XML contains some records that must be inserted and some updates

Hello

I get an XML that contains documents that I need to insert into a table of ABC.

It's some of the files in the xml file are already present in the table ABC (that's why I get unique key constraint error). I update the records instead of inserting it.

How to differentiate records of new records and create a separate xml and update in the ABC table?

Hello

Use a MERGE statement, whose source is the result of a XMLTable.

Something like that, in pseudocode:

MERGE INTO target_table t
USING (
  SELECT x.pk_id, x.col1, x.col2, ...
  FROM XMLTable(
         '/root/record'
         passing my_xml_doc
         columns pk_id number       path 'ID'
               , col1  varchar2(30) path 'COL1'
               , col2  varchar2(30) path 'COL2'
               , ...
       ) x
) src
ON ( t.pk_id = src.pk_id )
WHEN MATCHED THEN UPDATE
 SET t.col1 = src.col1
   , t.col2 = src.col2
   , ...
WHEN NOT MATCHED THEN INSERT
 (pk_id, col1, col2, ...)
 VALUES ( src.pk_id, src.col1, src.col2, ...)
;

Tags: Oracle Development

Similar Questions

Help with code to insert, and then update the form table.
I have a form where a user can add a new record in a table. Here is the procedure for recording button that works very well. But I call another procedure for updating the table with an "if" statement "The SQL code that I use works fine to get the data I'm looking for when I run it in SQL, but the 'if' statement updates all records in the table to ' n ' when some should be"Y". I am new to Oracle forms development (3 months) so it's probably very simple, but I'm stuck and need to get this finished form. Any help is greatly appreciated.

SAVE BUTTON TRIGGER PROCEDURE:*.

PROCEDURE SAVE_MM_1099_ADD IS

MM1099_NO NUMBER (10);
NUMBER OF THE YEAR (4);
NUMBER OF NOTICE_SEQ_NO (10);
NOTICE_NAME VARCHAR2 (30);
TIN_FROM_DB VARCHAR2 (23);
TSDI_DB VARCHAR2 (1);
IRS_NTYPE VARCHAR2 (10);
IRS_NDATE VARCHAR2 (10);
BU_WH VARCHAR2 (1);
NO_TIN_NAME VARCHAR2 (1);
DUP_IRS_NOTICE VARCHAR2 (1);
G1_IRS_NOTICE_3YR_2BS VARCHAR2 (1);

B_SAVE_ADD boolean;

Start

IF MM1099_NO IS NOT NULL THEN

insert into MM_1099
(
MM1099_NO,
YEAR,
NOTICE_SEQ_NO,
NOTICE_NAME,
TIN_FROM_DB,
TSDI_DB,
IRS_NTYPE,
IRS_NDATE,
BU_WH,
NO_TIN_NAME,
DUP_IRS_NOTICE,
G1_IRS_NOTICE_3YR_2BS
)
values
(
MM1099_NO,
YEAR,
NOTICE_SEQ_NO,
NOTICE_NAME,
TIN_FROM_DB,
TSDI_DB,
IRS_NTYPE,
IRS_NDATE,
BU_WH,
NO_TIN_NAME,
DUP_IRS_NOTICE,
G1_IRS_NOTICE_3YR_2BS
);

: system.message_level: = '5';
Commit_FORM;
: system.message_level: = '0';

END IF;

UPDATE_NUMOCCURTIN;

END;

PROCEDURE CALLED:*.

PROCEDURE UPDATE_NUMOCCURTIN IS

CURSOR c1 (NUMBER NUMOCCURTIN) IS
SELECT THE YEAR, TIN_FROM_DB, COUNT (TIN_FROM_DB) AS NUMOCCURTIN
OF MM_1099
TIN_FROM_DB GROUP, YEAR
SEEN (COUNT (TIN_FROM_DB) > 0);

CR1 c1% ROWTYPE;
number of v_recs;
NUMBER OF NUMOCCURTIN;

BEGIN
v_recs: = 0;

OPEN c1 (NUMOCCURTIN);

loop

Fetch c1 into cr1;
When the % notfound c1 or c1% rowcount output > v_recs;

IF cr1. NUMOCCURTIN > 1 THEN

UPDATE MM_1099
SET DUP_IRS_NOTICE = 'Y ';

ON THE OTHER

UPDATE MM_1099
SET DUP_IRS_NOTICE = 'N';

: system.message_level: = '5';
commit;
: system.message_level: = '0';

END IF;

v_recs: = v_recs + 1;

END LOOP;

CLOSE c1;

END;

For any help or suggestion will be greatly appreciated.
IF cr1. NUMOCCURTIN > 1 THEN

UPDATE MM_1099
SET DUP_IRS_NOTICE = 'Y ';

ON THE OTHER

UPDATE MM_1099
SET DUP_IRS_NOTICE = 'N';

: system.message_level: = '5';
commit;
: system.message_level: = '0';

END IF;

There is no WHERE condition in your update, so always update you all THE RECORDS in your table. I guess it must be something like:
```
UPDATE MM_1099 SET
  DUP_IRS_NOTICE = 'Y';
 WHERE TIN_FROM_DB=cr1.TIN_FROM_DB
   AND YEAR=cr1.YEAR
```
Some general order questions about your code:
Where your variables for your insert statement are filled? looking at your code they will always be NULL.
Be careful when using the variable exactly as the names of column names. This may cause unexpected behavior (for example when it is used in the UPDATE statements).
Why engage in the other branch, but not in the branch so?

Find a record that does not exist and displays a message

When I run the following query why used, it displays the word 'FAILURE '.

SELECT 
  CASE 
    WHEN table_name IN('DBA_OBJECTS') THEN 'PASS' 
    ELSE 'FAIL' 
  END TABLE_EXISTS 
 FROM dictionary 
WHERE table_name = 'DBA_OBJECT'
;

Ben

Hey Benton,

The query filter (WHERE table_name = 'DBA_OBJECT') cause zero record is returned.

To trick you can use aggregate function that always return the single record even if the record is found or not. If no record is found, NULL is returned.

SQL> SELECT CASE WHEN MAX(table_name) IN('DBA_OBJECTS') THEN 'PASS' ELSE 'FAIL' END TABLE_EXISTS -- OR NVL2(MAX(table_name),'PASS','FAIL') TABLE_EXISTS
  2  FROM dictionary
  3  WHERE table_name = 'DBA_OBJECT'
  4  AND ROWNUM = 1
  5  /

TABLE_EXISTS
------------
FAIL

{code}

Insert and update server behaviors in the same form

I have a php form that is inserted into a table and then update another table. I have the form of inserting the record first and then I want to update the other table. I'm starting to learn and understand PHP, but I don't know how to have the update function to run after that it inserts the records. I guess that's something like if you insert this disc, then update this record and then go to the redirect page. I know this goes to redirect page after it inserts the record, but I don't know how to tell him not to go to the $insertGoTo. I want him to run the following update function and then go to the $insertGoTo and not the $updateGoTo. Thanks in advance for any help.

if ((isset($_POST["MM_insert"])) && ($_POST["MM_insert"] == "form1")) {
  $insertSQL = sprintf("INSERT INTO employee (CompId, empfirstname, empmiddleint, emplastname, EmpType, EmpStatus) VALUES (%s, %s, %s, %s, %s, %s)",
                       GetSQLValueString($_POST['CompId'], "int"),
                       GetSQLValueString($_POST['empfirstname'], "text"),
                       GetSQLValueString($_POST['empmiddleint'], "text"),
                       GetSQLValueString($_POST['emplastname'], "text"),
                       GetSQLValueString($_POST['EmpType'], "text"),
                       GetSQLValueString($_POST['EmpStatus'], "text"));
  mysql_select_db($database_dotweb, $dotweb);
  $Result1 = mysql_query($insertSQL, $dotweb) or die(mysql_error());
  $insertGoTo = "Roster.php?CompId=" . $_POST['CompId'] . "";
  if (isset($_SERVER['QUERY_STRING'])) {
    $insertGoTo .= (strpos($insertGoTo, '?')) ? "&" : "?";
    $insertGoTo .= $_SERVER['QUERY_STRING'];
  }
  header(sprintf("Location: %s", $insertGoTo));
}
if ((isset($_POST["MM_update"])) && ($_POST["MM_update"] == "form1")) {
  $updateSQL = sprintf("UPDATE company SET CompanyType=%s WHERE CompId=%s",
                       GetSQLValueString($_POST['CompanyType'], "text"),
                       GetSQLValueString($_POST['CompId'], "int"));
  mysql_select_db($database_dotweb, $dotweb);
  $Result1 = mysql_query($updateSQL, $dotweb) or die(mysql_error());
  $updateGoTo = "purchases.php";
  if (isset($_SERVER['QUERY_STRING'])) {
    $updateGoTo .= (strpos($updateGoTo, '?')) ? "&" : "?";
    $updateGoTo .= $_SERVER['QUERY_STRING'];
  }
  header(sprintf("Location: %s", $updateGoTo));
  
}

PHP code is executed from top down and is controlled by the conditional statements ('if' blocks). To achieve what you want, you need to combine the conditional statements that control the insert and update server behaviors and to move the redirect to the end of the code. Amend as follows:

if ((isset($_POST["MM_insert"])) && ($_POST["MM_insert"] == "form1")) {
  $insertSQL = sprintf("INSERT INTO employee (CompId, empfirstname, empmiddleint, emplastname, EmpType, EmpStatus) VALUES (%s, %s, %s, %s, %s, %s)",
                       GetSQLValueString($_POST['CompId'], "int"),
                       GetSQLValueString($_POST['empfirstname'], "text"),
                       GetSQLValueString($_POST['empmiddleint'], "text"),
                       GetSQLValueString($_POST['emplastname'], "text"),
                       GetSQLValueString($_POST['EmpType'], "text"),
                       GetSQLValueString($_POST['EmpStatus'], "text"));
  mysql_select_db($database_dotweb, $dotweb);
  $Result1 = mysql_query($insertSQL, $dotweb) or die(mysql_error());

  $updateSQL = sprintf("UPDATE company SET CompanyType=%s WHERE CompId=%s",
                       GetSQLValueString($_POST['CompanyType'], "text"),
                       GetSQLValueString($_POST['CompId'], "int"));
  mysql_select_db($database_dotweb, $dotweb);
  $Result1 = mysql_query($updateSQL, $dotweb) or die(mysql_error());  

  $insertGoTo = "Roster.php?CompId=" . $_POST['CompId'] . "";

  if (isset($_SERVER['QUERY_STRING'])) {
    $insertGoTo .= (strpos($insertGoTo, '?')) ? "&" : "?";
    $insertGoTo .= $_SERVER['QUERY_STRING'];
  }
  header(sprintf("Location: %s", $insertGoTo));
}

Insertion of text records that contain the '& '.
Hello
I'm you insert records in my table in the field NAME (varchar2) PLACES, and the record that I'm trying to insert is as "Barrie & District Park. I didn't know if I wrote my statement correctly.

Update SITES
name = q '(Barrie & District Park)' where...;

Thank you!
Try:

SET DEFINE OFF

Hide the form field based on the name that contains some characters

I think it's easy.
We're looking for help with a script that allows you to hide all instances of text fields that contain a certain value. We have a running feature that applies to the same text fields, however, each text field should be labeled with a number at the end also running with this is a custom save button that flattens the page and hides some elements before the record on. The hide feature works pretty simple with "this.getField... = display.hidden; The problem is we do not want to repeat this step for each unique instance of fields that share the same name.
We must look to each text box that has a similar title, so for example if we had several fields like that... "my-text-field-01", "my-text-field-02" and "my-text-field-03" we want the script to find the "my-text field" name and hide all instances of that. ' "."
Hope this makes sense, I'm sorry, our programming skills are still amateur. Thanks in advance for any help.

You can do if you use a hierarchical naming convention. For example, instead of using something like: my-text - field.1, my-text - field.2, etc..

You can then hide as well as:
```
// Hide all of the "my-text-field" fields
getField("my-text-field").display = display.hidden;
```

Problem updating a table of clusters that contain some gauges

Hello

I have problems when I write to an array of clusters that contain some gauges.

I wrote an example program to illustrate the problem.

[I create a digital, digital picture].

I have complete 4 elements of the array with the data.

If I shoot each element with a function Index Array and write the data in 4 indicator groups independently, I have no problem.

If I have the wiring of the table of 4 elements in an array of clusters, the needles on the gauges redraw correctly.

However, simply by moving the mouse on the element table causes the display to redraw this element.

I played a bit with synchronous display and reporter Panel updates, but they do not seem to affect this behavior.

Any thoughts?

A picture of the problem and the VI are attached.

When something like this pops up I usually try covering the incriminated with a transparent 2D image control. Controls overlapping sometimes cause problems, sometimes they solve them.

When I insert a compatible game must be installed and the alert arrives to allow another comes in and says "the file 'Localization.xml' could not be loaded."

When I insert a compatible game must be installed and the alert arrives to allow another comes in and says "the file 'Localization.xml' could not be loaded."

Hi Scarface0721,

Welcome to Microsoft Answers Forums.

We would like to get a better understanding of this issue, so we can better help not only you but other users with similar problems.

(a) game in which you try to read?

(b) have you been able to install and play the game before?

(c) you are trying to install the game to your computer from the disk or directly trying to play from the disc?

Method 1:
Try to install and play the game in a different computer.

Method 2:
Try another disc and verify if you are able to install the game.

Method 3:
You can also post your request in the following forums site.

http://forums.gamesforwindows.com/

Thanks and regards.

Thahaseena M
Microsoft Answers Support Engineer.
Visit our Microsoft answers feedback Forum and let us know what you think.

I'm having a problem with my Opnegl ver 1.1, must be updated 1.3 for some reason that I can't seem to update or to locate in the profile of my hard drive... ERRRR... of suggestions?

I'm having a problem with my Opnegl ver 1.1, must be updated 1.3 for some reason that I can't seem to update or to locate in the profile of my hard drive... ERRRR... of suggestions?

Update your video card drivers. Bring:

1. site of the device mftr. ; OR
2. site of the card mother mftr. If the material is on board; OR
3. website of the OEM manufacturer for your specific machine if you have a PC OEM (HP, Dell, Sony, etc.).

Read the installation instructions on the Web site where you get the drivers.

To find out what hardware is in your computer:

1. read all documentation that you got when you purchased the computer.
2. If the computer is OEM, access the site Web of the OEM for your specific model machine and look at the specs (you'll be there to get the drivers anyway)
3. download, install and run a program of inventory of the free system like Belarc Advisor or System Information for Windows.

http://www.belarc.com/free_download.html - Belarc Advisor
http://www.gtopala.com/ - system information for Windows MS - MVP - Elephant Boy computers - Don ' t Panic!

I have a table of the adf, I added a column that contains a button that I created, when I click it must remove this row in the table, but it is not, please help

I have a table of the adf, I added a column that contains a button that I created, when I click it must remove this row in the table, but it is not, please help

I don't understand. You use vo and eo for you to use business components.

Again, this kind of code call in trouble.

You must post the changes to make them visible to the eo find vo. You must then run the query for the changes in the business layer strips then you must update the iterator he table is based on.

In your code I see that happen, hooch maybe because it is more often than not formatted and undocumented.

My advice is to do a small test case that you can manage with easy sql. Once you get it to run transfer you the results to the actual application.

Timo

How to filter the records that contains the Spanish character?

Hello

I had an obligation to always records based on the Spanish character.

Kindly guide me for below.

To filter the records that contains Spanish characters?

For example, I had a value in the column name as 'Sureshn '.

My query should return the above folder that contains the Spanish character "N".

Please do the necessary help / advice on that.

Thank you
Orahar

I don't know what you want, because if you want a? ¢ Analysys and Sureshn, then my previous query is sufficient. However, maybe something like that.

SQL> ed
Wrote file afiedt.buf

  1  WITH Sample_Data AS (SELECT 'â?¢ Analysys' str FROM DUAL UNION ALL
  2     SELECT 'SureshÑ' str FROM DUAL UNION ALL
  3     SELECT 'ABCD' str FROM DUAL UNION ALL
  4     SELECT 'WXYZ' str FROM DUAL UNION ALL
  5     SELECT 'Saubhik' str FROM DUAL
  6     )
  7     SELECT str AS "Contains other than English" from Sample_Data
  8*    WHERE REGEXP_LIKE(str,'[^[a-z,A-Z,0-9]]*')
SQL> /

Contains other
--------------
â?¢ Analysys
SureshÑ

SQL> ed
Wrote file afiedt.buf

  1  WITH Sample_Data AS (SELECT 'â?¢ Analysys' str FROM DUAL UNION ALL
  2     SELECT 'SureshÑ' str FROM DUAL UNION ALL
  3     SELECT 'ABCD' str FROM DUAL UNION ALL
  4     SELECT 'WXYZ' str FROM DUAL UNION ALL
  5     SELECT 'Saubhik' str FROM DUAL
  6     )
  7     SELECT str AS "Contains English" from Sample_Data
  8*    WHERE NOT REGEXP_LIKE(str,'[^[a-z,A-Z,0-9]]*')
SQL> /

Contains Engli
--------------
ABCD
WXYZ
Saubhik

Before file is deleted, the record must be insert into another table
Hi friends,

I have to create tabular form to delete some record. Using this tabular form I deleted successfully.
But I want to when I delete a record before deleting record that he should go to another table.

How can I do that.

Ed
Hello

You create before you remove the trigger table.
http://download.Oracle.com/docs/CD/B10500_01/server.920/a96524/c18trigs.htm

BR, Jari

Create a package that contains a record type

Hello
I have a class called table as follows

Name Null? Type

-----
--------
-----

-----
CLASS_ID NOT NULL NUMBER 4
CLASS_NAME VARCHAR2 (100)
NUMBER OF SEAT_NUMBER
NUMBER OF STUD_ID

(stud_id) is a foreign table for an another table called student.

IWant to create a package of class table contained a record of the table of class type
This includes all columns in the table of class
but I get the warning: -.
package created with compilation error.
Please can you help.
It is my code: -.

sql&gt;create or replace package classpackage as
type tclass is record(
rclass class.class%type,
rclass_id class.class_id%type,
rclass_name class.class_name%type,
rseat_number class.seat_number%type,
rstud_id class.stud_id%type);
end classpackage;
/

Why don't test you it?

satyaki>
satyaki>
satyaki>select * from v$version;

BANNER
----------------------------------------------------------------
Oracle Database 10g Enterprise Edition Release 10.2.0.3.0 - Prod
PL/SQL Release 10.2.0.3.0 - Production
CORE    10.2.0.3.0      Production
TNS for 32-bit Windows: Version 10.2.0.3.0 - Production
NLSRTL Version 10.2.0.3.0 - Production

Elapsed: 00:00:00.43
satyaki>
satyaki>
satyaki>create table class
  2   (
  3      CLASS_ID     NUMBER(4)     NOT NULL,
  4      CLASS_NAME   VARCHAR2(100),
  5      SEAT_NUMBER  NUMBER,
  6      STUD_ID      NUMBER
  7   );

Table created.

Elapsed: 00:00:04.75
satyaki>
satyaki>create or replace package classpackage
  2  as
  3     type tclass is record
  4       (
  5          rclass_id     number(4),
  6          rclass_name   varchar2(100),
  7          rseat_number  number,
  8          rstud_id      number
  9       );
 10  end classpackage;
 11  /

Package created.

Elapsed: 00:00:01.82
satyaki>
satyaki>
satyaki>
satyaki>create or replace package classpackage
  2  as
  3     type tclass is record
  4      (
  5         rclass_id class.class_id%type,
  6         rclass_name class.class_name%type,
  7         rseat_number class.seat_number%type,
  8         rstud_id class.stud_id%type
  9      );
 10  end classpackage;
 11  /

Package created.

Elapsed: 00:00:00.30
satyaki>
satyaki>
satyaki>

Kind regards.

LOULOU.

Change in the merger of insert and create a procedure for updating the records after
Hello

I have a table that contains about 50 M lines of partition, and every day I have an application that performs a merge to insert records statement about 100 k per day and it's 300 M updates a day.

Because I both update and insert the application performs a merge.
I need to make at least 500 M, updated day by day, and for this reason this solution doesn't have enough performance for what I'm asking.

I think another solution that I will try to implement now, and I just posted this in an attempt to get help on the option that I may have to do.

I want to start putting in place:
-Locations is updated 'last_record_date' using 'KEY1' and 'Key2' "KEY3";
-Change the application to only inserts into a temporary table.
-Have two procedures, one that takes the new records from the temporary table and inserts it into the final table, and another that selects records which is updated and perform an update.
-This two procedures will choose the records and delete them from the temporary table (or mark them as updated\inserted).

The volume is very large, so performance is a challenge! That's why I'm writing this, because a solution with enough performance for the intermediate stage will be difficult.
My final table is partitioned by date and the date is the area that I need to update (instead of create daily partitions in order to reduce the frequency of the movement of the line, I create monthly partitions)!
It is currently my table definition:
```
CREATE TABLE MY_TEST_TABLE
    (KEY1                           VARCHAR2(50 BYTE) ,
    KEY2                         VARCHAR2(50 BYTE) NOT NULL,
    KEY3                           VARCHAR2(50 BYTE) ,
    last_record_date                    VARCHAR2(50 BYTE) NOT NULL)
  PCTFREE     10
  INITRANS    1
  MAXTRANS    255
   STORAGE   (
    BUFFER_POOL DEFAULT
  )
  NOCACHE
  MONITORING
  ENABLE ROW MOVEMENT
  PARTITION BY RANGE (last_record_date)
  (
  PARTITION MY_TEST_TABLE_201112 VALUES LESS THAN ('20120101')
  PCTFREE     10
  INITRANS    1
  MAXTRANS    255
  NOLOGGING,
  PARTITION MY_TEST_TABLE_201201 VALUES LESS THAN ('20120201')
  PCTFREE     10
  INITRANS    1
  MAXTRANS    255
  NOLOGGING,
  PARTITION MY_TEST_TABLE_201202 VALUES LESS THAN ('20120301')
  PCTFREE     10
  INITRANS    1
  MAXTRANS    255
  NOLOGGING,
  PARTITION def VALUES LESS THAN (MAXVALUE)
  PCTFREE     10
  INITRANS    1
  MAXTRANS    255
  NOLOGGING
  )
  NOPARALLEL
/



-- Constraints for MY_TEST_TABLE

ALTER TABLE MY_TEST_TABLE
ADD CHECK ("KEY1" IS NOT NULL)
/

ALTER TABLE MY_TEST_TABLE
ADD CHECK ("KEY2" IS NOT NULL)
/

ALTER TABLE MY_TEST_TABLE
ADD CHECK ("KEY3" IS NOT NULL)
/
ALTER TABLE MY_TEST_TABLE
ADD CONSTRAINT pk_MY_TEST_TABLE PRIMARY KEY (KEY1, KEY2, KEY3)
USING INDEX
  PCTFREE     10
  INITRANS    2
  MAXTRANS    255
/
```
Thank you
Ricardo Tomas
naoseionome wrote:
Hello

My version is:
Oracle Database 11 g Enterprise Edition Release 11.2.0.2.0 - 64 bit Production

It's the DOF I used to create the database.

The merger is common during every day for about 10 M records.

There are several methods to do the update, but they do sequencially to avoid deadlocks (because I can't control the source and each process can update the same keys.

The update uses "KEY1, KEY2, KEY3" to update the value 'LAST_RECORD_DATE' have impact by being date or varchar2?

I'm still waiting to receive permissions in order to follow up the matter, but in the meantime, in that I am trying to implement this work around in order to accelerate a development in the case where it would be necessary! Comments only I have at the moment, it is that he is not able to process the total amount of records timely (order book is generated).

Thank you
Ricardo

Well, the first thing we do is to change the type of data type varchar to a date. The main impact is that a date is not a string, it's a date. If you are adding data processes in different date formats (which is possible when you use a character to store information) you're going to be very sad (good luck converting back an effective date).

An example of how this can happen easily.
```
ME_TUBBZ?select to_char(sysdate) from dual;

TO_CHAR(SYSDATE)
-----------------------------
12-jan-2012 10 55:33

1 row selected.

Elapsed: 00:00:00.01
ME_TUBBZ?alter session set nls_date_format = 'yyyy-mon-dd';

Session altered.

Elapsed: 00:00:00.01
ME_TUBBZ?select to_char(sysdate) from dual;

TO_CHAR(SYSDATE)
--------------------
2012-jan-12

1 row selected.

Elapsed: 00:00:00.00
```
Assuming that the client passes the column as an Oracle DATE will make an (implicit) conversion to store the data as a string, which will be based on the current settings of the NLS for the session. This allows also for someone to completely non-jour the information in the column (super bad). It's just really bad form to use strings to store dates, numbers, anything that is not naturally a string.

Based on the description of your problem, I can't imagine how your partitioning scheme performs one action in another that require more work in order to process the updates (which are most of your needs). Partitioning does not equate to increase performance. It is a tool that, like every tool has its place, but I do not think that this is justified for what you do (certainly not how you currently use it anyway).

I say no partitioning can not be useful for you, but I don't think that on the column that you are currently using is 'good '.

Something I would like to personally study (must be able to compare this approach to your course, if you need parameters for comparison) if I were you would be using a Table Index organized for it. It will take more time to perform the operations insertion, but that is a small part of all the work that you do for this process. For updates, it should eliminate a lot of IO, assuming that you are doing an INDEX seek followed by a table access by ROWID to perform updates. This recommendation assumes that everything you said is true. You have 4 columns in your table, of which 3 are components of the primary key and the 4th is a date (note I did not string :)) and that you refresh mainly, not fit.) Also that you have no 'secondary' on this entity index (and none are needed).

I would investigate this approach before the solution of temporary table that you are currently studying.

See you soon,.

My Contacts Windows Live is 452,1 MB and contains 16 files that extend into the size of 28.1 MB to 29.2 MB. Can one be deleted and if so, which?

Contacts Windows Live is 452,1 MB

I was sent here from the "Windows Live Solution Center" to ask my 'Q', below:

My Contacts Windows Live is 452,1 MB and contains 16 files that extend into the size of 28.1 MB to 29.2 MB.
Why so many large files containing essentially the same... Contacts?
Do we really need 28 + MB to store contacts?
Two of the files have been accessed today, while others have not been accessed for 6 days.
Can one be deleted and if so, which?

XPPro SP3 32 bit

--
happytilton®

@happytilton®

Since you are using Windows XP, your version of Windows Live Messenger be WLMsgr 2009.

- so working in online mode (b) - Messenger requires a Live ID access code to use.

The article I wrote and quoted above explained that WLM (Windows Live Mail), which is also applicable to the WLMsgr 2009 versions contain most of the information to determine what to remember as...

1. how to use the registry to determine the names of digital files of the database of Contacts

2A explained that 2009 has two functional databases for each Live ID (a primary and a shadow)

-both are needed. used 2009 the 'shade' as a catchment area to ensure synchronization occurs in both directions, when changes are made in both locations (local in Messenger or online at contacts.live.com)

Please read and understand the best article

http://liveunplugged.WordPress.com/2011/02/04/Windows-Live-Mail-contacts-database/

... especially the section dedicated to explain the Contacts 2009 databases.

Once you determine the digital records of your single Live ID database (shadow and primary) and the program created 'default' you can check the size of each. If you feel that your registry database shows other numeric names (created using other Live ID or perhaps due to an error by typing a user name in the Messenger sign-in box) and these folders can be deleted safely. Note: you may need to configure Windows Explorer to show hidden and system files to see when viewing in Windows Explorer.

In addition, you can also, as your contacts are stored online at contacts.live.com, after closing of Messenger remove the any local digital database called folders (for the primary shadow and if is the default). WLMsgr will recreate them on next signon and rebuild any local database folders and files based on what it sync (downloads) of the Live Server to contacts.live.com.

-Before the deletion of all files, it would be wise to connect to contacts.live.com and check that all your contacts are indeed present and represent for and for saving documents to archives/purposes export contacts to a csv file using the web UI option... Once completed then disconnection from the web UI.

-Do not, and I it cannot emphasize enough strongly, do not delete anything from the web interface before you delete the local databases on your hard drive (you can always return to contacts.live.com later to manage the contacts (remove, clean the unused or duplicates, merge duplicates etc.) if necessary after WLMsgr recreates local databases.)

Once done your registry key should show that three databases total (assuming you are using only a Live ID in WLMsgr)-the main and shadow for that Live ID and a default value (if it is created by the program).

You should have enough information to move forward or to ask additional questions.

As I mentioned earlier that, at a minimum, an empty list of contacts would be at least some multiple approx. 24 MB. Since you use WLMsgr 2009, and it uses a primary and shadow, the minimum size of about for you with no contact would be approximately 72MB (24 MB for each - primary, shadow and by default). As well as the contacts are added, the primary databases and shadows will be size.

Other topics presented in this thead element (but not significant) - comments below to close this loop

S ' for the use of your Hotmail account in Outlook. This use is not related or significant to the database used by WLMsgr. Contacts in Outlook (when the mail is configured to use Pop3 or DeltaSync with Outlook Hotmail Connector) are stored in the Outlook profile contacts, and quite separate from the WLMsgr database (i.e. neither Outlook or WLMsgr program communicates or synchronize or share data with each other to keep the feature)

FYI... not associated with 2009, but 2011 (for those who read along. WLM 2011 no longer uses the approach of two (primary and shadow) file for databases local contact so databases are smaller and also more robust in 2011 and better synchronize. Finally, with respect to the use of 2009, wysiwyg... no other upgrades for 2009 are planned (the code was frozen for more than a year, the latest version (never) released in May 2010).

Good luck.

.. .Winston

msft mvp mail

XML contains some records that must be inserted and some updates

Similar Questions

Maybe you are looking for