Using the match/merge operator

Hi guys,
I have created a simple table with 4 columns:
EMPLOYEES
- employee_id
- emp_firstname
- emp_lastname
- emp_telephone
I have placed 5 entries in the table:
11, jane, doe, 80980980
12, william, gates,6876876868
13, john, anderson, 545646546
14, michelle, millo, 340580334
15, john, anderson, 545646546
Row 13 and 15 are duplicates, except for the employee_id. I'm trying to use the match-merge operator to check for matches on emp_firstname, emp_lastname and emp_telephone. I would like to merge the rows matching on firstname, lastname and telephone number, and would like to keep the lowest employee_id as PK.
I have tried to use the match-merge operator with the following settings;
Input connections:
- employee_id
- emp_firstname
- emp_lastname
- emp_telephone
Merge output
- employee_id
- emp_firstname
- emp_lastname
- emp_telephone
Xref output
none
Match bins:
- emp_firstname
- emp_lastname
- emp_telephone
Match rules:
MA_0, all_match
Merge rules:
emp_firstname: all match
emp_lastname: all match
emp_telephone: all match
When I debug the match merge operator I get " 4 in, 4 out", but I expected to have "5 in, 4 out". Am I using this match/merge operator the right way?
thx for you help!
michiel
null

Hi Robert,
I tried on 3 database installations (all 10.2). there is no parameter DB_BLOCK_BUFFERS set. And i do not assume that there is a reasn for increasing the db_cache_size - means: why i did not get any error for that presumed case?
btw: can you explain why i should increase db_cache_size?
thanks,
Andreas

Similar Messages

Pl/sql procedure for match-merge operator

I have used the match-merge operator in OWB. I want to implement the same match-merge operation using a pl/sql procedure. Please help.

Yes sure. Lets say, we have a table tab1, which has the following structure and data:
First_name Last_name Address City
JACK ANDREWS XYZ NY
ANDREWS JACK NY
Now, i want to write a procedure which matches and merges the above two records, into a single record.The matching is done with the help of the first_name and last_name columns.

Use the findByMultipleParameters bo operation in application service

Hi
I am trying to use the default findByMultipleParameters operation of a business object in an application service.
I want to use the application service exposed as a webservice within webdynpro.
But how can I pass all the parameters i a way, that it is not necessary to fill all fields in webdynpro? Is there any example code?
Any help would be appreciated.
Bernd

Hi Ketan,
You can implement findAll or findById in application service. First while defining the operations set the output parameter to be a structure which contains that BO(You create yourself a complex structure for it-there you can set cardinality).
Then in the application service under that operation using the service Bean of the BO access the Bo's corresponding methods(either findById or findAll). Get that output and set it in the output parmeter and return it.
public com.caf.types.FindAllAmendmentsResMsg findAllAmendments(
               @javax.jws.WebParam(name="inputParameter1")
               com.hcl.xapps.ipr.caf.types.FindAllAmendmentsReqMsg inputParameter1) {
          FindAllAmendmentsResMsg response = new FindAllAmendmentsResMsg();
          try {
               List<Amendment> amendmentList = new ArrayList<Amendment>();
               amendmentList = (List<Amendment>) this.getAmendmentService().findAll();
               response.setAmendmentList(amendmentList);
          } catch (CAFFindException cfe) {
          return response;
The above is an example for findAll.
Hope it helps you
Regards,
Srinivasan Subbiah

How do I use the Match Pattern Function to exclude only 0.000?

Hi,
I'm trying to use the mattch pattern function to find the first string in a table thats is >0. My table looks like:
1,0.000000,0.000 %2007/01/13 00:16:19 196281
1,0.000000,0.000 %2007/01/13 00:16:22 196282
1,0.831262,0.000 %2007/01/13 00:17:20 196375
2,0.811154,0.000 %2007/01/13 00:17:20 196375
If I us the paremeter "1,[~0]" It doesn't find the line 1,0.831262,0.000... which is the one that I want. I also tried :1,[0-9].+[~0] and that didn't work either. the problem is that the first digit after to 1, isn't allways going to go from 0 to 0.0 sometimes it might go from 0 to 2.??.
Thanks for the help
Matt

"Matt361" <[email protected]> wrote in message news:[email protected]..
Hi,
 
I'm trying to use the mattch pattern function to find the first string in a table thats is >0. My table looks like:
1,0.000000,0.000 %2007/01/13 00:16:19 196281
1,0.000000,0.000 %2007/01/13 00:16:22 196282
1,0.831262,0.000 %2007/01/13 00:17:20 196375
2,0.811154,0.000 %2007/01/13 00:17:20 196375
If I us the paremeter "1,[~0]" It doesn't find the line 1,0.831262,0.000... which is the one that I want. I also tried :1,[0-9].+[~0] and that didn't work either. the problem is that the first digit after to 1, isn't allways going to go from 0 to 0.0 sometimes it might go from 0 to 2.??.
 
Thanks for the help
Matt
 
Hi,
1,[~0] matches a "1" a "," and then any character that is not "0".
1,[0-9].+[~0] matches a "1" a "," and then any character that is "0-9", any number of anything!! (1 or more) and then anything that is not "0". Note that you have to escape a . to match a ".". Like this "\.".
There is no way to check if there is anything other then a "0" in the match, from within the match pattern function. So I think you won't be able to find a pattern that does the trick.
Why not use a whileloop to find the first item? Or use the Spreadsheet String To Array, and then compare the desired row or column with the string "0.000000"? (Or replace all ,0.000000, by a string like ",NULL,", then match the pattern?)
In LabVIEW 8 there is a new Match Regular Expression function. Haven't tried it, but it should be much more powerfull then the Match Pattern function. But also much more complex.
Regards,
Wiebe.

I am trying to use the T. S. operator interface I need to add pass fail and error

Hello:
I am using Labview 6.1 and Test Stand 2.0
I have made some modification to a copy of the T.S. Labview operator interface. How do I get the pass/fail or error from test stand to be pass to LabView so I can tell the operator that the unit pass or fail the test. I do not want to use the Test UUTs. I want this to be very low operator response.
Message Edited by Toro on 02-28-2006 01:52 PM

Not sure if this is available in the earlier versions of TestStand, but I think so.
You could post a UI Message, from your execution entry point in the process model, to your User Interface indicating the pass/fail/error status when a test run is complete. You will need to handle the UI Message in your Operator Interface. See TestStand II Customisation course notes for handling UI Messages.
The data passed in the UI Message can be anything you like, the UI Message handler in the User Interface will have to process the message and update the display as required.
Hope this helps
Steve
Message Edited by SercoSteve on 03-01-2006 10:02 AM
There are 10 types of people in the world that understand binary, those that do and those that don't.

Why do I receive "4.3.2 connection rate limit exceeded" error message using the mail merge extension and what can be done about it?

I am using TB 31.3.0 with Mail Merge 3.9.1. I routinely send an email to 435 members of an volunteer emergency responders group that I coordinate. I do so using a .csv list with mail merge. While there were no problems in the past, more recently the mail merge function will hang after sending a varied number of messages successfully and I get the following error message:
"An error occurred sending mail. The mail server sent an incorrect greeting: 4.3.2 connection rate exceeded.."
There is an "OK" button in the error message pop up that can be clicked to resume the mail merge process. It would appear that if I click the "OK" button immediately whenever the error message is received I must do so frequently and some members do not receive their email. If I delay clicking the "OK" button, I found that I only needed to click the "OK" button twice to send to all the members.
An on line search suggests this is the result of some sort of throttling by my ISP, Sonic.net. There is also this comment: "If you are receiving this error, you are likely using mailing list software which cannot decipher the temporary fail codes. If so, you will need to set your software to slow down its delivery rate and/or reduce the number of active connections per remote host."
I am not super technical. Is it realistic to think that I can tweak mail merge to that I do not have to babysit my email to this group?

then I suggest you send using a mail provider that is not actively trying to block your outgoing mail. Or use a yahoo / google groups mailing list feature so you only send a single mail.
Or use a free account from the likes of http://www.ymlp.com/ who limit free account mailing lists to 1000 subscribers. ( Googled them this morning)

Using the variable ? operator in an XPath Expression

I'm trying to using the variable operator (?) in an Oracle prepared statement used in a query where clause. Here is the query:
select xt.APPLICATION_NAME, xt.VERSION, xt.EVENT_TYPE, xt.SENDING_SITE, xt.RECEIVING_SITE, xt.EVENT_ID
      from AUDITED_EVENT_XML_MIN e,
         XMLTable(XMLNAMESPACES('http://gov/va/med/datasharing/audit/endpoint/audit' AS "xae"),
                '/xae:auditable-event'
                PASSING e.xml_event_content
                COLUMNS
                APPLICATION_NAME VARCHAR2(255) PATH 'xae:application-name',
                VERSION          VARCHAR2(255) PATH 'xae:version',
                EVENT_TYPE       VARCHAR2(255) PATH 'xae:event-type',
                SENDING_SITE     VARCHAR2(255) PATH 'xae:xml-event-content/*/MSH/MSH.4/HD.2',
                RECEIVING_SITE   VARCHAR2(255) PATH 'xae:xml-event-content/*/MSH/MSH.6/HD.2',
                EVENT_ID         VARCHAR2(255) PATH 'xae:xml-event-content/*/MSH/MSH.10') xt
            where existsNode(E.XML_EVENT_CONTENT, '/xae:auditable-event[xae:event-type=?]','xmlns:xae="http://gov/va/med/datasharing/audit/endpoint/audit"') = 1This code works when the ? is replaced with "aValue". I need to use the ? so that my Java client can pass a variable value into the query.
Is this a supportable feature or am I doing something wrong?
I didn't see any examples in the Oracle XML DB Developers Guide where there was a ? in an XPath expression.
I also tried "?" with the same issue...
Here is the version info:
SQL*Plus: Release 11.1.0.6.0 - Production on Mon Sep 10 18:41:55 2012
Copyright (c) 1982, 2007, Oracle. All rights reserved.
Connected to:
Oracle Database 11g Enterprise Edition Release 11.1.0.7.0 - 64bit Production
With the Partitioning, OLAP, Data Mining and Real Application Testing optionsAny assistance would be greatly appreciated.
Thanks in advance...

I incorporated this suggestion and it worked very well. Thanks!
Now I have converted my solution to use a structured index. As such, I want to write the where clauses in terms of column names rather than XPath expressions.
I hoped it was as simple as changing the code that generates the XPath expression to generate a SQL expression. After making that change, I'm now getting this error:
Caused by: java.sql.SQLSyntaxErrorException: ORA-00920: invalid relational operator
     at oracle.jdbc.driver.SQLStateMapping.newSQLException(SQLStateMapping.java:91)
     at oracle.jdbc.driver.DatabaseError.newSQLException(DatabaseError.java:112)
     at oracle.jdbc.driver.DatabaseError.throwSqlException(DatabaseError.java:173)
     at oracle.jdbc.driver.T2CConnection.checkError(T2CConnection.java:686)
     at oracle.jdbc.driver.T2CConnection.checkError(T2CConnection.java:617)
     at oracle.jdbc.driver.T2CPreparedStatement.executeForDescribe(T2CPreparedStatement.java:559)
     at oracle.jdbc.driver.OracleStatement.executeMaybeDescribe(OracleStatement.java:1077)
     at oracle.jdbc.driver.OracleStatement.doExecuteWithTimeout(OracleStatement.java:1186)
     at oracle.jdbc.driver.OraclePreparedStatement.executeInternal(OraclePreparedStatement.java:3381)
     at oracle.jdbc.driver.OraclePreparedStatement.executeQuery(OraclePreparedStatement.java:3425)
     at oracle.jdbc.driver.OraclePreparedStatementWrapper.executeQuery(OraclePreparedStatementWrapper.java:1490)
     at gov.va.med.datasharing.audit.dao.jdbc.runner.MapQueryRunner.run(MapQueryRunner.java:47)
     at gov.va.med.datasharing.audit.web.dao.jdbc.AuditableEventsQueryDAOImpl.getAuditableEventsByDate(AuditableEventsQueryDAOImpl.java:150)It seemed to me that if it worked for the XPath expression that it should also work for the SQL expression.
The static SQL is:
SELECT PK_ID, EVENT_TYPE, OUTCOME, SENDING_SITE, RECEIVING_SITE, CREATED_TIME, EVENT_ID, PATIENT_ID FROM
(SELECT rownum as rn, PK_ID, EVENT_TYPE, OUTCOME, SENDING_SITE, RECEIVING_SITE, CREATED_TIME, EVENT_ID, PATIENT_ID FROM
     (SELECT aet.AUDITABLE_EVENT_XML_PK_ID AS PK_ID, xt.EVENT_TYPE, xt.OUTCOME, xt.SENDING_SITE, xt.RECEIVING_SITE, aet.CREATED_TIME as CREATED_TIME, xt.EVENT_ID, xt.PATIENT_ID
        FROM AUDITABLE_EVENT_XML aet,
          XMLTable(XMLNAMESPACES('http://gov/va/med/datasharing/audit/endpoint/audit' AS "xae"),
            '/xae:auditable-event'
            PASSING aet.xml_event_content
            COLUMNS
            EVENT_TYPE       VARCHAR2(255) PATH 'xae:event-type',
            SENDING_SITE     VARCHAR2(255) PATH 'xae:sending-site',
            RECEIVING_SITE   VARCHAR2(255) PATH 'xae:receiving-site',
            EVENT_ID         VARCHAR2(255) PATH 'xae:event-id',
            PATIENT_ID           VARCHAR2(255) PATH 'xae:patient-id',
            OUTCOME                VARCHAR2(255) PATH 'xae:outcome') xt
        WHERE aet.CREATED_TIME BETWEEN ? AND ?
             AND ?
      ) ORDER BY CREATED_TIME desc
    ) WHERE rn > ? and rn <= ?So I'm thinking that Oracle must perform strong checking in SQL, not allowing the '=' operator to be substituted. If I run this query in Toad, replacing the ? with the SQL, it runs fine...
Can someone tell me whether or not this is a legal substitution?
The SQL that runs in Toad is:
SELECT PK_ID, EVENT_TYPE, OUTCOME, SENDING_SITE, RECEIVING_SITE, CREATED_TIME, EVENT_ID, PATIENT_ID FROM
(SELECT rownum as rn, PK_ID, EVENT_TYPE, OUTCOME, SENDING_SITE, RECEIVING_SITE, CREATED_TIME, EVENT_ID, PATIENT_ID FROM
     (SELECT aet.AUDITABLE_EVENT_XML_PK_ID AS PK_ID, xt.EVENT_TYPE, xt.OUTCOME, xt.SENDING_SITE, xt.RECEIVING_SITE, aet.CREATED_TIME as CREATED_TIME, xt.EVENT_ID, xt.PATIENT_ID
        FROM AUDITABLE_EVENT_XML aet,
          XMLTable(XMLNAMESPACES('http://gov/va/med/datasharing/audit/endpoint/audit' AS "xae"),
            '/xae:auditable-event'
            PASSING aet.xml_event_content
            COLUMNS
            EVENT_TYPE       VARCHAR2(255) PATH 'xae:event-type',
            SENDING_SITE     VARCHAR2(255) PATH 'xae:sending-site',
            RECEIVING_SITE   VARCHAR2(255) PATH 'xae:receiving-site',
            EVENT_ID         VARCHAR2(255) PATH 'xae:event-id',
            PATIENT_ID         VARCHAR2(255) PATH 'xae:patient-id',
            OUTCOME             VARCHAR2(255) PATH 'xae:outcome') xt
        WHERE aet.CREATED_TIME BETWEEN TO_DATE ('9/19/2011 12:00:01 AM','MM/DD/YYYY HH:MI:SS PM')
                                     AND TO_DATE ('12/20/2012 12:00:00 AM','MM/DD/YYYY HH:MI:SS PM')
            AND (OUTCOME=0 or OUTCOME=1)
      ) ORDER BY CREATED_TIME desc
    ) WHERE rn > 0 and rn <= 5;Any assistance will be appreciated,
Thanks in advance.

Why are my LabView Dialog Boxes no longer modal when using the Example LabVIEW Operator Interfaces?

Hello,
I have a step in my sequence that pops up a modal LabVIEW dialog box. My LabVIEW vi uses the TestStand - Start Modal Dialog.vi and TestStand - Stop Modal Dialog.vi as described in the example and it works as advertised when I'm using the TestStand Development environment however when I use the sample Operator Interfaces distributed with TestStand the Dialog boxes lose focus. I have the VI Properties->Window Appearance set to Dialog and have followed all of the instructions I could find for making a dialog modal. I've also reviewed all the documentation I could get my hands on including previous posts on this board and didn't see anything obvious.
Does anyone have any suggestions as to what I may be missing? Has anyone else experienced this problem and come up with the solution?
I am using LabVIEW 8.0 along with TestStand 3.5.
Thanks,
RG

Hi RG,
Running the example modal dialog in the teststand3.5 examples folder and using labVIEW8.0, I notice that when the Dialog is displayed its not active, but it is modal to the OI. Clicking on the Dialog panel makes it active. If you try to click on the OI panel behind, you can not. Is this what you are seeing?
I am using the Full labVIEW IO as supplied with TestStand and the example VI in the modal dialog folder is as supplied.
The LabVIEW Adapter is still configured to use the active labVIEW developement environment.
If I run the same example, only this time using LabVIEW7.1, the Dialog panel is active.
There is a patch to take it to labVIEW8.01, at this moment in time I am not sure what the update fixes.
There does seem to be a problem with TestStand3.5 (engine) and LabVIEW 8.0.
Regards
Ray Farmer
Regards
Ray Farmer

Using the "Data Merge Tool"

Hi everyone,
I have a 2 questions about the data merge tool. I understand how to use a CSV file to replace dynamically any text I desire.
1. Can I use this same tool to replace pictures? Say, like if I create baseball cards for an entire team.
2. If I use it for creating business cards, but I only need to create ONE card at a time... I like how the "Records per document" option that optimizes the page size to fit as many as possible. Ok, to the question; How can I print one record at a time, multiple times (say 20 per page) and include art for the other side of the page (double side print, that does not contain variable data, just a background)? I just need to make sure the tool knows how to line the art up, so that when it is cut, the front and back art is aligned properly.
Thank you for your time.
Rafa.

Hi Lori
Many thanks the binding was set to 'none'.
Regards
Dave

HT4060 Even though the iPad is not charging, will it still use the USB for operating power?

I was wondering if there is enough power for operating (to conserve battery), even though there is not enough power for charging?

The quickest way (and really the only way) to charge your iPad is with the included 10W USB Power Adapter. iPad will also charge, although more slowly, when attached to a computer with a high-power USB port (many recent Mac computers) or with an iPhone Power Adapter (5W). When attached to a computer via a standard USB port (most PCs or older Mac computers) iPad will charge very slowly (but iPad indicates not charging). Make sure your computer is on while charging iPad via USB. If iPad is connected to a computer that’s turned off or is in sleep or standby mode, the iPad battery will continue to drain.
Apple recommends that once a month you let the iPad fully discharge & then recharge to 100%.
How to Calibrate Your Mac, iPhone, or iPad Battery
http://www.macblend.com/how-to-calibrate-your-mac-iphone-or-ipad-battery/
At this link http://www.tomshardware.com/reviews/galaxy-tab-android-tablet,3014-11.html , tests show that the iPad 2 battery (25 watt-hours) will charge to 90% in 3 hours 1 minute. It will charge to 100% in 4 hours 2 minutes. The new iPad has a larger capacity battery (42 watt-hours), so using the 10W charger will obviously take longer. If you are using your iPad while charging, it will take even longer. It's best to turn your new iPad OFF and charge over night. Also look at The iPad's charging challenge explained http://www.macworld.com/article/1150356/ipadcharging.html
Also, if you have a 3rd generation iPad, look at
Apple: iPad Battery Nothing to Get Charged Up About
http://allthingsd.com/20120327/apple-ipad-battery-nothing-to-get-charged-up-abou t/
Apple Explains New iPad's Continued Charging Beyond 100% Battery Level
http://www.macrumors.com/2012/03/27/apple-explains-new-ipads-continued-charging- beyond-100-battery-level/
New iPad Takes Much Longer to Charge Than iPad 2
http://www.iphonehacks.com/2012/03/new-ipad-takes-much-longer-to-charge-than-ipa d-2.html
Apple Batteries - iPad http://www.apple.com/batteries/ipad.html
Extend iPad Battery Life (Look at pjl123 comment)
https://discussions.apple.com/thread/3921324?tstart=30
New iPad Slow to Recharge, Barely Charges During Use
http://www.pcworld.com/article/252326/new_ipad_slow_to_recharge_barely_charges_d uring_use.html
Tips About Charging for New iPad 3
http://goodscool-electronics.blogspot.com/2012/04/tips-about-charging-for-new-ip ad-3.html
Prolong battery lifespan for iPad / iPad 2 / iPad 3: charging tips
http://thehowto.wikidot.com/prolong-battery-lifespan-for-ipad
In rare instances when using the Camera Connection Kit, you may notice that iPad does not charge after using the Camera Connection Kit. Disconnecting and reconnecting the iPad from the charger will resolve this issue.
 Cheers, Tom

I have an Imac using the old Classic operating system. The printer I was using died and I need a new printer that will work on the OS 9.1/9.2 operating system. Does the install disk that comes with the printer allow the computer to communicate with the

My printer died and I need a printer that is compatable with OS 9.1/9.2. Some printers say they are compatible with OS 9.1/9.2 but do they need a driver or is it included with the install disk that comes with the printer?

As long as the printer is accurately described as being compatible with OS 9.1/9.2, if it needs to have software (such as drivers) installed that software should either be included (usual) or downloadable from the net(usually from the manufacyrer's website).
Not all non-Apple printers need drivers installed. Some, such as many Postscript ones that connect via ethernet, are quite comfortable using the LaserWriter 8 printware that's part of OS 9. Some of those, though, may need to have a model-specific PPD (Postscript Printer Description) installed, a small document that informs the printware of the capabilities of the printer.

URGENT:Problem in a mapping with 8 tables in JOIN and using the DEDUP op.

I have an urgent problem with a mapping: I must load data from 8 source tables (joined togheter) in a target table.
Some data from 1 of the 8 tables have to be deduplicated, so I created a sort of staging table where I inserted the "cleaned" data, I mean, the deduplicated ones.
I made it to make all the process faster.
Then, I joined the 8 tabled, writing the join conditions in the operator properties, and connected the outputs into the fields of the target table.
But..it does not work because this kind of mapping create a cartesian product.
Then, I tried again with another mapping builted up in this way: after the joiner operator, I used the Match-Merge Operator and I load all data into a staging table exactly alike the target one, except for the PK (because it is a sequence). Then, I load the data from this staging table into the target one, and, of course, I connect to the target table also the sequence (the primary key). The first loading works fine (and load all the data as I expected).
For the next loadings,I scheduled a pre-mapping process that truncate the staging table and re-load the new data.
But..it does not work. I mean, It doesn't update the data previously loaded (or inser the new ones), but only insert the data, not considering the PK.
So, my questions are as follow:
1) Why loading the data directly from the joiner operator into the fact table doesn't work? Why does it generate a cartesian product??
2) The "escamotage" to use the Match-Merge operator is correct? I have to admit that I didn't understand very well the behaviour of this operator...
3) And, most of all, HOW CAN I LOAD MY DATA? I cannot find a way out....

First of all, thanks for the answer!
Yes, I inserted the proper join condition and in fact I saw that when WB generates the script it considers my join but, instead of using the fields in a single select statement, it builts up many sub-selects. Furthermore, it seems as it doesn't evaluate properly the fields coming from the source tables where I inserted the deduplicated data...I mean, the problems seems not to be the join condition, but the data not correctly deduplicated..

Use of Sequence in Match-Merge

Dear All,
I have a source table containing name and address data. I want to do householding, which means I need to split the data using a match-merge operator. The address data is the merged output, and the name data is the cross-reference output. All well and good so far.
I want eventually to load the outputs into two levels of a Customer dimension (although for testing right now, I'm just using two ordinary target tables). I know that I will need a 'business key' to link the merged and the cross-reference outputs (as the two dimensional levels). I decide that I shall create a column in both outputs called HH_SEQ, and load this Household Sequence number from a Sequence (which I prepared earlier).
In Merge Rules, for the Attribute HH_SEQ, I use a Rule Type of Sequence, and select the appropriate Sequence from the drop-down list box. I map this attribute in the MERGE outgroup to the appropriate column in the target merged table, and MM_HH_SEQ in the XREF outgroup to the appropriate column in the target cross-reference table.
Here is the issue: to get HH_SEQ to be available as a potential Output Attribute in either Merge Output or Cross-Reference Output, it must first be present as a Source Attribute. Which means I had to add it to Input Attributes first. But of course it is not on my input source table! And thus I get a "VLD-1000 Input parameter INGRP1.HH_SEQ does not have input" warning.
Either I'm just not understanding how to use the Sequence rule type properly (the documentation is pretty atrocious, mind), or there is some technique required here that is just not apparent from several days of experimentation.
All help gratefully received!
Regards,
Donna
p.s. it does actually compile and run just fine!

Hi Donna
I think you are using it in the correct way! I've seen other examples like this from the MM creators.
The warning can be ignored, I think it is there to highlight a potential issue if it was being used for a scenario other than what you are doing.
Cheers
David

10GR2 - Match Merge - How can I Rank a list of values?

Hi
I am trying to use 10GR2 to take a set of multiple records and merge them into a single record. e.g.
ID Value
1 val1
1 val2
etc
I want to merge these into one record and give val1 priority over val2 etc. So the output record for ID 1 will have a value = val1.
I am trying to use a match-merge operator using rank, but am not sure how to use the Rank Rule Editor.
Can anyone please help with how I format the expression to rank these values? Or if this is the correct way to do this in 10GR2?
Thanks
GB

You can use this approach:
http://www.oracle.com/technetwork/developer-tools/jdev/listbindingvalue-088449.html
Or this:
https://blogs.oracle.com/shay/entry/adding_lov_to_non_database

How to perform merge operation using Data services designer?

I am new here and new to Data Services Desinger as well.
And currently, I want to do a merge operation using query transform or table comparison transform. But sadly, neither one of them is giving me what I want.
So, wonder if you could help clarify if either one of them could achieve MERGE operation between and source and target tables?
if positive, then how?
PS:
by MERGE, i mean, if data exists in target table, then only do update, else, do insert. same with the standard MERGE operation of oracle.
If you need any more info, let me know!
Thanks in advance.

You can do this.
You don't need the Tabe_Comparison transform actually.
Just query the data you need and output the flow from your query to your target table.
Then, you'll have to check the "Auto correct load" option: this will insert new lines and updates old ones. One checked, you can ckeck the "Auto correct load" checkbox.
By default, BODS will know a line already exists based uppon the primary key columns of your target table. If you want to change the columns used to determine if a line should be inserted or updated, use the "Use input key" checkbox (and define some PK column in your query).
Hope this helps
Guillaume

Using the match/merge operator

Similar Messages

Maybe you are looking for