Hi,
I'm using NHibernate L2 caching with AppFabric and using the AppFabric
caching provider written by Simon Taylor on GitHub.
This works fine most of the time but every so often (after 5 minutes
or so in the test environment) I'll get a crash within the provider.
It's difficult to replicate in d
Mohammed,
Thanks for the reply.We have thought of registering a session as the
"active" one so that lower level entites use this. It will work, much
like it will work if we pass the current session into the entity
method. Unfortunately it depends on the developer remembering to
register a session
Hi,
I'm updating a framework we have around NHibernate and we're moving it
from a pattern based around Open Session In View (and we only use one
session per thread for all access) to one that allows us to open and
maintain multiple sessions per thread. We use this session in an
"ambient" manner, i
And just to clarify, this isn't a complaint at anyone. NH is a free
product that is as good as anything else on the market, if not better,
and it's worked on unpaid by the developers, so those of us who use it
should be grateful for this and that NH is still an active product.
On the flip side, th
Scott,
To jump in here. I wrote the article you referred to in your reply to
Fabio, and tbh, I wouldn't consider it official NHibernate
documentation. It was written a long time ago as an example of the
things you could do with the events,when documentation on the
interfaces as opposed to the old
Mohamed,
As an aside, many thanks for the blog entry. I saw the Nuget feed and
was instantly confused as to why I shouldn't install it until I came
here :) May I suggest that this should actually be a wiki entry on
nhforge?
Fabio: . or some helper text be put in the Nuget feed description
(e.
I believe that unless you disable lazy loading in the mapping (which
isn't the best of ideas) then a Get will not "get" the object tree
from the database in one hit, unless you edit the collection mapping
to do this, e.g. fetch="join" etc. See
http://nhforge.org/blogs/nhibernate/archive/2009/04/13/
> Are you using System Transactions and/or explicit NH transactions?
Explicit NH Transactions
> What isolation level are you using?
Can't remember, I'm not at my desk atm. The default for SQL Server
2005
> Are you sharing sessions across threads?
No, I don't believe so
> Could you provide the
Hi,
I'm having a few issued with deadlocks that seem to eventually cause
Zombied connections. I'm using NH 2.1.2GA and SQL Server 2005.
My scenario is thus:
I have a multi-threaded process that scans a database table and
creates a bunch of inserts. All inserts are logicially distinct sets
of dat
Ok, that's a reasonable way forward for now, thanks!
--
You received this message because you are subscribed to the Google Groups
"nhusers" group.
To post to this group, send email to nhusers@googlegroups.com.
To unsubscribe from this group, send email to
nhusers+unsubscr...@googlegroups.com.
F
HI Jason, thanks for the reply.
An upgrade to NH3 isn't really possible yet. We use quite a bit of NH-
Linq and I've read the new provider fails where the last one succeeds,
which would be a bit of a regression test problem. I'm waiting for NH
3.1 before I consider upgrading.
I'm a little confuse
Hi,
I'm using NH 2.1.2GA and am attempting to reproduce the leaky
connection bug with Transaction Scope. I'm looking at this as our
existing architecure doesn't use the recommended pattern of
TransactionScope -> Open Session -> Start NH transaction, and although
we get no issues, I've always known
with an example hilo definition of:
CREATE TABLE [dbo].[hibernate_unique_key](
[next_hi] [int] NOT NULL,
[Entity] [varchar](50) NOT NULL
) ON [PRIMARY]
GO
and an example mapping...
2
Entity='Advert'
.
you need a row for each e
We use the where parm after a migration from identity. It's pretty
straightforward.
The HiLo table definition is:
CREATE TABLE [dbo].[hibernate_unique_key](
[next_hi] [int] NOT NULL,
[Entity] [varchar](50) NOT NULL
) ON [PRIMARY]
GO
and an example mapping is:
... however if you did want to do this then you'd need some framework
to manage the sessions for you.
the best way may be to wrap each Create within a method that manages
the session for you e.g. (semi pseudo-code)
Parallel.ForEach(() => { ServiceLayer.Create(entity); };
ServiceLayer.Create(T en
On Sep 5, 11:02 pm, José F. Romaniello wrote:
> I think this is the place to open issues/bugs:http://connect.microsoft.com/
Which, to be fair, was in the original poster's mail and it states MS
will not fix it. Therefore, as much as that decision is BS, the onus
is on the NH community to work-aro
Frans...
> How does NH now set the parameter type for a Time or Date parameter?
> As the value is always of type System.DateTime..
It doesn't.
http://216.121.112.228/browse/NH-2207
I for one would wish NH would get this sorted. I *know* it's MS
problem and it's their fault but sometimes produc
All,
I have an HQL query built up over several steps that works quite well.
However, I now need to add some locking hints to the query via
SetLockMode.
I can't get this to work as the alias keeps failing. My HQL is
something like this (NH 2.1)
SELECT Count(*) FROM Transaction MT
WHERE MT.From.L
Hmm... I am using a single HiLo table with a lookup column to identify
the entity. I wonder if that's getting confused somehow?
--
You received this message because you are subscribed to the Google Groups
"nhusers" group.
To post to this group, send email to nhus...@googlegroups.com.
To unsubscr
Hi all,
I've converted a project from SQL Server identity primary key columns
to HiLo. So far all seems to work ok.
However there is one part of my application that performs a large
number of inserts across a wide range of data. This data is logically
isolated so I've split the code into threads
nvm, it had been mapped as a List not a Set. The only one in the
application the be different :p
--
You received this message because you are subscribed to the Google Groups
"nhusers" group.
To post to this group, send email to nhus...@googlegroups.com.
To unsubscribe from this group, send email
ok... an update to this.
I have a new object. This object has a child collection which is an
intitialised empty iesi hashedset.
Perform a Save on this parent and NH crashes:
Unable to cast object of type
'Iesi.Collections.Generic.HashedSet`1[iTrace.Domain.Entities.LocationOrderDay]'
to type
'Sys
>"forward" well... "forward" is a mere hope in some cases.
lol. That's is certainly true. We can hope though...
>I saw somebody trying to go from NH to EF and "forward" is not exactly his
>feeling.
Well, EF is a non starter at the moment but it's something that senior
managers keep questionin
Oskar
:
>Maybe we mean different thing with "repository", but I consider the repository
>concept (not necessarily the implementation) to be part of the domain model.
No I don't think we do. My domain layer uses BCL collections and
accesses my repository through an interface. The only rule my
repo
Fabio - I know what you mean and I'm coming to the state of mind of
"I'm a bit fed up of trying to seperate repository from domain so just
go with it".
However the problem is that although iesi is open source and is not an
NH specific dll, it's not a BCL collection. So far I've strictly
enforced d
Fabio,
Thank you for the more helpful update and link to JIRA. Perhaps Ayende
can update that blog entry to say that actually you CAN'T use HashSet
and must still use the Iesi HashedSet.
Although it will be a bit of an OMG change for .NET 4.0, it would be
nice as (hopefully) it will enable greate
on the session object, try this:
entity =
session.GetSessionImplementation().PersistenceContext.Unproxy(entity);
--
You received this message because you are subscribed to the Google Groups
"nhusers" group.
To post to this group, send email to nhus...@googlegroups.com.
To unsubscribe from this
Thanks for the "answers".
As I said in my post, it also fails if I use HashSet AS PER AYENDE'S
POST
Unable to cast object of type
'System.Collections.Generic.HashSet`1[iTrace.Domain.Entities.Operation.MovementTransactionItem]'
to type 'Iesi.Collections.ISet'.
You telling me we MUST use the IESI
All,
I've come across an odd situation with NH and the Version settings. My
App uses generic ICollections and map these to sets. It works fine
without any problems.
However, if i add the Version setting to the mapping file I get
intermittent problems such as:
Unable to cast object of type
'Syste
All,
I have a parent object "Location" that has a set of child objects
"Account". 9/10 times these are treated as distinct entities so I have
logically placed them in seperate "domain aggregates".
This means I have cascade="none" from the location to the account set.
However, when I create a loc
>Let me guess: you're using identity.
>
> Diego
Can I ask if this makes a difference to the pattern? I'm NOT using NH
txn's as, like most people, I ran the tests, it worked and so, far
have not had any problems. I'm using GUID.COMB and FlushMode.Never
though. IMO this whole issue needs clearing
All,
I'm trying to track the the object causing this error:
"NHibernate.TransientObjectException : object references an unsaved
transient instance - save the transient instance before flushing.
Type: Entities.Allocation.OrderItemAllocation, Entity:
Entities.Allocation.OrderItemAllocation".
I'm t
Thanks Paulo. My immediate reply was to John.
That seems to work ok - looks like NH is more intelligent than I give
it credit for. Will this still prevent ghosts, i.e. NH loads an
object, does a cast to a different type, so assumes it changed when in
fact it hasn't?
--
You received this me
Cheers, found this and yes, I've looked on google. I'm mapping several
other enums to properties perfectly well using the method listed in my
post (using NH 2.1.2GA btw). I'm not sure, but I think the post you
listed was only necessary to NH 1.2 GA?
My issue not with mappings enums but with mappin
Hi,
I'm trying to map a System.DayOfWeek enum to a database int field. My
mapping as as follows:
When NH verifies the mapping I get a TypeInitializationException and a
message of:
"Could not determine type for: System.DayOfWeek, System, for columns:
NHibernate.Mapping.Column(FirstDayOfTheWeek
It would be great, but too soon. Many people want a full Linq provider
who are not able to upgrade to .NET 4.0 straight away.
IMO on way would be a release of 3.0.0 as .NET 3.5 and an immediate
branch to (e.g. 4.0.0) for a .NET 4.0 version. Maintain 3.x in .NET
3.5 as a bug patch version alongside
Anything in particular?
--
You received this message because you are subscribed to the Google Groups
"nhusers" group.
To post to this group, send email to nhus...@googlegroups.com.
To unsubscribe from this group, send email to
nhusers+unsubscr...@googlegroups.com.
For more options, visit this g
If you're mocking data then now, probably no point in an interface -
except maybe unless you wish to enforce a specific layer boundary that
can be replaced/mocked. Even then, mocking POCO's seem a bit
pointless.
However if you wish to avoid the anemic data model (i.e. pure POCO's
with behavour in
Fabio pt2. Thanks for the links btw... it isn't that tricky at all now
I can see an example. I misunderstood your comment re: configuration
blocks.
NH saves the day again!
--
You received this message because you are subscribed to the Google Groups
"nhusers" group.
To post to this group, send e
Thanks Fabio - I much prefer the answer "yes", even if its tricky ;)
If nothing else, I get to learn more about the inside of NH, which can
only be a good thing. Looking forward to NH 3.0.0 :)
--
You received this message because you are subscribed to the Google Groups
"nhusers" group.
To post
>AFAIK, one limitation of using proxy I encounter is that access to data
>model limited through public property. For example, I can't inject data when
>loading an instance through its private field.
True, you can't as the interface doesn't have the property you need.
You can do things like check i
Thanks for the opinions.
Yes, I'd thought of both of those options already. However my
extension is to NHibernate behaviour, i.e. If this configuration
setting = true then extend NHibernate so that it does not perform
implicit dirty checking (as per the NH Add Ins process Fabio wrote)
otherwise us
Hi,
I'm trying to add custom properties to the hibernate.cfg file to
control additional behaviour that we have extended in NH.
Unfortunately the NH XSD forces the Configure() step to fail as it
doesn't recognise my configuration.
I'd like to extend the configure class to, perhaps, intercept the
c
If you're going to interface everything then you might want to
consider making your proxies from the interface instead of the
concrete class. It works, but it's not without its problems as the
proxy doesn't inherit from the concrete class but from its own
representation of the interface (at least i
Fabio,
Nobody doubts your contribution to the NH ecosystem but if the key
contributors to NH (you, Ayende, Tuna etc.) don't use NH Forge where
is the momentum to drive people to the site, making them more aware of
it, and then contributing there own posts?
If cross posting is the best we can hope
All,
I have a system setup to generate proxies based on an interface. This
causes me a problem when I use the standard collection compare methods
(e.g. Contains(obj), Remove(obj) etc.)
If I pass a proxy'd object to a collection Remove method (where the
collection is a generic type of the concrete
Sorry, too tired to read the full post and work out what's going on
for you but look at this thread that I raised with similar questions:
http://groups.google.com/group/nhusers/browse_thread/thread/fce0810bee14d5b3#
A proxy will always be a proxy unless you force replace it (I think)
--
You rec
That may be the SQL produced but I would expect you not to get
duplicate items in the collection?
If so, then that's the way it works. You could change your mapping (I
think) to do two seperate selects in one database round trip. I'm
sorry, but I can't remember the mapping option to do that.
If y
You can also use the hibernate.cfg file to have a list of assemblies,
each containing the embedded resources you need. I'm not sure if there
is a way for NH to read the hbm at runtime, but I haven't looked too
hard tbh.
--
You received this message because you are subscribed to the Google Groups
Thanks again Diego.
It is a framework but the main issues I'm having is when I have an
interface that (e.g.) restricts properties to getters but the concrete
class has an internal setter to allow controller methods to hook
things together. These controller methods take in an interface but
need to
To extend this slightly, is there an NHibernate IsProxy method?
I've looked at the NHibernateProxyHelper but nothing in there. The
above almost gave me what I needed but in some edge cases I may have
an initialised proxy that will pass an NHibernateUtil.IsInitialized
check but still not cast to a
Thank you Diego
:)
--
You received this message because you are subscribed to the Google Groups
"nhusers" group.
To post to this group, send email to nhus...@googlegroups.com.
To unsubscribe from this group, send email to
nhusers+unsubscr...@googlegroups.com.
For more options, visit this group
By unwrap I assume you don't mean an NH function called "unwrap" but I
need to write my own converter to take the proxy to a concrete class -
i.e. create a concrete and copy the properties over?
I'm using NH 2.1.3GA so I guess the new lazy="no-proxy" isn't
available.
There is nothing in NH to do
All,
I'm using proxies based on an interface, e.g.
When I do a Session.Load (or anything that creates a proxy) then the
proxy is based on the interface, IThing. Any code such as this:
Thing obj = (Thing)proxy;
will fail as an "IThing" proxy from NHibernate does not inherit from
"Thing".So f
lol, wish google had Edit.
and now a workaround.
IEntity is a { get; set } for assigning the ID during hydration from
the DB.
If I add " new ID { get; }" to IDomain, the problem goes away.
Unfortunate, but less of a problem than my whole application stack
being screwed :)
Still sounds
Ok, recreated it and will raise a JIRA. However, it's a bit of an
"edge"case (but unfortunately fundamental to our application stack) so
before I do, any comments appreciated...
To get it to fail I have the following:
public interface IEntity (containing the ID)
public interface IDomain : IEntity
Fabio, thanks for the reply.
I'm in the process of creating a failing test now with just NH, but
I've just checked the JIRA and noticed NH-2069, from NH 2.1.1. Looks
like the same issue.
In case it isn't, can you confirm which version of NUnit (or other)
you use - I want to use the Assert.Throws
btw, I'm really confused now. This blog: "http://
blogs.hibernatingrhinos.com/nhibernate/Default.aspx" and this section
"First and Second Level caching in NHibernate" states:
"When using the session.Load(id) method NHibernate only instantiates a
proxy for the given entity. As long as we only acce
Hi,
I think this is a day for working through my misconceptions about NH.
Can someone confirm this is correct behaviour?
1) I get a key from the UI (Web page so I store the key only for
performance)
2) I create a proxy in NH using Session.Load in my UI. I assume the
row matching the ID exists bec
>The generic IList add IndexOf(9, RemoveAt(), Insert() and Item[]
>compared to ICollection<>. All of those deal with index, and if you
>really need any of those, it should probably be mapped as a
>instead of set or bag.
and of those the only useful one I regularly come across is
IndexOf(), w
You'd have to change your mapping so that NH knew to use the anonymous
property (the property accessor would be set to "backfield") as the
second is not using the warehouseList you've created but instead is
telling .NET create the backing variable for you and hide it behind
the scenes.
That's how
I use bag's with IList and don't have any performance problems that
i've noticed. In my mapping I also have a key column to map to the SQL
table primary key.
However in NH 2.1.x you can have ICollection instead and still use
sets. If there is anything you need from the IList specific to your
appli
I refer you to my previous post.
As far as I'm concerned, the HQL I provided works fine with your
mappings as I've listed the SQL that is generated. Are you doing
anything wierd? (old version of NH, using NH Fluent or something?)
--
You received this message because you are subscribed to the Goog
Are you sure?
Using your mappings (except I use a GUID.COMB generator instead of
Native) My SQL equivalent is:
select audit0_.Id as Id0_, audit0_.Active as Active0_,
audit0_.DateAudited as DateAudi3_0_
from audits audit0_ inner join audit_auditTypes audittypes1_
on audit0_.Id=audittypes1_.auditId
Oh, and unless you have no choice, try not to use the native
generator. Try HiLo or something. Native will be pain sooner or
later...
--
You received this message because you are subscribed to the Google Groups
"nhusers" group.
To post to this group, send email to nhus...@googlegroups.com.
To un
Something like;
SELECT au FROM Audit au JOIN au.AuditTypes at
WHERE at.id IN (:ids)
I think
--
You received this message because you are subscribed to the Google Groups
"nhusers" group.
To post to this group, send email to nhus...@googlegroups.com.
To unsubscribe from this group, send email to
/bump
Any comments? Why is the unnecessary (as I understand it) UPDATE
being generated by NH?
--
You received this message because you are subscribed to the Google Groups
"nhusers" group.
To post to this group, send email to nhus...@googlegroups.com.
To unsubscribe from this group, send email t
All,
I'm trying to sort out performance on a 1.2 NH app I've inherited. It
works on the "Aggregate root" concept, i.e. a parent entity owns
children and when the parent it cascades any changes down to the
children automatically.
The amount of UPDATE/DELETE/INSERT generated against objects that ar
Hmm.. Don't use Fluent so can't help you there. If you are sure ALL
NHibernate files are in the bin directory that the test runs from,
then I'm at a bit of a loss to explain why your error is "can't find
NHibernate.dlls"
--
You received this message because you are subscribed to the Google Groups
Not sure what you mean tbh but
Maybe a Multi-query? 1st query gets the count of rows (in case it
changed between page interactions) the second to do the paging itself.
Get the results, do some calculations from the 1st result to work out
the index position the items from the second query would
1) Are the NHibernate DLL's in the bin directory that the Unit test
code runs in?
2) Is the hibernate .cfg file in the bin directory (and/or settings in
test harness config file)?
This are the usual scenarios as your test project probably will not
reference NH directly, therefore the DLL's will no
Thanks all for the quick response, and thanks for the example Diego.
Oskar -> Is there a reason why you'd have seperate tables over
multiple rows in a single table? I believe the HiLo algorithm locks by
row?. Is it for performance reasons (i.e. lock by table instead)?
I assume the hbm per entity
All,
I know this isn't the "correct" way to do this but I need to know how
to use HiLo and have the hilo table store a row for each entity in my
system.
My reason for this is that I am migrating an existing DB that uses 32-
bit integers with Identity over to NHibernate (currently 1.2 but we
aim t
Although I'm aware of this "feature" for ADO exceptions, I'm curious
as to why. I know this debate has been had before but I've never seen
it come to a conclusion.
Why is the session unusable after a "restartable" database errors such
as timeout, deadlock, connection failure etc ? These are scenar
Thanks fabio.
I thought of that, but I'm concerned that (using your example 1000
limit):
a) I need to sort the data via an order by. If I'm going to keep the
first 1000 then SQL will need to retrieve all the possible rows, then
sort them to get the top 1000. There may be a lot of work for SQL to
Although I haven't solved it yet, I'm thinking through the same issues
for our own architectures here.
I think I'm going to go down the route of MVP combined with a Business
Conversation (http://fabiomaulo.blogspot.com/2008/12/implementing-
conversation-per-business.html) so that the business conv
Hi,
I'm wondering if there is a high performance way of getting the result
count from a SQL statement (SQL Server) before actually loading the
rows into the entity objects, via an HQL query.
i.e.
Do complicated query - result set = 45000 rows
Get count. If count < 1000 load entities
At the mome
All,
I'm really struggling to come up with the correct HQL for an object
configuration such as:
Transaction (singular)
-> TransactionItems (collection of TransactionItem)
--> SendAccount (singular property of a Transaction Item)
--> Entries (collection of Entry)
Thanks Fabio. Obvious in the end...:p
Usually I don't use it, but I'm trying to reproduce a bug and want to
force a database deadlock situation.
--
You received this message because you are subscribed to the Google Groups
"nhusers" group.
To post to this group, send email to nhus...@googleg
/bump
Anybody able to explain what the alias should be for an HQL based
IQuery?
--
You received this message because you are subscribed to the Google Groups
"nhusers" group.
To post to this group, send email to nhus...@googlegroups.com.
To unsubscribe from this group, send email to
nhusers+uns
/bump
. or any links to SetLockMode with a Query and what the alias
means? Google doesn't come up with much for Querys. There's quite a
lot for Criterion... does SetLockMode not actually work for a query?
The documentation says its IQuery.SetLockMode() whereas it is actually
IQuery.SetLockMo
All,
Trying to use SetLockMode with an IQuery. I don't fully understand
what the alias means (documentation isn't that clear, could do with an
example) but I think it refers to the object being locked.
I have the following statement:
IDomainClassChild child = session.CreateQuery("FROM DomainClas
set the substring outside the call?
NH-Linq can be a bit flaky sometimes
On Nov 16, 2:45 pm, Nieve wrote:
> Hello all,
> I'm using Linq-to-nh and when I do
> query.Where(user => names.Contains(user.Name.Substring(0, 2)));
> I get a null reference exception.
> Any idea how this should be done a
Hi,
I'm getting an odd error after installing our application into a
Windows Server 2008 R2 environment.
The application works fine running under a development environment
(i.e under Visual Studio 2008) but in QA we get an object reference
exception in the NH proxy generator:
Stack trace:
at N
I use a SQL generated version so I'm not sure. However, I'm guessing
you're using the "version" mapping type.This has an option of
generated="never" (meant to be the default). I'm assuming this means
let NH do it, not the DB, but I haven't tried it to know for sure.
--~--~-~--~~--
I'm assuming your version is set to be generated by the DB?
In this case, I don't think there is a way to turn off the select -
though I too wish there was a way to keep it for updates but a "don't
bother this time" for inserts. hopefully someone else will confirm
there is such a feature !
-
> "_ParentPersons" property that by default should return an empty collection
> if I don't want to load it explicitly in my DAO
My question would be, "why?". As in, why return an empty collection if
you don't want to load it explictly? The data either exists, or it
doesn't so, because Lazy is tu
I don't know about Oracle, sorry, but there was a similar bug I raised
recently with the SQL Server 2008 dialect DATETIME2. It seemed the
dialect wasn't generating the milliseconds properly as it hadn't
overridden the base class correctly. SQL did generate milliseconds as
I would expect for Timest
It's an issue I'm thinking over at the moment as well.
You could look at NH collection options such as "batch-size". This can
drastically reduce the number of round-trips to the DB. Lazy loading
as a default is a good idea, then remove it for those cases where it
doesn't make sense (e.g. a custom
Any NH guru's got any feedback on whether resubmitting without
starting a new session is acceptable in database exception cases such
as timeout or deadlock?
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups
"nhusers" gr
Ok, worked it out myself, with .NET 3.5 use anonymous methods.
e.g.
ISerial result = QueryProcessor.Linq(() => Session.Linq
().SingleOrDefault(s => s.SerialId == serialIdentifier));
and
public static T Linq(Func executeStatement)
{
int retry = 0;
T result = default(T);
d
All,
In my application I have a repository layer which wraps access to
database queries. As NH-Linq has come on the scene we've started using
his a lot more for new apps.
However, as Linq queries are deferred until first use, I'm not sure
how to deal with common DB errors that can be retried - i
Found it, I think.
Coding error - passing a proxy to the SetParameter method, not a Guid.
I should just learn not to doubt the NH developers and concentrate on
investigating my own team instead :)
--~--~-~--~~~---~--~~
You received this message because you are sub
Ok, this is repeatable now. I'll try to write a test case but it will
be in about a week.
NH Stack trace (removing product info)
at NHibernate.Impl.AbstractQueryImpl.GuessType(Type clazz)
at NHibernate.Impl.AbstractQueryImpl.SetParameter[T](String name, T
val)
Error: "Could not determine
I'll be honest, I'm a little confused trying to follow your example. I
can tell you've tried to be clear but it's late and I'm tired ;)
Any chance of you writing a test case to demonstrate the issue instead?
--~--~-~--~~~---~--~~
You received this message because y
Dario,
I've updated NH-1973 with the latest version of the test case.
I'll have a look in the NH code to see if I can work out how to fix
it, but I've only just downloaded the codebase and haven't even got it
to compile yet ;)
--~--~-~--~~~---~--~~
You received th
All,
I raised a bug (NH-1973) which someone kindly pointed out was a
limitation in SQL Server. My mapping file was mapping a .NET datetime
to a NH Timestamp field and millisecond precision was being lost.
After the update to the bug, I changed my test to the following
mapping instead:
Oskar. Yes I have, and my test fails to confirm it.
using (ISession session = this.OpenSession())
{
DomainClass entity = session.Get(1);
Assert.AreEqual(DateTime.Parse("2009-09-23 16:28:21.489"),
entity.Timestamp);
}
--~--~-~
Apologies, you're correct.
"Could not determine a type for class
Domain.Interfaces.Entities.IDevice"
That's all I can recall. Stupidly, I didn't log the stack trace.
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups
All,
Unfortunately I haven't been able to reproduce this to create a
failing test but I had the following error with NH 2.1.0GA:
"Could not determine a type for class"
where the reported class was an Interface declaration.
A snip of the mapping file is:
..
..
HQL that ca
1 - 100 of 170 matches
Mail list logo
