Saving Large XML as Table in Database

Hi All,
I have a requirement like I need to store large XML (when transformed to rows in table, more than 15K rows will be there),
Can I have any sample API's to do the same, (performance matters)
Thanks,

Check out the FAQ for the XMLDB forum. You should find some examples to get you started.
HTH,
Chris

Similar Messages

  • Error when saving large XML Forms - XFBuilder

    When creating large XML Forms (many entry fields, many links when metadata properties), the xfbuilder aborts the operation with an error message stating a networkcommunication failure / server response time-out. When some fields are deleted, the saving takes some time, but will prevail...
    Does anybody have experienced these problems as well and is there a solution so that larger xmlforms can be created?? Currently the xmlforms producing errors have over 100 entry fields.
    Rgds Caspar

    OK, I found it: when editing the Project Options, the server time out can be set. In NW04 edition this is set default to 120; in EP60 SP2 this value is empty. Increasing the value (e.g. 180) does the trick!

  • Import Large XML File to Table

    I have a large (819MB) XML file I'm trying to import into a table in the format:
    <ROW_SET>
    <ROW>
    <column_name>value</column_name>
    </ROW>
    <ROW>
    <column_name>value</column_name>
    </ROW>
    </ROW_SET>
    I've tried importing it with xmlsequence(...).extract(...) and ran into the number of nodes exceed maximum error.
    I've tried importing it with XMLTable(... passing XMLTYPE(bfilename('DIR_OBJ','large_819mb_file.xml'), nls_charset_id('UTF8'))) and I gave up after it ran for 15+ hours ( COLLECTION ITERATOR PICKLER FETCH issue ).
    I've tried importing it with:
    insCtx := DBMS_XMLStore.newContext('schemaname.tablename');
    DBMS_XMLStore.clearUpdateColumnList(insCtx);
    DBMS_XMLStore.setUpdateColumn(insCtx,'column1name');
    DBMS_XMLStore.setUpdateColumn(insCtx,'columnNname');
    ROWS := DBMS_XMLStore.insertXML(insCtx, XMLTYPE(bfilename('DIR_OBJ','large_819mb_file.xml'), nls_charset_id('UTF8')));
    and ran into ORA-04030: out of process memory when trying to allocate 1032 bytes (qmxlu subheap,qmemNextBuf:alloc).
    All I need to do is read the XML file and move the data into a matching table in a reasonable time. Once I have the data in the database, I no longer need the XML file.
    What would be the best way to import large XML files?
    Oracle Database 11g Release 11.2.0.1.0 - 64bit Production
    PL/SQL Release 11.2.0.1.0 - Production
    "CORE     11.2.0.1.0     Production"
    TNS for Linux: Version 11.2.0.1.0 - Production
    NLSRTL Version 11.2.0.1.0 - Production

    This (rough) approach should work for you.
    CREATE TABLE HOLDS_XML
            (xml_col XMLTYPE)
          XMLTYPE xml_col STORE AS SECUREFILE BINARY XML;
    INSERT INTO HOLDS_XML
    VALUES (xmltype(bfilename('DIR_OBJ','large_819mb_file.xml'), nls_charset_id('UTF8')))
    -- Should be using AL32UTF8 for DB character set with XML
    SELECT ...
      FROM HOLD_XML HX
           XMLTable(...
              PASSING HX.xml_col ...)How it differs from your approach.
    By using the HOLDS_XML table with SECUREFILE BINARY XML storage (which became the default in 11.2.0.2) we are providing a place for Oracle to store a parsed version of the XML. This allows the XML to be stored on disk instead of in memory. Oracle can then access the needed pieces of XML from disk by streaming them instead of holding the whole XML in memory and parsing it repeatedly to find the information needed. That is what COLLECTION ITERATOR PICKLER FETCH means. A lot of memory work. You can search on that term to learn more about it if needed.
    The XMTable approach then simply reads this XML from disk and should be able to parse the XML with no issue. You have the option of adding indexes to the XML, but since you are simply reading it all one time and tossing it, there is no advantage to indexes (most likely)

  • How to insert large xml data into database tables.

    Hi all,
    iam new to xml. i want to insert data in xml file to my database tables.but the xml file size is very large. performance is also one of the issue. can anybody please tell me the procedure to take xml file from the server and insert into my database tables.
    Thanks in advance

    Unfortunately posting very generic questions like this in the forum tends not to be very productive for you, me or the other people who read the forum. It really helps everyone if you take a little time to review existing posts and their answers before starting new threads which replicate subjects that have already been discussed extensively in previous threads. This allows you to ask more sensible questions (eg, I'm using this approach and encountering this problem) rather than extremely generic questions that you can answer yourself by spending a little time reviewing existings posts or using the forum's search feature.
    Also in future your might want to try being a little more specific before posting questions
    Eg Define "very large". I know of customers who thing very large is 100K, and customers who think 4G is medium. I cannot tell from your post what size your files are.
    What is the content of the file. Is it going to be loaded into a single record, or a a single table, or will it need to be loaded into multiple records in a single table or multiple records in multiple tables ?
    Do you really need to load the data into exsiting relational tables or could your application work with relational views of the XML Content.
    Finally which release of the database are you working with.
    Define performance. Is it reasonable to expect to process this kind of document on this machine (Make, memory, #number of CPUs, CPU Speed, number of discs) in this period of time.
    WRT to your original question. If you take a few minutes to search this forum you will find a very large number of threads with very similar titles to yours. These theads document a number of different approaches that can be used to solve this problem.
    I suggest you start by looking for threads that cover topics like DBMS_XMLSTORE, XMLTable(), Relational Views of XML content, loading XML content in relational tables.

  • Insert XML file into Relational database model without using XMLTYPE tables

    Dear all,
    How can I store a known complex XML file into an existing relational database WITHOUT using xmltypes in the database ?
    I read the article on DBMS_XMLSTORE. DBMS_XMLSTORE indeed partially bridges the gap between XML and RDBMS to a certain extent, namely for simply structured XML (canonical structure) and simple tables.
    However, when the XML structure will become arbitrary and rapidly evolving, surely there must be a way to map XML to a relational model more flexibly.
    We work in a java/Oracle10 environment that receives very large XML documents from an independent data management source. These files comply with an XML schema. That is all we know. Still, all these data must be inserted/updated daily in an existing relational model. Quite an assignment isn't it ?
    The database does and will not contain XMLTYPES, only plain RDBMS tables.
    Are you aware of a framework/product or tool to do what DBMS_XMLSTORE does but with any format of XML file ? If not, I am doomed.
    Constraints : Input via XML files defined by third party
    Storage : relational database model with hundreds of tables and thousands of existing queries that can not be touched. The model must not be altered.
    Target : get this XML into the database on a daily basis via an automated process.
    Cheers.
    Luc.

    Luc,
    your Doomed !
    If you would try something like DBMS_XMLSTORE, you probably would be into serious performance problems in your case, very fast, and it would be very difficult to manage.
    If you would use a little bit of XMLType stuff, you would be able to shred the data into the relational model very fast and controlable. Take it from me, I am one of those old geezers like Mr. Tom Kyte way beyond 40 years (still joking). No seriously. I started out as a classical PL/SQL, Forms Guy that switched after two years to become a "DBA 1.0" and Mr Codd and Mr Date were for years my biggest hero's. I have the utmost respect for Mr. Tom Kyte for all his efforts for bringing the concepts manual into the development world. Just to name some off the names that influenced me. But you will have to work with UNSTRUCTURED data (as Mr Date would call it). 80% of the data out there exists off it. Stuff like XMLTABLE and XML VIEWs bridge the gap between that unstructured world and the relational world. It is very doable to drag and drop an XML file into the XMLDB database into a XMLtype table and or for instance via FTP. From that point on it is in the database. From there you could move into relational tables via XMLTABLE methods or XML Views.
    You could see the described method as a filtering option so XML can be transformed into relational data. If you don't want any XML in your current database, then create an small Oracle database with XML DB installed (if doable 11.1.0.7 regarding the best performance --> all the new fast optimizer stuff etc). Use that database as a staging area that does all the XML shredding for you into relational components and ship the end result via database links and or materialized views or other probably known methodes into your relational database that isn't allowed to have XMLType.
    This way you would keep your realtional Oracle database clean and have the Oracle XML DB staging database do all the filtering and shredding into relational components.
    Throwing the XML DB option out off the window beforehand would be like replacing your Mercedes with a bicycle. With both you will be able to travel the distance from Paris to Rome, but it will take you a hell of lot longer.
    :-)

  • Hello Anybody, I have a question. Can any of you please suggest me how to make an xml file from the database table with all the rows? Note:- I am having the XSD Schema file and the resulted XML file should be in that XSD format only.

    Hello Anybody, I have a question. Can any of you please suggest me how to make an xml file from the database table with all the records?
    Note:- I am having the XSD Schema file and the resulted XML file should be in that XSD format only.

    The Oracle documentation has a good overview of the options available
    Generating XML Data from the Database
    Without knowing your version, I just picked 11.2, so you made need to look for that chapter in the documentation for your version to find applicable information.
    You can also find some information in XML DB FAQ

  • A query while  importing  an XML file into a Database Table

    Hi,
    I am Creating an ODI Project to import an XML file into a Database Table.With the help of the following link
    http://www.oracle.com/webfolder/technetwork/tutorials/obe/fmw/odi/odi_11g/odi_project_xml-to-table/odi_project_xml-to-table.htm
    I am facing a problem while creating Physical Schema for the XML Source Model.
    For the
    Schema(Schema)
    and
    Schema(Work Schema) field they have selected GEO_D.
    What is GEO_D here??
    or
    What I have to select here?
    1) schema of the xml file (NB:-I havn't created any .xsd or .dtd file for my .xml file)
    or
    2)my target servers schema
    Please tell me what I'll do??
    Thanks

    and
    Schema(Work Schema) field they have selected GEO_D.
    What is GEO_D here??This is the schema name which is specified in the XML file .
    What I have to select here?
    1) schema of the xml file (NB:-I havn't created any .xsd or .dtd file for my .xml file)Yes
    2)my target servers schema
    Please tell me what I'll do??
    Thanks

  • XML to table mapping in oracle database

    Hi,
    I am having 2 modules in my application:
    First module takes the XML file, parse it, fetches data, creates a table in database and stores the data into that table.
    Second module does vice versa. it takes a table, retrives data and write out an XML file.
    In first module I use DOM parser. But I want to use Oracle XSU for the second module. But it seems that I can not use oracle XSU and DOM togather.
    when I include xmlparserv2.jar in class path I can not run my first module.
    Is it possible that I can use dom for first module and XSU for second module?
    It gives me error like ..
    Exception in thread "AWT-EventQueue-0" java.lang.NoClassDefFoundError: oracle/xml/parser/v2/XMLText
    at oracle.xml.parser.v2.NonValidatingParser.<init>(NonValidatingParser.java:139)
    at oracle.xml.parser.v2.XMLParser.<init>(XMLParser.java:122)
    at oracle.xml.parser.v2.DOMParser.<init>(DOMParser.java:82)
    at oracle.xml.jaxp.JXDocumentBuilder.<init>(JXDocumentBuilder.java:71)
    at oracle.xml.jaxp.JXDocumentBuilderFactory.newDocumentBuilder(JXDocumentBuilderFactory.java:72)
    BUT if I remove xmlparserv2.jar from classpath then it works fine...
    I would really appreciate if anyone can throw light ...
    Thanks in advance
    rp

    Hi,
    I had somewhat a similar problem.
    Did you use oracle.xdb.XMLType.getDOM()?
    getDOM() needs xmlparserv2.jar, which only provides a DOM implementation that is DOM 1.0 and Core 2.0 compliant.
    In my case, when I used DOM 3.0 methods in org.w3c.dom, I ran into the "method not found" type of error.
    My solution was to use the DocumentBuilder that comes with the JDK to create the DOM instance in the code.
    P.

  • Create a large XML extract from relational tables - 10g

    The Oracle 10g documentation recommends using the various SQL functions such as XMLAGG,XMLELEMENT or DBMS_XMLGEN under chapter "16 Generating XML Data from the Database" for generating a XML extract file. The DBMS_XMLDOM seems like a useful structured package to create a XML formatted extract. Any idea why Oracle does not recommend the use of XMLDOM API ? Are there performance issues or likely de-support ?
    Whats the best way to produce a large XML extract?
    Thanks.

    mdrake wrote:
    No reason not to use it. It just tends to require a lot more coding than the other methods. Underneath the covers they all map into the same code-line. In general, to generate a given XML document you wlll write far fewer lines of code using the SQL/XML operators. Less Code written = less code to maintan = faster time to market.Less code to parse...?

  • Date insertion from large XML document (clob) into relation table very slow

    Hi Everybody!
    I'm working with Oracle 9.2.0.5 on Microsoft Windows Server 2003 Enterprise Edition.
    The server (a test server) is a Pentium 4 2.8 GHz, 1GB of RAM.
    I use a procedure called PARITOP_TRAITERXMLRESULTMASSE to insert the data contained in the pXMLDOC clob parameter in the table pTABLENAME. (You can see the format of the XML document below). The first step on this procedure is to verify that the XML document is not empty. If not, the procedure needs to add a node in the document, in every <ROW> tag. This added node is named “RST_ID”. It’s the foreign key of each record. I can retrieve the value of each <RST_ID> node in an other table in which the data has been previously added (by the calling procedure). When each of the <ROW> elements has been treated, the PARITOP_INSERTXML procedure is called. This procedure uses DBMS_XMLSAVE.INSERTXML to insert the data contained in the XML document in the specified table.
    (Below, you can see the code of my procedures.)
    With this information, can you tell me why this treatment is very very very slow with a large XML document and how I can improve it?
    Thank you for your help!
    Anne-Marie
    CREATE OR REPLACE PROCEDURE "PARITOP_TRAITERXMLRESULTMASSE" (
    pPRC_ID IN PARITOP_PARC.PRC_ID%TYPE,
    pRST_MONDE IN PARITOP_RESULTAT.RST_MONDE%TYPE,
    pXMLDOC IN CLOB,
    pTABLENAME IN VARCHAR2)
    AS
    Objectif :Insérer le contenu du XML passé en paramètre (pXMLDOC) à la table passée en paramètre (pTABLENAME)
    La table passée en paramètre doit être une table ayant comme clé étrangère le champs "RST_ID" .
    (Le noeud RST_ID est donc ajouté à tous les document XML. Ce rst_id est
    déterminé à partir de la table PARITOP_RESULTAT grâce à pPRC_ID et
    pRstMonde fournis en paramètre)
    result_doc CLOB;
    XMLDOMDOC XDB.DBMS_XMLDOM.DOMDOCUMENT;
    NODE_ROWSET DBMS_XMLDOM.DOMNODE;
    NODE_ROW DBMS_XMLDOM.DOMNODE;
    vUE_ID PARITOP_RESULTAT.UE_ID%TYPE;
    vRST_ID PARITOP_RESULTAT.RST_ID%TYPE;
    nodeList DBMS_XMLDOM.DOMNODELIST;
    BEGIN
    BEGIN
    vUE_ID := 0;
    vRST_ID := 0;
    XMLDOMDOC := DBMS_XMLDOM.NEWDOMDOCUMENT(pXMLDOC);
    IF NOT GESTXML_PKG.FN_PARITOP_DOCUMENT_IS_NULL(XMLDOMDOC) THEN
    NODE_ROWSET := DBMS_XMLDOM.item(DBMS_XMLDOM.GETCHILDNODES (DBMS_XMLDOM.MAKENODE(XMLDOMDOC)),0);
    for i in 0..dbms_xmldom.getLength(DBMS_XMLDOM.getchildnodes(NODE_ROWSET))-1 loop
    NODE_ROW := DBMS_XMLDOM.ITEM(DBMS_XMLDOM.GETCHILDNODES(NODE_ROWSET), i) ;
    nodeList := DBMS_XMLDOM.GETELEMENTSBYTAGNAME(DBMS_XMLDOM.makeelement(NODE_ROW) , 'UE_ID');
    IF vUE_ID <> DBMS_XMLDOM.GETNODEVALUE(DBMS_XMLDOM.GETFIRSTCHILD(DBMS_XMLDOM.ITEM(nodeList, 0))) THEN
    vUE_ID := DBMS_XMLDOM.GETNODEVALUE(DBMS_XMLDOM.GETFIRSTCHILD(DBMS_XMLDOM.ITEM(nodeList, 0)));
    --on ramasse le rst_id
    SELECT RST_ID INTO vRST_ID
    FROM PARITOP_RESULTAT RST
    WHERE RST.PRC_ID = pPRC_ID
    AND RST.UE_ID = vUE_ID
    AND RST.RST_MONDE = pRST_MONDE
    AND RST_A_SUPPRIMER = 0;
    END IF;
    GESTXML_PKG.PARITOP_ADDNODETOROW(XMLDOMDOC, NODE_ROW, 'RST_ID', vRST_ID);
    end loop;
    RESULT_DOC := ' '; --à garder, pour ne pas que ca fasse d'erreur lors du WriteToClob.
    dbms_xmldom.writeToClob(DBMS_XMLDOM.MAKENODE(XMLDOMDOC), RESULT_DOC);
    --Insertion du document XML dans la table "tableName"
    GESTXML_PKG.PARITOP_INSERTXML(RESULT_DOC, pTABLENAME);
    DBMS_XMLDOM.FREEDOCUMENT( XMLDOMDOC);
    end if;
    EXCEPTION
    […exception treatement…]
    END;
    END;
    The format of a XML clob is :
    <ROWSET>
    <ROW>
    <PRC_ID>193</PRC_ID>
    <UE_ID>8781</UE_ID>
    <VEN_ID>6223</VEN_ID>
    <RST_MONDE>0</RST_MONDE>
    <CMP_SELMAN>0</CMP_SELMAN>
    <CMP_INDICESELECTION>92.307692307692307</CMP_INDICESELECTION>
    <CMP_PVRES>94900</CMP_PVRES>
    <CMP_PVAJUSTE>72678.017699115066</CMP_PVAJUSTE>
    <CMP_PVAJUSTEMIN>72678.017699115095</CMP_PVAJUSTEMIN>
    <CMP_PVAJUSTEMAX>72678.017699115037</CMP_PVAJUSTEMAX>
    <CMP_PV>148000</CMP_PV>
    <CMP_VALROLE>129400</CMP_VALROLE>
    <CMP_PVRESECART>4790</CMP_PVRESECART>
    <CMP_PVRHAB>101778.01769911509</CMP_PVRHAB>
    <CMP_UTILISE>1</CMP_UTILISE>
    <CMP_TVM>1</CMP_TVM>
    <CMP_PVA>148000</CMP_PVA>
    </ROW>
    <ROW>
    <PRC_ID>193</PRC_ID>
    <UE_ID>8781</UE_ID>
    <VEN_ID>6235</VEN_ID>
    <RST_MONDE>0</RST_MONDE>
    <CMP_SELMAN>0</CMP_SELMAN>
    <CMP_INDICESELECTION>76.92307692307692</CMP_INDICESELECTION>
    <CMP_PVRES>117800</CMP_PVRES>
    <CMP_PVAJUSTE>118080</CMP_PVAJUSTE>
    <CMP_PVAJUSTEMIN>118080</CMP_PVAJUSTEMIN>
    <CMP_PVAJUSTEMAX>118080</CMP_PVAJUSTEMAX>
    <CMP_PV>172000</CMP_PV>
    <CMP_VALROLE>134800</CMP_VALROLE>
    <CMP_PVRESECART>0</CMP_PVRESECART>
    <CMP_PVRHAB>147180</CMP_PVRHAB>
    <CMP_UTILISE>1</CMP_UTILISE>
    <CMP_TVM>1</CMP_TVM>
    <CMP_PVA>172000</CMP_PVA>
    </ROW>
    </ROWSET>
    PARITOP_COMPARABLE TABLE :
    RST_ID NUMBER(10) NOT NULL,
    VEN_ID NUMBER(10) NOT NULL,
    CMP_SELMAN NUMBER(1) NOT NULL,
    CMP_UTILISE NUMBER(1) NOT NULL,
    CMP_INDICESELECTION FLOAT(53) NOT NULL,
    CMP_PVRES FLOAT(53) NULL,
    CMP_PVAJUSTE FLOAT(53) NULL,
    CMP_PVRHAB FLOAT(53) NULL,
    CMP_TVM FLOAT(53) NULL
    ROCEDURE PARITOP_INSERTXML (xmlDoc IN clob, tableName IN VARCHAR2)
    AS
    insCtx DBMS_XMLSave.ctxType;
    rowss number;
    BEGIN
    --permet d'insérer les champs du XML dans la table passée en paramètre.
    --il suffit que les champs XML aient le même nom que les champs de la table
    BEGIN
    insCtx := DBMS_XMLSave.newContext(tableName); -- get context handle
    DBMS_XMLSAVE.SETDATEFORMAT( insCtx, 'yyyy-MM-dd HH:mm:ss');--attention, case sensitive
    DBMS_XMLSAVE.setIgnoreCase(insCtx, 1);
    rowss := DBMS_XMLSAVE.INSERTXML(insCtx , xmlDoc);
    DBMS_XMLSave.closeContext(insCtx);
    EXCEPTION
    […]
    END;
    END;
    PROCEDURE PARITOP_ADDNODETOROW (
    XMLDOMDOC DBMS_XMLDOM.DOMDOCUMENT,
    NODE_ROW dbms_xmldom.DOMNode,
    NOM_NOEUD VARCHAR2,
    VALEUR_NOEUD VARCHAR2)
    AS
    --PERMET D'AJOUTER UN NOEUD AVEC 1 SEULE VALEUR DANS une ROW D'UN XML.
    --UTILE SURTOUT POUR LES CLÉS ÉTRANGÈRES
    domElemAInserer DBMS_XMLDOM.DOMELEMENT;
    NODE dbms_xmldom.DOMNode;
    NODE_TMP dbms_xmldom.DOMNode;
    BEGIN
    domElemAInserer := DBMS_XMLDOM.createElement(XMLDOMDOC, NOM_NOEUD) ;
    NODE := DBMS_XMLDOM.MAKENODE(domElemAInserer); --cast
    NODE := DBMS_XMLDOM.APPENDCHILD(NODE_ROW,NODE);
    NODE_TMP := DBMS_XMLDOM.MAKENODE(DBMS_XMLDOM.CREATETEXTNODE(XMLDOMDOC, VALEUR_NOEUD ) );
    NODE := DBMS_XMLDOM.APPENDCHILD(NODE,NODE_TMP );
    END;

    Hi Everybody!
    I'm working with Oracle 9.2.0.5 on Microsoft Windows Server 2003 Enterprise Edition.
    The server (a test server) is a Pentium 4 2.8 GHz, 1GB of RAM.
    I use a procedure called PARITOP_TRAITERXMLRESULTMASSE to insert the data contained in the pXMLDOC clob parameter in the table pTABLENAME. (You can see the format of the XML document below). The first step on this procedure is to verify that the XML document is not empty. If not, the procedure needs to add a node in the document, in every <ROW> tag. This added node is named “RST_ID”. It’s the foreign key of each record. I can retrieve the value of each <RST_ID> node in an other table in which the data has been previously added (by the calling procedure). When each of the <ROW> elements has been treated, the PARITOP_INSERTXML procedure is called. This procedure uses DBMS_XMLSAVE.INSERTXML to insert the data contained in the XML document in the specified table.
    (Below, you can see the code of my procedures.)
    With this information, can you tell me why this treatment is very very very slow with a large XML document and how I can improve it?
    Thank you for your help!
    Anne-Marie
    CREATE OR REPLACE PROCEDURE "PARITOP_TRAITERXMLRESULTMASSE" (
    pPRC_ID IN PARITOP_PARC.PRC_ID%TYPE,
    pRST_MONDE IN PARITOP_RESULTAT.RST_MONDE%TYPE,
    pXMLDOC IN CLOB,
    pTABLENAME IN VARCHAR2)
    AS
    Objectif :Insérer le contenu du XML passé en paramètre (pXMLDOC) à la table passée en paramètre (pTABLENAME)
    La table passée en paramètre doit être une table ayant comme clé étrangère le champs "RST_ID" .
    (Le noeud RST_ID est donc ajouté à tous les document XML. Ce rst_id est
    déterminé à partir de la table PARITOP_RESULTAT grâce à pPRC_ID et
    pRstMonde fournis en paramètre)
    result_doc CLOB;
    XMLDOMDOC XDB.DBMS_XMLDOM.DOMDOCUMENT;
    NODE_ROWSET DBMS_XMLDOM.DOMNODE;
    NODE_ROW DBMS_XMLDOM.DOMNODE;
    vUE_ID PARITOP_RESULTAT.UE_ID%TYPE;
    vRST_ID PARITOP_RESULTAT.RST_ID%TYPE;
    nodeList DBMS_XMLDOM.DOMNODELIST;
    BEGIN
    BEGIN
    vUE_ID := 0;
    vRST_ID := 0;
    XMLDOMDOC := DBMS_XMLDOM.NEWDOMDOCUMENT(pXMLDOC);
    IF NOT GESTXML_PKG.FN_PARITOP_DOCUMENT_IS_NULL(XMLDOMDOC) THEN
    NODE_ROWSET := DBMS_XMLDOM.item(DBMS_XMLDOM.GETCHILDNODES (DBMS_XMLDOM.MAKENODE(XMLDOMDOC)),0);
    for i in 0..dbms_xmldom.getLength(DBMS_XMLDOM.getchildnodes(NODE_ROWSET))-1 loop
    NODE_ROW := DBMS_XMLDOM.ITEM(DBMS_XMLDOM.GETCHILDNODES(NODE_ROWSET), i) ;
    nodeList := DBMS_XMLDOM.GETELEMENTSBYTAGNAME(DBMS_XMLDOM.makeelement(NODE_ROW) , 'UE_ID');
    IF vUE_ID <> DBMS_XMLDOM.GETNODEVALUE(DBMS_XMLDOM.GETFIRSTCHILD(DBMS_XMLDOM.ITEM(nodeList, 0))) THEN
    vUE_ID := DBMS_XMLDOM.GETNODEVALUE(DBMS_XMLDOM.GETFIRSTCHILD(DBMS_XMLDOM.ITEM(nodeList, 0)));
    --on ramasse le rst_id
    SELECT RST_ID INTO vRST_ID
    FROM PARITOP_RESULTAT RST
    WHERE RST.PRC_ID = pPRC_ID
    AND RST.UE_ID = vUE_ID
    AND RST.RST_MONDE = pRST_MONDE
    AND RST_A_SUPPRIMER = 0;
    END IF;
    GESTXML_PKG.PARITOP_ADDNODETOROW(XMLDOMDOC, NODE_ROW, 'RST_ID', vRST_ID);
    end loop;
    RESULT_DOC := ' '; --à garder, pour ne pas que ca fasse d'erreur lors du WriteToClob.
    dbms_xmldom.writeToClob(DBMS_XMLDOM.MAKENODE(XMLDOMDOC), RESULT_DOC);
    --Insertion du document XML dans la table "tableName"
    GESTXML_PKG.PARITOP_INSERTXML(RESULT_DOC, pTABLENAME);
    DBMS_XMLDOM.FREEDOCUMENT( XMLDOMDOC);
    end if;
    EXCEPTION
    […exception treatement…]
    END;
    END;
    The format of a XML clob is :
    <ROWSET>
    <ROW>
    <PRC_ID>193</PRC_ID>
    <UE_ID>8781</UE_ID>
    <VEN_ID>6223</VEN_ID>
    <RST_MONDE>0</RST_MONDE>
    <CMP_SELMAN>0</CMP_SELMAN>
    <CMP_INDICESELECTION>92.307692307692307</CMP_INDICESELECTION>
    <CMP_PVRES>94900</CMP_PVRES>
    <CMP_PVAJUSTE>72678.017699115066</CMP_PVAJUSTE>
    <CMP_PVAJUSTEMIN>72678.017699115095</CMP_PVAJUSTEMIN>
    <CMP_PVAJUSTEMAX>72678.017699115037</CMP_PVAJUSTEMAX>
    <CMP_PV>148000</CMP_PV>
    <CMP_VALROLE>129400</CMP_VALROLE>
    <CMP_PVRESECART>4790</CMP_PVRESECART>
    <CMP_PVRHAB>101778.01769911509</CMP_PVRHAB>
    <CMP_UTILISE>1</CMP_UTILISE>
    <CMP_TVM>1</CMP_TVM>
    <CMP_PVA>148000</CMP_PVA>
    </ROW>
    <ROW>
    <PRC_ID>193</PRC_ID>
    <UE_ID>8781</UE_ID>
    <VEN_ID>6235</VEN_ID>
    <RST_MONDE>0</RST_MONDE>
    <CMP_SELMAN>0</CMP_SELMAN>
    <CMP_INDICESELECTION>76.92307692307692</CMP_INDICESELECTION>
    <CMP_PVRES>117800</CMP_PVRES>
    <CMP_PVAJUSTE>118080</CMP_PVAJUSTE>
    <CMP_PVAJUSTEMIN>118080</CMP_PVAJUSTEMIN>
    <CMP_PVAJUSTEMAX>118080</CMP_PVAJUSTEMAX>
    <CMP_PV>172000</CMP_PV>
    <CMP_VALROLE>134800</CMP_VALROLE>
    <CMP_PVRESECART>0</CMP_PVRESECART>
    <CMP_PVRHAB>147180</CMP_PVRHAB>
    <CMP_UTILISE>1</CMP_UTILISE>
    <CMP_TVM>1</CMP_TVM>
    <CMP_PVA>172000</CMP_PVA>
    </ROW>
    </ROWSET>
    PARITOP_COMPARABLE TABLE :
    RST_ID NUMBER(10) NOT NULL,
    VEN_ID NUMBER(10) NOT NULL,
    CMP_SELMAN NUMBER(1) NOT NULL,
    CMP_UTILISE NUMBER(1) NOT NULL,
    CMP_INDICESELECTION FLOAT(53) NOT NULL,
    CMP_PVRES FLOAT(53) NULL,
    CMP_PVAJUSTE FLOAT(53) NULL,
    CMP_PVRHAB FLOAT(53) NULL,
    CMP_TVM FLOAT(53) NULL
    ROCEDURE PARITOP_INSERTXML (xmlDoc IN clob, tableName IN VARCHAR2)
    AS
    insCtx DBMS_XMLSave.ctxType;
    rowss number;
    BEGIN
    --permet d'insérer les champs du XML dans la table passée en paramètre.
    --il suffit que les champs XML aient le même nom que les champs de la table
    BEGIN
    insCtx := DBMS_XMLSave.newContext(tableName); -- get context handle
    DBMS_XMLSAVE.SETDATEFORMAT( insCtx, 'yyyy-MM-dd HH:mm:ss');--attention, case sensitive
    DBMS_XMLSAVE.setIgnoreCase(insCtx, 1);
    rowss := DBMS_XMLSAVE.INSERTXML(insCtx , xmlDoc);
    DBMS_XMLSave.closeContext(insCtx);
    EXCEPTION
    […]
    END;
    END;
    PROCEDURE PARITOP_ADDNODETOROW (
    XMLDOMDOC DBMS_XMLDOM.DOMDOCUMENT,
    NODE_ROW dbms_xmldom.DOMNode,
    NOM_NOEUD VARCHAR2,
    VALEUR_NOEUD VARCHAR2)
    AS
    --PERMET D'AJOUTER UN NOEUD AVEC 1 SEULE VALEUR DANS une ROW D'UN XML.
    --UTILE SURTOUT POUR LES CLÉS ÉTRANGÈRES
    domElemAInserer DBMS_XMLDOM.DOMELEMENT;
    NODE dbms_xmldom.DOMNode;
    NODE_TMP dbms_xmldom.DOMNode;
    BEGIN
    domElemAInserer := DBMS_XMLDOM.createElement(XMLDOMDOC, NOM_NOEUD) ;
    NODE := DBMS_XMLDOM.MAKENODE(domElemAInserer); --cast
    NODE := DBMS_XMLDOM.APPENDCHILD(NODE_ROW,NODE);
    NODE_TMP := DBMS_XMLDOM.MAKENODE(DBMS_XMLDOM.CREATETEXTNODE(XMLDOMDOC, VALEUR_NOEUD ) );
    NODE := DBMS_XMLDOM.APPENDCHILD(NODE,NODE_TMP );
    END;

  • Saving xml files in oracle database

    hi all
    anyone who could tell me how can i save xml files in the database?

    In order to provide an intelligent answer we need much more information. Please describe the problem you are trying to solve in more detail. The kind of information that would help is
    Database Version ?
    What is the source of the XML Files ?
    What is the size of the Files ?
    How are they stored prior to wanting to load them into the database, on the database server or on a client ?
    What is you preferred programming language ?
    What kind of querys to do you want to run against the files once they are stored in the database ?
    Note if your database version is pre 9i then the question can be better answered in the forum Technologies -> XML -> General
    I can move you post there if appropriate

  • Load XML file and transform to table in database

    Hi All,
    If I have a xml file as sample xml data
    <EMPLOYEES>
    - <EMP>
    <EMPNO>7369</EMPNO>
    <ENAME>SMITH</ENAME>
    <JOB>CLERK</JOB>
    <MGR>7902</MGR>
    <HIREDATE>17-DEC-80</HIREDATE>
    <SAL>800</SAL>
    </EMP>
    </EMPLOYEES>
    And I need to import to Emplyees table
    Create table Employees
    ( EMPNO number,
    ENAME varchar2(100),
    JOB varchar2(100),
    MGR number,
    HIREDATE date,
    SAL number
    How to import the xml file to the table?
    Regards,
    Hiko

    Well, if it's a complex XML you can use register a Schema and shred the XML to tables as described on the following thread:
    Re: XML file processing into oracle
    Or if it's fairly simple XML, you can just use XMLTABLE to flatten the data out...
    SQL> ed
    Wrote file afiedt.buf
      1  WITH xml_table AS
      2      (SELECT xmltype('
      3      <root>
      4        <child name="name1">
      5          <grandchild name="name11">
      6            <greatgrandchild name="name111"/>
      7            <greatgrandchild name="name112"/>
      8          </grandchild>
      9          <grandchild name="name12"/>
    10       </child>
    11       <child name="name2">
    12         <grandchild name="name21"/>
    13            <greatgrandchild name="name211"/>
    14         <grandchild name="name22"/>
    15       </child>
    16     </root>') object_value FROM dual)
    17     select po.child_name, gc.gchild_name, ggc.ggchild_name
    18     from   xml_table p
    19           ,xmltable('/root/child'
    20                     passing p.object_value
    21                     columns child_name  varchar2(100) path '@name'
    22                           ,gchild      xmltype       path 'grandchild'
    23                    ) po
    24           ,xmltable('/grandchild'
    25                     passing po.gchild
    26                     columns gchild_name varchar2(100) path '@name'
    27                            ,ggchild     xmltype       path 'greatgrandchild'
    28                   ) (+) gc
    29           ,xmltable('/greatgrandchild'
    30                     passing gc.ggchild
    31                     columns ggchild_name varchar2(100) path '@name'
    32*                  ) (+) ggc
    SQL> /
    CHILD_NAME           GCHILD_NAME          GGCHILD_NAME
    name1                name11               name111
    name1                name11               name112
    name1                name12
    name2                name21
    name2                name22and then insert the results into whatever table(s) you want.

  • XML data put into database

    I am new to oracle and new to xml... so if anyone has some suggestions on this topic I would appreciate it...
    we are looking at getting data from multiple locations and multiple databases and was looking at a way to uniformly add this to the database... I was exploring the XSU utility, but I had a few questions...
    We have quite a few tables that will need to be updated with data (around 100). Will we need an xml file for each table? is there a way to have an xml file so that the insert, update, and delete routines can be included in one file, or will this have to be three files... so potentially we would have 300 files to update the database?
    Thanks for your time and input... let me know if you need more information.

    Brian,
    A 100 tables would mean 100 XML-documents.
    But, if source and destination table definitions differ (e.g. in column names), you will need a transformation of the XML into the (canonical) format of the destination table.
    So you may have to define multiple xslt stylesheets per table for these transformations.
    And something else you may need to consider:
    processing a few large XML-documents takes far more time than processing many small XML-documents. Advice: keep the XML-documents small.

  • Insert XML file into Relational database model - no XMLTYPE!

    Dear all,
    How can I store a known complex XML file into an existing relational database WITHOUT using xmltypes in the database ?
    I read the article on DBMS_XMLSTORE. DBMS_XMLSTORE indeed partially bridges the gap between XML and RDBMS to a certain extent, namely for simply structured XML (canonical structure) and simple tables.
    However, when the XML structure will become arbitrary and rapidly evolving, surely there must be a way to map XML to a relational model more flexibly.
    We work in a java/Oracle10 environment that receives very large XML documents from an independent data management source. These files comply with an XML schema. That is all we know. Still, all these data must be inserted/updated daily in an existing relational model. Quite an assignment isn't it ?
    The database does and will not contain XMLTYPES, only plain RDBMS tables.
    Are you aware of a framework/product or tool to do what DBMS_XMLSTORE does but with any format of XML file ? If not, I am doomed.
    Cheers.
    Luc.
    Edited by: user6693852 on Jan 13, 2009 7:02 AM

    In case you decide to follow my advice, here's a simple example showing how to do this.. (Note the XMLTable syntax is the preferred approach in 10gr2 and later..
    SQL> spool testase.log
    SQL> --
    SQL> connect / as sysdba
    Connected.
    SQL> --
    SQL> set define on
    SQL> set timing on
    SQL> --
    SQL> define USERNAME = XDBTEST
    SQL> --
    SQL> def PASSWORD = XDBTEST
    SQL> --
    SQL> def USER_TABLESPACE = USERS
    SQL> --
    SQL> def TEMP_TABLESPACE = TEMP
    SQL> --
    SQL> drop user &USERNAME cascade
      2  /
    old   1: drop user &USERNAME cascade
    new   1: drop user XDBTEST cascade
    User dropped.
    Elapsed: 00:00:00.59
    SQL> grant create any directory, drop any directory, connect, resource, alter session, create view to &USERNAME identified by &PASS
    ORD
      2  /
    old   1: grant create any directory, drop any directory, connect, resource, alter session, create view to &USERNAME identified by &
    ASSWORD
    new   1: grant create any directory, drop any directory, connect, resource, alter session, create view to XDBTEST identified by XDB
    EST
    Grant succeeded.
    Elapsed: 00:00:00.01
    SQL> alter user &USERNAME default tablespace &USER_TABLESPACE temporary tablespace &TEMP_TABLESPACE
      2  /
    old   1: alter user &USERNAME default tablespace &USER_TABLESPACE temporary tablespace &TEMP_TABLESPACE
    new   1: alter user XDBTEST default tablespace USERS temporary tablespace TEMP
    User altered.
    Elapsed: 00:00:00.00
    SQL> connect &USERNAME/&PASSWORD
    Connected.
    SQL> --
    SQL> var SCHEMAURL varchar2(256)
    SQL> var XMLSCHEMA CLOB
    SQL> --
    SQL> set define off
    SQL> --
    SQL> begin
      2    :SCHEMAURL := 'http://xmlns.example.com/askTom/TransactionList.xsd';
      3    :XMLSCHEMA :=
      4  '<?xml version="1.0" encoding="UTF-8"?>
      5  <!--W3C Schema generated by XMLSpy v2008 rel. 2 sp2 (http://www.altova.com)-->
      6  <xs:schema xmlns:xs="http://www.w3.org/2001/XMLSchema" xmlns:xdb="http://xmlns.oracle.com/xdb" xdb:storeVarrayAsTable="true">
      7     <xs:element name="TransactionList" type="transactionListType" xdb:defaultTable="LOCAL_TABLE"/>
      8     <xs:complexType name="transactionListType"  xdb:maintainDOM="false" xdb:SQLType="TRANSACTION_LIST_T">
      9             <xs:sequence>
    10                     <xs:element name="Transaction" type="transactionType" maxOccurs="unbounded" xdb:SQLCollType="TRANSACTION_V"
    >
    11             </xs:sequence>
    12     </xs:complexType>
    13     <xs:complexType name="transactionType" xdb:maintainDOM="false"  xdb:SQLType="TRANSACTION_T">
    14             <xs:sequence>
    15                     <xs:element name="TradeVersion" type="xs:integer"/>
    16                     <xs:element name="TransactionId" type="xs:integer"/>
    17                     <xs:element name="Leg" type="legType" maxOccurs="unbounded" xdb:SQLCollType="LEG_V"/>
    18             </xs:sequence>
    19             <xs:attribute name="id" type="xs:integer" use="required"/>
    20     </xs:complexType>
    21     <xs:complexType name="paymentType"  xdb:maintainDOM="false" xdb:SQLType="PAYMENT_T">
    22             <xs:sequence>
    23                     <xs:element name="StartDate" type="xs:date"/>
    24                     <xs:element name="Value" type="xs:integer"/>
    25             </xs:sequence>
    26             <xs:attribute name="id" type="xs:integer" use="required"/>
    27     </xs:complexType>
    28     <xs:complexType name="legType"  xdb:maintainDOM="false"  xdb:SQLType="LEG_T">
    29             <xs:sequence>
    30                     <xs:element name="LegNumber" type="xs:integer"/>
    31                     <xs:element name="Basis" type="xs:integer"/>
    32                     <xs:element name="FixedRate" type="xs:integer"/>
    33                     <xs:element name="Payment" type="paymentType" maxOccurs="unbounded"  xdb:SQLCollType="PAYMENT_V"/>
    34             </xs:sequence>
    35             <xs:attribute name="id" type="xs:integer" use="required"/>
    36     </xs:complexType>
    37  </xs:schema>';
    38  end;
    39  /
    PL/SQL procedure successfully completed.
    Elapsed: 00:00:00.00
    SQL> set define on
    SQL> --
    SQL> declare
      2    res boolean;
      3    xmlSchema xmlType := xmlType(:XMLSCHEMA);
      4  begin
      5    dbms_xmlschema.registerSchema
      6    (
      7      schemaurl => :schemaURL,
      8      schemadoc => xmlSchema,
      9      local     => TRUE,
    10      genTypes  => TRUE,
    11      genBean   => FALSE,
    12      genTables => TRUE,
    13      ENABLEHIERARCHY => DBMS_XMLSCHEMA.ENABLE_HIERARCHY_NONE
    14    );
    15  end;
    16  /
    PL/SQL procedure successfully completed.
    Elapsed: 00:00:00.26
    SQL> desc LOCAL_TABLE
    Name                                                                   Null?    Type
    TABLE of SYS.XMLTYPE(XMLSchema "http://xmlns.example.com/askTom/TransactionList.xsd" Element "TransactionList") STORAGE Object-rela
    ional TYPE "TRANSACTION_LIST_T"
    SQL> --
    SQL> create or replace VIEW TRAN_VIEW
      2  as
      3  select
      4    extractvalue(x.column_value,'/Transaction/TradeVersion/text()') tradeversion,
      5    extractvalue(x.column_value,'/Transaction//text()') transactionid
      6  from
      7    local_table,
      8    table(xmlsequence(extract(OBJECT_VALUE,'/TransactionList/Transaction'))) x
      9  /
    View created.
    Elapsed: 00:00:00.01
    SQL> create or replace VIEW TRAN_LEG_VIEW
      2  as
      3  select
      4    extractvalue(x.column_value,'/Transaction/TransactionId/text()') transactionid,
      5    extractvalue(y.column_value,'/Leg/Basis/text()') leg_basis,
      6    extractValue(y.column_value,'/Leg/FixedRate/text()') leg_fixedrate
      7  from
      8    local_table,
      9    table(xmlsequence(extract(OBJECT_VALUE,'/TransactionList/Transaction'))) x,
    10    table(xmlsequence(extract(x.column_value,'/Transaction/Leg'))) y
    11  /
    View created.
    Elapsed: 00:00:00.01
    SQL> create or replace VIEW TRAN_LEG_PAY_VIEW
      2  as
      3  select
      4    extractvalue(x.column_value,'/Transaction/TransactionId/text()') transactionid,
      5    extractvalue(y.column_value,'/Leg/LegNumber/text()') leg_legnumber,
      6    extractvalue(z.column_value,'/Payment/StartDate/text()') pay_startdate,
      7    extractValue(z.column_value,'/Payment/Value/text()') pay_value
      8  from
      9    local_table,
    10    table(xmlsequence(extract(OBJECT_VALUE,'/TransactionList/Transaction'))) x,
    11    table(xmlsequence(extract(x.column_value,'/Transaction/Leg'))) y,
    12    table(xmlsequence(extract(y.column_value,'/Leg/Payment'))) z
    13  /
    View created.
    Elapsed: 00:00:00.03
    SQL> desc TRAN_VIEW
    Name                                                                   Null?    Type
    TRADEVERSION                                                                    NUMBER(38)
    TRANSACTIONID                                                                   VARCHAR2(4000)
    SQL> --
    SQL> desc TRAN_LEG_VIEW
    Name                                                                   Null?    Type
    TRANSACTIONID                                                                   NUMBER(38)
    LEG_BASIS                                                                       NUMBER(38)
    LEG_FIXEDRATE                                                                   NUMBER(38)
    SQL> --
    SQL> desc TRAN_LEG_PAY_VIEW
    Name                                                                   Null?    Type
    TRANSACTIONID                                                                   NUMBER(38)
    LEG_LEGNUMBER                                                                   NUMBER(38)
    PAY_STARTDATE                                                                   DATE
    PAY_VALUE                                                                       NUMBER(38)
    SQL> --
    SQL> create or replace VIEW TRAN_VIEW_XMLTABLE
      2  as
      3  select t.*
      4    from LOCAL_TABLE,
      5         XMLTable
      6         (
      7            '/TransactionList/Transaction'
      8            passing OBJECT_VALUE
      9            columns
    10            TRADE_VERSION  NUMBER(4) path 'TradeVersion/text()',
    11            TRANSACTION_ID NUMBER(4) path 'TransactionId/text()'
    12         ) t
    13  /
    View created.
    Elapsed: 00:00:00.01
    SQL> create or replace VIEW TRAN_LEG_VIEW_XMLTABLE
      2  as
      3  select t.TRANSACTION_ID, L.*
      4    from LOCAL_TABLE,
      5         XMLTable
      6         (
      7            '/TransactionList/Transaction'
      8            passing OBJECT_VALUE
      9            columns
    10            TRANSACTION_ID NUMBER(4) path 'TransactionId/text()',
    11            LEG            XMLType   path 'Leg'
    12         ) t,
    13         XMLTABLE
    14         (
    15           '/Leg'
    16           passing LEG
    17           columns
    18           LEG_NUMBER     NUMBER(4) path 'LegNumber/text()',
    19           LEG_BASIS      NUMBER(4) path 'Basis/text()',
    20           LEG_FIXED_RATE NUMBER(4) path 'FixedRate/text()'
    21         ) l
    22  /
    View created.
    Elapsed: 00:00:00.01
    SQL> create or replace VIEW TRAN_LEG_PAY_VIEW_XMLTABLE
      2  as
      3  select TRANSACTION_ID, L.LEG_NUMBER, P.*
      4    from LOCAL_TABLE,
      5         XMLTable
      6         (
      7            '/TransactionList/Transaction'
      8            passing OBJECT_VALUE
      9            columns
    10            TRANSACTION_ID NUMBER(4) path 'TransactionId/text()',
    11            LEG            XMLType   path 'Leg'
    12         ) t,
    13         XMLTABLE
    14         (
    15           '/Leg'
    16           passing LEG
    17           columns
    18           LEG_NUMBER     NUMBER(4) path 'LegNumber/text()',
    19           PAYMENT        XMLType   path 'Payment'
    20         ) L,
    21         XMLTABLE
    22         (
    23           '/Payment'
    24           passing PAYMENT
    25           columns
    26           PAY_START_DATE     DATE      path 'StartDate/text()',
    27           PAY_VALUE          NUMBER(4) path 'Value/text()'
    28         ) p
    29  /
    View created.
    Elapsed: 00:00:00.03
    SQL> desc TRAN_VIEW_XMLTABLE
    Name                                                                   Null?    Type
    TRADE_VERSION                                                                   NUMBER(4)
    TRANSACTION_ID                                                                  NUMBER(4)
    SQL> --
    SQL> desc TRAN_LEG_VIEW_XMLTABLE
    Name                                                                   Null?    Type
    TRANSACTION_ID                                                                  NUMBER(4)
    LEG_NUMBER                                                                      NUMBER(4)
    LEG_BASIS                                                                       NUMBER(4)
    LEG_FIXED_RATE                                                                  NUMBER(4)
    SQL> --
    SQL> desc TRAN_LEG_PAY_VIEW_XMLTABLE
    Name                                                                   Null?    Type
    TRANSACTION_ID                                                                  NUMBER(4)
    LEG_NUMBER                                                                      NUMBER(4)
    PAY_START_DATE                                                                  DATE
    PAY_VALUE                                                                       NUMBER(4)
    SQL> --
    SQL> set long 10000 pages 100 lines 128
    SQL> set timing on
    SQL> set autotrace on explain
    SQL> set heading on feedback on
    SQL> --
    SQL> VAR DOC1 CLOB
    SQL> VAR DOC2 CLOB
    SQL> --
    SQL> begin
      2    :DOC1 :=
      3  '<TransactionList>
      4    <Transaction id="1">
      5      <TradeVersion>1</TradeVersion>
      6      <TransactionId>1</TransactionId>
      7      <Leg id="1">
      8        <LegNumber>1</LegNumber>
      9        <Basis>1</Basis>
    10        <FixedRate>1</FixedRate>
    11        <Payment id="1">
    12          <StartDate>2000-01-01</StartDate>
    13          <Value>1</Value>
    14        </Payment>
    15        <Payment id="2">
    16          <StartDate>2000-01-02</StartDate>
    17          <Value>2</Value>
    18        </Payment>
    19      </Leg>
    20      <Leg id="2">
    21        <LegNumber>2</LegNumber>
    22        <Basis>2</Basis>
    23        <FixedRate>2</FixedRate>
    24        <Payment id="1">
    25          <StartDate>2000-02-01</StartDate>
    26          <Value>10</Value>
    27        </Payment>
    28        <Payment id="2">
    29          <StartDate>2000-02-02</StartDate>
    30          <Value>20</Value>
    31        </Payment>
    32      </Leg>
    33    </Transaction>
    34    <Transaction id="2">
    35      <TradeVersion>2</TradeVersion>
    36      <TransactionId>2</TransactionId>
    37      <Leg id="1">
    38        <LegNumber>21</LegNumber>
    39        <Basis>21</Basis>
    40        <FixedRate>21</FixedRate>
    41        <Payment id="1">
    42          <StartDate>2002-01-01</StartDate>
    43          <Value>21</Value>
    44        </Payment>
    45        <Payment id="2">
    46          <StartDate>2002-01-02</StartDate>
    47          <Value>22</Value>
    48        </Payment>
    49      </Leg>
    50      <Leg id="22">
    51        <LegNumber>22</LegNumber>
    52        <Basis>22</Basis>
    53        <FixedRate>22</FixedRate>
    54        <Payment id="21">
    55          <StartDate>2002-02-01</StartDate>
    56          <Value>210</Value>
    57        </Payment>
    58        <Payment id="22">
    59          <StartDate>2002-02-02</StartDate>
    60          <Value>220</Value>
    61        </Payment>
    62      </Leg>
    63    </Transaction>
    64  </TransactionList>';
    65    :DOC2 :=
    66  '<TransactionList>
    67    <Transaction id="31">
    68      <TradeVersion>31</TradeVersion>
    69      <TransactionId>31</TransactionId>
    70      <Leg id="31">
    71        <LegNumber>31</LegNumber>
    72        <Basis>31</Basis>
    73        <FixedRate>31</FixedRate>
    74        <Payment id="31">
    75          <StartDate>3000-01-01</StartDate>
    76          <Value>31</Value>
    77        </Payment>
    78      </Leg>
    79    </Transaction>
    80  </TransactionList>';
    81  end;
    82  /
    PL/SQL procedure successfully completed.
    Elapsed: 00:00:00.01
    SQL> insert into LOCAL_TABLE values ( xmltype(:DOC1))
      2  /
    1 row created.
    Elapsed: 00:00:00.01
    Execution Plan
    Plan hash value: 1
    | Id  | Operation                | Name        | Rows  | Bytes | Cost (%CPU)| Time     |
    |   0 | INSERT STATEMENT         |             |     1 |   100 |     1   (0)| 00:00:01 |
    |   1 |  LOAD TABLE CONVENTIONAL | LOCAL_TABLE |       |       |            |          |
    SQL> insert into LOCAL_TABLE values ( xmltype(:DOC2))
      2  /
    1 row created.
    Elapsed: 00:00:00.01
    Execution Plan
    Plan hash value: 1
    | Id  | Operation                | Name        | Rows  | Bytes | Cost (%CPU)| Time     |
    |   0 | INSERT STATEMENT         |             |     1 |   100 |     1   (0)| 00:00:01 |
    |   1 |  LOAD TABLE CONVENTIONAL | LOCAL_TABLE |       |       |            |          |
    SQL> select * from TRAN_VIEW_XMLTABLE
      2  /
    TRADE_VERSION TRANSACTION_ID
                1              1
                2              2
               31             31
    3 rows selected.
    Elapsed: 00:00:00.03
    Execution Plan
    Plan hash value: 650975545
    | Id  | Operation          | Name                           | Rows  | Bytes | Cost (%CPU)| Time     |
    |   0 | SELECT STATEMENT   |                                |     3 |   168 |     3   (0)| 00:00:01 |
    |   1 |  NESTED LOOPS      |                                |     3 |   168 |     3   (0)| 00:00:01 |
    |*  2 |   TABLE ACCESS FULL| SYS_NTGgl+TKyhQnWoFRSrCxeX9g== |     3 |   138 |     3   (0)| 00:00:01 |
    |*  3 |   INDEX UNIQUE SCAN| SYS_C0010174                   |     1 |    10 |     0   (0)| 00:00:01 |
    Predicate Information (identified by operation id):
       2 - filter("SYS_NC_TYPEID$" IS NOT NULL)
       3 - access("NESTED_TABLE_ID"="LOCAL_TABLE"."SYS_NC0000800009$")
    Note
       - dynamic sampling used for this statement
    SQL> select * from TRAN_LEG_VIEW_XMLTABLE
      2  /
    TRANSACTION_ID LEG_NUMBER  LEG_BASIS LEG_FIXED_RATE
                 1          1          1              1
                 1          2          2              2
                 2         21         21             21
                 2         22         22             22
                31         31         31             31
    5 rows selected.
    Elapsed: 00:00:00.04
    Execution Plan
    Plan hash value: 1273661583
    | Id  | Operation           | Name                           | Rows  | Bytes | Cost (%CPU)| Time     |
    |   0 | SELECT STATEMENT    |                                |     5 |   560 |     7  (15)| 00:00:01 |
    |*  1 |  HASH JOIN          |                                |     5 |   560 |     7  (15)| 00:00:01 |
    |   2 |   NESTED LOOPS      |                                |     3 |   159 |     3   (0)| 00:00:01 |
    |*  3 |    TABLE ACCESS FULL| SYS_NTGgl+TKyhQnWoFRSrCxeX9g== |     3 |   129 |     3   (0)| 00:00:01 |
    |*  4 |    INDEX UNIQUE SCAN| SYS_C0010174                   |     1 |    10 |     0   (0)| 00:00:01 |
    |*  5 |   TABLE ACCESS FULL | SYS_NTUmyermF/S721C/2UXo40Uw== |     5 |   295 |     3   (0)| 00:00:01 |
    Predicate Information (identified by operation id):
       1 - access("SYS_ALIAS_1"."NESTED_TABLE_ID"="SYS_ALIAS_0"."SYS_NC0000800009$")
       3 - filter("SYS_NC_TYPEID$" IS NOT NULL)
       4 - access("NESTED_TABLE_ID"="LOCAL_TABLE"."SYS_NC0000800009$")
       5 - filter("SYS_NC_TYPEID$" IS NOT NULL)
    Note
       - dynamic sampling used for this statement
    SQL> select * from TRAN_LEG_PAY_VIEW_XMLTABLE
      2  /
    TRANSACTION_ID LEG_NUMBER PAY_START  PAY_VALUE
                 1          1 01-JAN-00          1
                 1          1 02-JAN-00          2
                 1          2 01-FEB-00         10
                 1          2 02-FEB-00         20
                 2         21 01-JAN-02         21
                 2         21 02-JAN-02         22
                 2         22 01-FEB-02        210
                 2         22 02-FEB-02        220
                31         31 01-JAN-00         31
    9 rows selected.
    Elapsed: 00:00:00.07
    Execution Plan
    Plan hash value: 4004907785
    | Id  | Operation            | Name                           | Rows  | Bytes | Cost (%CPU)| Time     |
    |   0 | SELECT STATEMENT     |                                |     9 |  1242 |    10  (10)| 00:00:01 |
    |*  1 |  HASH JOIN           |                                |     9 |  1242 |    10  (10)| 00:00:01 |
    |*  2 |   HASH JOIN          |                                |     5 |   480 |     7  (15)| 00:00:01 |
    |   3 |    NESTED LOOPS      |                                |     3 |   159 |     3   (0)| 00:00:01 |
    |*  4 |     TABLE ACCESS FULL| SYS_NTGgl+TKyhQnWoFRSrCxeX9g== |     3 |   129 |     3   (0)| 00:00:01 |
    |*  5 |     INDEX UNIQUE SCAN| SYS_C0010174                   |     1 |    10 |     0   (0)| 00:00:01 |
    |*  6 |    TABLE ACCESS FULL | SYS_NTUmyermF/S721C/2UXo40Uw== |     5 |   215 |     3   (0)| 00:00:01 |
    |*  7 |   TABLE ACCESS FULL  | SYS_NTelW4ZRtKS+WKqCaXhsHnNQ== |     9 |   378 |     3   (0)| 00:00:01 |
    Predicate Information (identified by operation id):
       1 - access("NESTED_TABLE_ID"="SYS_ALIAS_1"."SYS_NC0000900010$")
       2 - access("SYS_ALIAS_1"."NESTED_TABLE_ID"="SYS_ALIAS_0"."SYS_NC0000800009$")
       4 - filter("SYS_NC_TYPEID$" IS NOT NULL)
       5 - access("NESTED_TABLE_ID"="LOCAL_TABLE"."SYS_NC0000800009$")
       6 - filter("SYS_NC_TYPEID$" IS NOT NULL)
       7 - filter("SYS_NC_TYPEID$" IS NOT NULL)
    Note
       - dynamic sampling used for this statement
    SQL>Out of interest why are you so against using XMLType...
    Edited by: mdrake on Jan 13, 2009 8:25 AM

  • Performance problems loading an XML file into oracle database

    Hello ODI Guru's,
    I am trying to load and XML file into the database after doing simple business validations. But the interface takes hours to complete.
    1. The XML files are large in size >200 Mb. We have an XSD file for the schema definition instead of a DTD.
    2. We used the external database feature for loading these files in database.
    The following configuration was used in the XML Data Server:
    jdbc:snps:xml?f=D:\CustomerMasterData1\CustomerMasterInitialLoad1.xml&d=D:\CustomerMasterData1\CustomerMasterInitialLoad1.xsd&re=initialLoad&s=CM&db_props=oracle&ro=true
    3. Now we reverse engineer the XML files and created models using ODI Designer
    4. Similar thing was done for the target i.e. an Oracle database table as well.
    5. Next we created a simple interface with one-to-one mapping from the XSD schema to the Oracle database table and executed the interface. This execution takes more than one hour to complete.
    6. We are running ODI client on Windows XP Professional SP2.
    7. The Oracle database server(Oracle 10g 10.2.0.3) for the target schema as well as the ODI master and work repositories are on the same machine.
    8. I tried changing the following properties but it is not making much visible difference:
    use_prepared_statements=Y
    use_batch_update=Y
    batch_update_size=510
    commit_periodically=Y
    num_inserts_before_commit=30000
    I have another problem that when I set batch_update_size to value greater that 510 I get the following error:
    java.sql.SQLException: class org.xml.sax.SAXException
    class java.lang.ArrayIndexOutOfBoundsException said -32413
    at com.sunopsis.jdbc.driver.xml.v.a(v.java)
    The main concern is why should the interface taking so long to execute.
    Please send suggestions to resolve the problem.
    Thanks in advance,
    Best Regards,
    Nikunj

    Approximately how many rows are you trying to insert?
    One of the techniques which I found improved performance for this scenario was to extract from the xml to a flat file, then to use SQL*LOADER or external tables to load the data into Oracle.

Maybe you are looking for