be happening? Better yet, can anyone suggest how to fix it?
Regards,
Richard
It could be that the default 'connection-release-mode' configuration setting is the cause of the problems.
A while ago, I ran into a similar issue, and I found that changing the connection.release-mode to 'on_close' (instead of the default after_transaction) solved the issue.
More information can be found on my blog
edit: as I'm thinking of it, perhaps it can be solved without changing the release-mode as well; what happens if you use a transaction to save your event ?
The default release-mode is after transaction, so I'm thinking; perhaps when you use an explicit transaction, the connection will only be closed after the transaction. The question offcourse is, will NHibernate try to retrieve the primary key that has been given to the object inside this transaction, or will it use another transaction ...
If it does not work, then changing the release-mode will solve your problem as well, but it is maybe not the best option.
I think the best option/solution, is to use an explicit transaction first, and see if this solves the problem...