Insert node xmltype

Similar Questions

• XQuery insert node and assign the value based on an XPATH expression

  Hello

  I am trying to write a function that receives an XML as input parameters and return updated XML output

  -nodes are removed from the input XML code

  -a node must be inserted and the node value must be 1 or 0, based on an XPATH expression

  I discovered how to remove nodes, I discovered how to insert a new node, but I can not set the value of the conditionally inserted node an XPATH expression.

  Here's my current procedure and below a sample of one XML of entry. I use Oracle 12 c.

  CREATE OR REPLACE FUNCTION STRIP_XML
  (
    IN_XML IN SYS.XMLTYPE
  ) RETURN SYS.XMLTYPE AS
  p_result XMLType;
  BEGIN
    select
        xmlquery(
           'declare default element namespace "http://mad.evs.com/search"; (: :)
            copy $d := .
            modify (
    delete node $d//MainCategory/@logId,
             delete node $d/MainCategory/@id,
  -- a lot more of those delete node
   -- insert a node, but the value 1 is conditional
             insert node <DMZ>1</DMZ> after $d/MainCategory/SDataSection/EventDate)
            return $d'
          passing in_xml
          returning content
          ) into p_result
          from dual;
  
  
    RETURN p_result;
  END STRIP_XML;
  

  The condition for the value of is

  
  

  HASPATH (//DigitalAssets/DigitalAsset [@ available = "true" and @videoFormatId = "11"] / VideoLocations/Videorental [@typeId = "8"]) then 1 else 0

  
  

  It is complex because DigitalAssets/DigitalAsset is a collection. Here is an example of an XML to entry

  
  

  
  

  <MainCategory xmlns="http://mad.evs.com/search" id="9" logId="3349" logType="3">
    <Name>Sport</Name>
    <Serie id="163" externalId="557">
    <TitleAKA>UCL 2006/07</TitleAKA>
    <DigitalAssets available="true" som="20:28:49:05" dur="00:02:46:04" videoDurationMinutes="3">
       <DigitalAsset available="true" som="20:28:49:05" dur="00:02:46:04" videoDurationMinutes="3" videoFormatId="3">
          <VideoLocations>
              <VideoLocation id="3" path="003349MA.mxf" typeId="1" locationId="1" priority="0"/>
              <VideoLocation id="3" path="003349MA.mxf" typeId="2" locationId="1" priority="0"/>
              <VideoLocation id="3" path="003349MA.mxf" typeId="5" locationId="1" priority="0"/>
           </VideoLocations>
        </DigitalAsset>
        <DigitalAsset available="true" som="20:28:49:05" dur="00:02:46:04" videoDurationMinutes="3" videoFormatId="11">
           <VideoLocations>
              <VideoLocation id="101" path="003349MA.mpg" typeId="1" locationId="1" priority="0"/>
              <VideoLocation id="101" path="003349MA.mpg" typeId="2" locationId="1" priority="0"/>
              <VideoLocation id="101" path="003349MA.mpg" typeId="8" locationId="1" priority="0"/>
           </VideoLocations>
        </DigitalAsset>
    </DigitalAssets>
    <SDataSection xmlns="http://mad.evs.com/search">
      <EventDate>2006-08-09</EventDate>
      <LogType>3</LogType>
    </SDataSection>
  </MainCategory>
  

  
  

  In this case, because HASPATH expression could be set to true, the value of the added node must be 1.

  
  

  Any help or advice how I should fight against that would be appreciated. I have no experience with XML and XQuery, I create my function through from trial and error of the doc.
  

  
  

  Thanks and regards, Pierre
  

  Hi Pierre,.

  You can use the if-then-else statement, like this:

  Insert the node

  {

  If ($d/MainCategory/DigitalAssets/DigitalAsset[@available="true' and @videoFormatId = '11'] / VideoLocations/Videorental [@typeId = '8'])

  then 1

  0 otherwise

  }

  After $d, MainCategory, SDataSection, EventDate

  XSLT can be an alternative to the XQuery Update in this case.

• [11g] ORA-31061/ORA-19202 - how to insert data xmltype table/validation of the schema (just well formed)

  Hello

  Oracle Database 11 g Enterprise Edition Release 11.2.0.3.0 - 64 bit Production

  PL/SQL Release 11.2.0.3.0 - Production

  CORE Production 11.2.0.3.0

  AMT for Linux: Version 11.2.0.3.0 - Production

  NLSRTL Version 11.2.0.3.0 - Production

  Here's the demo program:

  ==============================

  create a global temporary table GTT_Test(data xmltype) validation delete rows;

  - and run this

  declare

  v_data xmltype.

  Start

  -It does not work

  /*

  v_data: = xmltype ("< itf:meta > < node / > < / itf:meta > '");

  */

  - but it works

  v_data: = xmltype)

  XMLDATA = > ' < my: meta > < node / > < / my: meta > ',

  schema = > null,

  validated = > 0,

  correct = > 1);

  - but try to insert into the table in the same way fails again:

  insert into GTT_Test (data)

  values (xmltype)

  XMLDATA = > ' < my: meta > < node / > < / my: meta > ',

  schema = > null,

  validated = > 0,

  correct = > 1));

  end;

  ==============================

  
  

  It translates errors ORA above and suspicion: "prefix 'my' is not declared".

  
  I need to create the TWG somehow different to accept without diagram validation?

  
  

  -Thanks!

  
  

  Kind regards

  Frank
  

  Hello.

  Try to declare to your XML namespace.

  for examplehttp://youaddress">

  "If you don't have a schema, try xmlns: my =" "(espace en blanc)."

• Add the xmltype attribute node

  How can I add an attribute node to xml in an xmltype?
  Say that I

  declare
    t_xml xmltype;
  begin
    t_xml := xmltype( '<x a="1"/>' );
  end;

  How can I manipulate this variable t_xml become

  <x a="1" b="2"/>

  Anton

  Depends on the version.

  Before 11.2.0.3:

  SQL> with t as (  select xmltype('') as t_xml from dual )
    2  select insertchildxml(
    3           t_xml
    4         , '/x'
    5         , '@b'
    6         , '2'
    7         )
    8  from t;
  
  INSERTCHILDXML(T_XML,'/X','@B'
  --------------------------------------------------------------------------------
  
   
  

  On 11.2.0.3, by using XQuery Update:

  select /*+ no_xml_query_rewrite */
         xmlquery(
          'copy $d := .
           modify ( insert node attribute b {$val} into $d/x )
           return $d'
          passing t_xml
                , '2' as "val"
          returning content
         )
  into t_xml
  from dual
  ;
  

  (not tested, but should be close)

  Marco Gralike wrote:
  You should be able to build up the little together 'back' in a dynamic method, but for now, don't ask me how (mark)?

  The generic approach would require market the entire tree down recursively to the copy existing nodes and add the new in the right place.
  Although XQuery (without involving XQUF) can do it, it's more of a task for XSLT:

  SQL> with t as (  select xmltype('') as t_xml from dual )
    2  select xmltransform(
    3           t_xml
    4         , xmlparse(document
    5  '
    6    
    7    
    8    
    9      
   10        
   11      
   12    
   13    
   14      
   15        
   16      
   17    
   18  '
   19           )
   20         , q'{att="'2'"}'
   21         )
   22  from t;
  
  XMLTRANSFORM(T_XML,XMLPARSE(DO
  --------------------------------------------------------------------------------
  
  
   
  

• XMLTYPE string buffer too small

  I do the following:

  DECLARE

  xmlValue CLOB

  : = ' < HTNG_HotelRoomStatusSearchRS > < success > < / success > < errors > < error Type = '0' doc 'None' = > < / error > < / errors >

  < RoomInformationList > < TPA_Extensions > < TPA_Extension >

  < UNITI > < / UNITI > < / TPA_Extension > < / TPA_Extensions >

  < / RoomInformationList > < / HTNG_HotelRoomStatusSearchRS > ';

  XMLTYPE x_xml;

  NewValue VARCHAR2 (2000): = ' < UNIT_NUM UNIT = "ABCD" / > ";

  BEGIN

  x_xml: = xmltype (xmlvalue);

  FOR I IN (SELECT col1

  OF xml_modify

  )

  LOOP

  SELECT XMLQUERY (' copy $i: $p1 = edit)

  (for $j to $i / / TPA_Extensions/TPA_Extension/States)

  back insert node $p2 before $j)

  Return $i "PASSAGE x_xml AS"p1", xmltype (i.col1) AS 'p2' RETURN CONTENT)

  IN x_xml

  FROM DUAL;

  END LOOP;

  x_xml: = x_xml. DELETEXML ('/ / STATES ');

  Dbms_output.put_line (x_xml.getstringval ());

  END;

  BTW, there is very little effective insertion of nodes like this in a loop.

  Build the fragment in a go with XMLAGG on the XML_MODIFY table, and then insert that fragment using XQuery Update.

  You can even remove the UNITI node within the same expression:

  DECLARE
  
    x_xml  xmltype := xmltype(
    '
  
  
  ');
  
    x_units  xmltype;
  
  BEGIN
  
    select xmlagg(xmlparse(content col1))
    into x_units
    from xml_modify
    --where rownum < 10
    ;
  
    SELECT XMLQUERY(
             'copy $i := $p1
              modify (
                for $j in $i/HTNG_HotelRoomStatusSearchRS/RoomInformationList/TPA_Extensions/TPA_Extension/UNITI
                return (
                  insert node $p2 before $j
                , delete node $j
                )
              )
              return $i'
             PASSING x_xml AS "p1"
                   , x_units AS "p2"
             RETURNING CONTENT
           )
    INTO x_xml
    FROM DUAL;
  
    --x_xml := x_xml.DELETEXML ('//UNITI');
    --DBMS_OUTPUT.put_line(x_xml.getclobval(1,2));
  
  END;
  /
  

• XMLQUERY Multiple inserts

  I have the following:

  DECLARE

  xmlValue CLOB

  : = ' < HTNG_HotelRoomStatusSearchRS > < success > < / success > < errors > < error Type = '0' doc 'None' = > < / error > < / errors >

  < RoomInformationList > < TPA_Extensions > < TPA_Extension >

  "" < UNIT_NUM UNIT = "pm_unit_num" UNIT_TYPE = "pm_unit_type_desc" HSK_DESC = "CLN" ROOM_STATUS = "EMPY" STATUS_CODE = "HERE" ARRIVAL_DATE = DEPARTURE_DATE "now" = "THEN" RESV_TYPE ="RT" DRI_PROP_ID = "DPI" RESV_NUM = "RN" CREATE_DATE = "CD" MC_PROPERTY_CODE = "MPC" HOTEL_DATE ="HD" SERVICE_CODE = 'SC' > < / UNIT > < / TPA_Extension > < / TPA_Extensions >

  < / RoomInformationList > < / HTNG_HotelRoomStatusSearchRS > ';

  XMLTYPE x_xml;

  CLOB y_xml;

  BEGIN

  x_xml: = xmltype (xmlvalue);

  BECAUSE me in 1... 2

  LOOP

  SELECT XMLQUERY (' copy $tmp: = change the node the insert)

  < UNIT_NUM UNIT = "ABCD" / >

  After $tmp / / TPA_Extensions/TPA_Extension/UNIT

  return $tmp' PASSING x_xml RENVOYANT CONTENT)

  IN x_xml

  FROM DUAL;

  y_xml: = x_xml.getstringval (); -Thought that the reset would fix it. I think it has to do with more than one < UNIT / >

  x_xml: = xmltype (y_xml);

  END LOOP;

  Dbms_output.put_line (x_xml.getstringval ());

  END;

  When I run the present for an insert, it works very well. When I try 2 times I get this error.

  ORA-18104: XUTY0005 - invalid target for INSERTION expression

  ORA-06512: at line 17 level

  It worked.

  DECLARE

  xmlValue CLOB

  := '

  ';

  XMLTYPE x_xml;

  NewValue VARCHAR2 (2000): = '';

  BEGIN

  x_xml: = xmltype (xmlvalue);

  BECAUSE me in 1... 10

  LOOP

  SELECT XMLQUERY (' copy $i: $p1 = edit)

  (for $j to $i / / TPA_Extensions/TPA_Extension/States)

  back insert node $p2 before $j)

  Return $i "PASSAGE x_xml AS"p1", xmltype (newvalue) AS"p2"RETURN CONTENT)

  IN x_xml

  FROM DUAL;

  NewValue: = '';

  END LOOP;

  x_xml: = x_xml. DELETEXML ('/ / STATES ');

  Dbms_output.put_line (x_xml.getstringval ());

  END;

• Replacement of the nodes with duplicates

  I can't find out how to write an XQuery query which can essentially 'jump' a replacement when I said to. I don't know who makes no sense, so here's a better explanation.
  
  I have an XML file that is stored in an Oracle XML DB that looks like this:
  

  <bookstore>
      <book title="Analytical Chemistry" publish_date="198710">
         <author>...
         <publisher>....
         <etc.>
      </book>
      <book title="Particle Physics" publish_date="199202">
         <author>...
         <publisher>....
         <etc.>
      </book>
      <book title="Applied Biophysics" publish_date="201005">
         <author>...
         <publisher>....
         <etc.>
      </book>
  </bookstore>

  We get the update commands that look like the same thing, with a 'command' attached to add the book to the store or drop:
  

  <bookstore>
      <book title="Analytical Chemistry" publish_date="199210">
         <order>merge</order>
         <author>...
         <publisher>....
         <etc.>
      </book>
      <book title="Particle Physics" publish_date="199202">
         <order>drop</order>
         <author>...
         <publisher>....
         <etc.>
      </book>
      <book title="Chaos Theory" publish_date="199102">
         <order>merge</order>
         <author>...
         <publisher>....
         <etc.>
      </book>
  </bookstore>

  The rule is should I replace or add a book in the store if I have a book of the same title, and the order is "merge", or drop it if the order is to 'drop '. I have a request XQuery Update function now makes what looks like this:
  

  declare updating function local:book_merge($old, $book_merge)
  {
       for $orderEntr in $book_merge/book
       let $chgCd := $orderEntr/order
       let $orderTitle := $orderEntr/@title
          let $oldBooks := $old/bookstore
       let $oldBookEntr := $oldBooks/book[@title=$orderTitle]
       return
          if( $chgCd = "drop" and exists($oldBookEntr) )
          then delete node $oldBookEntr
          else if( exists($oldBookEntr) )
          then replace node $oldBookEntr with $orderEntr
          else insert node $orderEntr into $oldBooks
  };

  This works very well for commands like above, but once from time to time, we receive a command like this:
  

  <bookstore>
     <book title="Analytical Chemistry" publish_date="199210">
         <order>merge</order>
         <author>...
         <publisher>....
         <etc.>
      </book>
      <book title="Analytical Chemistry" publish_date="199210">
         <order>merge</order>
         <author>...
         <publisher>....
         <etc.>
      </book>
      <book title="Particle Physics" publish_date="199202">
         <order>drop</order>
         <author>...
         <publisher>....
         <etc.>
      </book>
      <book title="Chaos Theory" publish_date="199102">
         <order>merge</order>
         <author>...
         <publisher>....
         <etc.>
      </book>
  </bookstore>

  Notice how the first entry is duplicated. In this case, the requirement must only enter the book in the library once. Can someone give me an idea of how I would go about implementing that? I am struggling to understand how to ask the right question, so please forgive me.

  You can add a predicate that will filter the next (or previous) duplicates of a given input:

  where not($orderEntr/following-sibling::book[@title = $orderTitle])
  

  Full test case:

  Connected to:
  Oracle Database 11g Enterprise Edition Release 11.2.0.3.0 - 64bit Production
  With the Partitioning, OLAP, Data Mining and Real Application Testing options
  
  SQL> create table tmp_xml of xmltype;
  
  Table created.
  
  SQL> insert into tmp_xml values (
    2  xmltype('
    3      
    4         XXX
    5         
    6      
    7      
    8         YYY
    9         
   10      
   11      
   12         ZZZ
   13         
   14      
   15  ')
   16  );
  
  1 row created.
  
  SQL> var orders clob
  SQL> begin
    2
    3   :orders := '
    4      
    5         merge
    6         XXX
    7         123
    8      
    9      
   10         merge
   11         XXX
   12         123
   13      
   14      
   15         drop
   16         YYY
   17         
   18      
   19      
   20         merge
   21         ZZZ-2
   22         
   23      
   24  ';
   25
   26  end;
   27  /
  
  PL/SQL procedure successfully completed.
  
  SQL> set long 5000
  SQL> set pages 100
  SQL>
  SQL> select /*+ no_xml_query_rewrite */
    2         xmlserialize(document
    3           xmlquery(
    4           'copy $d := $old
    5            modify (
    6              for $orderEntr in $orders/bookstore/book
    7              let $chgCd := $orderEntr/order
    8              let $orderTitle := $orderEntr/@title
    9              let $oldBooks := $d/bookstore
   10              let $oldBookEntr := $oldBooks/book[@title = $orderTitle]
   11              where not($orderEntr/following-sibling::book[@title = $orderTitle])
   12              return
   13                 if( $chgCd = "drop" and exists($oldBookEntr) )
   14                   then delete node $oldBookEntr
   15                 else if( exists($oldBookEntr) )
   16                        then replace node $oldBookEntr with $orderEntr
   17                      else insert node $orderEntr into $oldBooks
   18            )
   19            return $d'
   20           passing object_value as "old"
   21                 , xmlparse(document :orders) as "orders"
   22           returning content
   23           )
   24           indent
   25         ) as result
   26  from tmp_xml ;
  
  RESULT
  --------------------------------------------------------------------------------
  
    
      merge
      XXX
      123
    
    
      ZZZ
      
    
    
      merge
      ZZZ-2
      
    
  
  

  It's another question, you probably already have the answer: I guess you must also remove the nodes of the newly merged books?

• HOW to add an attribute to the existing node?

  Hello
  
  I have a question - is possible to add a NEW attribute to existing xml node registered in xmltype?
  
  I found no such a possibility.
  
  I need to use the update! :)
  
  I have: < tag > < tag2 / > < tag / >
  
  I want to have < tag value = 1 > < tag2 / > < tag / >
  
  THANKS IN ADVANCE! :)

  With insertChildXML:

  SQL> with t as (
    2    select xmltype('') doc
    3    from dual
    4  )
    5  select insertChildXML(doc, '/tag', '@value', 1)
    6  from t
    7  ;
  
  INSERTCHILDXML(DOC,'/TAG','@VA
  --------------------------------------------------------------------------------
  
   
  

  Alternatively, in 11.2.0.3, updated XQuery:

  select xmlquery(
  'copy $d := $doc/tag
   modify ( insert node attribute value {$val} into $d )
   return $d'
   passing t.doc as "doc"
         , 1 as "val"
   returning content
  )
  from t
  ;
  

• upsert xmltype data xpath - missing something obvious?

  Version: 10.2.0 * 2 * EA
  
  I'm doing a upsert (update / insert) of an xmltype for a given xpath.
  
  The xml data is passed as an xmltype, xpath is a string (varchar2), the new value is a string, varchar2.
  
  If the node specified by the xpath expression exists it is trivial using updatexml; of course if the node is not nothing happens. I can verify the existence of nodes using is node, but I just take off, trying to figure out how to insert the node.
  
  All the APIs I can find who to insert XML (APPENDCHILDXML, INSERTXMLBEFORE, INSERTCHILDXML) do not take an xpath but require an xmltype. Which means that if I'm reading this right I have no other choice to the parse varchar2 string xpath to create an xmltype (ugh!).
  
  I can't find any API that takes an xpath as a varchar2 and a value and gives me an xmltype. Ideally I want same, here an API that returns the XML updated which took the set xpath to update (or inserted if it did not exist) with the new value.
  
  I'm not an XML expert but I'm pretty handy with SQL and PL/SQL. For me this seems incredibly heavy just to do something as mundane as an upsert, so I hope I'm missing something obvious!
  
  Can soemone please put me everything right with a pointer or an example?
  
  On top of the XML uses namespaces, but they are not registered, there is no record of current schema.

  Marco Gralike wrote:
  Could not be done through XQuery, although the version of database doesn't really help...

  It would be possible with a dynamic, but 'painful' XQuery with a static expression expression because it would require parsing the XPath expression and rebuild the entire document of entry with the necessary adaptations.
  I think XSLT would be more effective in this case.

  Perhaps, one day, when the database is XQuery Update Facility, we will be able to do:

  copy $a := $doc/Address
  modify (
    if ($a/zip)
      then replace value of node $a/zip with "12345"
      else insert node element zip {"12345"} into $a
  )
  return $a
  

  user12083137 wrote: for me, from a SQL background, I can't believe this simple case is simply not covered.

  Perhaps not as simple as it sounds. :)

  The feature can be simulated by an IF/THEN/ELSE logic, or a DELETE/INSERT sequence.
  For example:

  case when existsNode(doc, xpath) = 1
    then updateXML(
           doc
         , xpath || '/text()'
         , somevalue
         )
    else appendchildxml(
           doc
         , substr(xpath, 1, instr(xpath, '/', -1)-1)
         , xmlelement(
             evalname(substr(xpath, instr(xpath, '/', -1)+1))
           , somevalue
           )
         )
  end
  

  or,

  appendChildXML(
    deleteXML(doc, xpath)
  , substr(xpath, 1, instr(xpath, '/', -1)-1)
  , xmlelement(
      evalname(substr(xpath, instr(xpath, '/', -1)+1))
    , somevalue
    )
  )
  

• Why the nodes are placed on web sites that I discovered with fire fox but not other browsers

  When you view the web page with fire fox is someone changing words and inserting nodes on the site. This does not happen with chrome or explorer. When I inspect the element it is HTML not allowed more written on the site. I have remove the node (a commercial that redirects the browser), but they pop up.

  You can check the suspicious extensions or recently installed unknowns.

  • https://support.Mozilla.org/KB/troubleshooting+extensions+and+themes

  Make a check of malware with several malware scanning of programs on the Windows computer.
  Please scan with all programs, because each program detects a different malicious program.
  All of these programs have free versions.

  Make sure that you update each program to get the latest version of their databases before scanning.

  • Malwarebytes' Anti-Malware:
    http://www.Malwarebytes.org/MBAM.php
  • AdwCleaner:
    http://www.bleepingcomputer.com/download/adwcleaner/
    http://www.Softpedia.com/get/antivirus/removal-tools/AdwCleaner.shtml
  • SuperAntispyware:
    http://www.SUPERAntiSpyware.com/
  • Microsoft security scanner:
    http://www.Microsoft.com/security/scanner/en-us/default.aspx
  • Windows Defender:
    http://Windows.Microsoft.com/en-us/Windows/using-Defender
  • Spybot Search & Destroy:
    http://www.safer-networking.org/en/index.html
  • Kasperky free Security Scan:
    http://www.Kaspersky.com/security-scan

  Alternatively, you can write a check for an infection rootkit TDSSKiller.

  • Utility anti-rootkit TDSSKiller:
    http://support.Kaspersky.com/5350?El=88446

  See also:

  • "Spyware on Windows": http://kb.mozillazine.org/Popups_not_blocked
  • https://support.Mozilla.org/KB/troubleshoot-Firefox-issues-caused-malware

• Update of the data in the Table using XMLTYPE DATA

  I did insertions using XMLTYPE data but have never done it and update. Can someone give me some advice?

  PROCEDURE ADD_LABORDER_CODES)

  IN_ORDERCODESXML IN CLOB DEFAULT NULL,

  Number of OUT OUT_AFFECTEDROWS

  )

  AS

  X SYS. XMLTYPE;

  BEGIN

  X: = SYS. XMLTYPE. CREATEXML (IN_ORDERCODESXML);

  INSERT INTO MAINT_LABORD_CODES)

  INSERT INTO MAINT_LABORD_CODES)

  LABORD_CODE_ID,

  COMPENDIUM_ID,

  ORDER_CODE,

  ORDER_DESC,

  ACTIVE,

  TIMESTAMP,

  MODIFIED_BY)

  SELECT MLOCDS_SEQ. NEXTVAL,

  EXTRACTVALUE (VALUE (MLOC), '/ ORDERCODE/COMPENDIUM_ID') AS COMPENDIUM_ID,

  EXTRACTVALUE (VALUE (MLOC), '/ ORDERCODE/ORDER_CODE') AS ORDER_CODE,

  EXTRACTVALUE (VALUE (MLOC), '/ ORDERCODE/ORDER_DESC') AS ORDER_DESC,.

  EXTRACTVALUE (VALUE (MLOC), '/ ORDERCODE/LOINC_CODE') AS LOINC_CODE,

  EXTRACTVALUE (VALUE (MLOC), '/ ORDERCODE/ACTIVE') AS ACTIVE.

  EXTRACTVALUE (VALUE (MLOC), '/ ORDERCODE/TIMESTAMP') AS TIMESTAMP.

  EXTRACTVALUE (VALUE (MLOC), '/ ORDERCODE/MODIFIED_BY') AS MODIFIED_BY

  TABLE (XMLSEQUENCE (EXTRACT(X,'/ORDERCODES/ORDERCODE'))) NMCO;

  OUT_AFFECTEDROWS: = NUMBER OF ROWS SQL %;

  EXCEPTION

  WHILE OTHERS THEN

  dbms_output.put_line (SQLERRM);

  RAISE_APPLICATION_ERROR (-20001, SQLERRM);

  END;

  Example of use of the FUSION-

  If the line exists in the target table (based on the COMPENDIUM_ID and ORDER_CODE values), the UPDATE is, if not to INSERT:

  declare
  
    in_ordercodesxml  clob :=
    '
    
      500
      696231
      ABO Group & Rh Type
      NULL
      12345
      Y
      2014-08-13
      1
    
  ';
  
  begin
  
    merge into maint_labord_codes t
    using (
      select compendium_id
           , order_code
           , order_desc
           , loinc_code
           , active
           , timestamp
           , modified_by
      from xmltable('/ORDERCODES/ORDERCODE'
             passing xmltype(in_ordercodesxml)
             columns COMPENDIUM_ID    number(10)    path 'COMPENDIUM_ID'
                   , ORDER_CODE       varchar2(50)  path 'ORDER_CODE'
                   , ORDER_DESC       varchar2(250) path 'ORDER_DESC'
                   , LOINC_CODE       varchar2(10)  path 'LOINC_CODE'
                   , ACTIVE           varchar2(1)   path 'ACTIVE'
                   , TIMESTAMP        date          path 'TIMESTAMP'
                   , MODIFIED_BY      number(10)    path 'MODIFIED_BY'
           )
    ) x
    on (     t.compendium_id = x.compendium_id
         and t.order_code = x.order_code )
    when matched then update
      set t.order_desc = x.order_desc
        , t.loinc_code = x.loinc_code
        , t.active     = x.active
        , t.timestamp  = x.timestamp
        , t.modified_by = x.modified_by
    when not matched then insert
    (
      labord_code_id
    , compendium_id
    , order_code
    , order_desc
    , loinc_code
    , active
    , timestamp
    , modified_by
    )
    values (
      mlocds_seq.nextval
    , x.compendium_id
    , x.order_code
    , x.order_desc
    , x.loinc_code
    , x.active
    , x.timestamp
    , x.modified_by
    );
  
  end;
  /
  

  Also note that I used XMLTABLE instead of TABLE/XMLSEQUENCE, which is much easier to use (and not deprecated in the latest versions).

  You have not precisely the date format in the TIMESTAMP element so I assumed a conform to W3C.

  If you have a problem with this part, return to a projection of VARCHAR2 and use TO_DATE with actual size.

• Insert the XML document with values of characters &lt; or &gt;

  Hello world

  Used Oracle: version 11.2.0.3.0

  I want to insert an XML document in a binary XML field. But for an element, the value includes characters such as > or <.

  Here's an example query:

  Update / * + no_xml_query_rewrite * / ms2 set mstag = XMLQuery (' copy $tmp: = .modify (for $i in $tmp/tags/fields[@id="MS"]))

  back insert nodes (< value > TEST\n & lt; Address1 & gt; < / value >) in $i) return $tmp' from mstag content back) where psid = 5900;

  But because we're using the character & in a SQL query, this request has failed. We invited to specify the 2 values.

  So, what is the way of giving the < Address1 > value to the value of the element in the example?

  Thanks in advance

  The command is not really "fail" in itself. It is just the tool that you are using recognizes '&' characters and treats the following identifiers as replacement variables.

  What tool do you use?

  For example, in SQL * Plus you can disable this behavior through the SET DEFINE OFF or SET SCAN OFF command.

• Addition of several xml using Insertxmlbefore in a single query. Help, please.

  Oracle version

  Oracle Database 11 g Enterprise Edition Release 11.2.0.4.0 - 64 bit Production

  PL/SQL Release 11.2.0.4.0 - Production

  CORE Production 11.2.0.4.0

  AMT for Linux: Version 11.2.0.4.0 - Production

  NLSRTL Version 11.2.0.4.0 - Production

  I have an xml where I need to insert a new xml basedupon the partyID.

  for each partyid in the need to xmoutput to insert representative data.

  with t as (
  select 1 source_no , xmltype('<report>
                     <partyReported partyId="1">
                       <name> TEST123</name>
                    </partyReported>
                    <partyReported partyId="2">
                       <name> TEST456</name>
                    </partyReported>
                 </report>')
                 xmloutput
                 from dual
              )
            ,
            t_member as (
             select 1 candidate_no, xmltype('<Representative>
                                         <name> rep123 </name>
                                </Representative>') rep_xml from dual
            union all
            select 2 , xmltype('<Representative>
                                         <name> rep456 </name>
                                </Representative>') from dual
          )
          , t_rep_member as(
          SELECT
            source_no
          , X.*
          , xmloutput
       FROM
            t
          , XMLTABLE ('/report/partyReported' passing xmloutput COLUMNS candidate_no INTEGER PATH '@partyId', candidate_idx FOR ordinality ) AS X
       )
       , all_data
       as (
       select sourcE_no, a.candidate_no, xmloutput, rep_xml from t_rep_member a, t_member b
       where a.candidate_no = b.candidate_no
       )
       select source_no, MergeRepXml(repxml(candidate_no,rep_xml,xmloutput)) final_xml from all_data
       group by sourCe_no;
  
  

  source_no, candidate_no, candidate_idx, xmloutput
  
  1    1    1    "<report>                   <partyReported partyId="1">                     <name> TEST123</name>                  </partyReported>                   <partyReported partyId="2">                     <name> TEST456</name>                  </partyReported>                </report>"
  1    2    2    "<report>                   <partyReported partyId="1">                     <name> TEST123</name>                  </partyReported>                   <partyReported partyId="2">                     <name> TEST456</name>                  </partyReported>                </report>"
  
  
  
  

  I solved this problem, but I get an error message when I try to do 1000 of them.

  ERROR:

  ORA-22813: value of the operand is greater than the limits of the system

  create or replace
  type RepXml as object( candidate_no number, rep_xml xmltype, output xmltype)
  /
  create or replace
  type RepXmltab as table of RepXml
  /
  
  
  
  
  
  create or replace
  type MERGEREP as object
  (
    -- string varchar2(4000),                  -- deleted
    val_table  RepXmltab ,           -- added
  
  
    static function ODCIAggregateInitialize
      ( sctx in out MERGEREP )
      return number ,
  
    member function ODCIAggregateIterate
      ( self  in out MERGEREP ,
        value in     RepXml
      ) return number ,
  
    member function ODCIAggregateTerminate
      ( self        in  MERGEREP,
        returnvalue out xmltype,
        flags in number
      ) return number ,
  
    member function ODCIAggregateMerge
      ( self in out MERGEREP,
        ctx2 in     MERGEREP
      ) return number
  );
  
  
  create or replace
  type body MERGEREP
  is
  
    static function ODCIAggregateInitialize
    ( sctx in out MERGEREP )
    return number
    is
    begin
  
       sctx := MERGEREP( repxmltab(null,null,null) );
       --val_table:= repxmltab(null);
  
      return ODCIConst.Success ;
  
    end;
  
    member function ODCIAggregateIterate
    ( self  in out MERGEREP ,
      value in     RepXml
    ) return number
    is
    begin
  
      self.val_table.extend;
      self.val_table(self.val_table.count) := value;
  
      return ODCIConst.Success;
    end;
  
    member function ODCIAggregateTerminate
    ( self        in  MERGEREP ,
      returnvalue out xmltype ,
      flags       in  number
    ) return number
    is
  
      v_data  xmltype ;
  
    begin
  
  v_data:= null;
  
  for x in (
       select candidate_no,rep_xml,output
        from   table(val_table)
        )
        loop
        v_data:= x.output;
  
        end loop;
             
     
      for x in
      (
  select candidate_no,rep_xml,output
        from   table(val_table)
        )
      loop
      select insertxmlbefore(v_data,'/report/partyReported[@partyId="'||x.candidate_no||'"]', x.rep_xml) into v_data from dual;
  null;
      end loop;
  
      returnValue := ( v_data) ;
  
  v_data:= null;
  
      return ODCIConst.Success;
  
    end;
  
    member function ODCIAggregateMerge
    ( self in out MERGEREP ,
      ctx2 in     MERGEREP
    ) return number
    is
    begin
  
      for i in 1 .. ctx2.val_table.count
      loop
        self.val_table.extend;
        self.val_table(self.val_table.count) := ctx2.val_table(i);
      end loop;
  
  
  
      return ODCIConst.Success;
  
          end;
    end;
  /
  
  create or replace
  function MergeRepXml
    ( input RepXml )
    return xmltype
    deterministic
    parallel_enable
    aggregate using MERGEREP
  ;
  /
  
  

  With my limited knowledge, I tried to write recursive with clause and achieve the expected results, but many lines are displayed.

  WITH
       t1 AS
       ( SELECT * FROM aps_extendedoutput WHERE source_no = 261177
       )
       --     select * from t1;
     , t AS
       (SELECT
            source_no
          , X.*
          , xmloutput
       FROM
            t1
          , XMLTABLE ('/report/role/partyReported' passing xmloutput COLUMNS candidate_no INTEGER PATH '@partyId', candidate_idx FOR ordinality ) AS X
       )
       --     select * from t;
     , all_data AS
       (SELECT
            /*+ materialize */
            t.candidate_no
          , rep_xml
          , source_no
          , t.candidate_no
          , xmloutput
       FROM
            t
          , aps_reppartyxml t_p
       WHERE
            t.candidate_no = t_p.candidate_no
       )
       --   select * from all_data;
     ,recursive_data(candidate_i, xmloutput, source_no, candidate_no) AS
       (SELECT
            1 candidate_i
            , xmloutput
          ,  source_no
          , 0 candidate_no
       FROM
           t1
       UNION ALL
       SELECT
            candidate_i + 1
            , insertxmlbefore(rd.xmloutput,'/report/role/partyReported[@partyId="'
            ||ad.candidate_no
            ||'"]', ad.rep_xml) xmloutput
          , ad.source_no
          , rd.candidate_no
       FROM
            all_Data ad
          , recursive_data rd
       WHERE
            ad.sourcE_no        = rd.source_no and
             candidate_i     < 3
       )
  SELECT *  FROM recursive_data;
  
  

  for example

  SQL> create table t_member as (
    2  select 1 candidate_no, xmltype('
    3                                rep123 
    4                      ') rep_xml from dual
    5  union all
    6  select 2 , xmltype('
    7                                rep456 
    8                      ') from dual
    9  );
  
  Table created.
  
  SQL>
  SQL> set long 5000
  SQL>
  SQL> with t (source_no, xmloutput) as (
    2    select 1
    3         , xmltype('
    4                      
    5                          TEST123
    6                      
    7                      
    8                          TEST456
    9                      
   10                   ')
   11    from dual
   12  )
   13  select t.source_no
   14       , xmlserialize(document
   15           xmlquery(
   16             'copy $d := .
   17              modify (
   18                for $i in $d/report/partyReported
   19                  , $j in fn:collection("oradb:/DEV/T_MEMBER")/ROW
   20                where $j/CANDIDATE_NO = $i/@partyId
   21                return insert node $j/REP_XML/* before $i
   22              )
   23              return $d'
   24             passing t.xmloutput
   25             returning content
   26           )
   27           indent
   28         ) as final_xml
   29  from t ;
  
   SOURCE_NO FINAL_XML
  ---------- --------------------------------------------------------------------------------
           1 
               
                  rep123 
               
               
                  TEST123
               
               
                  rep456 
               
               
                  TEST456
               
             
  

• Oracle XML INSERTCHILDXML

  Hello

  I want to add a child node of XML in an XML column in the dependence, when the top element has a special attribute value. The XML Document looks like this:

  < root_element >

  <>node

  < name subnode = 'a' / >

  < name subnode = 'b' / >

  < name subnode = 'c' / >

  < / node >

  < / root_elemnt >

  If I want to insert a child node for the subnode with the name of c, so that it looks like after:

  < root_element >

  <>node

  < name subnode = 'a' / >

  < name subnode = 'b' / >

  < name subnode = 'c' >

  < new_element / >

  < / subnode >

  < / node >

  < / root_elemnt >

  I tried the following SQL statement, but it did not work:

  update from table_a a SET a.DOCUMENT =

  INSERTCHILDXML (a.DOCUMENT, ' / root_element/node/subnode/@name=c 'new_element", XMLTYPE ('blabla < new_element > < / new_element > '))

  where...;

  The statement is executable, I get no errors-XPATH-Expression...

  I hope you can help me.

  Heiko

  INSERTCHILDXML (a.DOCUMENT, ' / root_element/node/subnode/@name=c 'new_element", XMLTYPE ('Bond'))

  Besides what you have posted is not syntactically correct, you also have the parent XPath expression wrong.

  You want a predicate, not a step of the attribute.

  Here is the correct statement using insertChildXML:

  update table_a a
  set a.doc =
      insertchildxml(
        a.doc
      , '/root_element/node/subnode[@name="c"]'
      , 'new_element'
      , xmlparse(document 'blabla')
      )
  where id = 1 ;
  

  However, Oracle XML DML functions such as insertChildXML are deprecated.

  Use XQuery Update:

  SQL> drop table table_a purge;
  
  Table dropped
  
  SQL> create table table_a (id int, doc xmltype);
  
  Table created
  
  SQL>
  SQL> insert into table_a values (1, xmlparse(document
    2  '
    3       
    4            
    5            
    6            
    7       
    8  '));
  
  1 row inserted
  
  SQL> commit;
  
  Commit complete
  
  SQL>
  SQL> update table_a a
    2  set a.doc =
    3      xmlquery(
    4        'copy $d := .
    5         modify ( insert node $new into $d/root_element/node/subnode[@name="c"] )
    6         return $d'
    7        passing a.doc
    8              , xmlparse(document 'blabla') as "new"
    9        returning content
   10      )
   11  where id = 1 ;
  
  1 row updated
  
  SQL>
  SQL> select xmlserialize(document doc) from table_a where id = 1;
  
  XMLSERIALIZE(DOCUMENTDOC)
  --------------------------------------------------------------------------------
  
    
      
      
      
        blabla
      
    
  
  

• XQuery and relational queries with the RAW data

  So, I have a table like this setting:
  

  CREATE TABLE book_frag
  (
    book_id RAW(16),
    xpath_id INT,
    last_modified_ts TIMESTAMP,
    created_date TIMESTAMP,
    book_data XmlType,
    PRIMARY KEY(book_id, xpath_id),
    FOREIGN KEY (book_id) REFERENCES book(book_id),
    FOREIGN KEY (xpath_id) REFERENCES book_xpath(xpath_id)
  );
  
  CREATE TABLE book_xpath
  (
    xpath_id INT,
    book_element_name VARCHAR2(32),
    PRIMARY KEY(xpath_id)
  );

  I am trying a query which will regroup the different XMLTypes book_frag into one. If I use this query, it works, but it is slow and sometimes times out:
  

  SELECT 
      e.book_id,
      
      e.title_txt,
      e.subj_txt,
      
      e.target_mkt,
      e.target_lcl_mkt,
      
      e.catg,
      e.lang_code,
      e.out_of_stock_reas,
      
      e.sucsr_title,
      e.sucsr_sub_txt,
      
      e.orig_copy_ref,
      e.orig_subj_txt,
      
      XMLQuery(
      'declare namespace invtdata="http://www.mrbook.com/InventoryData";
      
       copy $invtData := $orig/invtdata:inventory
       
       modify
       (
         for $bkRow in fn:collection("oradb:/BOOKSCHEMA/BOOK_FRAG")
         let $bk := $mpfRow/ROW/BOOK_DATA/node()
         where xs:string($bkRow/ROW/ID)=$bookId
         return insert node $bk into $invtData
       ) return $invtData'
       PASSING XmlType('<inventory xmlns="http://www.mrbook.com/InventoryData" />') AS "orig",
       CAST(e.book_id AS VARCHAR2(64)) AS "bookId" RETURNING CONTENT
       ) AS invt_data
  FROM
    entity e
  WHERE e.title_txt = 'Finite Variable Analysis' OR e.subj_txt = 'Finite Mathematics';

  However, if I try this, it works at all.
  

  SELECT 
      e.book_id,
      
      e.title_txt,
      e.subj_txt,
      
      e.target_mkt,
      e.target_lcl_mkt,
      
      e.catg,
      e.lang_code,
      e.out_of_stock_reas,
      
      e.sucsr_title,
      e.sucsr_sub_txt,
      
      e.orig_copy_ref,
      e.orig_subj_txt,
      
      XMLQuery(
      'declare namespace invtdata="http://www.mrbook.com/InventoryData";
      
       copy $invtData := $orig/invtdata:inventory
       
       modify
       (
         for $bkRow in fn:collection("oradb:/BOOKSCHEMA/BOOK_FRAG")
         let $bk := $mpfRow/ROW/BOOK_DATA/node()
         where $bkRow/ROW/ID=$bookId
         return insert node $bk into $invtData
       ) return $invtData'
       PASSING XmlType('<inventory xmlns="http://www.mrbook.com/InventoryData" />') AS "orig",
       e.id AS "bookId" RETURNING CONTENT
       ) AS invt_data
  FROM
    entity e
  WHERE e.title_txt = 'Finite Variable Analysis' OR e.subj_txt = 'Finite Mathematics';

  The error I get with this query is:

  ORA-00932: inconsistent datatypes: expected - got BINARY
  00932. 00000 -  "inconsistent datatypes: expected %s got %s"
  *Cause:    
  *Action:
  Error at Line: 52 Column: 5

  Is it possible to leave the XQuery to use the RAW data type (16) without casting?

  KnightOfBlueArmor wrote:
  Is it possible to leave the XQuery to use the RAW data type (16) without casting?

  You are overloading the things, no need to XQuery Update in this case.
  As far as I understand it, all you need is XMLAgg:

  SELECT e.book_id
       , e.title_txt
       , e.subj_txt
       , XMLElement("inventory"
         , XMLAttributes('http://www.mrbook.com/InventoryData' as "xmlns")
         , XMLAgg( b.book_data )
         ) as inv_data
  FROM entity e
       LEFT OUTER JOIN book_frag b
                    ON b.book_id = e.book_id
  WHERE e.title_txt = 'Finite Variable Analysis'
     OR e.subj_txt = 'Finite Mathematics'
  GROUP BY e.book_id
         , e.title_txt
         , e.subj_txt
  ;
  

  (the OUTER JOIN may or may not be necessary, depending on the relationship between the two tables)

  Published by: odie_63 on March 27, 2013 23:16