I like the idea (and have implemented something similar, although with less
smartness), but there are a few things to consider:

1. Allowing only mapped classes to be bulked can be a bit limiting.
2. The fallback should use plain inserts. Using Session.Save gives a very
different behavior.
3. On e.g MSSQL it is possible to retrieve post insert identifiers.
4. It may be a good idea to disallow Session bound instances to be bulked,
or the pending Session.Flush may perform unnecessary/bad things.

Also, it would be nice if bulking was a feature of the persisters. "I have
1000 pending insert of type X which is bulk insertable. I'll use that."

/G



2014-10-29 2:00 GMT+01:00 Ricardo Peres <[email protected]>:

> Hi,
>
> I submitted a pull request for https://nhibernate.jira.com/browse/NH-3675.
> This is for supporting bulk inserts in NHibernate through the native
> ADO.NET provider. For example, SqlClient uses SqlBulkCopy,
> Oracle.DataAccess uses OracleBulkCopy, etc.
> I would like to hear from you about this. The API is very simple:
>
> session.BulkInsert(new Entity[] { new Entity { }, ... });
>
> I only implemented support for SqlClient and Oracle.DataAccess, other
> drivers get a dummy implementation that just does regular saves as a
> fallback. AFAIK, at least MySQL also supports this, but it does so through
> CSV files. Maybe someone else can provide support for this.
>
> RP
>
> --
>
> ---
> You received this message because you are subscribed to the Google Groups
> "nhibernate-development" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to [email protected].
> For more options, visit https://groups.google.com/d/optout.
>

-- 

--- 
You received this message because you are subscribed to the Google Groups 
"nhibernate-development" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to