You are viewing a plain text version of this content. The canonical link for it is here.
Posted to torque-user@db.apache.org by Martin Tilsted <ti...@daimi.au.dk> on 2008/11/29 17:51:48 UTC
Unsigned integers
How do i handle unsigned integers with torque (Using MySql 5.0)?
I use maven torque:jdbc to generate my schema.xml file, but i have a
field(Ip) that is an unsigned int in the database,
but that generate a
<column name="Ip" type="INTEGER"/>
in the schema file.
So I have 2 questions:
1: How do I get maven torque:jdbc to generate a schema for an unsigned
int? Is this even possible? I could not find the documentation for the
schema.xml format.
2: If i keep the signed integer version in my schema, but the field is
unsigned in the database, what happens if call setInt() with a negative
value.?
Martin
---------------------------------------------------------------------
To unsubscribe, e-mail: torque-user-unsubscribe@db.apache.org
For additional commands, e-mail: torque-user-help@db.apache.org
RE: Unsigned integers
Posted by Greg Monroe <mo...@dukece.com>.
FWIW, I seem to remember jdbc task issues with getting foreign
key constraints, indices, and recognizing autoincrement columns.
Plus, I think there is a fair amount of variation in what you get
depending on the DB you're connecting to. I've always considered
the XML from this as a starting point to avoid typing a lot ;) but
not the end product. But then I use my XML to generate the SQL
statements for replicating my Application DBs, so it has to be full
featured with all that stuff.
That said, you might want to take a look at the DDL UtilS project:
http://db.apache.com/ddlutils
This is a "sister" project to torque that uses the same schema
(with some exceptions around external-schema tags). They have
been the focus of getting the jdbc task more robust. In fact,
the Torque 4.0 design discussions talked about converting the
tasks relating to DB creation, etc to use the ddlutils project.
FWIW, they have a nice set of easy to use classes to do a lot of
the same tasks via code. They have a nice feature that can compare
an existing DB to an XML schema (or another DB) and generate the
SQL code needed to update the destination DB to match an source
schema/DB. I've been using this one to generate version
update SQL scripts. NOTE that this class is kind of "militant"
in that if there are table NOT in the XML, it will generate
drop statements for them. But with a little quick manual editing,
I can produce an upgrade script in about 30 minutes or less,
including testing.
Finally, you asked about Torque 4.0 features, etc. Here are a
couple of resources that you might want to check out:
http://wiki.apache.org/db-torque/NextRelease
This is the result of an extended discussion about Torque 4.0 that
took place starting in Nov 2006 (and on into early Feb or so). If
you truly want to understand what the Wiki is talking about, I
suggest going thru the torque-dev list archives and reading the
threads related to Torque 4.0.
> -----Original Message-----
> From: Thomas Fischer [mailto:fischer@seitenbau.net]
> Sent: Wednesday, December 03, 2008 9:46 AM
> To: Apache Torque Users List
> Subject: Re: Unsigned integers
>
> > > IMHO torque:jdbc is no good as a totally automatic way to create a
> > > schema.xml. It is good as a first guess, but nothing more, because
> the
>
> > > informations from the jdbc driver are not complete enough to extract
> a
>
> > > good schema.
> > > If it worked for you so far, you were lucky.
> > >
> > Or maybe Torque have lots of features I don't know about.
> >
> > But what kind of features in my database could be a problem for
> torque:jdbc?
> >
> > I love torque, because I can just define my database in sql and then
> run
>
> > maven torque:jdbc to get a schema.xml file, that maps
> > all the tables to classes, with primary and foreign keys, and allow me
> > to retrieve data from the database.
>
> The problem might be that the mapping is different from what you expect.
> I
> do not have an example ready because I do not use torque:jdbc myself, but
> searching the mailing list should find some complaints.
>
> Thomas
DukeCE Privacy Statement:
Please be advised that this e-mail and any files transmitted with
it are confidential communication or may otherwise be privileged or
confidential and are intended solely for the individual or entity
to whom they are addressed. If you are not the intended recipient
you may not rely on the contents of this email or any attachments,
and we ask that you please not read, copy or retransmit this
communication, but reply to the sender and destroy the email, its
contents, and all copies thereof immediately. Any unauthorized
dissemination, distribution or copying of this communication is
strictly prohibited.
---------------------------------------------------------------------
To unsubscribe, e-mail: torque-user-unsubscribe@db.apache.org
For additional commands, e-mail: torque-user-help@db.apache.org
Re: Unsigned integers
Posted by Thomas Fischer <fi...@seitenbau.net>.
> > IMHO torque:jdbc is no good as a totally automatic way to create a
> > schema.xml. It is good as a first guess, but nothing more, because the
> > informations from the jdbc driver are not complete enough to extract a
> > good schema.
> > If it worked for you so far, you were lucky.
> >
> Or maybe Torque have lots of features I don't know about.
>
> But what kind of features in my database could be a problem for
torque:jdbc?
>
> I love torque, because I can just define my database in sql and then run
> maven torque:jdbc to get a schema.xml file, that maps
> all the tables to classes, with primary and foreign keys, and allow me
> to retrieve data from the database.
The problem might be that the mapping is different from what you expect. I
do not have an example ready because I do not use torque:jdbc myself, but
searching the mailing list should find some complaints.
Thomas
Re: Unsigned integers
Posted by Martin Tilsted <ti...@daimi.au.dk>.
Thomas Fischer wrote:
>>> That said, the best bet would be to manually define the column
>>> type to be an UNSIGNED INTEGER in the DB and then use BIGINT
>>> as the type in the schema. This will make the Record object
>>> use the long type and avoid having large binary int values
>>> become negative.
>>>
>>>
>> That sound like a good solution. The only small problem is that I
>> generate the schema.xml directly from the database,
>> so next time i change the database and need a new schema.xml file, I
>> will have to change the schema.xml file again.
>>
>
> IMHO torque:jdbc is no good as a totally automatic way to create a
> schema.xml. It is good as a first guess, but nothing more, because the
> informations from the jdbc driver are not complete enough to extract a
> good schema.
> If it worked for you so far, you were lucky.
>
Or maybe Torque have lots of features I don't know about.
But what kind of features in my database could be a problem for torque:jdbc?
I love torque, because I can just define my database in sql and then run
maven torque:jdbc to get a schema.xml file, that maps
all the tables to classes, with primary and foreign keys, and allow me
to retrieve data from the database.
Martin
>> Oh and to be total off-topic: Does anyone have a description/status for
>> "torque 4.0" I am thinking about things such as
>> "New features","Status"(Is anyone really working on it at all) and
>> "expected release date").
>>
>
> Hm, yes, "we are thinking about it" is a good description of the status.
> The problem is that everybody was very busy with other stuff the last half
> of a year. But now with the economy crisis coming, probably there will be
> more time in the near future.
>
> Thomas
>
Re: Unsigned integers
Posted by Thomas Fischer <fi...@seitenbau.net>.
> > That said, the best bet would be to manually define the column
> > type to be an UNSIGNED INTEGER in the DB and then use BIGINT
> > as the type in the schema. This will make the Record object
> > use the long type and avoid having large binary int values
> > become negative.
> >
> That sound like a good solution. The only small problem is that I
> generate the schema.xml directly from the database,
> so next time i change the database and need a new schema.xml file, I
> will have to change the schema.xml file again.
IMHO torque:jdbc is no good as a totally automatic way to create a
schema.xml. It is good as a first guess, but nothing more, because the
informations from the jdbc driver are not complete enough to extract a
good schema.
If it worked for you so far, you were lucky.
> Oh and to be total off-topic: Does anyone have a description/status for
> "torque 4.0" I am thinking about things such as
> "New features","Status"(Is anyone really working on it at all) and
> "expected release date").
Hm, yes, "we are thinking about it" is a good description of the status.
The problem is that everybody was very busy with other stuff the last half
of a year. But now with the economy crisis coming, probably there will be
more time in the near future.
Thomas
Re: Unsigned integers
Posted by Martin Tilsted <ti...@daimi.au.dk>.
> That said, the best bet would be to manually define the column
> type to be an UNSIGNED INTEGER in the DB and then use BIGINT
> as the type in the schema. This will make the Record object
> use the long type and avoid having large binary int values
> become negative.
>
That sound like a good solution. The only small problem is that I
generate the schema.xml directly from the database,
so next time i change the database and need a new schema.xml file, I
will have to change the schema.xml file again.
But I guess I will just make a small script, to do a string replace on
the schema.xml file, after running maven torque:jdbc so it's not a big
problem :}
And thanks for the help.
Oh and to be total off-topic: Does anyone have a description/status for
"torque 4.0" I am thinking about things such as
"New features","Status"(Is anyone really working on it at all) and
"expected release date").
Martin
---------------------------------------------------------------------
To unsubscribe, e-mail: torque-user-unsubscribe@db.apache.org
For additional commands, e-mail: torque-user-help@db.apache.org
RE: Unsigned integers
Posted by Greg Monroe <mo...@dukece.com>.
The UNSIGNED type is not in the DTD because it is not part
of any SQL Standard. The SQL standard for this type of data
is to use one of the following column types: BIT, BINARY, or
VARBINARY column type (with possible CHAR and DECIMAL options).
Unfortunately, it looks like Village doesn't properly support
BIT (it assumes a 1 byte Boolean.. need to fix this now that
we have control of Village...). And BINARY and VARBINARY
are mapped to BLOBS which aren't very efficient.
That said, the best bet would be to manually define the column
type to be an UNSIGNED INTEGER in the DB and then use BIGINT
as the type in the schema. This will make the Record object
use the long type and avoid having large binary int values
become negative.
As to ensuring the proper values get placed into this, you can
either depend on JDBC or SQL to throw an exception if the value
is too large (e.g. a negative number of type long SHOULD be
considered as too many bits). Or you can just add some checks
to the set<Column>(long) method in the stub.
Hope this makes sense..
> -----Original Message-----
> From: Martin Tilsted [mailto:tiller@daimi.au.dk]
> Sent: Saturday, November 29, 2008 11:52 AM
> To: Apache Torque Users List
> Subject: Unsigned integers
>
> How do i handle unsigned integers with torque (Using MySql 5.0)?
>
> I use maven torque:jdbc to generate my schema.xml file, but i have a
> field(Ip) that is an unsigned int in the database,
> but that generate a
> <column name="Ip" type="INTEGER"/>
> in the schema file.
> So I have 2 questions:
>
> 1: How do I get maven torque:jdbc to generate a schema for an unsigned
> int? Is this even possible? I could not find the documentation for the
> schema.xml format.
>
> 2: If i keep the signed integer version in my schema, but the field is
> unsigned in the database, what happens if call setInt() with a negative
> value.?
>
> Martin
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: torque-user-unsubscribe@db.apache.org
> For additional commands, e-mail: torque-user-help@db.apache.org
>
DukeCE Privacy Statement:
Please be advised that this e-mail and any files transmitted with
it are confidential communication or may otherwise be privileged or
confidential and are intended solely for the individual or entity
to whom they are addressed. If you are not the intended recipient
you may not rely on the contents of this email or any attachments,
and we ask that you please not read, copy or retransmit this
communication, but reply to the sender and destroy the email, its
contents, and all copies thereof immediately. Any unauthorized
dissemination, distribution or copying of this communication is
strictly prohibited.