BEx queries - one or more SQL statements generated?

Iu2019m a BusinessObjects man and BEx is new to meu2026When you execute a BEx query against a cube u2013 how does BEx manage that query and what gets send to the backend DB (e.g. Oracle)?
I understand the cube is one table in Oracle?, but other tables (dimensions?) will also be needed to return all the data for a given report/query.
Does BEx pass that as one query to Oracle, or is it performed as separate queries, with BEx pulling the result sets together?

Hi Durgesh, Susan
Thank you to both of you for your response to Ed on this thread. The background to these questions on how SQL statements are generated from BEx queries that we are working with Ed in a project to test the capability of accelerating the performance of BEx queries via the following process described below. (Please excuse any terminology errors as like I do not have a background in SAP):
1. Export fact table InfoObjects relevant to queries for InfoCubes to be accelerated by defining as Open Hub destinations, creating mappings and a data transfer process. Trigger export from SAP BI via a Process Chain and then load data to external database via ETL tool. Master data (dimension tables) remain in SAP BI
This step was completed successfully, including exporting of data.
2. Create a Universal Data Connect (UDC) source system to map to the exported fact tables.
3. Create a virtual provider mapping to the UDC source defined in step 2 to enable access to external table via JDBC.
4. Execute BEx report, selecting the query properties 'Cache inactive' and 'Reading all data at once', specifying restrictions (variables) in the report which would result in a small subset of data being returned from the fact table if applied in the SQL passed to the external database.
We hoped that SAP BI would push down predicates which could be applied to the fact table into the SQL sent to the external database via the virtual provider, but instead an SQL query selecting all columns with no WHERE clause was generated with the result that a 36 million row results set was sent across a network to join with dimension data, rather than a few rows.
Is this how SAP BI works in version 7.0 or are there configuration options which can change this behaviour? Is so, if all tables (including master data (dimensions) were exported to external database would SAP BI then transmit the database queries as joins to the external database?
Sorry this was a bit long, and thank you in advance for any advice you can give.
Stuart.

Similar Messages

Sql statement generator for java.

Is there a (api's ) tool which generates sql statements given a database table, which can be embeeded into the application.

Dear Anand,
Try this one:
CASE balance WHEN 0 THEN 'Completed' ELSE CASE WHEN DATEDIFF(DD, EndDate, GetDate()) >= 0 THEN 'Progress' ELSE 'Pending' END END
I have assumed you don't have negative balance. It that is not true, give me the complete case.
Thanks,
Gordon
Thanks,
Gordon

More sql statements at once

Hello,
Can I send more commands (statements) at one calling separating them with semicolon?
For example:
.exec("INSERT INTO myTable (col1) VALUES (val1); INSERT INTO myTable (col1) VALUES (val2)"
I tried it through odbc and oci driver, but was unsuccessful.
In case using MS SQL this works fine.
Is there any chance to be able to do this in case Oracle?
Thanks in advance.
kind regards,
d.a.

>
Can I send more commands (statements) at one calling separating them with semicolon?
.exec("INSERT INTO myTable (col1) VALUES (val1); INSERT INTO myTable (col1) VALUES (val2)"Yes (see below), but you shouldn't for the above statements. You should use just one statement with bind variables, and execute it twice (or more). To make it even faster, you can even specify the bind variables via arrays to OCI to have only one round-trip to the DB.
Is there any chance to be able to do this in case Oracle?The Oracle way to do this is via an PL/SQL anonymous block:
.exec("BEGIN stmt1; stmt2; END;")
But again in this particular case, bind variable is the way. Read Tom Kyte on the subject, nobody explains it better than him. --DD

Logging/viewing SQL statements generated by the SQL Developer IDE?

I know it is possible to view the SQL statements that the SQL Developer IDE generates to fetch result-sets from the database whenever one performs any action - say expanding the "Tables" node for a database, for example. I would appreciate if someone could point out how this functionality could be enabled.
Thanks,
RG

In 1.0, you could use the AddVMOption -Daudit.trace=db_api option inside the sqldeveloper.conf file.
In 1.1 however, I don't get output inside the console anymore (for expanding trees, etc.).
Kris, has this option been replaced by another one? Please let us in on all the fun ;-)
K.

How to view the SQL statement generated during execution of the BW Query?

Dear Experts,
I am trying to retrieve data in a SAP BW Query from a Non-SAP system.
I think if I am able to see the complete SQL statement that was generated when I executed the BW Query, I may be able to use it to retrieve the data.
Do you know how and where I can see the SQL statement of a BW Query's SQL statement?
I tried RSRT options to execute the Query but still could not find the SQL statement.
Thanks in advance.
Regards,
Shunhui.

hi
goto rsrt
give your query name
select execute + debug option
in the debug option under data manager check the check box "display SQL/BIA query
selcet continue
you can see the sql statement
thanq

MS Access queries different to Java SQL statements

Ok, i dont understand something:
My program is connected to a MSAccess datasource. The query in MSAccess is
SELECT *
FROM Food_Data
WHERE Description Like 'Cheese*';
...but this wont work with a Java SQl query. I have to use this instead:
String query = "SELECT * FROM Food_Data " +
"WHERE Description LIKE 'Cheese%'";
Surely java would just passes the string query to MSAccess. But no. Java needs '%' as the wildcard, and MSAccess needs '*' as the wildcard.
Can anyone please explain this to me?

Well % is the standard wildcard character for SQL as defined by the ANSI standard. MS Access just doesn't conform to this standard.
Having said that, I'm still kind of astonished that it doesn't work :-)
The JDBC driver should pass the statement "as is" to the underlying DBMS, but obviously the JDBC/ODBC bridge does some "quirking". But maybe it
s the ODBC datasource. I can see a "ExtendedAnsiSQL" setting in one of my Access ODBC settings, maybe that changes something.

DBIF_RSQL_INVALID_RSQL The maximum size of an SQL statement was exceeded

Dear,
I would appreciate a helping hand
I have a problem with a dump I could not find any note that I can help solve the problem.
A dump is appearing at various consultants which indicates the following.
>>> SELECT * FROM KNA1                     "client specified
559                  APPENDING TABLE IKNA1
560                  UP TO RSEUMOD-TBMAXSEL ROWS BYPASSING BUFFER
ST22
What happened?
    Error in the ABAP Application Program
    The current ABAP program "/1BCDWB/DBKNA1" had to be terminated because it has
    come across a statement that unfortunately cannot be executed.
Error analysis
    An exception occurred that is explained in detail below.
    The exception, which is assigned to class 'CX_SY_OPEN_SQL_DB', was not caught
     and
    therefore caused a runtime error.
    The reason for the exception is:
    The SQL statement generated from the SAP Open SQL statement violates a
    restriction imposed by the underlying database system of the ABAP
    system.
    Possible error causes:
     o The maximum size of an SQL statement was exceeded.
     o The statement contains too many input variables.
     o The input data requires more space than is available.
     o ...
    You can generally find details in the system log (SM21) and in the
    developer trace of the relevant work process (ST11).
    In the case of an error, current restrictions are frequently displayed
    in the developer trace.
SQL sentence
550     if not %_l_lines is initial.
551       %_TAB2[] = %_tab2_field[].
552     endif.
553   endif.
554 ENDIF.
555 CASE ACTION.
556   WHEN 'ANZE'.
557 try.
>>> SELECT * FROM KNA1                     "client specified
559                  APPENDING TABLE IKNA1
560                  UP TO RSEUMOD-TBMAXSEL ROWS BYPASSING BUFFER
561    WHERE KUNNR IN I1
562    AND   NAME1 IN I2
563    AND   ANRED IN I3
564    AND   ERDAT IN I4
565    AND   ERNAM IN I5
566    AND   KTOKD IN I6
567    AND   STCD1 IN I7
568    AND   VBUND IN I8
569    AND   J_3GETYP IN I9
570    AND   J_3GAGDUMI IN I10
571    AND   KOKRS IN I11.
572
573   CATCH CX_SY_DYNAMIC_OSQL_SEMANTICS INTO xref.
574     IF xref->kernel_errid = 'SAPSQL_ESCAPE_WITH_POOLTABLE'.
575       message i412(mo).
576       exit.
577     ELSE.
wp trace:
D *** ERROR => dySaveDataBindingValue: Abap-Field= >TEXT-SYS< not found [dypbdatab.c 510]
D *** ERROR => dySaveDataBindingEntry: dySaveDataBindingValue() Rc=-1 Reference= >TEXT-SYS< [dypbdatab.c 430]
D *** ERROR => dySaveDataBinding: dySaveDataBindingEntry() Rc= -1 Reference=>TEXT-SYS< [dypbdatab.c 137]
Y *** ERROR => dyPbSaveDataBindingForField: dySaveDataBinding() Rc= 1 [dypropbag.c 641]
Y *** ERROR => ... Dynpro-Field= >DISPLAY_SY_SUBRC_TEXT< [dypropbag.c 642]
Y *** ERROR => ... Dynpro= >SAPLSTPDA_CARRIER< >0700< [dypropbag.c 643]
D *** ERROR => dySaveDataBindingValue: Abap-Field= >TEXT-SYS< not found [dypbdatab.c 510]
D *** ERROR => dySaveDataBindingEntry: dySaveDataBindingValue() Rc=-1 Reference= >TEXT-SYS< [dypbdatab.c 430]
D *** ERROR => dySaveDataBinding: dySaveDataBindingEntry() Rc= -1 Reference=>TEXT-SYS< [dypbdatab.c 137]
Y *** ERROR => dyPbSaveDataBindingForField: dySaveDataBinding() Rc= 1 [dypropbag.c 641]
Y *** ERROR => ... Dynpro-Field= >DISPLAY_FREE_VAR_TEXT< [dypropbag.c 642]
Y *** ERROR => ... Dynpro= >SAPLSTPDA_CARRIER< >0700< [dypropbag.c 643]
I thank you in advance
If you require other information please request

Hi,
Under certain conditions, an Open SQL statement with range tables can be reformulated into a FOR ALL ENTRIES statement:
    DESCRIBE TABLE range_tab LINES lines.
    IF lines EQ 0.
      [SELECT for blank range_tab]
    ELSE.
      SELECT .. FOR ALL ENTRIES IN range_tab ..
      WHERE .. f EQ range_tab-LOW ...
      ENDSELECT.
    ENDF.
Since FOR ALL ENTRIES statements are automatically converted in accordance with the database restrictions, this solution is always met by means of a choice if the following requirements are fulfilled:
1. The statement operates on transparent tables, on database views or on a projection view on a transparent table.
2. The requirement on the range table is not negated. Moreover, the range table only contains entries with range_tab-SIGN = 'I'
and only one value ever occurs in the field range_tab OPTION.
This value is then used as an operator with operand range_tab-LOW or range_tab-HIGH.In the above example, case 'EQ range_tab-LOW' was the typical case.
3. Duplicates are removed from the result by FOR ALL ENTRIES.This must not falsify the desired result, that is, the previous Open SQL statement can be written as SELECT DISTINCT.
For the reformulation, if the range table is empty it must be handled in a different way:with FOR ALL ENTRIES, all the records would be selected here while this applies for the original query only if the WHERE clause consisted of the 'f IN range_tab' condition.
FOR ALL ENTRIES should also be used if the Open SQL statement contains several range tables.Then (probably) the most extensive of the range tables which fill the second condition is chosen as a FOR ALL ENTRIES table.
OR
What you could do in your code is,
prior to querying;
since your select options parameter is ultimately an internal range table,
1. split the select-option values into a group of say 3000 based on your limit,
2. run your query against each chunck of 3000 parameters,
3. then put together the results of each chunk.
For further reading, you might want to have a look at the Note# 13607 as the first suggestion is what I read from the note.

Update SQL statement in JDBC sender (system fields)

All,
Is it possible to update more than 1 field via the update sql statement ?
Also, is it possible to use system fields ?
Something like this
UPDATE database.table SET processed='Y', date = sy-datum
Regards, Michel

Hi
use sysdate as suggested above
Check your generated SQL query format is correct
At runtime you can find the genereated sql statements by doing configuraitons in Receiver JDBC adapter.
In the JDBC Receiver adapter you have the Advanced Properties .
Over there enter the following
left column logSQLStatement
right column true
To see the query created ..
Login to adapter monitoring ..select the relevant jdbc adapter.
Now when any message is processed by the jdbc adapter in adapter monitoring at that time you will see a message link. When you click on that link a new window will open. In that window if you click on page down you will get to see the sql statement generated by the jdbc adapter.

SQL Statement error

Hi All,
I was trying to execute a standard function module in which there is a sql statement where IN operator is used.
The no of choices given for that field by the IN operator is 1065.
The error displayed is the SQL statement generated from the SAP Open SQL statement violates a restriction imposed by the underlying database system of the ABAP system.
But when I am reducing the no of choices for that filed to 1000 the function modules executes smoothly.
Now from all these I can understand that there is some restriction imposed by underlying database for the no of choices by IN operator in select statements. I want a way to override that restriction.
The function module I was executing is 'K_PRCTR_VALUES_GET'. The error hits at line no 162 when the table t_selopt contains 1065 line items. Now this function modules gives us all the profit centers based on Profit-center group. So according to the current scenario for the respective profit-center group there are 1065 no. of profit centers. So there is no way that we can reduce the no. of profit centers for profit center group rather I have to find the other way out.
I will really appreciate the inputs. Please help me out in with this problem. Points will be awarded for good answer.
Thanks in advance.
Regards,
Piyush Kothari

Indeed there is a limit to the SELECT-OPTIONS / RANGES with the IN operator. But unfortunately there is no way bypass this limitation.
The only thing you can do (sort of workaround) is to call the function module several times for x-number of profit centers (if possible in your particular case).
It might be an option to talk to one of your basis administrators, perhaps they have another solution for this. But from ABAP side there is nothing you can do about it.

PHP - Multiple sql statements

I am using several different sql statements to be used
depending on which variables the user has chosen on the previous
page. Sometimes more than one variable will be selected
(checkboxes) so more than one of the sql statements will be run,
but I want all the results to go into the same $num variable.
I have tried this with the following scipt but it doesn't
work. How can I modify this to make it work???
Many Thanks
if ($buying_and_merchandising){
$query = "SELECT * FROM bwjobs WHERE location LIKE
'$location' AND description LIKE '$keyword' AND title LIKE
'$keyword' AND buying_and_merchandising =
'$buying_and_merchandising' AND visible = 'yes' ORDER BY
posted_date DESC";
if ($design){
$query2 = "SELECT * FROM bwjobs WHERE location LIKE
'$location' AND description LIKE '$keyword' AND title LIKE
'$keyword' AND design = '$design' AND visible = 'yes' ORDER BY
posted_date DESC";
if ($executive){
$query3 = "SELECT * FROM bwjobs WHERE location LIKE
'$location' AND description LIKE '$keyword' AND title LIKE
'$keyword' AND executive = '$executive' AND visible = 'yes' ORDER
BY posted_date DESC";
if ($human_resources){
$query4 = "SELECT * FROM bwjobs WHERE location LIKE
'$location' AND description LIKE '$keyword' AND title LIKE
'$keyword' AND human_resources = '$human_resources' AND visible =
'yes' ORDER BY posted_date DESC";
if ($manufacturing){
$query5 = "SELECT * FROM bwjobs WHERE location LIKE
'$location' AND description LIKE '$keyword' AND title LIKE
'$keyword' AND manufacturing = '$manufacturing' AND visible = 'yes'
ORDER BY posted_date DESC";
if ($retail_operations){
$query6 = "SELECT * FROM bwjobs WHERE location LIKE
'$location' AND description LIKE '$keyword' AND title LIKE
'$keyword' AND retail_operations = '$retail_operations' AND visible
= 'yes' ORDER BY posted_date DESC";
if ($sales_and_marketing){
$query7 = "SELECT * FROM bwjobs WHERE location LIKE
'$location' AND description LIKE '$keyword' AND title LIKE
'$keyword' AND sales_and_marketing = '$sales_and_marketing' AND
visible = 'yes' ORDER BY posted_date DESC";
if ($technical){
$query8 = "SELECT * FROM bwjobs WHERE location LIKE
'$location' AND description LIKE '$keyword' AND title LIKE
'$keyword' AND technical = '$technical' AND visible = 'yes' ORDER
BY posted_date DESC";
if ($visual_merchandising){
$query9 = "SELECT * FROM bwjobs WHERE location LIKE
'$location' AND description LIKE '$keyword' AND title LIKE
'$keyword' AND visual_merchandising = '$visual_merchandising' AND
visible = 'yes' ORDER BY posted_date DESC";
$result = @mysql_query ($query);
$result2 = @mysql_query ($query2);
$result3 = @mysql_query ($query3);
$result4 = @mysql_query ($query4);
$result5 = @mysql_query ($query5);
$result6 = @mysql_query ($query6);
$result7 = @mysql_query ($query7);
$result8 = @mysql_query ($query8);
$result9 = @mysql_query ($query9);
$num = mysql_num_rows ($result, $result2, $result3, $result4,
$result5, $result6, $result7, $result8, $result9);

You could just combine all into one query statement as
follows:
$query = "SELECT * FROM bwjobs WHERE location LIKE
'$location' AND description LIKE '$keyword' AND title LIKE
'$keyword' AND (buying_and_merchandising =
'$buying_and_merchandising' OR design = '$design' OR executive =
'$executive' OR human_resources = '$human_resources' OR
manufacturing = '$manufacturing' OR retail_operations =
'$retail_operations' OR sales_and_marketing =
'$sales_and_marketing' OR technical = '$technical' OR
visual_merchandising = '$visual_merchandising' ) AND visible =
'yes' ORDER BY posted_date DESC";

Query to find the the sql statements

Hi,
Oracle Version: 10.2.0.3
Can any one please help me how to find which sql statements generate the error "OPEN CURSORS EXCEED ".
Thanks & Regards,
Poorna Prasad.S

1) export the table, drop the table, create the table definition in the new
tablespace, and then import the data (imp ignore=y).
2) Create a new table in the new tablespace with the "CREATE TABLE x AS
SELECT * from y" command:
CREATE TABLE temp_name TABLESPACE new_tablespace AS SELECT * FROM real_table;
Then drop the original table and rename the temporary table as the original:
DROP TABLE real_table;
RENAME temp_name TO real_table;
3) Use the one mentioned in previous reply.
After #1 or #2 is done, be sure to recompile any procedures that may have been
invalidated by dropping the table.

Convert SQL statements to JAVA Code ?

Using SQL queries Can I convert SQL statements to JAVA Code ?
Edited by: user11238895 on Jun 7, 2009 10:54 PM
Edited by: user11238895 on Jun 7, 2009 11:12 PM

Me very new to Oracle.
can we convert SQL Queries to JAVA Code using simple built in SQL queries ?
Which SWISS SQL Tool does [Converting SQL Queries to JAVA Code]

How to generate synthetic rows (raw(16) guid cols) in one SQL statement?

We're populating a table containing two GUID columns:
create table object
( object_guid raw(16) primary key
, project_guid raw(16)
)All object GUIDs are unique (thus the PK), and each object belongs to a given project (should be a FK to some project table). We want N objects / rows, belonging to only 100 projects, i.e. 1% of the rows of the object table belong to the project #1, 1% to #2, etc...
Right now we're using about 25 lines of C++/OCI code to do that (one query doing a "select sys_guid() from dual", and using the generated GUIDs to do inserts into object), but I suspect it's possible to do this using a single SQL statement using mysterious connect by or some other Oracle SQL magic. (our OCI code does quite a few round-trips to do the equivalent).
Would anyone please demonstrate how to generate the rows as explained above, and possibly describe how it works for the non-initiated?
Thanks, --DD
PS: I'm sure it can be done in PL/SQL as well, but I'm interested in a SQL version if one's possible.

I've found two ways, both taking a few SQL statements, but somehow I think this ought to be possible without intermediary tables... I'm sure there's a better way.
#1drop   table project_tmp;
create table project_tmp
as select rownum pid, sys_guid() guid from dual
connect by level <= 100;
drop   table object_tmp;
create table object_tmp
as select mod(rownum, 100) + 1 pid, sys_guid() guid from dual
connect by level <= 1000;
drop   table object;
create table object
as select o.guid object_guid, p.guid project_guid
from object_tmp o, project_tmp p
where o.pid = p.pid;
drop table project_tmp;
drop table object_tmp;#2:drop table project;
create table project
as select mod(rownum, 100) + 1 prj_id, sys_guid() guid from dual
connect by level <= 100;
drop table object;
create table object
as select mod(rownum, 100) + 1 prj_id, sys_guid() object_guid from dual
connect by level <= 1000;
alter table object add project_guid raw(16);
update object o set o.project_guid = (select guid from project p where p.prj_id = o.prj_id);
drop table project;
alter table object drop column prj_id;Verification:select count(distinct project_guid) from object;
select project_guid, count(OBJECT_GUID) from object group by project_guid;

? in SQL Queries and not using prepared statements

Using EclipseLink 1.1.1
Prepared Statements are disabled
In our production server something went wrong and one of our Read Queries started erroring. A DatabaseException was thrown and we log the "getQuery.getSQLString()" statement. We found this query in the logs:
SELECT t1.ID, t1.NAME, t1.DESCRIPTION, t1.EXTREFID, t1.STYLESHEET, t1.DDSVNVERSION, t1.FIRSTNAME, t1.LASTNAME, t1.EMAILADDR, t1.PHONENUMBER, t1.ADDRESS, t1.ADDRESS2, t1.CITY, t1.STATE, t1.POSTALCODE, t1.COUNTRY, t1.ADMINACCTNAME, t1.HASDOCUMENTS, t1.HASTIMEDNOTIFICATIONS, t1.STATUS, t1.ENTRYDATE, t1.EVALEXPDATE, t1.LASTREMINDDATE, t1.FULLUSERS, t1.LIMUSERS, t1.REQUSERS, t1.ISENTERPRISE, t1.EXPDATE, t1.ISDISABLED, t1.DISABLEDDATE, t1.NEEDLICENSEAGREEMENT, t1.ISWARNINGDISABLED, t1.LOCALE, t1.TIMEZONE, t1.CURRENCY, t1.DOMAIN, t1.DOCUMENTSIZE, t1.EXTRADOCUMENTSTORAGE, t1.ONDEMANDOPTIONS, t1.SSOTYPE, t1.RESELLERID, t1.ACCOUNTREPID, t1.LASTUSAGEREPORTDATE, t1.NEXTUSAGEREPORTDATE, t1.USAGEREPORTATTEMPTS FROM T_SSOOPTIONS t0, T_CUSTOMERS t1 WHERE *((((t0.SSOENABLED = ?) AND (t1.SSOTYPE IN (?, ?))) AND (UPPER(t1.DOMAIN) = ?)) AND (t0.CUSTOMERID = t1.ID))*
Notice the values weren't entered into the where clause. We had to bounce the application to fix the problem. I've never seen this before. I've added more debugging statements to the code - so if this happens again in the future I'll have more information to report on. In the mean time I'm wondering if anyone else has every seen a problem of this nature.

Database error due to invalid SQL statement.
I don't have a stack, we were catching the exception and not printing the stack :(
Like I mentioned in my first post, I added more debugging code (e.printStackTrace()). I understand this is hard to track down without more information. I was just hoping you guys had seen something like this before and had any insight. Like I mentioned before: this is on our production server. I've never seen this type of error before. That particular server (we run in a cluster mode) had been up for several days and then started generating that error. IT bounced the node and everything went back to normal. We have been using toplink for about 5 years now and have never seen this problem, until August 3rd 2009. The only thing that has changed recently is our migration from toplink 10 to EclipseLink. I was wondering if anyone knows if anything had changed in EclipseLink/toplink 11 with the generation of SQL queries.
I'll keep looking. There is more debugging code in there now. Since the error was "Database error due to invalid SQL statement" this implies the SQL was generated, exited that part of the code and was sent to the db where it failed. I'm afraid the printStackTrace won't help if this error happens again.

Multiple Oracle SQL statements in one Add Command

I am creating a report that needs a bunch of processing(SQL DDL statements) before the final select statement is generated for the report.
I am connecting to Oracle database however Crystal Report only allows me to give one SQL statement in one Add Command.
Is there a way to give multiple statements in one "Add Command" for the report ?
Thanks.

you can add more than one "Add Command" in the same report, and you can also treat them as views or table, so you can link them to each others and so on
good luck

BEx queries - one or more SQL statements generated?

Similar Messages

Maybe you are looking for