Though I've never used oraclecursor with iBatis, I'll dare to note that
from the image attached I see that parametermap property for the
ORALECURSOR is the last one in the map.
But it must be the first one (order parameter map properties must match
order of ?,?... ).
On Fri, Aug 21, 2009 at 8:
>
> #3 - I think I'm the only person who does moderation on the ibatis
> lists and don't want another one to manage. (NOTE: If someone wants to
> do that with me, let me know, it's easy, but when I'm out of the
> office the "please subscribe" emails pile up).
>
I'll be out of the office till Septe
No.
Lol
On Fri, Aug 21, 2009 at 3:23 PM, Larry Meadors wrote:
> Because it's a bad idea.
>
> Happy?
>
> Larry
>
>
> On Fri, Aug 21, 2009 at 12:47 AM, Nicholoz Koka
> Kiknadze wrote:
> > No reply...
> >
> > Seems separate list idea is soo absurd
No reply...
Seems separate list idea is soo absurd that no one bothers to explain why ;)
On Tue, Aug 18, 2009 at 10:00 PM, Nicholoz Koka Kiknadze wrote:
> I've enjoyed much using iBatis and the mailing list was always very
> helpful.
>
> Well, now and then I have to search f
ig/SQLMapConfig.xml"/>
>
> />
>
>
>
> com.ibatis.sqlmap.engine.transaction.external.ExternalTransactionConfig
>
>
>
>
> false
>
>
>
&
I've enjoyed much using iBatis and the mailing list was always very helpful.
Well, now and then I have to search for answers in the mailing list and
usually find the answer ;) So far so good. But... I've used mailing list for
another framework - Tapestry for three major versions (3/4/5) and always
Nichol,
>
> Thanks for your response. We are already using same transaction while
> calling these procedures.
> But we are not getting expected results.
>
>
>
>
> Nicholoz Koka Kiknadze wrote:
> >
> > You are quite right. Your problem is that calls to tho
You are quite right. Your problem is that calls to those two procedures are
not in a single transaction. Include them in a single transaction and proc2
will see package level variables set in proc1. Note that executing those
procs on a single connection object is not enough (i.e. you need
transacti
Maybe I'm totally wrong, but I'm almost sure it must be possible to force
database server to RE-generate execution plans even for cached statements.
I've never used DB2 but googling a little I found:
http://publib.boulder.ibm.com/infocenter/dzichelp/v2r2/index.jsp?topic=/com.ibm.db29.doc.ugref/db2
is the connection between ibatis and mysql
>
>
>
>
>
> *From:* Nicholoz Koka Kiknadze [mailto:kikna...@gmail.com]
> *Sent:* Tuesday, July 14, 2009 9:52 PM
> *To:* user-java@ibatis.apache.org
> *Subject:* Re: problem with insert
>
>
>
> Is it po
Is it possible that you just did not set your log file encoding? Kind of
log4j.appender.someappender.Encoding=UTF-8
On Tue, Jul 14, 2009 at 8:40 PM, Odelya YomTov wrote:
> Hi!
>
> I am trying to insert data in Hebrew into the database
>
> The table is set to utf-8.
>
> But in the log file I
Have you tried enclosing field names containing # with *CDATA *? Maybe it
will help...
On Thu, Jun 25, 2009 at 2:56 PM, wrote:
> Hi,
>
> in our DB2 database, some columns have a # in their name, e.g. ID#
> This conflicts now with the iBatis placeholder character #, e.g.
>
> select * from tab
Turn on logging (see iBatis docs) to see what does the generated SQL
statement look like.
On Fri, Jun 19, 2009 at 7:34 AM, BruceKKK wrote:
>
> i met ora-01795 error, so i need separate list to several size limited
> lists,
> I am trying use nested iterators to iterate over a List of List using
I think the only reliable way to check whether connection is valid is to
execute something using the connection (so it requires roundtrip to db).
However
a) this execution happens only once for your transaction (which may include
numbers of CRUD operations etc)
b) Validation query is chosen to be s
I think what happens to open (pooled) connection upon JVM exit depends on
JDBC driver.
IMO if you were using SimpleDataSource of iBatis, you'd have to call
forceCloseAll() to close all pooled connections at the end of execution of
your app (I think I have even used that years ago). Not sure about
> IBatis creates this select:
>
> to_date('2009-05-15 23.59.59', '-mm-dd hh24.mi.ss')
Well, there's no to_date function in your XML, so iBatis can not create it.
It only prepares statement (ps) with
AND b.ins_time >= ?
AND b.ins_time <= ?
and calls ps.setDate
Could you re-check your xm
It may be database dependent. Here's extract from oracle docs
http://download.oracle.com/docs/cd/B10501_01/java.920/a96654/oraperf.htm
After a batch exception, the update counts array can be retrieved using the
getUpdateCounts() method of the BatchUpdateException object. This returns an
int array
> Have I
> generate for every session an own SqlMapClient or can they share one. I
> found no example
> which shows it.
Share. It's thread safe.
Well, let DBA optimize query, but from the iBatis side maybe it makes sense
to increase fetchSize. See e.g.
http://markmail.org/message/3vy7p455yr5qkzbk#query:ibatis%20fetchsize+page:1+mid:tgs7lyqlyetn2jfv+state:results
On Wed, Apr 29, 2009 at 2:01 AM, Chris O'Connell <
oconn...@gorillachicago.co
To flush on demand you can just call
SqlMapClient.flushDataCache(String cacheId).
HTH
On Tue, Apr 28, 2009 at 10:32 AM, WhyDontYouSlide wrote:
>
> Hello,
>
> I was asking myself if It's possible to force flushing of the cache without
> execute a "real" statement
>
> for example:
>
> the tradi
Try $thresholdValue$
On Mon, Apr 27, 2009 at 5:43 PM, Alex Sherwin
wrote:
> #thresholdValue# will put '100' in the SQL that is created (note: with
> single quotes, making it a non-number value)
>
> I assume what you want is to use $thresholdValue$, which will explicitly
> insert that value from
Maybe this would help
http://markmail.org/message/3r7dx4uixs273kxu
On Tue, Apr 21, 2009 at 1:33 PM, sjpn wrote:
>
> could it be possible that ibatis is return back an empty cached resultset?
> if
> so, how can i disable this cache or trigger it to clear?
> --
> View this message in context:
>
http://opensource.atlassian.com/confluence/oss/display/IBATIS/Oracle+REF+CURSOR+Solutions
On Tue, Apr 21, 2009 at 12:19 AM, sjpn wrote:
>
> Hi,
> I want to use ibatis as the standard for a large project. The client I'm
> working for does everything in stored procs for performance and transaction
Follow Nathan's suggestion.
>Look up ping query in the developers manual. It will validate the query
before it uses it
In fact it will validate connection, discard the old one if necessary and
opens/return fresh one.
On Tue, Apr 14, 2009 at 11:13 PM, yhliu wrote:
> Thanks for your reply.
>
No I think. iBatis is not aware whether your VO-s have changed and can not
decide whther to run updates or not.
Why don't you override the Object.hashcode() method on your VO and
> call it and save the value when you retrieve the record. Call it again when
> the user submits their chang
In the manual we read:
The *jdbcType* element is used to specify the JDBC type of the parameter.
The value must be one of the types listed in java.sql.Types (VARCHAR,
INTEGER, etc.) Generally the *jdbcType* element is needed if there is a
possibility that the value could be NULL, or to specify t
>
>
> 2009/3/25 Nicholoz Koka Kiknadze :
> > I think you can use any DataSource (including ProxoolDataSource). Just
> use
> > full class name in your transactionManager's dataSource element and IMO
> it
> > should work.
> >
> > On Mon, Mar 23, 2009
I think you can use any DataSource (including ProxoolDataSource). Just use
full class name in your transactionManager's dataSource element and IMO it
should work.
On Mon, Mar 23, 2009 at 3:39 AM, Mike.G wrote:
> Hi, everyone, ibatis can use proxool connection pool? how?
>
> thanks
>
> Mike.G
>
Found this:
http://www.nabble.com/Statement-Caching-Question-td20437084.html
Hope it helps
GL
2009/3/25 Daniel Sánchez Gómez
> Hi,
>
> I have a problem with open cursors using Ibatis in an Oracle data base.
>
> When I open a transaction with a lot of querys and before the transaction
> ends,
Meant ojdbc6.jar of course
On Sat, Mar 21, 2009 at 2:08 AM, Nicholoz Koka Kiknadze
wrote:
> URL seems fine. Try using ojdb*6*.jar, I'm using it without problems..
>
> GL
>
>
> On Fri, Mar 20, 2009 at 9:35 PM, Sergey Livanov
> wrote:
>
>> driver , url, use
URL seems fine. Try using ojdb*6*.jar, I'm using it without problems..
GL
On Fri, Mar 20, 2009 at 9:35 PM, Sergey Livanov wrote:
> driver , url, user & pwd.
> jdbc.ORAdriverClassName=oracle.jdbc.driver.OracleDriver
> jdbc.ORAurl=jdbc:oracle:thin:@192.168.1.15:1521:ORCL
> jdbc.ORAusername=server
Wow, took less than 24 hours ;).
On Fri, Mar 20, 2009 at 5:23 PM, Jeff Butler wrote:
> Support for this is now available in SVN.
>
> Jeff Butler
>
> On Fri, Mar 20, 2009 at 6:57 AM, Jeff Butler
> wrote:
> > Interesting idea.
> >
> > Ibator does not currently capture database field comments from
; > Kengkaj
> >
> > On Fri, Mar 20, 2009 at 6:13 PM,
> > Leffingwell, Jonathan R CTR FRCSE, JAX 7.2.2
> >
> > wrote:
> >
> > What does that mean, "set
> > reference to it transient"?
> >
> >
> >
> >
> >
>
My first thought is that maybe StandardDaoManager can not be serializable as
it deals with database connections. Maybe makes sense to set reference to it
transient?
On Thu, Mar 19, 2009 at 2:55 PM, Leffingwell, Jonathan R CTR FRCSE, JAX
7.2.2 wrote:
> I am running a web app in IBM RAD 7 wit
>
>
> It's possible to display the executed SQL somehow ?
Sure, it's possible to log what statements are prepared, what parameters are
set to and what results are retrived. Check the
'Logging SqlMap Activity' section in developer guide.
tus ? "Some Default Value" : status;
>
> }
>
>
> Alin Popa wrote:
>
>> Thanks Nicholoz,
>>
>> This is what I've done. And I also have a facade that will abstract
>> for the user, the insert actions.
>> I thought that ibatis knows how to ha
> Why don't you just exclude this field in the ?
INSERT INTO vjobs(name, description) VALUES
(#name#,#description#)
GL
IMO because sometimes one need to insert NULL values too. So why don't you
add another insert statement to your xml that will omit the status field?
On Wed, Mar 18, 2009 a
Add it manually :) Yours is such a weird problem, some unique combination
of some circumstances. I bet there's some no-framework related problem, both
iBatis and iBator are so simple or rather transparent (like everything
genial)
On Wed, Feb 18, 2009 at 4:41 PM, sanv wrote:
>
> Is there any oth
I recall one has to call
com.ibatis.common.logging.LogFactory.selectLog4JLogging()
before sqlmap is initialized
On Fri, Feb 6, 2009 at 4:25 PM, Stanley, Eric wrote:
> Richard,
>I know its being read. As I mentioned below, all other
> components of the app are using it just fine. My app,
I have used it with TimesTen 6 / 7 with JDK6. Had no iBatis related
problems.
On Thu, Jan 29, 2009 at 1:39 AM, Avitzur Alon wrote:
> Hi,
>
> I am thinking of using iBATIS in my application in order to access the DB.
> I have two questions:
>
> 1. Does this version support JDK 1.6?
>
> 2. Does i
Maybe it makes sense to switch to DBCP as it has testOnBorrow setting (needs
a validation query of course) and the DBCP docs saying:
testOnBorrow: The indication of whether objects will be validated before
being borrowed from the pool. If the object fails to validate, it will be
dropped from the p
0 == Doesn't matter to me
Absolutely
e you're still using a "pool", but your max, min, idle, and
>> active connections are all equal (i.e. 16). Otherwise, how do you allocate
>> connections to the incoming requests?
>>
>> Cheers,
>> Clinton
>>
>>
>> On Tue, Jan 20, 2009
re all equal (i.e. 16). Otherwise, how do you allocate
> connections to the incoming requests?
>
> Cheers,
> Clinton
>
>
> On Tue, Jan 20, 2009 at 12:33 PM, Nicholoz Koka Kiknadze <
> kikna...@gmail.com> wrote:
>
>> Ours is an application that requires guaranteed
Ours is an application that requires guaranteed response times under 50 ms,
so:
1) We dropped using any kind of pool, so that
2) number of constantly open connections equals to the number of processors
(16)
3) I know you were asking about pool, but still I dared to respond with this
no-pool varia
Thank you Jeff for this wonderful tool!
Plugins making classes serializable and adding equals/hashcode are exactly
what I wished for ;)
May I ask whether standalone version is fully functional - initially it
was not, but then I was using Eclipse anyway.
Merry Christmas and Happy New Year
On
Please enable logging to see what gets executed and specifically what value
for #jobPostingId# is passed.
On Thu, Dec 4, 2008 at 9:15 PM, Michael He <[EMAIL PROTECTED]> wrote:
>
> Nothing happens without deleting anything
>
> Micheal
>
>
> Kai Grabfelder-2 wrote:
> >
> > hm looks like a bug to m
2b) Scenario: copy/pasting from manual:
__
The element also allows an optional attribute *
commitRequired* that can be *true* or *false*. Normally iBATIS will not
commit transactions unless an insert, update, or delete operation has been
performed. *This is true even if yo
>
>
>
>
>
>
>
>
>
> INSERT INTO MSGS ( SY_ID1, SY_ID2, MS_COUNT, MS_TITLE)
> VALUES (#id1#,#id2,#messageCount#,#title#)
>
>
>
> UPDATE MSGS SET MS_COUNT=#messageCount#, MS_TITLE=#title# WHERE
> SY_ID1=#id1# AND SY_
On Wed, Nov 26, 2008 at 8:05 AM, Alejandro D. Garin <[EMAIL PROTECTED]>wrote:
> Hi Ingmar
>
>
>> provide an instance of Person as parameter, witch contains the values to
>> store.
>>
>>
>> UPDATE Person SET
>>name = #name:VARCHAR#,
>>lastname = #lastname:VARCHAR#
>> WHERE id = #id:INTEG
>
>
>
>> When I deploy everthing to a Linux box I get the weirdest errors from the
>> Driver.
>> --- Cause: java.sql.SQLException: [unixODBC][INTERCHAIN][UNIMS-ODBC]
>> [UNIMS]
>>
>
Well, maybe we can blame differences in ODBC layer? I've have never used
ODBC (or ODBC-JDBC bridge) on linux, any cha
>Or can somebody suggest another solution?
Maybe ask your DBA to increase max open cursors allowed? ;)
Honestly, I never quite understood iBATIS statement cacheing performance
gains with oRACLE (which IMO maintains stetement cache on its own). Are you
sure disabling it throughout your application
oops, misspelled, should be
gl ;)
On Tue, Nov 11, 2008 at 5:30 AM, <[EMAIL PROTECTED]> wrote:
>
> XML does not support < and > operator as it is. You have to use
>
>
> Hope this will help you
>
> Regards
> Ankit
>
> -Original Message-
> From: Mehak [mailto:[EMAIL PROTECTED]
> Sent:
Look into the sqlmaps guide. There are condidtions like
On Tue, Nov 11, 2008 at 1:37 AM, Mehak <[EMAIL PROTECTED]> wrote:
>
> Hi
> Please help me with this simpl problem regarding using operators in compare
> value.
> I have a condition something like:
>lastUpdatedDate > givenTime, then
If I get it right you need to execute same statements to multiple (two)
databases. Have a look at http://ha-jdbc.sourceforge.net/. I have NOT used
it, but it seems to be an easy and transparent way to achieve what you need.
Should not you use CDATA sections to get quotation marks intact in your SQL
? And like Jeff pointed out, look at the generated statements in your log
miles_davis is out also, what a pity ;)
On Sat, Aug 30, 2008 at 12:03 PM, <[EMAIL PROTECTED]> wrote:
> I will be out of the office starting 08/29/2008 and will not return until
> 09/08/2008.
>
> I will respond to your message when I return. Please contact Alan May if
> you need your questions add
To be honest in the first place I'd ask DBA to use partitioning if your db
engine supports that, and if not, to create union USERS query to run my
queries against.
AFAIK the root of the problem is that when iBatis handles resultset of left
joining master table with child table the child table columns may be all
NULL and it's not quite clear whether child table indeed contained
corresponding row with NULL elements or there was no result found in the
child tabl
Indeed the link is broken. But the subversion one works:
http://svn.apache.org/repos/asf/ibatis/trunk/java/ibatis-2/ibatis-2-docs/en/
Anyway usually docs in SVN are more up to date
Peace
On Mon, Aug 11, 2008 at 11:04 PM, Reuben Firmin <[EMAIL PROTECTED]>wrote:
> I'm looking for the Ibatis Ma
Have a look at this thread:
http://www.mail-archive.com/user-java@ibatis.apache.org/msg11828.html
Good luck
On Wed, Aug 6, 2008 at 4:47 PM, Brian Parkinson <[EMAIL PROTECTED]> wrote:
> Hello:
>
> So, I have things up and running under Java and iBATIS, but running into
> some starvation issue, a
shMap();
>m.put("this", "this");
>m.put("is", "is");
>m.put("a", "a");
>m.put("test", "test");
>for(Object o:m.values()) System.out.println(o);
>}
>
Hi,
I get results using resultClass="java.util.HashMap" and I expected that if I
iterate values of the HashMap the order will correspond to the order in my
fields. However it does not. Is that normal behavior?
In other words I have 'Select f1, f2, f3 from Blah', and iterating in the
HashMap value
No as long as you want to avoid new object creation. But
tr1 = getSqlMapper().queryForObject("selectTransferByRefId",tr1)
do I miss anything?
On Fri, Jun 27, 2008 at 7:56 PM, aleksey2 <[EMAIL PROTECTED]> wrote:
>
> Transfer tr1 = new Transfer();
> tr.setId(123);
> Transfer tr2 = getSqlMapper(
> wildcard parameter like qu*
Should not it be Like qu% ?
Turn on logging for connection and iBatis and check what SQL is actually
executed and whether it returns anything when executed from TOAD.
On Wed, Jun 18, 2008 at 3:13 AM, <[EMAIL PROTECTED]> wrote:
>
> Dear sir(s):
>
> Curren
First of all I'd add catch block to log errors to your try/finally. In your
code you do multiple inserts, so if any of them fails your whole transaction
will roll back. I can see no other possible scenario :(
Here's extract from iBatis docs:
*Normally iBATIS will not commit transactions unless an insert, update, or
delete operation has been
performed. This is true even if you explicitly call the commitTransaction()
method. This behavior
creates problems in some cases. If you want iBATIS to always commi
Jeff,
Just for an argument ;) -- you are understating powers of iBatis ;) iBatis
can return list of objects running multiple susbselects for each row
( ) - something you'd have to do manually in
pure JDBC. In fact, I'd go further allowing inserting object with list
properties in a single sqlmap me
> It is a trivial matter to write these methods
Exactly Jeff, that's why I hoped it would be a trivial addition to sqlmap
API.
I'd be happier not to write those trivial things over an over but rather
keep them in one place. As for transactions, obviously one can revert back
to looping in applicati
Thanks for an opinion Richard
Typically, there isn't a need to insert several items into the same table at
> once
How about shopping cart - most typical application ;) Actually I have
applicants with lists of faculties and exams. Every list property of newly
created object needs to be inserted,
ot;,List)
batchUpdate("updateSingleSomeObject",List)
Think it will make iBatis more consistent: we have queryForList, i.e.
getting list of objects in a single line of code, so it's quite natural to
expect that one should be able to persist the list in a single call.
Wonder what others think about it.
Nicholoz Koka Kiknadze
seconds maybe.
>
>
> On Tue, May 27, 2008 at 4:28 PM, Nicholoz Koka Kiknadze <
> [EMAIL PROTECTED]> wrote:
>
>> Can your query raise exception that's handled elsewhere, so that
>> session.close() is not called? Try this:
>>
>>
>> SqlMapSes
I'm inserting images into MS Sql (table field type is image i.e.
variable-length binary data) without problems with no special configuration.
I have an Image class with two properties
String imageID
byte[] imageBytes
and the insert looks like
insert into images (ImageID,ima
Can your query raise exception that's handled elsewhere, so that
session.close() is not called? Try this:
SqlMapSession session = sqlMap.openSession();
List list = null;
try {
list = (List) session.queryForList("queryId");
}
catch (Exception e)
{
As of Oracle 9 I'm certain that batching CallableStatements was not
implemented (it did not throw exception but in fact did not batch statements
).
Well as Oracle 10g 11g docs state that
> batch api is effective only only for preparedStatement
seems nothing has changed in these releases.
> I th
Well, here's thread discussing same problem, and it seems private setId()
should work
http://www.mail-archive.com/user-java@ibatis.apache.org/msg08629.html
I noticed there are only two tables involved, could you try with some
different database engine ? If you still get similar exception, one should
start inspecting your code/iBatis code, if not, definitely it's a driver/db
server problem (try different driver)...
Well, as much as I know there's no support for wildcards :(
>are there reasons against implementing them?
Guess no, maybe just restricted developer resources? IMO, after getting rid
of much (too much, too too much) of manual coding with the help of abator
(now it's ibator) I think that it could a
getting list of objects in a single call, so it's quite natural to expect
that one should be able to persist the list in a single call.
Wonder what others think about it.
Nicholoz Koka Kiknadze
Sorry, 'it'll save your headache' should read 'it'll save you from your
headache' ;)
On Thu, Apr 17, 2008 at 1:06 AM, Koka Kiknadze <[EMAIL PROTECTED]> wrote:
> Is there a better way of dealing with this?
I go for Spring framework's declarative tra
> Is there a better way of dealing with this?
I go for Spring framework's declarative transactions. Spring has support for
iBatis and definitely it'll save your headache.
Thanks for continuing improving the framework that's already cool.
May I ask about IBATIS-375? It seems fix has been contributed and I hoped it
to make through into the next release. The thing that its already there in
iBatis .NET makes me a bit jealous ;)
I've done some performance measurements too over an year ago and seems I've
pinpointed (with the help of Clinton) where one can loose dozen of
percents with Oracle driver, take a look at the thread
http://mail-archives.apache.org/mod_mbox/ibatis-user-java/200701.mbox/[EMAIL
PROTECTED]
GL
Imo no :(
And you'd better check the batch capabilities of iBatis, though maybe
somewhere soon Jeff will implement and autogenerate it in Abator? Sounds
practical?
Think you can limit how many symbols the user can enter to some reasonable
value. If you can limit it, say to 20, you can use something like
Select * from
Select * from table where column LIKE '$value$%'
i.e. malicious user will have to use 20 closing par
> I haven't implemented it yet, but I can't see a reason why it wouldn't
work.
I can not see either, but, to speed up such searches one usually indexes
search column. While index will work well for Column LIKE 'bla%', I doubt it
will be of any use with ANY_FUNCTION(Column...)...
Abator.
>
> Hope that helps...
> Jeff Butler
>
>
>
>
>
> On Dec 26, 2007 6:07 AM, Koka Kiknadze <[EMAIL PROTECTED]> wrote:
>
> > Merry Christmas,
> >
> > At last found time to start playing with Abator. Wow, after the first
> > ru
dn't do
> that just to satisfy Abator.
>
> Hope that helps...
> Jeff Butler
>
>
>
>
>
> On Dec 26, 2007 6:07 AM, Koka Kiknadze <[EMAIL PROTECTED]> wrote:
>
> > Merry Christmas,
> >
> > At last found time to start playing with Abator. Wow
Merry Christmas,
At last found time to start playing with Abator. Wow, after the first run
it's already clear it'll save tons of time.
The things I found a bit annoying are the following:
I am used to table names like: Employees, Cities
And the corresponding classes: Employee, City
Is it possibl
>
>
> It seems not give me what I expect as mentioned in the question.
>
Can not think of any explanation except that maybe you do not have an ORDER
BY clause in your select, so that each time you execute queryForList the
records are not returned in the same guaranteed order.
Jeff,
Think the question was not about including OR-s in the SQL but executing SQL
only if OR-ed expression is true. Anyway, the right place of OR-s seems to
be in java, but it seems to me one can achieve the same adding the following
to the desired SQL (using appropriate syntax for specific datab
Well, it's ojdbc14.jar that wed are using for cple of years...
On 11/2/07, Hemant. Kamatgi <[EMAIL PROTECTED]> wrote:
>
> We had used ojdbc.jar to configure our datasource. We changed it to
> classes12.zip and then the bug was resolved. Thanks for the suggestions,
> anyways.
>
> -Original Me
What was wrong with replys you got to this question recenly?
Like this one from Larry Meadors:
>I think I'd do it in the SQL - that way the rest of the app is clean.
>>I have no control over the structure of the db
>I think I'd do it in the SQL - that way the rest of the app is clean.
+1
And hope that some day you'll gain some control over the structure of the
db ;)
On 10/9/07, Claus Hausberger <[EMAIL PROTECTED]> wrote:
>
>
> do I always have to specify all column names in the query that are also in
> the ResultMap ?
>
Yes.
So I guess you are reusing the resultMap and in some queries some columns
may be absent (i.e. you do not care about corresponding java
Could you post your Select that causes error and ResultMap?
If you omit column that's present in the ReusltMap I guess you'll get
invalid column name exception...
How about underlying connection pool? Imho one needs to close connection
On 10/5/07, Larry Meadors <[EMAIL PROTECTED]> wrote:
>
> As with anything, you can set them to null just to be safe - if you do
> that, the GC *may* get them sooner than if you don't.
>
> If you don't set them to null, they'
Well, as long as error is after
DEBUG ResultSet - {rset-12} ResultSet
DEBUG SimpleDataSource - Returned connection 33534746 to pool.
it seems to me that error occured after executing PreparedStetement,
retrieving resultset and even returning connection to pool. Next comes
mapping resultset to
Can you provide the log? Maybe error is raised not at the query execution
but while mapping
resultset to java object, i.e. you have typo in columnname in resultMap
On 10/5/07, Claus Hausberger <[EMAIL PROTECTED]> wrote:
>
> Jeff, thanks. but that didn't help.
>
> I made a mistake describing my pro
Here is an ugly way:
Wrap sequence selects in database procedures with out param returning
sequence value. Calling procedures have same syntax in both database
flavours...
1 - 100 of 178 matches
Mail list logo
