Questions on select statement in the data model

Hi,
In the select statement, the order by statement seems only to work at the lowest level of the select statement. Is this true?
What is the the purpose of the group by statement in the select statement?
As far as I understand, the grouping is catered for in the data model.....
Regards,
Kin

From the oracle report docs:
The order of column values in a default group is
determined by the ORDER BY clause of the query for
SQL queries and by the sort column for Express queries.
For column values in user-created groups, however, you
must use Break Order to specify how to order the break
column's values.So order by in your sql query is affected by groups in your data model.
The purpose of a group by statement in your select statement is to group values for aggregation purposes, like
select department_id,
       min(salary),
       max (salary)
from employees
group
    by department_id
order
    by department_id;Grouping in the data model of a report is a little different than a group by clause, but not entirely.

Similar Messages

How to use multiple selection parameters in the data model

Hi, after have looked all the previous threads about how to use multiple selection parameters , I still have a problem;
I'm using Oracle BI Publisher 10.1.3.3.2 and I'm tried to define more than one multiple selection parameters inside the data template;
Inside a simple SQL queries they work perfectly....but inside the data template I have errors.
My data template is the following (it's very simple...I am just testing how the parameters work):
<dataTemplate name="Test" defaultPackage="bip_departments_2_parameters">
<parameters>
<parameter name="p_dep_2_param" include_in_output="false" datatype="character"/>
<parameter name="p_loc_1_param" include_in_output="false" datatype="character"/>
</parameters>
<dataTrigger name="beforeReport" source="bip_departments_2_parameters.beforeReportTrigger"/>
<dataQuery>
<sqlStatement name="Q2">
<![CDATA[
select deptno, dname,loc
from dept
&p_where_clause
]]>
</sqlStatement>
</dataQuery>
<dataStructure>
<group name="G_DEPT" source="Q2">
<element name="deptno" value="deptno"/>
<element name="dname" value="dname"/>
<element name="loc" value="loc"/>
</group>
</dataStructure>
</dataTemplate>
The 2 parameters are based on these LOV:
1) select distinct dname from dept (p_dep_2_param)
2) select distinct loc from dept (p_loc_1_param)
and both of them have checked the "Multiple selection" and "Can select all" boxes
The package I created, in order to use the lexical refence is:
CREATE OR REPLACE package SCOTT.bip_departments_2_parameters
as
p_dep_2_param varchar2(14);
p_loc_1_param varchar2(20);
p_where_clause varchar2(100);
function beforereporttrigger
return boolean;
end bip_departments_2_parameters;
CREATE OR REPLACE package body SCOTT.bip_departments_2_parameters
as
function beforereporttrigger
return boolean
is
l_return boolean := true;
begin
if (p_dep_2_param is not null) --and (p_loc_1_param is not null)
then
p_where_clause := 'where (dname in (' || replace (p_dep_1_param, '''') || ') and loc in (' || replace (p_loc_1_param, '''') || '))';
else
p_where_clause := 'where 1=1';
end if;
return (l_return);
end beforereporttrigger;
end bip_departments_2_parameters;
As you see, I tried to have only one p_where_clause (with more than one parameter inside)....but it doesn't work...
Using only the first parameter (based on deptno (which is number), the p_where_clause is: p_where_clause := 'where (deptno in (' || replace (p_dep_2_param, '''') || '))';
it works perfectly....
Now I don't know if the problem is the datatype, but I noticed that with a single parameter (deptno is number), the lexical refence (inside the data template) works.....with a varchar parameter it doesn't work....
So my questions are these:
1) how can I define the p_where_clause (inside the package) with a single varchar parameter (for example, the department location name)
2) how can I define the p_where_clause using more than one parameter (for example, the department location name and the department name) not number.
Thanks in advance for any suggestion
Alex

Alex,
the missing thing in your example is the fact, that if only one value is selected, the parameter has exact this value like BOSTON. If you choose more than one value, the parameter includes the *'*, so that it looks like *'BOSTON','NEW YORK'*. So you need to check in the package, if there's a *,* in the parameter or not. If yes there's more than one value, if not it's only one value or it's null.
So change your package to (you need to expand your variables)
create or replace package bip_departments_2_parameters
as
p_dep_2_param varchar2(1000);
p_loc_1_param varchar2(1000);
p_where_clause varchar2(1000);
function beforereporttrigger
return boolean;
end bip_departments_2_parameters;
create or replace package body bip_departments_2_parameters
as
function beforereporttrigger
return boolean
is
l_return boolean := true;
begin
p_where_clause := ' ';
if p_dep_2_param is not null then
if instr(p_dep_2_param,',')>0 then
p_where_clause := 'WHERE DNAME in ('||p_dep_2_param||')';
else
p_where_clause := 'WHERE DNAME = '''||p_dep_2_param||'''';
end if;
if p_loc_1_param is not null then
if instr(p_loc_1_param,',')>0 then
p_where_clause := p_where_clause || ' AND LOC IN ('||p_loc_1_param||')';
else
p_where_clause := p_where_clause || ' AND LOC = '''||p_loc_1_param||'''';
end if;
end if;
else
if p_loc_1_param is not null then
if instr(p_loc_1_param,',')>0 then
p_where_clause := p_where_clause || 'WHERE LOC in ('||p_loc_1_param||')';
else
p_where_clause := p_where_clause || 'WHERE LOC = '''||p_loc_1_param||'''';
end if;
end if;
end if;
return (l_return);
end beforereporttrigger;
end bip_departments_2_parameters;
I've written a similar example at http://www.oracle.com/global/de/community/bip/tipps/Dynamische_Queries/index.html ... but it's in german.
Regards
Rainer

Use a multiple selection parameter in the data model

Hi
i have created a parameter with the property multiple selection.
This parameter i use in the Data Model where area,
if i just mark one value the report works well if i mark more the one
the report run into a error.
WHERE PSNR IN :ma
the xdo looks like
<?xml version = '1.0' encoding = 'utf-8'?>
<report version="1.1" xmlns="http://xmlns.oracle.com/oxp/xmlp" defaultDataSourceRef="tisoware">
<title>BDE Nettozeit</title>
<properties>
<property name="showControls" value="true"/>
<property name="online" value="true"/>
<property name="parameterColumns" value="2"/>
<property name="openLinkInNewWindow" value="true"/>
</properties>
<dataModel defaultDataSet="BDERUECK">
<dataSet id="BDERUECK">
<sql dataSourceRef="tisoware">
<![CDATA[SELECT (SUBSTR(DATUM,1,6)) AS DATUM, SUM(NTTOZEIT) AS ZEIT,
TRUNC(SUM(NTTOZEIT)/60) AS STUNDE
FROM BDERUECK
WHERE (DATUM >= :datev) AND (DATUM <= :dateb) AND (KZDELETE = 'N') AND FIRMA = '1' AND ((PSNR IN :ma))
GROUP BY (SUBSTR(DATUM,1,6))]]>
</sql>
<input id="datev" value="${datev}" dataType="xsd:string"/>
<input id="dateb" value="${dateb}" dataType="xsd:string"/>
<input id="ma" value="${ma}" dataType="xsd:string"/>
</dataSet>
</dataModel>
<valueSets>
<valueSet id="date">
<sql dataSourceRef="tisoware">
<![CDATA[SELECT DISTINCT DATUM
FROM BDERUECK]]>
</sql>
</valueSet>
<valueSet id="employee">
<sql dataSourceRef="tisoware">
<![CDATA[select      PERSTAMM.PSPERSNR as PSPERSNR,
      PERSTAMM.PSNR as PSNR
from      TISOWARE.PERSTAMM PERSTAMM
where    PERSTAMM.FIRMA = '1'
order by PERSTAMM.PSPERSNR]]>
</sql>
</valueSet>
</valueSets>
<parameters>
<parameter id="datev" defaultValue="20050103">
<select label="Datum von : " valueSet="date" multiple="false" all="false" refreshParameters="false"/>
</parameter>
<parameter id="dateb" defaultValue="20060930">
<select label="Datum bis : " valueSet="date" multiple="false" all="false"/>
</parameter>
<parameter id="ma" defaultValue="152">
<select label="Mitarbeiter : " valueSet="employee" multiple="true" all="false"/>
</parameter>
</parameters>
<templates default="BDE Nettozeit">
<template label="BDE Nettozeit" type="rtf" url="BDE Nettozeit.rtf"/>
</templates>
</report>

Alex,
the missing thing in your example is the fact, that if only one value is selected, the parameter has exact this value like BOSTON. If you choose more than one value, the parameter includes the *'*, so that it looks like *'BOSTON','NEW YORK'*. So you need to check in the package, if there's a *,* in the parameter or not. If yes there's more than one value, if not it's only one value or it's null.
So change your package to (you need to expand your variables)
create or replace package bip_departments_2_parameters
as
p_dep_2_param varchar2(1000);
p_loc_1_param varchar2(1000);
p_where_clause varchar2(1000);
function beforereporttrigger
return boolean;
end bip_departments_2_parameters;
create or replace package body bip_departments_2_parameters
as
function beforereporttrigger
return boolean
is
l_return boolean := true;
begin
p_where_clause := ' ';
if p_dep_2_param is not null then
if instr(p_dep_2_param,',')>0 then
p_where_clause := 'WHERE DNAME in ('||p_dep_2_param||')';
else
p_where_clause := 'WHERE DNAME = '''||p_dep_2_param||'''';
end if;
if p_loc_1_param is not null then
if instr(p_loc_1_param,',')>0 then
p_where_clause := p_where_clause || ' AND LOC IN ('||p_loc_1_param||')';
else
p_where_clause := p_where_clause || ' AND LOC = '''||p_loc_1_param||'''';
end if;
end if;
else
if p_loc_1_param is not null then
if instr(p_loc_1_param,',')>0 then
p_where_clause := p_where_clause || 'WHERE LOC in ('||p_loc_1_param||')';
else
p_where_clause := p_where_clause || 'WHERE LOC = '''||p_loc_1_param||'''';
end if;
end if;
end if;
return (l_return);
end beforereporttrigger;
end bip_departments_2_parameters;
I've written a similar example at http://www.oracle.com/global/de/community/bip/tipps/Dynamische_Queries/index.html ... but it's in german.
Regards
Rainer

PowerPivot removed table replicated into the data model on refresh

Hey
I am facing the following problem which is absolutely persistent and annoying in nature. I am using a data model created with PowerPivot with connections to three different SQL servers and 25 tables. The bug is that every time I refresh my data, a table
that was initially added to the data model (and later removed) is replicated in equivalent number of tables that use this connection, which is currently around 12 tables, so I get 12 duplicates of the same unnecessary table on every refresh, named with the
name of the schema on the SQL server that is followed by a running number. Removing these from the data model on every refresh simply takes too much time on a daily basis. The additional undesired table which is created is what is defined in the SQL query
of the connection string of this connection.
I tried to solve this problem by zipping the Excel file and modifying this SQL statement present in the connection string in the connections.xml file in a way that it would not return anything, but this always leads into a corrupted workbook.
After encountering this I have managed to avoid this problem by making sure that the initial table I'm adding to the data model will remain in the model, but I really don't want to rebuild this current model. Is there any fix for this issue without
rebuilding the data model?
Thanks for all the help in advance!

I am using 64-bit Office 2013 with the recent SP1 upgrade and the included PowerPivot plug-in on a Windows 8.1 machine.
I always create the connection by using "From SQL server" then fill in the server and log in details and use a ready SQL statement that I have written by using SQL Server Management Studio to import the initial table. For all the preceding tables
using the same server and schema, I always select it from existing connections and just paste the SQL queries required for any additional tables to the "Write a query that will specify the data to import", rather than using the table import tool.
I am not sure if it could have any effect that I've been using the type connection "From SQL server" although the database is really running on Azure (there is a possibility to select From Other sources > Microsoft SQL Azure, but the functionality
seems the same).
I don't really understand what I should edit from the existing connections section. Just to clarify, apparently the Connection String parameters can be edited from PowerPivot, but not the Command Text which can be viewed from Excel Data > Connections
> Properties, but is shown in gray with a message "Some properties cannot be changed because this connection was modified using the PowerPivot add-in."
In the data model I am using two sets of identical tables, but connected to two different servers that are running a database 1:1 in structure. Initially the problem considered only the other of these, but I accidentally reproduced it by, again, editing
the SQL query in the table that was initially imported when that connection was created (PowerPivot > Design > Table Properties). Now when refreshing I get tables that are structured as in what the "Connection Text" part of the connection would
produce, which replicated for the equivalent amount of additional tables using the same connection, so now I'm getting 12 tables (the number of tables using the same connection) each named "*schemaname1* Query", "*schemaname1* Query2",
"*schemaname1* Query3" "*schemaname2* Query" and so on.
Personally I can definitely see a pattern here. If there isn't a table matching what has been specified in the "Command Text" that was described when the connection was created, then it for some reason runs this query anyway on every table that
is using the connection.

Powerpivot Error on Refresh -- "We couldn't get data from the data model..."

I'm using Excel 2013 and Windows 8.1. I have a spreadsheet I've been using for over a year, and I've just started getting this error message when I try to refresh the data.
"We couldn't get data from the Data Model. Here's the error message we got:
The 'attributeRelationship' with 'AttributeID' - 'PuttDistCat9' doesn't exist in the collection"
Any idea how I can fix this problem? I haven't changed anything related to that particular attribute. All the data is contained in separate sheets in the workbook, so there are no external sources of data.
Thanks.
Jean

Thanks for all the suggestions.
I found a slightly older version of the spreadsheet that still refreshes properly, so I don't think I have any issues with the version of Excel or Power Query. (I've had this same error before, and I believe I applied the hotfix at that time.)
I think this problem started after I updated a number of the date filters in the pivot tables. I haven't made any changes to the data model, and the only updates I've made were to add data (which I do all the time), and to change the date filters on
the pivot tables.
As suggested, I added a new pivot table querying one table (the table with the attribute that shows up in the error message), and it worked fine. I can also refresh this pivot table.
Then I tried adding a pivot table which went against several tables in the data model (including the table in question). The pivot table seemed to return that data properly. However, when I tried to refresh it, I got the same error message ("we
couldn't get data from the data model...").
Dany also suggested running the queries one at a time to see which one is in error. Without checking all the pivot tables, it appears that any which use the table "HolePlayedStrokes" generate the error (this is the table with the attribute
mentioned in the error message). Pivot Tables without that particular table seem to refresh OK. Unfortunately, that is the main table in my data model, so most of the pivot tables use it.
Any other suggestions? I'd be happy to send a copy of the spreadsheet.
Thanks for all the help.
Jean

"An error occured while working on the Data Model in the workbook" on some workbooks published to Power BI site

Hello,
I am using the Power BI for Office 365, and I have published several Excel 2013 workbooks having Power Pivot Data Models.
I have a problem on some of the workbooks, once a slicer is selected, I get the error: "An error occurred while working on the Data Model in the workbook" and the slicers do not affect the charts.
Some workbooks work perfectly fine. I am using the same user for all workbooks when creating and publishing. I tried with small workbooks less than < 10 MB size and larger workbooks > 10 MB. There is no rule, some workbooks larger than 10 MB work perfectly
with the slicers effecting the charts, and some don't. Similarly for smaller size.
Any ideas of how I can debug the cause of the issue?
Appreciate any feedback,
Thanks,
Grace

Hi Grace,
I assume that the experience in the Excel client is working fine, right?
Are you getting a correlation id with the error?
Please send us a bit more information / samples to reproduce over email to
this address.
thanks,
Guy
GALROY

Why do I get the symbol data link in the query_name of the data model

If I use an inline subquery within the SELECT of a data model query, the symbol data link appears in the query_name of the data model.
This causes the data link between two queries not to be used even if there is an existing link between the two. The trace file doesn't
have that join condition and so the child query in which the inline subquery was added fetches many redundant data.
Any idea what that symbol represents and how to get rid of it. It also appears during other times, not only when we use inline subquery
within a SELECT. Not sure if it can cause any harm to the report.
The problem is seen in Report Builder 6.0.8.19.0. Not sure if the problem is with other releases also.
Any suggestions are welcome.

I am living with the symbol in the query box as well, because my report only has a single query. My query also has several inline subqueries. I had never seen this before last week, spent a day re-reading the manuals and couldn't figure it out.
My post is not a solution, but a "me too" echo that could use an answer.
Troy Tremble
If I use an inline subquery within the SELECT of a data model query, the symbol data link appears in the query_name of the data model.
This causes the data link between two queries not to be used even if there is an existing link between the two. The trace file doesn't
have that join condition and so the child query in which the inline subquery was added fetches many redundant data.
Any idea what that symbol represents and how to get rid of it. It also appears during other times, not only when we use inline subquery
within a SELECT. Not sure if it can cause any harm to the report.
The problem is seen in Report Builder 6.0.8.19.0. Not sure if the problem is with other releases also.
Any suggestions are welcome.

Is it proper for an application module has 300 views in the data model?

Based on my understanding, all view object instances are initialized at the same time as the application module.
It follows that when an application module is associated with a user session, the queries of each view object in the data model are executed.
ie. if the data model has 325 view object instances, 325 queries will be executed.
This doesn't seem efficient:
1. As each user session only need to access a subset of the view objects (for examples 35 views), the rest of view objects need not be instantiated for that session.
2. Performance may be slow on executing all 325 queries at one time, even worse when some queries are complex queries.
Am I correct on this?
In this case, will it be better to instantiate the view objects using the view object definition at runtime instead of adding all of them to the application module at design time?

No, that's not the default behaviour.
Assuming you're talking JDev 11, the AM on it's General tab as a Tuning option where you can set the loading of the VOs to Lazy or Non-Lazy, Lazy is the default. In turn each VO on their General tab also has a Tuning option of similar nature.
Online documentation says:
Lazy LoadingSelect to choose lazy loading. In this case, the application module instance is created without any of its components. Your view object and nested application module instances will not load until the first time they are used. For example, when lazy loading is enabled, a view object is instantiated only when the application accesses it. This will make start-up time faster.
Non-lazy Loading
Select to choose non-lazy loading. Your view object and nested application module instances will load when the application starts up. For example, suppose the application module you are editing defines three view objects and two nested application modules (which appear in the Application Module Instance section of the Data Model page over the application module overview editor). When lazy loading is disabled, the application module instance is created together with all three view objects and both nested application modules.>
So in essence as you describe for each session, if we're using the Lazy setting, even though you have 325 VOs within an AM, if the user within the session only visits 35 queries, only 35 queries should fire.
Does this help?
CM.

How to know if the data model qry is not returning any data

Hi
If my report doesn't have data (meaning if the data model qry doesn't retreive any data), I wan't to show this text: "There is no data returned."
My question is how would I know that the qry is not returning any data. I know I can create a text field, and write a trigger on it, to set it True or False depending on the fact that data is being returned or not.
Thanks
Shalu

Hi,
In Data Model create a summary column for function=count on one of your columns.
Then for BeforeReport trigger you can write
function BeforeReport return boolean is
begin
if :CS_RPT_NODATA > 0 then
return (TRUE);
end if;
SRW.MESSAGE(300,'There is no data');
return (FALSE);
end;
Regards, Gicu

Benefit of using store procedure instead of select statement to pull data into biztalk

I was wondering why store procedure is more beneficial than using select statement to pull data into biztalk?

In addition to the above two points, in case if there is a change in logic of stored procedure, you only need to modify the stored proc and the applications calling/using it may be left intact.
Also, stored procedures are complied code so performance is better and safe too.
Thanks,
Prashant
Please mark this post accordingly if it answers your query or is helpful.

The Data Model connection could not be created

Hello,
Using powerquery, I am trying to access daily logs conveniently, the are over 24m records in a month.
I have selected the columns i need, and checked the box ' load data to Model', Once the query has ran it returns the error, "The Data Model connection could not be created".
Thank you for any help on this :)

This is a tough one because as an add-in, we can't get much information from the API for failures like this. Our general recommendation for people working with big data sets is to use 64-bit Excel. I realize that's a big hammer, but it might be your only
option if you need to work with a 24 million row data set.
Another option (which you may have already done) is to filter and aggregate that dataset as much as possible before adding it back into Excel. The analysis done inside the Power Query editor is less likely to negatively affect the working set on your machine.
(It does depend how the query is built and where the data is come from. If we have to pull all the data into memory to do some of the operations then it won't help at all.)

Can i copy the data model from report?

hi all
I use oracle reports 6i newlly
I want to make a copy of the data model in one report and paste it to an other report
as they have almost the same structure with little modifications
how can i do that ?
Regards

hi all
I use oracle reports 6i newlly
I want to make a copy of the data model in one report
and paste it to an other report
as they have almost the same structure with little
modifications
how can i do that ?
RegardsCan't you save the Report1 as Report2 (by doing this you will have your Data Model and everything from Report1 will be copied to Report2) and do those little modifications in the Report2.
OR
Take the SQL statement from the Report1 and use that in the Report2.

Importing editable data that can be compared to the data model?

I will describe my objective and how I've solved it currently without Powerpivot. Then I will ask if there's a way to improve upon my solution by using Powerpivot.
The topic is project allocation forecasting. Currently I import two copies of the same dataset from a work management system in to Excel. One copy is then modified iteratively and produces graphs used in predict resources on various projects
in the future. We work through many "what if" modifications to this data. Once we believe we have the data correct, I then compare the modified data to the second copy of the data, to see what we changed.
With Powerpivot, much of what I coded in VBA for Excel previously is extremely simple (and better.) Additional advantage is I don't have to actually import the data in to Excel (after first exporting it from the work management system.) But I
lost the ability to modify the data over and over, and then compare it back to its original source.
Finally to my question... is there a better way to import the data in a way that it will be editable, and then compare that data back to the data existing in the data model?

well, you cannot modify any data once it is loaded to power pivot - unfortunately
the only thing i could think of is to use Excel linked tables which are then updated in power pivot once you modified them
depending on how you want to change your data this may be an option for you
Gerhard Brueckl
blogging @ http://blog.gbrueckl.at
working @ http://www.pmOne.com

Can "trigger" management be addressed in the data modeler?

Please add an easier way to add/modify triggers.
I reverse-engineered my model into the data modeler and my triggers all migrated correctly to the model but, for the life of me, had no idea how to view, modify or create triggers. After extensive research, I found that they can be accessed in the left-hand Browser by drilling some 8 levels to view the triggers for each table - why? Why can't this functionality be added to properties context menu for the table. I'm not a big advocate for triggers but if I need to work on one, the current methodology is tedious at best.
Of course, if there's some other way to do this, please inform...
Accessing Triggers via the Browser
Design -> Relational Model -> APEX_DEMO -> Physical Model -> Oracle Database 11g -> Tables -> ACTIVITY_LOG -> Triggers

Thanks Joop and Phillip...
While I occasionally use SQL Developer, I currently use PL/SQL Developer (Allround Automations) as my development tool of choice and bringing a 3rd tool as SQL Developer into the mix, makes no sense to me.
I completely agree that managing a trigger via a development tool is much easier but one occasionally wants to quickly add a trigger either manually or via a template and this is not easy currently.
While functions, procedures, package, etc. are all important to a data model, the existence of most triggers are completely dependent on a table and are a critical component in any model that uses them. The ability to access and modify them easily in Data Modeler is important and shouldn't be an afterthought as is perceived here. If trigger access cannot be added to the Table Properties (as noted in the image below), at the very least, please provide a shortcut methodology to do so.

Excel 2013 PowerPivot Error - "PowerPivot is unable to load the Data Model"

I'm attempting to use the PowerPivot add-on in Excel. When clicking 'Manage Data Model' or attempting to 'Load to data model,' I receive the error 'We couldn't load the Data Model. This may be because the Data Model in this workbook is damaged.',
followed by 'PowerPivot is unable to load the Data Model.' This happens irrespective of data source type. I have Excel 2013 32-bit, PowerPivot 32-bit, running on Windows 7 OS 64-bit. I am not running an SSAS instance on my machine.
Any suggestions?

Hi
Have you ever tried to repair your office?
Also we may refer to the following blog :
http://blogs.technet.com/b/the_microsoft_excel_support_team_blog/archive/2013/11/12/powerpivot-for-excel-2013-errors-after-october-update-kb-2825655.aspx
This issue seems to be caused by the October 2013 update (KB 2825655) for Excel 2013. Try to remove it or restore the system to an earlier point and check the result.
Regards
Tylor Wang
TechNet Community Support

Questions on select statement in the data model

Similar Messages

Maybe you are looking for