Hi all,
I got the code to compile and run fine... I am digging through the sql2xml
task, and think that there may be a bug with using the sql2xml task and DDL
exported from MS SQL Server using Enterprise Manager. I am using SQL Server
2000 on Windows 2000 Advanced Server.
When I export my "Stores" table, my DDL looks like the:
CREATE TABLE dbo.STORES (
STORE_ID int IDENTITY (1,1) NOT NULL,
STORE_NAME varchar(50) COLLATE SQL_Latin1_General_CP1_CI_AS NOT NULL
) ON PRIMARY
GO
The differences between what MSSQL Server exports and what the
SQLToAppData.java expects are:
1) IDENTITY value. The store_id column is the primary key, and is an
autoincrementing column. The issue seems to be that the SQLToAppData class
in the Create_Table_Column(Table tbl) method expects the column to be either
the name of the column or PRIMARY/FOREIGN/UNIQUE. However, it is none of
these, it is IDENTITY. Now, from what I understand about IDENTITY columns,
they are a combination of PRIMARY and UNIQUE. They have to be unique in the
database, and they are typically the primary key. So what I am now doing is
when we run into an identity column in the method Create_Table_Column_Data I
set the col to be auto increment and to be the primary key.
2) COLLATE value. We are using SQL Server's Natural Language Query
function, which I think creates the COLLATE SQL... columns. I added then to
the tokens as something to just be consumed.
3) ON PRIMARY GO values. I changed the Create_Table method for the while
loop to see if it finds either ";" or "ON" string. Then if the token is
"ON" then it consumes the PRIMARY and GO values.
4) dbo.STORES value. I changed what the tablename is set to to remove any
database owner/user information attached to the table name.
These changes made it product a schema.xml file that looks correct. I am
going to try and write up my notes on what I did for other people who are
trying to integrate Turbine and Torque into existing MS SQL Server setups.
What is the best format to do this in? I looked at the .xml versions of
some of the documentation, is that I how I should format my document?
Here is the diff between CVS and what I did:
cvs diff SQLToAppData.java (in directory
C:\java\temp\jakarta-turbine-torque\src\java\org\apache\torque\engine\databa
se\transform)
Index: SQLToAppData.java
===================================================================
RCS file:
/home/cvspublic/jakarta-turbine-torque/src/java/org/apache/torque/engine/dat
abase/transform/SQLToAppData.java,v
retrieving revision 1.1.1.1
diff -r1.1.1.1 SQLToAppData.java
28,30c28,30
< * 4. The names "Apache" and "Apache Software Foundation" and
< * "Apache Turbine" must not be used to endorse or promote products
< * derived from this software without prior written permission. For
---
> * 4. The names "Apache" and "Apache Software Foundation" and
> * "Apache Turbine" must not be used to endorse or promote products
> * derived from this software without prior written permission. For
34c34
< * "Apache Turbine", nor may "Apache" appear in their name, without
---
> * "Apache Turbine", nor may "Apache" appear in their name, without
79c79
< *
---
> *
149c149
< throw new ParseException (name + " at [ line: " + token.getLine()
+
---
> throw new ParseException (name + " at [ line: " + token.getLine()
+
182a183,185
> if (tableName.indexOf(".") >= 0){
> tableName = tableName.substring(tableName.lastIndexOf(".")+1);
> }
185,186c188,189
< //tbl.setIdMethod("none");
< while (!token.getStr().equals(";"))
---
>
> while (!token.getStr().equals(";") &&
!token.getStr().equals("ON"))
190c193,198
<
---
>
> if (token.getStr().equals("ON")){
> next(); // eat PRIMARY
> next(); // eat GO
> }
>
>
437c448
< if (!token.getStr().toUpperCase().equals("NULL"))
---
> if (!token.getStr().toUpperCase().equals("NULL"))
441a453,468
> else if(token.getStr().toUpperCase().equals("IDENTITY"))
// MS SQL Server specific
> {
> col.setAutoIncrement(true);
> col.setPrimaryKey(true);
> while (!token.getStr().equals(")"))
> {
> // skip until )
> next();
> }
> next();
> }
> else if(token.getStr().toUpperCase().equals("COLLATE"))
// MS SQL Server specific
> {
> next(); // eat the SQL_Latin1_General_CP1_CI_AS stuff
> next();
> }
445c472
< if (!token.getStr().toUpperCase().equals("KEY"))
---
> if (!token.getStr().toUpperCase().equals("KEY"))
480a508
>
*****CVS exited normally with code 1*****
-
Thanks for all the help,
Eric
-----Original Message-----
From: Pugh, Eric [mailto:[EMAIL PROTECTED]]
Sent: Wednesday, September 26, 2001 10:53 AM
To: '[EMAIL PROTECTED]'
Subject: Build Torque fails...
Hi,
I am trying to run the torque ant task sql2xml and for a simple table, it
seems to run forever, on my NT box the processor hits 90% utilization! So I
am trying to build torque, and running into problems..
I checked out the jakarta-turbine-torque library, and got all the src code
down. In the readme file it mentions a build.properties.sample that I
should rename build.properties, but that doesn't seem to exist. There also
is no build/ directory. The build.xml and default.properties files are at
the root.
So I am creating my own, but where is the easiest place to get some of the
harder to find libs like commons-colletions.jar and
commons-util-0.1-dev.jar? Should they have come when I downloaded the
source code? I found the commons-util-0.1 in the path [Apache] /
jakarta-turbine-torque / lib / Attic . The note from jvanzyl is that there
shouldn't be jars in CVs: "- no jars in cvs, even if it is a PITA right now.
we have to move in the right ... " Why is that? It seem to make building
the source harder for new bies?
At any rate, i guess I will keep poking around trying to find the various
jars...
Thank you for putting up with all the newbie questions...
Eric
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
