squirrel-sql-develop — A subscriber-only list for those interested in developing SQuirreL and Plugins.

[Squirrel-sql-develop] Performance HowTo?

[cookiebox <ms...@in...>]

Hi,

I've a complex infrastructure with an Oracle-DB, an ApplicationServer and
Client. The AppSrv Queries via "Default-JDBC" are very slow. So I've tried
to run the same Query via Squirrel and it's very fast. How is it possible? I
tried to debug Squirrel an can't find any differences. Same driver (class),
Methods aso.

After that I looked at the ServerTrace. There are differences :

slow query summery (15 sec.):

        1  session in tracefile.
      19  user  SQL statements in trace file.
     146  internal SQL statements in trace file.
     165  SQL statements in trace file.
      42  unique SQL statements in trace file.
    1911  lines in trace file.
      22  elapsed seconds in trace file.


fast (Squirrel) summery (0.7 sec):

       1  session in tracefile.
      17  user  SQL statements in trace file.
      44  internal SQL statements in trace file.
      61  SQL statements in trace file.
      18  unique SQL statements in trace file.
     891  lines in trace file.
       0  elapsed seconds in trace file.

Squirrel reaches nearly the same speed like a "native" call at DB.
Need (JDBC-)Help :confused:

Best Regards!
-- 
View this message in context: http://old.nabble.com/Performance-HowTo--tp29057260p29057260.html
Sent from the squirrel-sql-develop mailing list archive at Nabble.com.

Re: [Squirrel-sql-develop] When will Squirrel Memory Issues be Addressed?

[Robert M. <rob...@gm...>]

I don't know if anyone else feels this way, but bringing back the
entire contents of the database and browsing them to find what you are
looking for doesn't seem very helpful or even efficient.  SQuirreL
does provide the ability to limit the results so that a user can
browse a subset of all of the data when they have no idea what they
are looking for.  That mechanism seems to work pretty well at limiting
the RAM footprint.

Rob

On 6/2/10, bcarle1 <bca...@dr...> wrote:
>
> Hi all,
>
> I recently switched from the Sql Server Client application to the Squirrel
> Client.  And while I think Squirrel is a very good multi-lingual sql
> graphical client, I do have some issues with how Squirrel handles memory.
>
> Why on earth does squirrel keep all the results of a query in RAM?  In a
> production environment where you might run a query with results of a Gig or
> more, this really eats up your computer's memory fast.  Why doesn't Squirrel
> write those results to disk and then only pull them if they are being
> viewed.  This is how Microsoft's Sql Client works and is able to run a query
> with 2 GB of results, yet only take up less than 300 MB of RAM.  It writes
> it all to disk until the user tries to view it.
>
> I don't know if this is in the pipeline for you fellows or not, but it
> should be.
>
> Any comments?
> --
> View this message in context: http://old.nabble.com/When-will-Squirrel-Memory-Issues-be-Addressed--tp28759538p28759538.html
> Sent from the squirrel-sql-develop mailing list archive at Nabble.com.
>
>
> ------------------------------------------------------------------------------
> ThinkGeek and WIRED's GeekDad team up for the Ultimate
> GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the
> lucky parental unit.  See the prize list and enter to win:
> http://p.sf.net/sfu/thinkgeek-promo
> _______________________________________________
> Squirrel-sql-develop mailing list
> Squ...@li...
> https://lists.sourceforge.net/lists/listinfo/squirrel-sql-develop
>

[Squirrel-sql-develop] When will Squirrel Memory Issues be Addressed?

[bcarle1 <bca...@dr...>]

Hi all,

I recently switched from the Sql Server Client application to the Squirrel
Client.  And while I think Squirrel is a very good multi-lingual sql
graphical client, I do have some issues with how Squirrel handles memory.  

Why on earth does squirrel keep all the results of a query in RAM?  In a
production environment where you might run a query with results of a Gig or
more, this really eats up your computer's memory fast.  Why doesn't Squirrel
write those results to disk and then only pull them if they are being
viewed.  This is how Microsoft's Sql Client works and is able to run a query
with 2 GB of results, yet only take up less than 300 MB of RAM.  It writes
it all to disk until the user tries to view it.

I don't know if this is in the pipeline for you fellows or not, but it
should be.  

Any comments?
-- 
View this message in context: http://old.nabble.com/When-will-Squirrel-Memory-Issues-be-Addressed--tp28759538p28759538.html
Sent from the squirrel-sql-develop mailing list archive at Nabble.com.

Re: [Squirrel-sql-develop] [Squirrel-sql-users] Trouble on starting Squirrel 3.1 on Win XP after fresh installation

[Robert M. <rob...@gm...>]

Jørgen,

Unfortunately, this change ( %CLASSPATH% -> "%CLASSPATH%" ) breaks
anyone with correctly quoted paths in their CLASSPATH environment
variable (see http://sourceforge.net/tracker/index.php?func=detail&aid=2964850&group_id=28383&atid=393414).
 My testing revealed that quoting and already quoted path effectively
removes the original quotes around the path with spaces.  It seems to
me that the best approach would be to split the CLASSPATH variable
value into paths, and test each path for spaces to determine whether
or not it needs quoting, thereby fixing the CLASSPATH variable as it
is defined to be something acceptable to the JVM.  Writing batch files
is not something I do all the time, so I've never encountered the
syntax that would accomplish this task.  If someone would like to
propose a patch that does the above (using only commands that can be
found on any clean install of Windows XP SP2/Vista/7), I would be more
than happy to test it out and commit and prepare a new release if it
works properly.   For now, until we have a bullet-proof way of
"fixing" the quotes on the CLASSPATH, I will be removing the reference
to it from the batch file that launches SQuirreL on Windows.

Rob

On Tue, Apr 20, 2010 at 6:29 AM, Jørgen Nørgaard <jn...@an...> wrote:
> Hi,
>
> If you have spaces in path names in your java class path this may help you:
>
>
> Change line 37 (probably) to
>       SET SQUIRREL_CP=%TMP_CP%;"%CLASSPATH%"
> from
>       SET SQUIRREL_CP=%TMP_CP%;%CLASSPATH%
>
>
> The bat-file on windows have changed between quoting and not quoting the
> CLASSPATH environment variable.
>
>
>> Hi Christian,
>>
>> adding a "@rem" in front of the final "exit"-line of the
>> "squirrel-sql.bat" should keep the cmd-shell open and so you should be
>> able to read the messages.
>> What kind of databases do you want to connect to and which JDBC-drivers
>> do you use? I haven't seen the effect, that connecting to an alias will
>> fail
>> after "Test" of the alias has worked successfully. Do you get an error
>> message when you try to connect to one of the aliases? Do you use the
>> "Auto Logon"
>> option of the alias?
>>
>> Hope that helps
>>
>> Cheers
>>
>> Andreas
>>
>> Christian Hausknecht schrieb:
>>> -----BEGIN PGP SIGNED MESSAGE-----
>>> Hash: SHA1
>>>
>>> Hello,
>>>
>>> I just downloaded SQuirreL 3.1 and installed it. I have trouble to start
>>> it via the delivered BAT file. I read some other threads and uncommented
>>> the default start-Method and activated the one for "older" OS.
>>>
>>> The problem is, that the CMD-Shell-Window allways closes after printing
>>> out messages. So I am not even able to read the error message. Even if I
>>> start a fresh CMD-Shell, navigate to the install folder and start the
>>> BAT-file manually.
>>>
>>> Would be great if somebody can help me with that.
>>>
>>> The jar starts well btw. But I cannot connect to any Database, even if
>>> Squirrel claims that everythong is ok when testing an alias.
>>>
>>> Greets,
>>> Chris
>>> -----BEGIN PGP SIGNATURE-----
>>> Version: GnuPG v1.4.10 (MingW32)
>>> Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
>>>
>>> iEYEARECAAYFAkvMNiMACgkQx1N5zUp+BXKkCgCdG+rc0zGNjVnfZHr08QSxrERO
>>> 7SoAn0SbTjkcPBGyaipcGc4B+i3oEEOD
>>> =20+B
>>> -----END PGP SIGNATURE-----
>>>
>>> ------------------------------------------------------------------------------
>>> Download Intel&#174; Parallel Studio Eval
>>> Try the new software tools for yourself. Speed compiling, find bugs
>>> proactively, and fine-tune applications for parallel performance.
>>> See why Intel Parallel Studio got high marks during beta.
>>> http://p.sf.net/sfu/intel-sw-dev
>>> _______________________________________________
>>> Squirrel-sql-users mailing list
>>> Squ...@li...
>>> https://lists.sourceforge.net/lists/listinfo/squirrel-sql-users
>>>
>>>
>>
>>
>> ------------------------------------------------------------------------------
>> Download Intel&#174; Parallel Studio Eval
>> Try the new software tools for yourself. Speed compiling, find bugs
>> proactively, and fine-tune applications for parallel performance.
>> See why Intel Parallel Studio got high marks during beta.
>> http://p.sf.net/sfu/intel-sw-dev
>> _______________________________________________
>> Squirrel-sql-users mailing list
>> Squ...@li...
>> https://lists.sourceforge.net/lists/listinfo/squirrel-sql-users
>>
>
>
> --
> jørgen nørgaard
>
>  |\      _,,,---,,_
> /,`.-'`'    -.  ;-;;,_
> |,4-  ) )-,_. ,\ (  `'-'
> '---''(_/--'  `-'\_)
>
>
>
>
>
>
>
> ------------------------------------------------------------------------------
> Download Intel&#174; Parallel Studio Eval
> Try the new software tools for yourself. Speed compiling, find bugs
> proactively, and fine-tune applications for parallel performance.
> See why Intel Parallel Studio got high marks during beta.
> http://p.sf.net/sfu/intel-sw-dev
> _______________________________________________
> Squirrel-sql-users mailing list
> Squ...@li...
> https://lists.sourceforge.net/lists/listinfo/squirrel-sql-users
>

Re: [Squirrel-sql-develop] Need your help about some background of your community

[Robert M. <rob...@gm...>]

Andreas,

See below for my responses ...

On Wed, Apr 7, 2010 at 7:09 PM, Andreas Kaltenecker
<and...@st...> wrote:
>
> 1) Since when does the project exists?

The project was registered on 2001-05-31.   I have no idea how much
longer before that Colin was working on it.

>
> 2) Who were the founder of the project?

Colin Bell.  I think we do a pretty good job at advertising that :)

>
> 3) Are they still active and in charge?

No, unfortunately Colin moved on and only occasionally stops by to
check-in.  He had his reasons for doing so and the current development
team respects that.

>
> 4) Was there something like a transition in the core team? Was the
> responsibility taken over by somebody else?

There were a half-dozen main contributors to the project besides
Colin.  Most of them have moved on.  Now it's mostly one old-timer and
a few of us rookies.

>
> 5) Is somebody in charge of answering Bug reports or who does that job?
>

I think it's fair to say that one veteran and one rookie respond to
all of the bug reports (at least the ones we have some clue about
solving).

> 6) How many people have access to the code repository?

By SourceForge's count : 24 committers
(https://sourceforge.net/project/memberlist.php?group_id=28383)
Unfortunately commit access doesn't imply active contributions.  Many
of our committers have moved on.

>
> 7) To understand the question: Right now I am using information out of
> the Bug Tracking System. What I can see so far is that in January to
> June 2005 there was kind of a change in the structure because a
> developer who used to be very central/important disappeared. Is it
> possible that there was some change that year?
>

Hmmm... I guess the disappearing developer would be Colin.  Prior to
that time, I can think of no more active project member than him.  He
actually left in late 2004 though, not in June 2005.  Perhaps we can
reconcile this with the fact that I came aboard as a committer in
about June 2005 and started tackling tickets that were formerly
assigned to Colin (who was on sabbatical at that point).  So I
probably re-assigned to myself bug tickets that had been auto-assigned
to Colin.  SourceForge's tracker has this feature that allows you to
configure the bug tracker to auto-assign new bugs to a given team
member if the bug ticket author did not.  For more accuracy, you may
need to adjust your research algorithm to tickets that are being
worked.  Unfortunately, SourceForge gives little hint about bugs that
are being worked on.  Probably the easiest thing to look for would be
that if there was a comment made by the assignee, then it is being
worked by that person.  If not, then it could just be an auto-assigned
ticket.

I have to say, when I saw your email subject I was thinking, here we
go, another online survey to fill out.  It turns out that open-source
software development is a popular thesis topic these days; as it
should be, of course :)  But I can tell by your questions, you have
done some research already and that is refreshing.  Keep up the good
work and let us know how we can help you again.

Cheers,

Rob

[Squirrel-sql-develop] Need your help about some background of your community

[Andreas K. <and...@st...>]

Hy everybody,

at first i have to excuse myself if it's the wrong place for my  
concerns but it's for scientific use ;)

I am a student of Information and Business Science and my diploma  
thesis deals with open source software development and a possibility  
to evaluate the health status of a project by analyzing the community  
behind. Probably you know something about social network analysis.  
That's what I am using. I am collecting information about your  
communication and coordination behavior out of the bug reporting  
system and CVS/subversion. To validate my ideas I would need some  
background information about your project. Hopefully somebody takes  
the time to answer my questions! Thank you in advance!

1) Since when does the project exists?

2) Who were the founder of the project?

3) Are they still active and in charge?

4) Was there something like a transition in the core team? Was the  
responsibility taken over by somebody else?

5) Is somebody in charge of answering Bug reports or who does that job?

6) How many people have access to the code repository?

7) To understand the question: Right now I am using information out of  
the Bug Tracking System. What I can see so far is that in January to  
June 2005 there was kind of a change in the structure because a  
developer who used to be very central/important disappeared. Is it  
possible that there was some change that year?


Don't hesitate if you've any questions.

Thank you for your help!

andreas

[Squirrel-sql-develop] SVN Migration complete

[Robert M. <rob...@gm...>]

All,

Our CVS repository has been migrated to Subversion.  Aside from the
obvious benefits of Subversion over CVS, this will help us preserve
history when we restructure the source code for Maven builds.  There
is no schedule for the Maven migration at the moment, but we will give
notice in advance of the switch to give you time to commit your
changes.  Obviously the switch to Subversion means that you must
abandon any of your CVS checkouts and re-checkout using a subversion
client like the following:

svn co https://squirrel-sql.svn.sourceforge.net/svnroot/squirrel-sql/trunk/sql12
sql12

Though the CVS repo is still writable, we will no longer release from
the CVS trunk.  Please treat the CVS repo as "read-only" until we make
it so.

Cheers,

Rob

[Squirrel-sql-develop] CVS to SVN migration

[Robert M. <rob...@gm...>]

All,

The CVS to SVN migration is complete.  It was completed as of Mar. 25th, so
any commits after that date to the CVS code base will have to be re-done on
the SVN trunk (best to hold off commits to both or make changes to both for
the next week or so).

To checkout do:

svn co https://squirrel-sql.svn.sourceforge.net/svnroot/squirrel-sql/trunk/sql12
sql12

To browse the source code you can use this

http://squirrel-sql.svn.sourceforge.net/viewvc/squirrel-sql/

or trac's "Browse Source" link here:

https://sourceforge.net/apps/trac/squirrel-sql/browser

If anyone has a large amount of code changes on existing file to commit let
me know offline and I can work with you.  The CVS trunk will remain open
for now for verification and syncing purposes, but hold off commits to both
until we have had a chance to validate the SVN trunk.

Rob

[Squirrel-sql-develop] Please use the Java-VM parameter -Dsun.java2d.noddraw=true in your SQuirreL start batch to prevent performance problems

[Owens, R. A. <RA...@in...>]

Hi SquirreL development support,

I just upgraded to version 3.1 and received the following error:
Please use the Java-VM parameter -Dsun.java2d.noddraw=true in your SQuirreL start batch to prevent performance problems

Would it be possible to show the correct syntax to include this parameter in my start batch.  I'm including my start batch script below.

I'm on Windows XP
Thank you in advance,

@echo off

IF "%JAVA_HOME%"=="" SET LOCAL_JAVA=java
IF NOT "%JAVA_HOME%"=="" SET LOCAL_JAVA=%JAVA_HOME%\bin\java

set basedir=%~f0
:strip
set removed=%basedir:~-1%
set basedir=%basedir:~0,-1%
if NOT "%removed%"=="\" goto strip
set SQUIRREL_SQL_HOME=%basedir%

@rem Check to see if we are running in a 1.6/1.7 JVM and inform the user if not and skip launch. versioncheck.jar
@rem is a special jar file which has been compiled with javac version 1.2.2, which should be able to be run by
@rem that version of higher.  The arguments to JavaVersionChecker below specify the minimum acceptable version
@rem (first arg) and any other acceptable subsequent versions.  <MAJOR>.<MINOR> should be all that is
@rem necessary for the version form.
"%LOCAL_JAVA%w" -cp "%SQUIRREL_SQL_HOME%\lib\versioncheck.jar" JavaVersionChecker 1.6 1.7
if ErrorLevel 1 goto ExitForWrongJavaVersion

@rem If the changelist.xml file isn't present or the downloaded update jars don't exist, skip launching the updater - these files are created by the
@rem software update feature inside of SQuirreL. So their absence, simply means the software update feature hasn't been accessed.
if not exist "%SQUIRREL_SQL_HOME%\update\changeList.xml" goto launchsquirrel
SET TMP_CP="%SQUIRREL_SQL_HOME%\update\downloads\core\squirrel-sql.jar"
if not exist %TMP_CP% goto launchsquirrel
dir /b "%SQUIRREL_SQL_HOME%\update\downloads\core\*.*" > %TEMP%\update-lib.tmp
FOR /F %%I IN (%TEMP%\update-lib.tmp) DO CALL "%SQUIRREL_SQL_HOME%\addpath.bat" "%SQUIRREL_SQL_HOME%\update\downloads\core\%%I"
SET UPDATE_CP=%TMP_CP%
SET UPDATE_PARMS=--log-config-file "%SQUIRREL_SQL_HOME%\update-log4j.properties" --squirrel-home "%SQUIRREL_SQL_HOME%" %1 %2 %3 %4 %5 %6 %7 %8 %9
"%LOCAL_JAVA%w" -cp %UPDATE_CP% -Dlog4j.defaultInitOverride=true -Dprompt=true net.sourceforge.squirrel_sql.client.update.gui.installer.PreLaunchUpdateApplication %UPDATE_PARAMS%

:launchsquirrel
@rem build SQuirreL's classpath
set TMP_CP="%SQUIRREL_SQL_HOME%\squirrel-sql.jar"
dir /b "%SQUIRREL_SQL_HOME%\lib\*.*" > %TEMP%\squirrel-lib.tmp
FOR /F %%I IN (%TEMP%\squirrel-lib.tmp) DO CALL "%SQUIRREL_SQL_HOME%\addpath.bat" "%SQUIRREL_SQL_HOME%\lib\%%I"
SET SQUIRREL_CP=%TMP_CP%;"%CLASSPATH%"
echo "SQUIRREL_CP=%SQUIRREL_CP%"

SET TMP_PARMS=--log-config-file "%SQUIRREL_SQL_HOME%\log4j.properties" --squirrel-home "%SQUIRREL_SQL_HOME%" %1 %2 %3 %4 %5 %6 %7 %8 %9

@rem Run with a command window.
@rem "%LOCAL_JAVA%" -cp %TMP_CP% net.sourceforge.squirrel_sql.client.Main %TMP_PARMS%

@rem To add translation working directories to your classpath edit and uncomment this line:
@rem start "SQuirreL SQL Client" /B "%LOCAL_JAVA%w" -Xmx256m -cp %TMP_CP%;<your working dir here> net.sourceforge.squirrel_sql.client.Main %TMP_PARMS%

@rem To change the language edit and uncomment this line:
@rem start "SQuirreL SQL Client" /B "%LOCAL_JAVA%w" -Xmx256m -cp %TMP_CP%;<your working dir here> -Duser.language=<your language here> net.sourceforge.squirrel_sql.client.Main %TMP_PARMS%

@rem Run with no command window. This may not work with older versions of Windows. Use the command above then.
start "SQuirreL SQL Client" /B "%LOCAL_JAVA%w" -Xmx256m -cp %SQUIRREL_CP% net.sourceforge.squirrel_sql.client.Main %TMP_PARMS%

@rem Run the executable jar file with or without a cmd window. However the
@rem classes from the %CLASSPATH% environment variable will not be available.
@rem "%LOCAL_JAVA%" -jar "%SQUIRREL_SQL_HOME%\squirrel-sql.jar" %TMP_PARMS%
@rem start "SQuirreL SQL Client" /B "%LOCAL_JAVA%w" -jar "%SQUIRREL_SQL_HOME%\squirrel-sql.jar" %TMP_PARMS%

:ExitForWrongJavaVersion
Exit

Richard Owens
Sr. Revenue Assurance Analyst
InterCall a subsidiary of West Corporation
-----------------------------------------------
(334) 768-7626 direct
(334) 768-7291 fax
ra...@in... email
www.intercall.com<../../../../cpetty/Application%20Data/Microsoft/Local%20Settings/Temp/www.intercall.com> company web site
-----------------------------------------------
Frost & Sullivan Recognizes InterCall with the 2008 North American Conferencing
Services Customer Value Enhancement Award. Learn more at www.intercall.com/fs2008<http://www.intercall.com/fs2008>

Re: [Squirrel-sql-develop] Classpath in Oracle Plug-in

[Robert M. <rob...@gm...>]

It's ok.  I tried Gerd's re-mix of your patch and it seemed to work
fine.  The magical Method.setAccessible(true) seems to make all the
difference :)

Cheers,

Rob

On Sun, Feb 7, 2010 at 7:15 PM, Neville Rowe <row...@ho...> wrote:
>
> Hi Robert,
>
>
>
> Okay, further investigation shows because I'm still using an Oracle 9 client (and an Oracle 9 db for a variety of silly reasons) this is a change between the Oracle 9 and Oracle 10 jdbc drivers. So my patch doesn't work for Oracle 10. And what is worse is because of a 'feature' of Java - where you have a private class implementing a method as public from public interface throws an IllegalAccessException -
>
> http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=4071593
>
> So the solution to this is to walk up the class hierarchy until you find the public superclass / interface.
>
>
>
>        private void setTimezoneForSession(ISession session, OraclePreferenceBean prefs)
>        {
>                if (!prefs.getInitSessionTimezone())
>                {
>                        if (s_log.isInfoEnabled())
>                        {
>                                s_log.info("setTimezoneForSession: user preference for init session timezone is disabled.  "
>
>                                      + "Local Timezone data types may not be displayed correctly.");
>                        }
>                        return;
>                }
>
>                Connection con = session.getSQLConnection().getConnection();
>                String timezoneStr = prefs.getSessionTimezone();
>                if (s_log.isInfoEnabled())
>                {
>                        s_log.info("setTimezoneForSession: attempting to set the session timezone to : " + timezoneStr);
>
>                }
>                boolean done = false;
>                Class here = con.getClass();
>                while (here.getName().startsWith("oracle.") && !done) {
>                        try {
>                                Method setSessionTimeZoneMethod = here.getMethod("setSessionTimeZone", String.class);
>                                if (setSessionTimeZoneMethod != null) {
>                                        setSessionTimeZoneMethod.invoke(con, timezoneStr);
>                                        done = true;
>                                }
>                        } catch (Exception e)
>                        {
>                        }
>                        if (! done) {
>                                here = here.getSuperclass();
>                        }
>                }
>                if (! done) {
>                        s_log.error("setTimezoneForSession: Could not set session timezone.");
>                }
>        }
>
>
>
> Sorry about that.
>
>
>
> Neville
>
>> Date: Sat, 6 Feb 2010 16:24:35 -0500
>> From: rob...@gm...
>> To: row...@ho...
>> CC: squ...@li...
>> Subject: Re: [Squirrel-sql-develop] Classpath in Oracle Plug-in
>>
>> Hi Neville,
>>
>> I don't see how
>> con.getClass().getName().equals("oracle.jdbc.OracleConnection") will
>> ever be true, since oracle.jdbc.OracleConnection is an interface, and
>> not an implementation. For example, my driver connection classname is
>> oracle.jdbc.driver.T4CConnection. The setSessionTimeZone method
>> appears to be implemented in
>> oracle.jdbc.driver.PhysicalConnection.setSessionTimeZone, which I
>> would guess is a superclass of T4CConnection (at least this is what is
>> returned from getMethod according to the debugger). The code that you
>> suggested gives me this exception:
>>
>> java.lang.IllegalAccessException: Class
>> net.sourceforge.squirrel_sql.plugins.oracle.OraclePlugin can not
>> access a member of class oracle.jdbc.driver.PhysicalConnection with
>> modifiers "public"
>>
>> Rob
>>
>>
>> On Sat, Feb 6, 2010 at 1:48 PM, Neville Rowe <row...@ho...> wrote:
>> >
>> > Hi there,
>> >
>> > In using the Oracle plug-in, I'm getting a regular 'class not found' for oracle.jdbc.OracleConnection. The issue is there is a check in the OraclePlugin.java to load that class to use reflection to do a set Time zone. So the current code from line 489 onwards looks like
>> >
>> > Class oraConClass = Class.forName("oracle.jdbc.OracleConnection");
>> > Method setSessionTimeZoneMethod = oraConClass.getMethod("setSessionTimeZone", String.class);
>> > if (setSessionTimeZoneMethod != null)
>> > {
>> >        setSessionTimeZoneMethod.invoke(con, timezoneStr);
>> >  } else
>> >  {
>> >        s_log.error("setTimezoneForSession: setSessionTimeZoneMethod returned by reflection was null.  "
>> >        + "Skipped setting session timezone");
>> > }
>> >
>> > The issue is if the jdbc driver is loaded through the extra classpath, then Class.forName doesn't find it.
>> >
>> > Wouldn't this be better....
>> >
>> >                        if (con.getClass().getName().equals("oracle.jdbc.OracleConnection")) {
>> >                                Method setSessionTimeZoneMethod = con.getClass().getMethod("setSessionTimeZone", String.class);                                 if (setSessionTimeZoneMethod != null)
>> >                                {
>> >                                        setSessionTimeZoneMethod.invoke(con, timezoneStr);
>> >                                } else
>> >                                {
>> >                                        s_log.error("setTimezoneForSession: setSessionTimeZoneMethod returned by reflection was null.  "
>> >                                                  + "Skipped setting session timezone");
>> >                                }
>> >                        }
>> >
>> > Suggestions ?
>> >
>> > Neville
>> >
>> > _________________________________________________________________
>> > Do you have a story that started on Hotmail? Tell us now
>> > http://clk.atdmt.com/UKM/go/195013117/direct/01/
>> > ------------------------------------------------------------------------------
>> > The Planet: dedicated and managed hosting, cloud storage, colocation
>> > Stay online with enterprise data centers and the best network in the business
>> > Choose flexible plans and management services without long-term contracts
>> > Personal 24x7 support from experience hosting pros just a phone call away.
>> > http://p.sf.net/sfu/theplanet-com
>> > _______________________________________________
>> > Squirrel-sql-develop mailing list
>> > Squ...@li...
>> > https://lists.sourceforge.net/lists/listinfo/squirrel-sql-develop
>> >
>>
>> ------------------------------------------------------------------------------
>> The Planet: dedicated and managed hosting, cloud storage, colocation
>> Stay online with enterprise data centers and the best network in the business
>> Choose flexible plans and management services without long-term contracts
>> Personal 24x7 support from experience hosting pros just a phone call away.
>> http://p.sf.net/sfu/theplanet-com
>> _______________________________________________
>> Squirrel-sql-develop mailing list
>> Squ...@li...
>> https://lists.sourceforge.net/lists/listinfo/squirrel-sql-develop
>
> _________________________________________________________________
> Do you have a story that started on Hotmail? Tell us now
> http://clk.atdmt.com/UKM/go/195013117/direct/01/
> ------------------------------------------------------------------------------
> The Planet: dedicated and managed hosting, cloud storage, colocation
> Stay online with enterprise data centers and the best network in the business
> Choose flexible plans and management services without long-term contracts
> Personal 24x7 support from experience hosting pros just a phone call away.
> http://p.sf.net/sfu/theplanet-com
> _______________________________________________
> Squirrel-sql-develop mailing list
> Squ...@li...
> https://lists.sourceforge.net/lists/listinfo/squirrel-sql-develop
>

Re: [Squirrel-sql-develop] Classpath in Oracle Plug-in

[Neville R. <row...@ho...>]

Hi Robert,

 

Okay, further investigation shows because I'm still using an Oracle 9 client (and an Oracle 9 db for a variety of silly reasons) this is a change between the Oracle 9 and Oracle 10 jdbc drivers. So my patch doesn't work for Oracle 10. And what is worse is because of a 'feature' of Java - where you have a private class implementing a method as public from public interface throws an IllegalAccessException - 

http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=4071593 

So the solution to this is to walk up the class hierarchy until you find the public superclass / interface.

 

        private void setTimezoneForSession(ISession session, OraclePreferenceBean prefs) 
        { 
                if (!prefs.getInitSessionTimezone()) 
                { 
                        if (s_log.isInfoEnabled()) 
                        { 
                                s_log.info("setTimezoneForSession: user preference for init session timezone is disabled.  "

                                      + "Local Timezone data types may not be displayed correctly."); 
                        } 
                        return; 
                } 

                Connection con = session.getSQLConnection().getConnection(); 
                String timezoneStr = prefs.getSessionTimezone(); 
                if (s_log.isInfoEnabled()) 
                { 
                        s_log.info("setTimezoneForSession: attempting to set the session timezone to : " + timezoneStr);

                } 
                boolean done = false; 
                Class here = con.getClass(); 
                while (here.getName().startsWith("oracle.") && !done) { 
                        try { 
                                Method setSessionTimeZoneMethod = here.getMethod("setSessionTimeZone", String.class); 
                                if (setSessionTimeZoneMethod != null) { 
                                        setSessionTimeZoneMethod.invoke(con, timezoneStr); 
                                        done = true; 
                                } 
                        } catch (Exception e) 
                        { 
                        } 
                        if (! done) { 
                                here = here.getSuperclass(); 
                        } 
                } 
                if (! done) { 
                        s_log.error("setTimezoneForSession: Could not set session timezone."); 
                } 
        } 

 

Sorry about that.

 

Neville
 
> Date: Sat, 6 Feb 2010 16:24:35 -0500
> From: rob...@gm...
> To: row...@ho...
> CC: squ...@li...
> Subject: Re: [Squirrel-sql-develop] Classpath in Oracle Plug-in
> 
> Hi Neville,
> 
> I don't see how
> con.getClass().getName().equals("oracle.jdbc.OracleConnection") will
> ever be true, since oracle.jdbc.OracleConnection is an interface, and
> not an implementation. For example, my driver connection classname is
> oracle.jdbc.driver.T4CConnection. The setSessionTimeZone method
> appears to be implemented in
> oracle.jdbc.driver.PhysicalConnection.setSessionTimeZone, which I
> would guess is a superclass of T4CConnection (at least this is what is
> returned from getMethod according to the debugger). The code that you
> suggested gives me this exception:
> 
> java.lang.IllegalAccessException: Class
> net.sourceforge.squirrel_sql.plugins.oracle.OraclePlugin can not
> access a member of class oracle.jdbc.driver.PhysicalConnection with
> modifiers "public"
> 
> Rob
> 
> 
> On Sat, Feb 6, 2010 at 1:48 PM, Neville Rowe <row...@ho...> wrote:
> >
> > Hi there,
> >
> > In using the Oracle plug-in, I'm getting a regular 'class not found' for oracle.jdbc.OracleConnection. The issue is there is a check in the OraclePlugin.java to load that class to use reflection to do a set Time zone. So the current code from line 489 onwards looks like
> >
> > Class oraConClass = Class.forName("oracle.jdbc.OracleConnection");
> > Method setSessionTimeZoneMethod = oraConClass.getMethod("setSessionTimeZone", String.class);
> > if (setSessionTimeZoneMethod != null)
> > {
> >        setSessionTimeZoneMethod.invoke(con, timezoneStr);
> >  } else
> >  {
> >        s_log.error("setTimezoneForSession: setSessionTimeZoneMethod returned by reflection was null.  "
> >        + "Skipped setting session timezone");
> > }
> >
> > The issue is if the jdbc driver is loaded through the extra classpath, then Class.forName doesn't find it.
> >
> > Wouldn't this be better....
> >
> >                        if (con.getClass().getName().equals("oracle.jdbc.OracleConnection")) {
> >                                Method setSessionTimeZoneMethod = con.getClass().getMethod("setSessionTimeZone", String.class);                                 if (setSessionTimeZoneMethod != null)
> >                                {
> >                                        setSessionTimeZoneMethod.invoke(con, timezoneStr);
> >                                } else
> >                                {
> >                                        s_log.error("setTimezoneForSession: setSessionTimeZoneMethod returned by reflection was null.  "
> >                                                  + "Skipped setting session timezone");
> >                                }
> >                        }
> >
> > Suggestions ?
> >
> > Neville
> >
> > _________________________________________________________________
> > Do you have a story that started on Hotmail? Tell us now
> > http://clk.atdmt.com/UKM/go/195013117/direct/01/
> > ------------------------------------------------------------------------------
> > The Planet: dedicated and managed hosting, cloud storage, colocation
> > Stay online with enterprise data centers and the best network in the business
> > Choose flexible plans and management services without long-term contracts
> > Personal 24x7 support from experience hosting pros just a phone call away.
> > http://p.sf.net/sfu/theplanet-com
> > _______________________________________________
> > Squirrel-sql-develop mailing list
> > Squ...@li...
> > https://lists.sourceforge.net/lists/listinfo/squirrel-sql-develop
> >
> 
> ------------------------------------------------------------------------------
> The Planet: dedicated and managed hosting, cloud storage, colocation
> Stay online with enterprise data centers and the best network in the business
> Choose flexible plans and management services without long-term contracts
> Personal 24x7 support from experience hosting pros just a phone call away.
> http://p.sf.net/sfu/theplanet-com
> _______________________________________________
> Squirrel-sql-develop mailing list
> Squ...@li...
> https://lists.sourceforge.net/lists/listinfo/squirrel-sql-develop
 		 	   		  
_________________________________________________________________
Do you have a story that started on Hotmail? Tell us now
http://clk.atdmt.com/UKM/go/195013117/direct/01/

Re: [Squirrel-sql-develop] Classpath in Oracle Plug-in

[Gerd W. <ger...@t-...>]

Neville,

thanks for the patch. I just committed it to CVS. I changed your code just a 
little tom make it work for oracle.jdbc.driver.T4CConnection too.

Gerd

Neville Rowe wrote:
> Hi there,
>  
> In using the Oracle plug-in, I'm getting a regular 'class not found' for oracle.jdbc.OracleConnection. The issue is there is a check in the OraclePlugin.java to load that class to use reflection to do a set Time zone. So the current code from line 489 onwards looks like
>  
> Class oraConClass = Class.forName("oracle.jdbc.OracleConnection");
> Method setSessionTimeZoneMethod = oraConClass.getMethod("setSessionTimeZone", String.class);
> if (setSessionTimeZoneMethod != null)
> {
>         setSessionTimeZoneMethod.invoke(con, timezoneStr);
>  } else
>  {
>         s_log.error("setTimezoneForSession: setSessionTimeZoneMethod returned by reflection was null.  "
>         + "Skipped setting session timezone");
> }
> 
> The issue is if the jdbc driver is loaded through the extra classpath, then Class.forName doesn't find it.
>  
> Wouldn't this be better....
>  
>                         if (con.getClass().getName().equals("oracle.jdbc.OracleConnection")) { 
>                                 Method setSessionTimeZoneMethod = con.getClass().getMethod("setSessionTimeZone", String.class);                                 if (setSessionTimeZoneMethod != null) 
>                                 { 
>                                         setSessionTimeZoneMethod.invoke(con, timezoneStr); 
>                                 } else 
>                                 { 
>                                         s_log.error("setTimezoneForSession: setSessionTimeZoneMethod returned by reflection was null.  "
>                                                   + "Skipped setting session timezone"); 
>                                 } 
>                         }
>  
> Suggestions ?
>  
> Neville
>  		 	   		  
> _________________________________________________________________
> Do you have a story that started on Hotmail? Tell us now
> http://clk.atdmt.com/UKM/go/195013117/direct/01/
> ------------------------------------------------------------------------------
> The Planet: dedicated and managed hosting, cloud storage, colocation
> Stay online with enterprise data centers and the best network in the business
> Choose flexible plans and management services without long-term contracts
> Personal 24x7 support from experience hosting pros just a phone call away.
> http://p.sf.net/sfu/theplanet-com
> _______________________________________________
> Squirrel-sql-develop mailing list
> Squ...@li...
> https://lists.sourceforge.net/lists/listinfo/squirrel-sql-develop
> 

Re: [Squirrel-sql-develop] Classpath in Oracle Plug-in

[Robert M. <rob...@gm...>]

Hi Neville,

I don't see how
con.getClass().getName().equals("oracle.jdbc.OracleConnection") will
ever be true, since oracle.jdbc.OracleConnection is an interface, and
not an implementation.  For example, my driver connection classname is
oracle.jdbc.driver.T4CConnection.  The setSessionTimeZone method
appears to be implemented in
oracle.jdbc.driver.PhysicalConnection.setSessionTimeZone, which I
would guess is a superclass of T4CConnection (at least this is what is
returned from getMethod according to the debugger).  The code that you
suggested gives me this exception:

java.lang.IllegalAccessException: Class
net.sourceforge.squirrel_sql.plugins.oracle.OraclePlugin can not
access a member of class oracle.jdbc.driver.PhysicalConnection with
modifiers "public"

Rob


On Sat, Feb 6, 2010 at 1:48 PM, Neville Rowe <row...@ho...> wrote:
>
> Hi there,
>
> In using the Oracle plug-in, I'm getting a regular 'class not found' for oracle.jdbc.OracleConnection. The issue is there is a check in the OraclePlugin.java to load that class to use reflection to do a set Time zone. So the current code from line 489 onwards looks like
>
> Class oraConClass = Class.forName("oracle.jdbc.OracleConnection");
> Method setSessionTimeZoneMethod = oraConClass.getMethod("setSessionTimeZone", String.class);
> if (setSessionTimeZoneMethod != null)
> {
>        setSessionTimeZoneMethod.invoke(con, timezoneStr);
>  } else
>  {
>        s_log.error("setTimezoneForSession: setSessionTimeZoneMethod returned by reflection was null.  "
>        + "Skipped setting session timezone");
> }
>
> The issue is if the jdbc driver is loaded through the extra classpath, then Class.forName doesn't find it.
>
> Wouldn't this be better....
>
>                        if (con.getClass().getName().equals("oracle.jdbc.OracleConnection")) {
>                                Method setSessionTimeZoneMethod = con.getClass().getMethod("setSessionTimeZone", String.class);                                 if (setSessionTimeZoneMethod != null)
>                                {
>                                        setSessionTimeZoneMethod.invoke(con, timezoneStr);
>                                } else
>                                {
>                                        s_log.error("setTimezoneForSession: setSessionTimeZoneMethod returned by reflection was null.  "
>                                                  + "Skipped setting session timezone");
>                                }
>                        }
>
> Suggestions ?
>
> Neville
>
> _________________________________________________________________
> Do you have a story that started on Hotmail? Tell us now
> http://clk.atdmt.com/UKM/go/195013117/direct/01/
> ------------------------------------------------------------------------------
> The Planet: dedicated and managed hosting, cloud storage, colocation
> Stay online with enterprise data centers and the best network in the business
> Choose flexible plans and management services without long-term contracts
> Personal 24x7 support from experience hosting pros just a phone call away.
> http://p.sf.net/sfu/theplanet-com
> _______________________________________________
> Squirrel-sql-develop mailing list
> Squ...@li...
> https://lists.sourceforge.net/lists/listinfo/squirrel-sql-develop
>

[Squirrel-sql-develop] Classpath in Oracle Plug-in

[Neville R. <row...@ho...>]

Hi there,
 
In using the Oracle plug-in, I'm getting a regular 'class not found' for oracle.jdbc.OracleConnection. The issue is there is a check in the OraclePlugin.java to load that class to use reflection to do a set Time zone. So the current code from line 489 onwards looks like
 
Class oraConClass = Class.forName("oracle.jdbc.OracleConnection");
Method setSessionTimeZoneMethod = oraConClass.getMethod("setSessionTimeZone", String.class);
if (setSessionTimeZoneMethod != null)
{
        setSessionTimeZoneMethod.invoke(con, timezoneStr);
 } else
 {
        s_log.error("setTimezoneForSession: setSessionTimeZoneMethod returned by reflection was null.  "
        + "Skipped setting session timezone");
}

The issue is if the jdbc driver is loaded through the extra classpath, then Class.forName doesn't find it.
 
Wouldn't this be better....
 
                        if (con.getClass().getName().equals("oracle.jdbc.OracleConnection")) { 
                                Method setSessionTimeZoneMethod = con.getClass().getMethod("setSessionTimeZone", String.class);                                 if (setSessionTimeZoneMethod != null) 
                                { 
                                        setSessionTimeZoneMethod.invoke(con, timezoneStr); 
                                } else 
                                { 
                                        s_log.error("setTimezoneForSession: setSessionTimeZoneMethod returned by reflection was null.  "
                                                  + "Skipped setting session timezone"); 
                                } 
                        }
 
Suggestions ?
 
Neville
 		 	   		  
_________________________________________________________________
Do you have a story that started on Hotmail? Tell us now
http://clk.atdmt.com/UKM/go/195013117/direct/01/

Re: [Squirrel-sql-develop] Fwd: DB2 for mainframe and Labels

[Robert M. <rob...@gm...>]

Excellent.  By the way, I have looked around to see what is involved
with acquiring a system on which to run DB2 for mainframe (z/OS,
OS/390, OS/400, etc).  It seems that there was a company/product
(FLEX-ES) that was authorized to sell emulation software that would
allow you to host a mainframe OS (and possibly DB2) on cheap hardware
(intel PC).  I read that IBM chose not to continue their license, and
now the only way to legally setup a test system is to buy expensive
hardware and monthly license to the OS.  I've seen Hercules w/ public
domain mainframe OS, but I guess that I probably would not be able to
run DB2 on it.  Am I missing something or is this the real deal with
IBM ?

Rob

On 2/1/10, Christian Gruenberg <chr...@di...> wrote:
> Hi Rob,
>
> I tried the snapshot. The update statements do their jobs. Thanks!
>
> Unfortunately our DB2-System is at version 9.1 and JDBC 4.0 is
> supported from DB2 version  9.5.
> @all: Can someone else try this workaround?
>
> Christian
>
>
> 2010/1/31 Robert Manning <rob...@gm...>:
> > Christian,
> >
> > I just applied the patch (minus the redundant DB2 custom types).  In
> > the end I decided to replace all occurrences with getColumnName since
> > I do not have a system against which to test.  Try the latest snapshot
> > :
> >
> > https://sourceforge.net/projects/squirrel-sql/files/3-snapshots/Snapshot-20100131_2249/squirrel-sql-Snapshot-20100131_2249-install.jar/download
> >
> > Let us know whether or not it addresses the issue.
> >
> > Rob
> >
> > On Thu, Jan 28, 2010 at 8:26 PM, Robert Manning
> > <rob...@gm...> wrote:
> >> Thanks for confirming that Christian - I suspected that but I wanted
> >> to have another look to be sure.  We will take care of this issue
> >> before the 3.1 release in February and when we do there will be a
> >> snapshot that you (and others who might be interested in the impact of
> >> this change) to try out.
> >>
> >> Rob
> >>
> >> On Thu, Jan 28, 2010 at 2:02 PM, Christian Gruenberg
> >> <chr...@di...> wrote:
> >>>>> I also think that if we do this in core, then probably
> >>>>> most if not all of Christian's custom DB2 datatypes are unnecessary.
> >>>>> I skimmed over them so it is possible that I missed something but they
> >>>>> seemed to be merely overriding the core implementation of
> >>>>> getWhereClauseValue() and the like in *DataTypeComponent.  I'll take
> >>>>> another look.
> >>>
> >>> You are right, they are unnecessary now if you'll corret the basic datatypes.
> >>>
> >>> Christain
> >>>
> >>
> >
>

Re: [Squirrel-sql-develop] Fwd: DB2 for mainframe and Labels

[Christian G. <chr...@di...>]

Hi Rob,

I tried the snapshot. The update statements do their jobs. Thanks!

Unfortunately our DB2-System is at version 9.1 and JDBC 4.0 is
supported from DB2 version  9.5.
@all: Can someone else try this workaround?

Christian


2010/1/31 Robert Manning <rob...@gm...>:
> Christian,
>
> I just applied the patch (minus the redundant DB2 custom types).  In
> the end I decided to replace all occurrences with getColumnName since
> I do not have a system against which to test.  Try the latest snapshot
> :
>
> https://sourceforge.net/projects/squirrel-sql/files/3-snapshots/Snapshot-20100131_2249/squirrel-sql-Snapshot-20100131_2249-install.jar/download
>
> Let us know whether or not it addresses the issue.
>
> Rob
>
> On Thu, Jan 28, 2010 at 8:26 PM, Robert Manning
> <rob...@gm...> wrote:
>> Thanks for confirming that Christian - I suspected that but I wanted
>> to have another look to be sure.  We will take care of this issue
>> before the 3.1 release in February and when we do there will be a
>> snapshot that you (and others who might be interested in the impact of
>> this change) to try out.
>>
>> Rob
>>
>> On Thu, Jan 28, 2010 at 2:02 PM, Christian Gruenberg
>> <chr...@di...> wrote:
>>>>> I also think that if we do this in core, then probably
>>>>> most if not all of Christian's custom DB2 datatypes are unnecessary.
>>>>> I skimmed over them so it is possible that I missed something but they
>>>>> seemed to be merely overriding the core implementation of
>>>>> getWhereClauseValue() and the like in *DataTypeComponent.  I'll take
>>>>> another look.
>>>
>>> You are right, they are unnecessary now if you'll corret the basic datatypes.
>>>
>>> Christain
>>>
>>
>

Re: [Squirrel-sql-develop] Fwd: DB2 for mainframe and Labels

[Robert M. <rob...@gm...>]

Christian,

I just applied the patch (minus the redundant DB2 custom types).  In
the end I decided to replace all occurrences with getColumnName since
I do not have a system against which to test.  Try the latest snapshot
:

https://sourceforge.net/projects/squirrel-sql/files/3-snapshots/Snapshot-20100131_2249/squirrel-sql-Snapshot-20100131_2249-install.jar/download

Let us know whether or not it addresses the issue.

Rob

On Thu, Jan 28, 2010 at 8:26 PM, Robert Manning
<rob...@gm...> wrote:
> Thanks for confirming that Christian - I suspected that but I wanted
> to have another look to be sure.  We will take care of this issue
> before the 3.1 release in February and when we do there will be a
> snapshot that you (and others who might be interested in the impact of
> this change) to try out.
>
> Rob
>
> On Thu, Jan 28, 2010 at 2:02 PM, Christian Gruenberg
> <chr...@di...> wrote:
>>>> I also think that if we do this in core, then probably
>>>> most if not all of Christian's custom DB2 datatypes are unnecessary.
>>>> I skimmed over them so it is possible that I missed something but they
>>>> seemed to be merely overriding the core implementation of
>>>> getWhereClauseValue() and the like in *DataTypeComponent.  I'll take
>>>> another look.
>>
>> You are right, they are unnecessary now if you'll corret the basic datatypes.
>>
>> Christain
>>
>

Re: [Squirrel-sql-develop] Fwd: DB2 for mainframe and Labels

[Robert M. <rob...@gm...>]

Christian,

After studying this a bit, it seems that DB2 is a bit complex with
regard to rsmd.getColumnName and rsmd.getColumnLabel.  First,
depending on the platform of DB2, a column label can mean different
things.  For example, in all platforms of DB2 (z/OS, LUW, etc.), in
some cases a select statement that uses a column alias, will yield the
alias name from getColumnLabel.  However, in mainframe DB2 (z/OS,
etc.), there is a "LABEL on " statement that is used to give columns a
"label" that in some configurations is returned from
rsmd.getColumnLabel.  There is a new data source property available
for DB2 JDBC 4.0 drivers that changes the semantics of these two API
methods slightly.

http://publib.boulder.ibm.com/infocenter/db2luw/v9r5/index.jsp?topic=/com.ibm.db2.luw.apdv.java.doc/doc/r0052607.html
(See useJDBC4ColumnNameAndLabelSemantics property)

If the driver conforms to JDBC 4.0, and DB2 server is mainframe, then
there are basically two configurations:
1. useJDBC4ColumnNameAndLabelSemantics = DB2BaseDataSource.YES :
"LABEL on" column labels are ignored; getColumnLabel = column alias in
select statement if provided.  If not provided, then it is the column
name.
2. useJDBC4ColumnNameAndLabelSemantics = DB2BaseDataSource.NO : "LABEL
on" column labels are returned from getColumnLabel.  Column aliases
are returned from getColumnName.

For older drivers the default behavior is item 2 above.  Item #1 would
potentially solve the issue that Christian is having without a code
change to SQuirreL.  I would be curious to see if that is correct, if
Christian would like to try.  According to the configuration
instructions in the document I cited, you would need to use a JDBC 4.0
driver and append the following to the JDBC URL:

;useJDBC4ColumnNameAndLabelSemantics=(com.ibm.db2.jcc.DB2BaseDataSource.YES)

I'm still planning to convert our usage of getColumnLabel to
getColumnName, but I would be curious to see if the above has any
effect on the problem with DB2 on the mainframe platform.

Rob

On Thu, Jan 28, 2010 at 8:26 PM, Robert Manning
<rob...@gm...> wrote:
> Thanks for confirming that Christian - I suspected that but I wanted
> to have another look to be sure.  We will take care of this issue
> before the 3.1 release in February and when we do there will be a
> snapshot that you (and others who might be interested in the impact of
> this change) to try out.
>
> Rob
>
> On Thu, Jan 28, 2010 at 2:02 PM, Christian Gruenberg
> <chr...@di...> wrote:
>>>> I also think that if we do this in core, then probably
>>>> most if not all of Christian's custom DB2 datatypes are unnecessary.
>>>> I skimmed over them so it is possible that I missed something but they
>>>> seemed to be merely overriding the core implementation of
>>>> getWhereClauseValue() and the like in *DataTypeComponent.  I'll take
>>>> another look.
>>
>> You are right, they are unnecessary now if you'll corret the basic datatypes.
>>
>> Christain
>>
>

Re: [Squirrel-sql-develop] Fwd: DB2 for mainframe and Labels

[Robert M. <rob...@gm...>]

Thanks for confirming that Christian - I suspected that but I wanted
to have another look to be sure.  We will take care of this issue
before the 3.1 release in February and when we do there will be a
snapshot that you (and others who might be interested in the impact of
this change) to try out.

Rob

On Thu, Jan 28, 2010 at 2:02 PM, Christian Gruenberg
<chr...@di...> wrote:
>>> I also think that if we do this in core, then probably
>>> most if not all of Christian's custom DB2 datatypes are unnecessary.
>>> I skimmed over them so it is possible that I missed something but they
>>> seemed to be merely overriding the core implementation of
>>> getWhereClauseValue() and the like in *DataTypeComponent.  I'll take
>>> another look.
>
> You are right, they are unnecessary now if you'll corret the basic datatypes.
>
> Christain
>

Re: [Squirrel-sql-develop] Fwd: DB2 for mainframe and Labels

[Christian G. <chr...@di...>]

>> I also think that if we do this in core, then probably
>> most if not all of Christian's custom DB2 datatypes are unnecessary.
>> I skimmed over them so it is possible that I missed something but they
>> seemed to be merely overriding the core implementation of
>> getWhereClauseValue() and the like in *DataTypeComponent.  I'll take
>> another look.

You are right, they are unnecessary now if you'll corret the basic datatypes.

Christain

Re: [Squirrel-sql-develop] Fwd: DB2 for mainframe and Labels

[Robert M. <rob...@gm...>]

Will do.

Rob

On Wed, Jan 27, 2010 at 3:46 AM, Gerd Wagner <ger...@t-...> wrote:
> Robert Manning wrote:
>>
>> Gerd,
>>
>> I definitely agree that getColumnName() seems to be the better choice
>> according to the documentation.  I also agree that the most desireable
>> code change is to replace getLabel with getColumnName wherever we are
>> dynamically building SQL to send to the database.  However, I was
>> reluctant to replace all occurrences since for display purposes we
>> should probably stick with getLabel (Users with databases that support
>> labels probably define them for their own purpose and we should
>> respect that).
>
> I looked at all the places where I changed label to column. As far as I can
> see your point concerns only to one of my changes, that is the one in
> MyTableModel.getColumnName(). I would have done that change too but I'm not
> so sure about it. I leave it to you to decide. See my changes in attachment.
>
>
>> I also think that if we do this in core, then probably
>> most if not all of Christian's custom DB2 datatypes are unnecessary.
>> I skimmed over them so it is possible that I missed something but they
>> seemed to be merely overriding the core implementation of
>> getWhereClauseValue() and the like in *DataTypeComponent.  I'll take
>> another look.
>
> Sorry, I didn't realize that. My attached patch includes Christians changes
> as well.
>
> Rob, would it be too much to ask you to do the final review of the code and
> then commit it? If you don't have the time or any further questions please
> tell me.
>
> Thanks
>
> Gerd
>
>
>
>>
>> Rob
>>
>>
>> On Mon, Jan 25, 2010 at 6:52 PM, Gerd Wagner <ger...@t-...>
>> wrote:
>>>
>>> Robert Manning wrote:
>>>>
>>>> Gerd,
>>>>
>>>> I think we should address this patch in our 3.1 release.  It seems
>>>> that a number of our DB2 on mainframe users run into this problem
>>>> according to this bug:
>>>>
>>>>
>>>>
>>>> https://sourceforge.net/tracker/index.php?func=detail&aid=2818117&group_id=28383&atid=393414
>>>>
>>>> It seems that it's for no good reason that we use
>>>> ColumnDisplayDefinition.getLabel() rather than getColumnName().  It
>>>> looks like Glenn chose to do that since the beginning and it was
>>>> simply used everywhere consistently.  It has occurred to me that this
>>>> could be addressed in several ways:
>>>>
>>>> 1. Modify the places where we build a where clause to use
>>>> getColumnName() instead of getLabel().(Maybe this breaks on other DBs
>>>> ?)
>>>> 2. Detect if the dialect is DB2 on Mainframe (z/OS, OS/400, etc.) and
>>>> call getColumnName instead of getLabel() to build where clause.
>>>> 3. Let the user configure an alias property to chose getColumnName()
>>>> vs. getLabel() for where clauses (default is getLabel() as it is now).
>>>>
>>>> Which solution above would be your preference (or any others) ?
>>>
>>> It seems funny to me that this problem was reported from DB2 on mainframe
>>> only up to now. If you look at the JDBC-API it appears to me that our way
>>> to
>>> use column labels must cause problems when a database supports them. My
>>> suggestion is choose the first of your ways because it seems the cleanest
>>> to
>>> me. If then anyone complains we should treat this as an exception from
>>> the
>>> standard way and see how we can cope with it.
>>>
>>> Perhaps because of the problem we should release 3.1 as a release
>>> candidate
>>> (rc1) first, wait a month or so to see what happens and then dot he final
>>> release.
>>>
>>> If you agree please tell me. I have all the necessary changes on my PC
>>> and
>>> will commit it together with the changes in Christian's patch. I even
>>> found
>>> some more places where the problem had to be fixed as well. Just look at
>>> the
>>> usages of ColumnDisplayDefinition.getLabel().
>>>
>>>
>>> Christian, thanks for pointing out the problem and sorry that we didn't
>>> realize the importance of your patch earlier.
>>>
>>> Gerd
>>>
>>>
>>>> Rob
>>>>
>>>>
>>>> ---------- Forwarded message ----------
>>>> From: Christian Gruenberg <gru...@ne...>
>>>> Date: Sat, Jan 2, 2010 at 8:31 AM
>>>> Subject: [Squirrel-sql-develop] DB2 for mainframe and Labels
>>>> To: squ...@li...
>>>>
>>>>
>>>> Hi,
>>>>
>>>> Db2 for mainframe support the labels. That means you can show
>>>> columnnames with a displayname. But DB2 doesn't support using the
>>>> labels in sql statements. To avoid this problem I have replaced the
>>>> getLabel calls with getColumnName. Unfortunately the replacement is
>>>> not only in the db2 plugin necessary but also in some general classes.
>>>>
>>>> You can find the patch here:
>>>>
>>>>
>>>> https://sourceforge.net/tracker/?func=detail&aid=2924887&group_id=28383&atid=393416
>>>>
>>>> I'm looking forward to some feedback,
>>>> Christian
>>>>
>>>>
>>>>
>>>> ------------------------------------------------------------------------------
>>>> This SF.Net email is sponsored by the Verizon Developer Community
>>>> Take advantage of Verizon's best-in-class app development support
>>>> A streamlined, 14 day to market process makes app distribution fast and
>>>> easy
>>>> Join now and get one step closer to millions of Verizon customers
>>>> http://p.sf.net/sfu/verizon-dev2dev
>>>> _______________________________________________
>>>> Squirrel-sql-develop mailing list
>>>> Squ...@li...
>>>> https://lists.sourceforge.net/lists/listinfo/squirrel-sql-develop
>>>>
>>>
>>
>
>

Re: [Squirrel-sql-develop] Fwd: DB2 for mainframe and Labels

[Gerd W. <ger...@t-...>]

Robert Manning wrote:
> Gerd,
> 
> I definitely agree that getColumnName() seems to be the better choice
> according to the documentation.  I also agree that the most desireable
> code change is to replace getLabel with getColumnName wherever we are
> dynamically building SQL to send to the database.  However, I was
> reluctant to replace all occurrences since for display purposes we
> should probably stick with getLabel (Users with databases that support
> labels probably define them for their own purpose and we should
> respect that).  
I looked at all the places where I changed label to column. As far as I can see 
your point concerns only to one of my changes, that is the one in 
MyTableModel.getColumnName(). I would have done that change too but I'm not so 
sure about it. I leave it to you to decide. See my changes in attachment.


> I also think that if we do this in core, then probably
> most if not all of Christian's custom DB2 datatypes are unnecessary.
> I skimmed over them so it is possible that I missed something but they
> seemed to be merely overriding the core implementation of
> getWhereClauseValue() and the like in *DataTypeComponent.  I'll take
> another look.
Sorry, I didn't realize that. My attached patch includes Christians changes as 
well.

Rob, would it be too much to ask you to do the final review of the code and then 
commit it? If you don't have the time or any further questions please tell me.

Thanks

Gerd



> 
> Rob
> 
> 
> On Mon, Jan 25, 2010 at 6:52 PM, Gerd Wagner <ger...@t-...> wrote:
>> Robert Manning wrote:
>>> Gerd,
>>>
>>> I think we should address this patch in our 3.1 release.  It seems
>>> that a number of our DB2 on mainframe users run into this problem
>>> according to this bug:
>>>
>>>
>>> https://sourceforge.net/tracker/index.php?func=detail&aid=2818117&group_id=28383&atid=393414
>>>
>>> It seems that it's for no good reason that we use
>>> ColumnDisplayDefinition.getLabel() rather than getColumnName().  It
>>> looks like Glenn chose to do that since the beginning and it was
>>> simply used everywhere consistently.  It has occurred to me that this
>>> could be addressed in several ways:
>>>
>>> 1. Modify the places where we build a where clause to use
>>> getColumnName() instead of getLabel().(Maybe this breaks on other DBs
>>> ?)
>>> 2. Detect if the dialect is DB2 on Mainframe (z/OS, OS/400, etc.) and
>>> call getColumnName instead of getLabel() to build where clause.
>>> 3. Let the user configure an alias property to chose getColumnName()
>>> vs. getLabel() for where clauses (default is getLabel() as it is now).
>>>
>>> Which solution above would be your preference (or any others) ?
>> It seems funny to me that this problem was reported from DB2 on mainframe
>> only up to now. If you look at the JDBC-API it appears to me that our way to
>> use column labels must cause problems when a database supports them. My
>> suggestion is choose the first of your ways because it seems the cleanest to
>> me. If then anyone complains we should treat this as an exception from the
>> standard way and see how we can cope with it.
>>
>> Perhaps because of the problem we should release 3.1 as a release candidate
>> (rc1) first, wait a month or so to see what happens and then dot he final
>> release.
>>
>> If you agree please tell me. I have all the necessary changes on my PC and
>> will commit it together with the changes in Christian's patch. I even found
>> some more places where the problem had to be fixed as well. Just look at the
>> usages of ColumnDisplayDefinition.getLabel().
>>
>>
>> Christian, thanks for pointing out the problem and sorry that we didn't
>> realize the importance of your patch earlier.
>>
>> Gerd
>>
>>
>>> Rob
>>>
>>>
>>> ---------- Forwarded message ----------
>>> From: Christian Gruenberg <gru...@ne...>
>>> Date: Sat, Jan 2, 2010 at 8:31 AM
>>> Subject: [Squirrel-sql-develop] DB2 for mainframe and Labels
>>> To: squ...@li...
>>>
>>>
>>> Hi,
>>>
>>> Db2 for mainframe support the labels. That means you can show
>>> columnnames with a displayname. But DB2 doesn't support using the
>>> labels in sql statements. To avoid this problem I have replaced the
>>> getLabel calls with getColumnName. Unfortunately the replacement is
>>> not only in the db2 plugin necessary but also in some general classes.
>>>
>>> You can find the patch here:
>>>
>>> https://sourceforge.net/tracker/?func=detail&aid=2924887&group_id=28383&atid=393416
>>>
>>> I'm looking forward to some feedback,
>>> Christian
>>>
>>>
>>> ------------------------------------------------------------------------------
>>> This SF.Net email is sponsored by the Verizon Developer Community
>>> Take advantage of Verizon's best-in-class app development support
>>> A streamlined, 14 day to market process makes app distribution fast and
>>> easy
>>> Join now and get one step closer to millions of Verizon customers
>>> http://p.sf.net/sfu/verizon-dev2dev
>>> _______________________________________________
>>> Squirrel-sql-develop mailing list
>>> Squ...@li...
>>> https://lists.sourceforge.net/lists/listinfo/squirrel-sql-develop
>>>
>>
> 

Re: [Squirrel-sql-develop] Fwd: DB2 for mainframe and Labels

[Robert M. <rob...@gm...>]

Gerd,

I definitely agree that getColumnName() seems to be the better choice
according to the documentation.  I also agree that the most desireable
code change is to replace getLabel with getColumnName wherever we are
dynamically building SQL to send to the database.  However, I was
reluctant to replace all occurrences since for display purposes we
should probably stick with getLabel (Users with databases that support
labels probably define them for their own purpose and we should
respect that).  I also think that if we do this in core, then probably
most if not all of Christian's custom DB2 datatypes are unnecessary.
I skimmed over them so it is possible that I missed something but they
seemed to be merely overriding the core implementation of
getWhereClauseValue() and the like in *DataTypeComponent.  I'll take
another look.

Rob


On Mon, Jan 25, 2010 at 6:52 PM, Gerd Wagner <ger...@t-...> wrote:
> Robert Manning wrote:
>>
>> Gerd,
>>
>> I think we should address this patch in our 3.1 release.  It seems
>> that a number of our DB2 on mainframe users run into this problem
>> according to this bug:
>>
>>
>> https://sourceforge.net/tracker/index.php?func=detail&aid=2818117&group_id=28383&atid=393414
>>
>> It seems that it's for no good reason that we use
>> ColumnDisplayDefinition.getLabel() rather than getColumnName().  It
>> looks like Glenn chose to do that since the beginning and it was
>> simply used everywhere consistently.  It has occurred to me that this
>> could be addressed in several ways:
>>
>> 1. Modify the places where we build a where clause to use
>> getColumnName() instead of getLabel().(Maybe this breaks on other DBs
>> ?)
>> 2. Detect if the dialect is DB2 on Mainframe (z/OS, OS/400, etc.) and
>> call getColumnName instead of getLabel() to build where clause.
>> 3. Let the user configure an alias property to chose getColumnName()
>> vs. getLabel() for where clauses (default is getLabel() as it is now).
>>
>> Which solution above would be your preference (or any others) ?
>
> It seems funny to me that this problem was reported from DB2 on mainframe
> only up to now. If you look at the JDBC-API it appears to me that our way to
> use column labels must cause problems when a database supports them. My
> suggestion is choose the first of your ways because it seems the cleanest to
> me. If then anyone complains we should treat this as an exception from the
> standard way and see how we can cope with it.
>
> Perhaps because of the problem we should release 3.1 as a release candidate
> (rc1) first, wait a month or so to see what happens and then dot he final
> release.
>
> If you agree please tell me. I have all the necessary changes on my PC and
> will commit it together with the changes in Christian's patch. I even found
> some more places where the problem had to be fixed as well. Just look at the
> usages of ColumnDisplayDefinition.getLabel().
>
>
> Christian, thanks for pointing out the problem and sorry that we didn't
> realize the importance of your patch earlier.
>
> Gerd
>
>
>>
>> Rob
>>
>>
>> ---------- Forwarded message ----------
>> From: Christian Gruenberg <gru...@ne...>
>> Date: Sat, Jan 2, 2010 at 8:31 AM
>> Subject: [Squirrel-sql-develop] DB2 for mainframe and Labels
>> To: squ...@li...
>>
>>
>> Hi,
>>
>> Db2 for mainframe support the labels. That means you can show
>> columnnames with a displayname. But DB2 doesn't support using the
>> labels in sql statements. To avoid this problem I have replaced the
>> getLabel calls with getColumnName. Unfortunately the replacement is
>> not only in the db2 plugin necessary but also in some general classes.
>>
>> You can find the patch here:
>>
>> https://sourceforge.net/tracker/?func=detail&aid=2924887&group_id=28383&atid=393416
>>
>> I'm looking forward to some feedback,
>> Christian
>>
>>
>> ------------------------------------------------------------------------------
>> This SF.Net email is sponsored by the Verizon Developer Community
>> Take advantage of Verizon's best-in-class app development support
>> A streamlined, 14 day to market process makes app distribution fast and
>> easy
>> Join now and get one step closer to millions of Verizon customers
>> http://p.sf.net/sfu/verizon-dev2dev
>> _______________________________________________
>> Squirrel-sql-develop mailing list
>> Squ...@li...
>> https://lists.sourceforge.net/lists/listinfo/squirrel-sql-develop
>>
>
>

Re: [Squirrel-sql-develop] Fwd: DB2 for mainframe and Labels

[Gerd W. <ger...@t-...>]

Robert Manning wrote:
> Gerd,
> 
> I think we should address this patch in our 3.1 release.  It seems
> that a number of our DB2 on mainframe users run into this problem
> according to this bug:
> 
> https://sourceforge.net/tracker/index.php?func=detail&aid=2818117&group_id=28383&atid=393414
> 
> It seems that it's for no good reason that we use
> ColumnDisplayDefinition.getLabel() rather than getColumnName().  It
> looks like Glenn chose to do that since the beginning and it was
> simply used everywhere consistently.  It has occurred to me that this
> could be addressed in several ways:
> 
> 1. Modify the places where we build a where clause to use
> getColumnName() instead of getLabel().(Maybe this breaks on other DBs
> ?)
> 2. Detect if the dialect is DB2 on Mainframe (z/OS, OS/400, etc.) and
> call getColumnName instead of getLabel() to build where clause.
> 3. Let the user configure an alias property to chose getColumnName()
> vs. getLabel() for where clauses (default is getLabel() as it is now).
> 
> Which solution above would be your preference (or any others) ?
It seems funny to me that this problem was reported from DB2 on mainframe only 
up to now. If you look at the JDBC-API it appears to me that our way to use 
column labels must cause problems when a database supports them. My suggestion 
is choose the first of your ways because it seems the cleanest to me. If then 
anyone complains we should treat this as an exception from the standard way and 
see how we can cope with it.

Perhaps because of the problem we should release 3.1 as a release candidate 
(rc1) first, wait a month or so to see what happens and then dot he final release.

If you agree please tell me. I have all the necessary changes on my PC and will 
commit it together with the changes in Christian's patch. I even found some more 
places where the problem had to be fixed as well. Just look at the usages of 
ColumnDisplayDefinition.getLabel().


Christian, thanks for pointing out the problem and sorry that we didn't realize 
the importance of your patch earlier.

Gerd


> 
> Rob
> 
> 
> ---------- Forwarded message ----------
> From: Christian Gruenberg <gru...@ne...>
> Date: Sat, Jan 2, 2010 at 8:31 AM
> Subject: [Squirrel-sql-develop] DB2 for mainframe and Labels
> To: squ...@li...
> 
> 
> Hi,
> 
> Db2 for mainframe support the labels. That means you can show
> columnnames with a displayname. But DB2 doesn't support using the
> labels in sql statements. To avoid this problem I have replaced the
> getLabel calls with getColumnName. Unfortunately the replacement is
> not only in the db2 plugin necessary but also in some general classes.
> 
> You can find the patch here:
> https://sourceforge.net/tracker/?func=detail&aid=2924887&group_id=28383&atid=393416
> 
> I'm looking forward to some feedback,
> Christian
> 
> ------------------------------------------------------------------------------
> This SF.Net email is sponsored by the Verizon Developer Community
> Take advantage of Verizon's best-in-class app development support
> A streamlined, 14 day to market process makes app distribution fast and easy
> Join now and get one step closer to millions of Verizon customers
> http://p.sf.net/sfu/verizon-dev2dev
> _______________________________________________
> Squirrel-sql-develop mailing list
> Squ...@li...
> https://lists.sourceforge.net/lists/listinfo/squirrel-sql-develop
> 

Re: [Squirrel-sql-develop] Fwd: DB2 for mainframe and Labels

[Robert M. <rob...@gm...>]

Christian,

The "process" for incorporating patches is very informal.  When we
have time, a committer will examine the patch to determine :
1) if it adequately addresses the issue reported, and
2) that it doesn't introduce any bugs

For database-specific patches such as yours, we like to test-drive it
against the target database (if possible) as well as other databases.
Your patch is also extensive, so we'll be looking for alternatives
that require fewer changes.  Thank you for identifying this issue with
DB2 on the mainframe and submitting the patch.  We'll let you know if
we have any questions.

Rob

On Sun, Jan 24, 2010 at 3:16 PM, Christian Gruenberg <gru...@ne...> wrote:
> Hi,
>
> can somebody explain, how the process for such a patch will go on?
>
> Regards,
> Christian
>
>
> ---------- Forwarded message ----------
> From: Christian Gruenberg <gru...@ne...>
> Date: 2010/1/2
> Subject: DB2 for mainframe and Labels
> To: squ...@li...
>
>
> Hi,
>
> Db2 for mainframe support the labels. That means you can show
> columnnames with a displayname. But DB2 doesn't support using the
> labels in sql statements. To avoid this problem I have replaced the
> getLabel calls with getColumnName. Unfortunately the replacement is
> not only in the db2 plugin necessary but also in some general classes.
>
> You can find the patch here:
> https://sourceforge.net/tracker/?func=detail&aid=2924887&group_id=28383&atid=393416
>
> I'm looking forward to some feedback,
> Christian
>
> ------------------------------------------------------------------------------
> Throughout its 18-year history, RSA Conference consistently attracts the
> world's best and brightest in the field, creating opportunities for Conference
> attendees to learn about information security's most important issues through
> interactions with peers, luminaries and emerging and established companies.
> http://p.sf.net/sfu/rsaconf-dev2dev
> _______________________________________________
> Squirrel-sql-develop mailing list
> Squ...@li...
> https://lists.sourceforge.net/lists/listinfo/squirrel-sql-develop
>