I'm new to Hibernate and maybe I'm off on a wrong track. Please set me straight.

The behavior I want in my application is I create a number of things and stick 'em in a database using Hibernate. If they're already there, skip the insert.

A field in the database table is "uniqueIdentifier" and I've set it to be "unique". In my Hibernate file, I have:
<property name="uniqueIdentifier" unique="true" />

My assumption was that Hibernate would deal with the issue if I attempt to insert a new record that already has been entered. Instead, I'm getting either an SQLException or a Hibernate exception when it tries to insert a record that's already in the database.

Am I misunderstanding the purpose of "unique"? Is this a way to keep duplicate non-primary-key values from getting into the database? If not, what's the 'hibernate way' of doing a series of inserts such that no two records in the database will end up with the same value in a particular field without throwing an exception (which seems to make the Session upset)?

Thanks in advance. hbm and sql statements are below.

Dave

=====================
<hibernate-mapping>
<class name="newsprint.MyNewsArticle" table="ARTICLE">
<id name="articleId" type="int" column="articleId" unsaved-value="0">
<generator class="identity"/>
</id>
<property name="rawHtml"/>
<property name="cleanText"/>
<property name="publicationId"/>
<property name="pubDate"/>
<property name="headline"/>
<property name="uniqueIdentifier" unique="true" />
</class>
</hibernate-mapping>


=====================
CREATE table article (
articleId int(11) PRIMARY KEY NOT NULL auto_increment,
publicationId int(11),
headline varchar(255),
rawHtml text,
cleanText text,
uniqueIdentifier varchar(255) unique,
pubDate date
);

I have the same problem, but no answer , does anyone know ?

Why not try saveOrUpdate?

Oh, I just relooked at this. I don't think the desired behaviour would be to skip records that duplicated a property just like the DB doesn't skip inserted records with a unique constraint.

I haven't really played with the unique attribute from Hibernates side but I assume it's an optimization within the session to trap the error before wasting the DB's time.

What's your business case for skipping these records? I would think it might get 'weird' if someone tried to insert a record and it just simply went into never never land, no?

Hi

I too wanted to use this functionality of unique multiple column.


As per Vampboy for calling saveorUpdate()

Consider i have record

item_id item_name item_price
1 computer 100
2 system 100


consider i update item_id 2 with item_name as computer to system.
then the values of the other two column are duplicated.
And i also want to tell you that i am not showing the customer about the item_id, that is only for my reference.
So according to this the value is duplicated.

I can prevent this only by unique constraint.
I have do this contraint in the Database but how can do it using Hibernate.

Any ideas to make multiple columns unique?