Re: [Gusdev-gusdev] Postgres GUS Scripts

[Jeetendra S. <so...@vb...>]

Hi Michael,

Thanks for your reply. I had another problem running the plugins using the
Postgresql version. I have generated all the objects by following the
normal build procedure. However, I am getting the following error while
running the plugin:

Error:  attempting to access a child 'GUS::Model::DoTS::NAEntry' of table
'GUS::Model::DoTS::ExternalNASequence' , but that table does not have a
child of that type. at
/home/soneja.local/gusHomePG/lib/perl/GUS/Model/GusRow.pm line 396, <GEN0>
line 163.

When I opened the automatically generated ExternalNASequence_Table.pm,
it's child list is empty.

Did I miss something during the build process ?

Thank you for your help.

Jeetendra.





>
> Hi Jeetendra:
>
> Jeetendra Soneja wrote:
>> Hi Michael,
>>
>>    Thanks for the updates. After downloading the latest code from the
>> CVS,
>> I was able to register a couple of plugins successfully. Just a small
>> change while registering GBParser, I had to comment out 'use
>> GUS::Model::DoTS::NAFeatureImp' line.
>>
>
> Right, and further, while this technically works in Oracle, it
> represents an incorrect usage of the data model:  The IMP tables should
> NEVER be used, either directly in SQL queries, nor through the object
> layer.  Instead, the appropriate superclass should be used
> (GUS::Model::DoTS::NAFeature in this case)
>
>> I am trying to run GBParser and am having the following problems :-
>>
>> --An oracle-specific call setOracleDateFormat() in the run mode of the
>> plugin was causing problems. (For the time being I have commented it)
>>
>
> In addition, it looks like GBParser uses Oracle's sysdate in setting the
> release date.
>
>> --It's having problems inserting rows into the tables. The value '' is
>> being inserted into the NUMERIC type columns. Postgresql doesn't seem to
>> interprete this as a null value, but an emptry string.
>>
>
> Thanks--- This will need to be resolved higher up in the object layer.
>
>>
>> Thank you once again for your help.
>>
>
> Thank you for all your testing!!  I have a growing list of issues and
> fixes for the Postgres version-- I'll keep this list updated as fixes
> are committed to CVS, and please continue to mail the list with your
> progress in getting this stuff working.
>
> --Mike
>
>> Jeetendra.
>>
>>
>>
>>
>>
>>
>>
>>>All,
>>>
>>>I've updated DbiDbHandle in CVS to address the capitalization issue
>>>(thanks to Angel for the pointer here).
>>>
>>>This should then allow plugins to be registered with postgres GUS
>>> schemas.
>>>
>>>Just to recap:
>>>
>>>1) You'll need to use the latest SQL scripts sent by me on 11/30, and
>>>the latest CVS (as of now!)
>>>2) You'll need to change line 34 of GusApplication such that the
>>>requiredDbVersion for Core is 3.0, not 3.
>>>3) You'll need to manually add rows as outlined in VBI's GUS
>>>installation document.  In addition, you'll need to add these rows:
>>>
>>>INSERT INTO Core.AlgorithmParamKeyType
>>>VALUES(0,'string',now(),1,1,1,1,1,0,1, 1, 1, 1);
>>>INSERT INTO Core.AlgorithmParamKeyType
>>>VALUES(1,'float',now(),1,1,1,1,1,0,1, 1, 1, 1);
>>>INSERT INTO Core.AlgorithmParamKeyType
>>>VALUES(2,'int',now(),1,1,1,1,1,0,1, 1, 1, 1);
>>>INSERT INTO Core.AlgorithmParamKeyType
>>>VALUES(3,'ref',now(),1,1,1,1,1,0,1, 1, 1, 1);
>>>INSERT INTO Core.AlgorithmParamKeyType
>>>VALUES(4,'boolean',now(),1,1,1,1,1,0,1, 1, 1, 1);
>>>INSERT INTO Core.AlgorithmParamKeyType
>>>VALUES(5,'date',now(),1,1,1,1,1,0,1, 1, 1, 1);
>>>
>>>Thanks,
>>>
>>>Mike
>>>
>>>
>>>Jeetendra Soneja wrote:
>>>
>>>>Hi all,
>>>>
>>>>The reason that I got errors that I posted earlier was that I did not
>>>> do
>>>>a
>>>>build. I had just created the Postgresql GUS schema and was trying to
>>>>register the gus application. However, after doing a build, things
>>>>worked
>>>>out fine.
>>>>
>>>>I would like to list some of the problems that I have faced till this
>>>>point in using the Postgresql version:-
>>>>
>>>>-- The rows included in the script gus-rows.pl need a minor
>>>>modification.
>>>>Values in the rows that are being inserted into core.databaseinfo need
>>>>to
>>>>be changed. The names of the schemas CORE, DOTS and SRES should be
>>>>changed
>>>>Core, DoTS, SRes respectively. Also the values for the version numbers
>>>>should be changed from 1.0 and 3.0 to 1 and 3 respectively.
>>>>
>>>>-- Currently, I am having trouble registering a plugin (any plugin). A
>>>>number of modules/subroutines are retrieving row as a hash-reference
>>>> and
>>>>access a value in a row using key, i.e. column name of the table. And
>>>>all
>>>>the keys used in the code are in upper case. This works in case of
>>>>oracle,
>>>>however, select statements in Postgresql return all the column names in
>>>>lower case. Therefore, in this case, it couldn't find the row for ga in
>>>>the core.algorithmimplementation table (although it's there since I
>>>> have
>>>>already registered ga). And when I change the key/column name to lower
>>>>case in GusApplication.pm code, it worked fine.
>>>>
>>>>I would appreciate any suggestion or comments.
>>>>
>>>>Thanks a lot,
>>>>Jeetendra.
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>>Hi Jeetendra,
>>>>>
>>>>>Comments below....
>>>>>
>>>>>Jeetendra Soneja wrote:
>>>>>
>>>>>
>>>>>>Hi Michael,
>>>>>>       I am glad that the PostgreSQL version has been released.
>>>>>>
>>>>>
>>>>>This is not an official release!!  These are preliminary scripts,
>>>>> which
>>>>>have had very little testing and are likely to change in the future
>>>>>(although probably not significantly).
>>>>>
>>>>>
>>>>>
>>>>>>       I had a question about using it. Could you tell me if any of
>>>>>>the
>>>>>>existing plugins (especially those in Common::Plugin) would work
>>>>>>with PostgreSQL version of GUS ?
>>>>>>
>>>>>
>>>>>No existing plugins have been tested with PGGUS.  Please report any
>>>>>successes or failures you have to the list.
>>>>>
>>>>>--Mike
>>>>>
>>>>>
>>>>>
>>>>>>Thanks,
>>>>>>Jeetendra.
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>>All,
>>>>>>>
>>>>>>>Postgres GUS scripts are attached.  The first, pggus.sql contains
>>>>>>> DDL
>>>>>>>for creating GUS tables, indexes, constraints, and sequences.  The
>>>>>>>second, gus-rows.sql, contains DML for populating the core.tableinfo
>>>>>>>and
>>>>>>>core.databaseinfo tables.
>>>>>>>
>>>>>>>Consider these early beta-- this is not part of an official release,
>>>>>>>they've had very limited testing, and they are unsupported.  In
>>>>>>>general
>>>>>>>the existing documentation should provide the information needed to
>>>>>>>create an instance with these and GUS with it, but there are
>>>>>>>derivations
>>>>>>>that will be necessary.
>>>>>>>
>>>>>>>More information on Postgres and GUS, including an official release,
>>>>>>>should be coming in the next several weeks.
>>>>>>>
>>>>>>>Thanks,
>>>>>>>
>>>>>>>Mike
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>
>>>
>>>-------------------------------------------------------
>>>SF email is sponsored by - The IT Product Guide
>>>Read honest & candid reviews on hundreds of IT Products from real users.
>>>Discover which products truly live up to the hype. Start reading now.
>>>http://productguide.itmanagersjournal.com/
>>>_______________________________________________
>>>Gusdev-gusdev mailing list
>>>Gus...@li...
>>>https://lists.sourceforge.net/lists/listinfo/gusdev-gusdev
>>>
>>
>>
>>
>
>
> -------------------------------------------------------
> SF email is sponsored by - The IT Product Guide
> Read honest & candid reviews on hundreds of IT Products from real users.
> Discover which products truly live up to the hype. Start reading now.
> http://productguide.itmanagersjournal.com/
> _______________________________________________
> Gusdev-gusdev mailing list
> Gus...@li...
> https://lists.sourceforge.net/lists/listinfo/gusdev-gusdev
>


-- 
Jeetendra Soneja
Research Associate
Virginia Bioinformatics Institute
1880 Pratt Dr., Building XV
Blacksburg, VA 24060, USA
Phone: (540)-231-2789
http://www.vbi.vt.edu