Mailing-List: contact user-help@cayenne.apache.org; run by ezmlm
Precedence: bulk
Reply-To: user@cayenne.apache.org
Received-SPF: pass (nike.apache.org: domain of siberian@siberian.org
 designates 209.85.216.185 as permitted sender)
MIME-Version: 1.0
Sender: siberian@siberian.org
In-Reply-To: <8f985b960910201710mec740b5x6f721895aa72497d@mail.gmail.com>
References: <507377.53458.qm@web38908.mail.mud.yahoo.com>
	 <8f985b960910201710mec740b5x6f721895aa72497d@mail.gmail.com>
Date: Tue, 20 Oct 2009 17:18:00 -0700
Message-ID: <479ffcd60910201718o7f64c4eat97aa8c09d65163b9@mail.gmail.com>
Subject: Re: Cayenne and H2 on the desktop
From: John Armstrong <siberian@gmail.com>
To: user@cayenne.apache.org
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

I implemented a DataSourceFactory to obtain this same functionality.
specifically around having a single executable JAR and derby not
finding its schema creation file. The DSF lets the system auto-detect
schema existence and use embedded ant to create it as required.

Your mileage may vary
John-

On Tue, Oct 20, 2009 at 5:10 PM, Mike Kienenberger <mkienenb@gmail.com> wro=
te:
> Do you want the database to be a single-user resource which is
> persisted between runs?
>
> If so, it should be as simple as using jdbc:h2:~/test/testdatabase as
> your jdbc url where "~/test/cistestdatabase" is the path to your
> database (it will be a directory). =A0Most likely, this is what you
> want.
>
> Using jdbc:h2:tcp://localhost/~/test/testdatabase will connect as a
> standard standalone database instance and assumes you started it
> standalone. =A0 This is now multiuser.
>
> The list for technical questions is a google group.
>
> http://groups.google.com/group/h2-database
>
>
> On Tue, Oct 20, 2009 at 8:00 PM, Eric Lazarus <ericllazarus@yahoo.com> wr=
ote:
>> Folks
>>
>> Thanks for you folks direction on using H2 as our database but we are a =
bit stuck on how to do it. We are moving toward using Cayenne 3.0M6 and H2 =
for the desktop version of an application that should be as idiot proof as =
possible. We want for the end user to not have to understand that there is =
a database there at all.
>>
>> Any suggestions about how we should configure things? For example, shoul=
d we configure H2 to run embedded if we don't want the end user to have to =
be aware of it? How DO we configure it that way? What should we be reading =
to figure out how to do that? I am not even sure we understand what it mean=
s to run "embeded." Does that mean that that database is in the same VM? Th=
at we don't go via JDBC at all?
>>
>> Also, we are trying to load a postgres dump into H2 and am failing becau=
se H2 does not support the COPY statement. I have tried enabling postgres c=
ompatibility mode by appending ';MODE=3DPostgreSQL' to my database url. Thi=
s results in a error message stating 'File not found 'backup.sql'. I tried =
putting my dump in a file by that name and got the same message.
>>
>> I also tried sticking 'SET MODE PostgreSQL ;' at the top of my SQL dump =
file. This had no effect the script still abended with a syntax error on th=
e COPY command.
>>
>> Where the best place to ask H2 questions? Is there an IRC room? A list s=
erve?
>>
>> thanks!
>>
>> Eric
>>
>> +1 (917) 589-6579
>>
>>
>>
>