y* be that the reason for the slow execution is outdated index
cardinality statistics. Check out "CALL
SYSCS_UTIL.SYSCS_INPLACE_COMPRESS_TABLE" either on nabble or in the
derby docs.
--
Thomas Nielsen
Hi Bradley,
Your problem is really not that bizarre. Using a order by clause in a
subquery is actually not allowed in the SQL standard (neither SQL-99 nor
SQL-2003) if you read the spec closely.
I can only speculate why other DBs have decided to deviate from the
standard. Derby attempts to adh
wrote:
Is there any expected release date for derby 10.3.2.2, especially since
it resolves this https://issues.apache.org/jira/browse/DERBY-3603
Best Regards
David
--
Thomas Nielsen
ing a index on the score column, and
tailoring your query so that the optimizer chooses that index for the
subquery select row retrieval, effectively ordering the result for you?
There's still no guarantee the optimizer *will* use the index though.
Cheers,
Thomas
--
Thomas Nielsen
give
the RC a quick spin to see if the problem is solved?
Cheers,
Thomas
[1] http://ndpsoftware.com/HibernateMappingCheatSheet.html
[2] http://wiki.apache.org/db-derby/DerbyTenFourRelease -->
ReleaseCandidates
--
Thomas Nielsen
caxo wrote:
I have to take a certain number of rows ordered by a field.
And the derby say me error: Syntax error: Encountered "ORDER"
SELECT * FROM (
SELECT ROW_NUMBER() OVER () AS R,
playerId,
playerName,
points
FROM ranks
ORDER BY points DESC
)
AS TR WHERE R <= 10
following properties
DOMAINKEY are 98% same.
NAMEs are all different.
CREATIONTIME 82% different.
Hi,
Derby's optimizer sometimes makes bad decisions because the cardinality
statistics are outdated. You could see if this thread helps you:
http://www.nabble.com/Re%3A-FW%3A-Advice-on-*very*-badly-performing-query-p14140691.html
--
Knut Anders
--
Thomas Nielsen
e scanned ?
How can you make the prepared statement use the same execution plan as
the statement so as to return the same type of performance?
The problem seems to be Derby specific as MSQL returns adequate
performance for ALT 1.
Thanks in advance,
--
Thomas Nielsen
Hi Pertti,
Pertti Ylijukuri wrote:
I have understood that in derby doesn't support offset in sql.
Correct. Derby does not support offset.
To my knowledge offset is actually not part of neither the SQL99 nor
SQL2003 specs. Derby (attempts to) strictly adhere to these specs, and
this is one of
Hi Robert,
To my knowledge there are no plans for a 10.3 maintenance release at
this point.
But, the derby 10.4 release candidate is just a few days away (and has
the fix for 3244).
Maybe you could give that a try to see whether it fixes the problems you
experience?
If that does not help,
Hi Ryan,
Without knowing what version of derby you are using, I dare say you've
hit DERBY-3301:
http://issues.apache.org/jira/browse/DERBY-3301
The bug is basically a too optimistic optimization, causing derby to
return the wrong result with nested "where [not] exists" queries.
DERBY-3301
Jeffrey,
With "M:" as your drive letter, I'm guessing you have the files on a
mapped network share? Are you using Samba?
There used to be file locking problems with Samba. Not sure whether they
have been fixed or not, nor what versions were affected/fixed. Google is
your friend here. The Samba
Mark Thornton wrote:
The more interesting question is whether those limits should remain
different or should they be brought into alignment with Java Double
(and add +-infinite and NaN as well).
It would probably be good to fully support the java Double datatype, but
see DERBY-3290 for some con
games are on Xbox 360. Click here for a special offer on an
Xbox 360 Console. Get it now!
<http://www.xbox.com/en-US/hardware/wheretobuy/>
--
Thomas Nielsen
Hi Jay,
You'll find a list of all the built-in functions, and specifics of each,
in the derby 10.3 reference manual
http://db.apache.org/derby/docs/10.3/ref/
specifically in the topic "Built-in functions":
http://db.apache.org/derby/docs/10.3/ref/rrefsqlj29026.html
Hope this helps!
Cheers
musky,
That statement does not mean Derby has a limit of 25 concurrent clients.
Like Brian said in his reply, Derby can handle thousands of connections,
and they can all work at the same time. The statement is most likely the
result of a simple "sanity test" the author of the article did, or
format with a
smaller page size?
My lovely spouse bought me a Sony PRS-505 ebook reader and 8.5x11 pages
come out frighteningly cramped (but still legible) on its 3.5x4.75 inch
screen. I was thinking that documents that I refer to regularly would be
a useful thing to put on it.
Thank you,
Donald
--
Thomas Nielsen
FYI:
I filed handling NaN, POSITIVE_INFINITY and NEGATIVE_INFINITY as an
improvement suggestion in jira:
https://issues.apache.org/jira/browse/DERBY-3290
BR,
Thomas
Thomas Nielsen wrote:
Continuing into the SQL 2003 foundation spec[3] I can't find any
explicit mention of NaN/INF
anguages - SQL - Part 2: Foundation (SQL/Foundation)"
Thomas Nielsen wrote:
Neither the JDBC 4.0 specification[1] nor "JDBC API Tutorial and
Reference"[2] mention handling of the NaN or INFINITE numbers at all
AFAICT.
As there is no mention of how to handle *jdbc* doubles (and
l*
double (and literals).
Cheers,
Thomas
[1] JSR-221, JDBC 4.0,
http://jcp.org/aboutJava/communityprocess/final/jsr221/index.html
[2] "JDBC API Tutorial and Reference, Third Edt.", Fisher, Ellis &
Bruce, Addison-Wesley 2003.
Andrew ``Bass'' Shcheglov wrote:
On Dec
on this?
Seems your only option at the moment is to add another column to your
table to flag NaN values.
Thomas
Thomas Nielsen wrote:
I quickly tried both of Andrews 1) and 2) suggestions. Results pasted
below.
Summary:
- both PreparedStatement.setDouble() and setFloat() throw the same
Ls to represent NaNs at database level.
Regards,
Andrew.
--
Thomas Nielsen
Hi Aneez,
The LIMIT syntax is not part of the SQL standard (yet at least...), and
may as such be considered a MySQL dialect.
Derby does not support LIMIT, but we are working on implementing.the
approach taken in the SQL standard using the ROW_NUMBER() function for
the upcoming 10.4 release.Y
ended only for Thomas. Please disregard :-))
Bernt M. Johnsen wrote (2007-11-16 09:56:00):
Thomas Nielsen wrote (2007-11-16 08:17:17):
It would probably be better to simply do
catch (Throwable e)
Neineinei. aldri be nybegynnere gjøre "catch Throwable". Det skal
man ligge lan
--
> Get easy, one-click access to your favorites. Make Yahoo! your
homepage.
>
Never miss a thing. Make Yahoo your homepage.
<http://us.rd.yahoo.com/evt=51438/*http://www.yahoo.com/r/hs>
--
Thomas Nielsen
opening too many connections
--
Thomas Nielsen
Dag H. Wanvik wrote:
I think this code is old DB2 compatibility code, which could possibly be
lifted now. Does anyone know?
IMHO this limitation should be lifted.
Thomas
--
Thomas Nielsen
Thomas Nielsen wrote:
Release testing (and automated nightly testing) is currently run on all
1.4, 1.5 and 1.6 JVMs AFAIK, given availablility.
If you are interested in detailed information on some of the many OS/JVM
combinations that were actually tested for the current 10.3.1.4 release
a prefered JVM to use in order to run
derby server?
which is the most suitable for this task?
--
Thomas Nielsen
nateDialect="org.hibernate.dialect.DerbyDialect"
Has anyone here hit this one before?
TIA
Nick
--
Thomas Nielsen
this?
DERBY-581 and DERBY-2998 are relevant, I believe.
If you are interested in helping us work on this, there are lots of
ways to help contribute to the effort, and we'd love to have more people
involved!
thanks,
bryan
--
Thomas Nielsen
Bryan Pendleton wrote:
1. Why does int pop = 0 == 0 ? 1 : null; even compile?
For what it's worth, it doesn't compile for me:
works with 1.5.0-13:
C:\Program Files\Java\jdk1.5.0_13\bin>javac c:\tmp\pop.java
C:\Program Files\Java\jdk1.5.0_13\bin>
even with 1.5.0-07:
C:\Program Files\Java\jdk1
and derby runs
inside the JVM on the CDC device.
You should probably start having a look at something like this page from
the reference guide for 10.3:
http://db.apache.org/derby/docs/10.3/ref/rrefjdbcjsr169.html
Hope this gets you started :)
Thomas
--
Thomas Nielsen
ilable from Sun that
you could try, 1.5.0_12:
http://java.sun.com/javase/downloads/index_jdk5.jsp
Hope this helps :)
Thomas
--
Thomas Nielsen
that no longer supported?
I found this on Google, does it help?
http://www.ibm.com/developerworks/db2/library/techarticle/dm-0409cline2/index.html
thanks,
bryan
--
Thomas Nielsen
anks for letting the community know :)
If you don't mind - what were your top three pains when moving from
MySQL to Derby?
Cheers,
Thomas
--
Thomas Nielsen
wether you run embedded or not?
Cheers,
Thomas
--
Thomas Nielsen
to attempt a create table or another
writing statement after connecting. As the jar'ed db is read-only, a
writing statement should be rejected if you're really connected to the
jar'ed db.
Thomas
Thomas Nielsen wrote:
Hi Ryan,
First of all, can you please state what version you a
- I am just not sure where
my tables are.
Also, if I change the URL to access the the database on the filesystem
(from the previous project's build) via the directory subprotocol,
everything works. So, I know the database was created correctly at one
point in time.
I have been struggling with this for a day now and I am at a loss. Is
there anything obvious I am doing wrong? Did I miss a configuration step
to make a embedded JAR database work? Any help would be great.
Thanks.
Ryan
--
Thomas Nielsen
John C. Turnbull wrote:
- the database engine *may* be able to precompile the stored procedure,
and reuse that precompiled block at a later invocation of the stored
procedure for a (very slight) performance gain. Impact depends on the
nastyness of the expression and the engine itself.
This is no
I don't think so.
But you might be able to do it anyway - grant access to the table(s)
only to a "sqlproc" (or whatever) user, and then execute the stored
procedure as that user?
Cheers
Thomas
--
Thomas Nielsen
FYI:
There's been some discussion about this issue on the derby-dev list lately.
Thomas
Kasper, Kay wrote:
hi Ole,
i have the same problem (derby version 10.2.2) and my opinion is, that the
server has stopped, but should not have done that. it should have rejected
the connection without a stop
Thomas Nielsen wrote:
The idea is to do something similar to a java System.exec(), which you
probably already know, but c-style.
Runtime.exec()
oh well...
T
--
Thomas Nielsen
rested in that:
"fork a new process running ij".
I virtually know nothing about that topic.
Could you please give me some simple
"HelloWorld" example on that?
2007/6/14, Thomas Nielsen <[EMAIL PROTECTED]
<mailto:[EMAIL PROTECTED]>>:
yarono,
If Tonys sugge
e functions (meaning
create DB and shutdown) via the ij prompt, run my code to perform table
creation, insertions, etc.. The shutdown and creation of the db's is done
through the ij prompt, while the rest of my code is in c.
I want the whole thing to be done in c.
Thomas Nielsen - Sun
ee is what you are looking for.
Hope this helps :)
Cheers,
Thomas
yarono wrote:
Is there a way to run the 'ij commands' in my code (written in c), just as I
execute SQL statements?
--
Thomas Nielsen
46 matches
Mail list logo