Html - XML parser

Hi, just wondering if anyone had any
recommendations with regards to a java
HTML->XML parser?
For an assignment i have i can either
code my own or integrate an already
existing one - even if i do end up
coding my own, it'd probably be useful
to see a working one as well.
Any suggestions as to which is the
best? By this i mean most efficient
and smallest.
Thanks,
Cam

Sorry to be a pain, i dont suppose
you have some sample/test code that
uses Jtidy to covert html into xml?
The sourceforge documentation didn't
really help me :/
Thanks,
Cam

Similar Messages

XML parser fails to convert html encoded text nodes

Under the strain of large documents this defect rears its ugly head more often. While parsing a text node containing html encoded chars i.e. < > &; etc...
The parser will seemingly forget to change the chars at random, 99.9% of the time everything is ok and the proper conversions take place:
< -> <
> -> >
&; -> &
Once an error occurs it is reproducible until the text node is changed ( values and/or order ) then it is a crapshoot again.
These tests were done using the default UTF-8 encoding, here is the exception thrown by the parser along with a portion of the text node before and after the first parsing.
Let me be clear, the parser actually succeeds the first time but the transformation of the HTML encoded pieces possibly fails. It is on the parsing of the text node value as its own document that the parser fails.
Error 0 : Error parsing XML string! Line :: 1, Column :: 65674
Error 1 : End tag does not match start tag 'Project'.
End tag does not match start tag 'Project'.
at oracle/xml/parser/v2/XMLError.flushErrors (XMLError.java:233)
at oracle/xml/parser/v2/NonValidatingParser.parseDocument (NonValidatingParser.java:248)
at oracle/xml/parser/v2/XMLParser.parse (XMLParser.java:117)
at pacificedge/xml/XMLUtilities.loadParserFromString (XMLUtilities.java:104)
Preprocessing ::
<Project stuff0="0" stuff1="0" stuff2="0" stuff3="1" stuff4="100167" stuff5="100213">
<StuffA>100213</StuffA>
<Name>I am a Name</Name>
<StartDate>1998-08-10</StartDate>
<FinishDate>2000-06-30</FinishDate>
<Path>Folder1\Folder2</Path>
</Project>
Post processing:
<Project stuff0="0" stuff1="0" stuff2="0" stuff3="1" stuff4="100167" stuff5="100213">
<StuffA>100213</StuffA>
<Name>I am a Name</Name>
<StartDate>1998-08-10</StartDate> <-- Error is raised here when the value of the text node is used as an xml document
<FinishDate>2000-06-30</FinishDate>
<Path>Folder1\Folder2</Path>
</Project>
Please investigate this. It is a chronic problem for us and possibly many others.
null

Sorry for the encoding issues in the message before here are the pertinent pieces hope this shows up correctly.
&;lt; -> &;#60;
&;gt; -> >
&;amp; -> &;#38;
Preprocessing ::
&;lt;Project stuff0="0" stuff1="0" stuff2="0" stuff3="1" stuff4="100167" stuff5="100213"&;gt;
&;lt;StuffA&;gt;100213&;lt;/StuffA&;gt;
&;lt;Name&;gt;I am a Name&;lt;/Name&;gt;
&;lt;StartDate&;gt;1998-08-10&;lt;/StartDate&;gt;
&;lt;FinishDate&;gt;2000-06-30&;lt;/FinishDate&;gt;
&;lt;Path&;gt;Folder1\Folder2&;lt;/Path&;gt;
&;lt;/Project&;gt;
Post processing:
<Project stuff0="0" stuff1="0" stuff2="0" stuff3="1" stuff4="100167" stuff5="100213">
<StuffA>100213</StuffA>
<Name>I am a Name</Name>
&;lt;StartDate>1998-08-10</StartDate> <-- Error is raised here when the value of the text node is used as an xml document
<FinishDate>2000-06-30</FinishDate>
<Path>Folder1\Folder2</Path>
</Project>
null

XML Parser puts HTML Tags each on its own line... BAD

Is there a way to tell Oracle XML Parser not to place each HTML tag on a separate line in the output stream? It appears that this is what is causing what I will call "dog-legging". The normal dotted line that surrounds an image on an HTML browser page has an extra protrusion in the lower-right corner under IE 4.x. A hyphen appers in the lower right corner under NN 4.x.
any information that you good folks could provide me on this matter would be greatly appreciated.
Thanks in advance,
Dave Reese

The way to avoid this is to specifically request that the HTML output be generated without indentation. Just include the
following at the top-level of your stylesheet inside the outermost <xsl:stylesheet> element:
<xsl:output method="html" indent="no"/>

Loading XML parser for PL/SQL

I am working my way thru 'Building Oracle XML Applications' by Steve Muench.
I am getting a problem when I try to install the XML parser for PL/SQL
The following errors occur when I try to do
loadjava -v -r -u xmlbook/xmlbook xmlparserv2.jar
initialization complete
loading : oracle/xml/parser/v2/XMLByteReader
creating : oracle/xml/parser/v2/XMLByteReader
Error while creating class oracle/xml/parser/v2/XMLByteReader
ORA-29545: badly formed class: User has attempted to load a class (oracle.xml.parser.v2.XMLByteReader) into a restricted package. Permission can be granted using dbms_java.grant_permission(<user>, LoadClassInPackage...
loading : oracle/xml/parser/v2/PrintDriver
creating : oracle/xml/parser/v2/PrintDriver
Error while creating class oracle/xml/parser/v2/PrintDriver
ORA-29545: badly formed class: User has attempted to load a class (oracle.xml.parser.v2.PrintDriver) into a restricted package. Permission can be granted using dbms_java.grant_permission(<user>, LoadClassInPackage...
resolving: oracle/xml/parser/v2/PrintDriver
Error while resolving class oracle/xml/parser/v2/PrintDriver
ORA-04043: object /a065289c_PrintDriver does not exist
resolving: org/w3c/dom/html/HTMLAnchorElement
Error while resolving class org/w3c/dom/html/HTMLAnchorElement
ORA-04043: object /564607d_HTMLAnchorElement does not exist
About 20 classes get loaded correctly. All the rest get ORA-29545 or ORA-04043 errors.
I expect the ORA-04043 errors are a result of the earlier errors.
The documentation seems to indicate that there is no need to do
dbms_java.grant_permission to load into your own schema.
I have NT4 (sp6)
Oracle 8.1.7
xmlparserv2.jar dated 24/8/2001 17:01 size 1,177,259
xmlparserv2_2027.jar gives the same errors.
Any help would be appreciated.
null

Was it definitely xmlparsev2.jar that you had to load?
I recently started going through some of these examples as well and found that the standard xml parser was there but that I needed to load the xml parser for PL/SQL. I did this using loadjava from a subdirectory within the xdk directory and everything worked fine.

Swapping XML Parser and XSLT to Xalan 2.7.0 - Not Working (OC4J 10.1.3)

Hi-
I'm trying to use the latest Xercies/Xalan classes in OC4J 10.1.3 as described in the How-To swap XML Parsers document:
http://www.oracle.com/technology/tech/java/oc4j/1013/how_to/how-to-swapxmlparser/doc/readme.html
What I can see happening is that the oracle.xml shared library is successfully 'turned off', but the xalan libraries are not added. Instead, the default xml classes that are distributed with the JDK become visible. For instance, using a slightly modified version of the index.jsp page from the how-to, I get this result:
------ Output from JSP page ----------------
TransformerFactory Instance: org.apache.xalan.processor.TransformerFactoryImpl
Transformer Instance: org.apache.xalan.transformer.TransformerIdentityImpl
Transformer Version: Xalan Java 2.4.1 (!!)
What I expect is for that last line to say version 2.7.0, which is the version of the xalan.jar included in my shared library (code to add that line to the how-to shown below).
I suspect what is happening is that the class loader is simply not letting a shared library override a system library - to do that you probably need to place the jar files in system endorsed directory.
Has anyone gotten this how-to to work - actually replacing the XML parser/transform classes with the latest Xalan classes? Are you sure it is seeing the current version you placed in the shared library?
Thanks,
Eric Everman
---- My modified getXSLTDetails() method in the index.jsp page of the how-to -------

<%@ page import="org.apache.xalan.Version" %>
public static String getXSLTDetails()
Transformer transformer=null;
TransformerFactory transformerfactory = TransformerFactory.newInstance();
     Version ver = null;
     String br = "<" + "br" + ">"; //otherwise the otn forum chocks on the break.
try
transformer = transformerfactory.newTransformer();
          ver = (Version) transformer.getClass().forName("org.apache.xalan.Version").newInstance();
          String ret_val =
               "TransformerFactory Instance: "+transformerfactory.getClass().getName() + br +
               "Transformer Instance: "+transformer.getClass().getName() + br;
          if (ver != null) {
               ret_val = ret_val + "Transformer Version: " + ver.getVersion() + br;
          } else {
               ret_val = ret_val + "Transformer Version not Available" + br;
          return ret_val;
catch (Exception e)
e.printStackTrace();
return e.getMessage();
}--------------------------------------------------------------------

Steve - Thanks for responding on this.
The Xalan SQL extension is built into Xalan. The most painless way to try it out is to run it via JEdit: www.jedit.org
JEdit a OS Java application with a fairly painless install process (I'm assuming you already have a current JRE installed). Once its installed, you'll need to install the XSLT plugin - in JEdit goto Plugins | Plugin Manager | Install and pick the XSLT plugin. You'll need the Oracle JDBC classes on your classpath - this can be done by copying the oracle_jdbc4.jar into the [JEdit install directory]/jars directory.
Restart to load that jar and you should be all set.
I included a sample XSLT page at the bottom of this post that is somewhat of a template transform for the SQL extension - its more complicated then it needs to be, but it does some nice things with the results of the query. Save it as a file and make the appropriate changes to the 'datasource' parameter near the top of the file.
Then in JEdit, open the file and make sure the XSLT plugin is visible (Plugins | XSLT | XSLT Processor Toggle - or alternately dock it in the window via global prefs). In the XSLT plugin: Allow the current buffer to be used as the source, Add that same file as a stylesheet via the '+' button, and pick a result file at the bottom. Then click the 'Transform XML' button.
Troubleshooting: I seem to remember having some classpath errors when I tried this on windows, but others have had it work w/o issues. I do remeber that the XSLT plugin had a popup window that gave a pretty good explaintion of the problem, if it occurs. Also, for some reason the XSLT plugin will not create a new file for the output, so its often best to create a file first, then choose it as the output. Of course, you can always run a transformation from the command line or w/in an applicatoin. Full docs on the Xalan SQL extension can be found at: http://xml.apache.org/xalan-j/extensionslib.html#sql
Here is my sample XSLT transform using the SQL extension:
<?xml version="1.0" encoding="UTF-8" ?>
<xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
          xmlns:sql="http://xml.apache.org/xalan/sql"
          xmlns:str="http://exslt.org/strings"
          xmlns:xalan="http://xml.apache.org/xalan"
          extension-element-prefixes="sql str xalan">
     <xsl:output indent="yes"/>
     <xsl:param name="driver">oracle.jdbc.OracleDriver</xsl:param>
     <xsl:param name="datasource">jdbc:oracle:thin:jqpublic/jqpublic@server_name:1521:dbname</xsl:param>
     <xsl:param name="jndiDatasource"></xsl:param>
     <xsl:param name="debug">true</xsl:param>
     <xsl:variable name="connection" select="sql:new()"/>
     <xsl:variable name="uppercase" select="'ABCDEFGHIJKLMNOPQRSTUVWXYZ'"/>
     <xsl:variable name="lowercase" select="'abcdefghijklmnopqrstuvwxyz'"/>
     
     <xsl:param name="query">
     select * from (
SELECT
          'One' FIELD_1,
          'Two' FIELD_2
     FROM DUAL
     ) where rownum < 100
     </xsl:param>
     
     <xsl:variable name="connection" select="sql:new()"/>
     <xsl:template match="/">
    <xsl:choose>
      <xsl:when test="$jndiDatasource != ''">
        <xsl:if test="not(sql:connect($connection, $jndiDatasource))">
          <xsl:message>Failed to connect to db via jndi connection</xsl:message>
          <xsl:comment>Failed to connect to db via jndi connection</xsl:comment>
          <xsl:if test="$debug = 'true'">
            <xsl:copy-of select="sql:getError($connection)/ext-error"/>
          </xsl:if>
        </xsl:if>
      </xsl:when>
      <xsl:otherwise>
        <xsl:if test="not(sql:connect($connection, $driver, $datasource))">
          <xsl:message>Failed to connect to db via driver/url connection</xsl:message>
          <xsl:comment>Failed to connect to db via driver/url connection</xsl:comment>
          <xsl:if test="$debug = 'true'">
            <xsl:copy-of select="sql:getError($connection)/ext-error"/>
          </xsl:if>
        </xsl:if>
      </xsl:otherwise>
    </xsl:choose>
          
          <xsl:variable name="table" select="sql:query($connection, $query)"/>
          <xsl:if test="not($table)">
               <xsl:message>Error in Query</xsl:message>
               <xsl:copy-of select="sql:getError($connection)/ext-error"/>
          </xsl:if>
          <page>
               
               <xsl:copy-of select="xalan:checkEnvironment()"/>
               
               <meta>
                    <cols>
                         <xsl:apply-templates select="$table/sql/metadata/column-header"/>
                    </cols>
               </meta>
               <rowset>
                    
                    <xsl:apply-templates select="$table/sql/row-set"/>
               </rowset>
          </page>
          <xsl:value-of select="sql:close($connection)"/>
     </xsl:template>
     <xsl:template match="row">
          <row>
               <xsl:apply-templates select="col"/>
          </row>
     </xsl:template>
     <xsl:template match="column-header">
          <col>
               <xsl:attribute name="type">
                    <xsl:value-of select="@column-typename"/>
               </xsl:attribute>
               <xsl:attribute name="display-name">
                    <xsl:call-template name="create-display-name"/>
               </xsl:attribute>
               <xsl:value-of select="@column-name"/>
          </col>
     </xsl:template>
     
     <xsl:template name="create-display-name">
          <xsl:variable name="col-name">
               <xsl:for-each select="str:tokenize(@column-name, '_')">
                    <xsl:value-of
                         select="concat(translate(substring(., 1, 1), $lowercase, $uppercase), translate(substring(.,2), $uppercase, $lowercase), ' ')"/>
               </xsl:for-each>
          </xsl:variable>
          <xsl:value-of select="substring($col-name, 1, string-length($col-name) - 1)"/>
     </xsl:template>
     
     <xsl:template match="col">
          <col>
               <xsl:attribute name="column-name"><xsl:value-of select="@column-name"/></xsl:attribute>
               <xsl:value-of select="."/>
          </col>
     </xsl:template>
</xsl:stylesheet>

Oracle.xml.parser.v2.SAXParser error

I'm reposting this error that I recently received one more time before removing the XML transforms from my Java projects and doing them in .NET.
Code that was working fine suddenly getting a transform error using JDeveloper.
Get the following:
javax.servlet.ServletException: javax.xml.transform.TransformerConfigurationException: XML-22000: (Fatal Error) Error while parsing XSL file (weblogic.xml.jaxp.RegistryXMLReader cannot be cast to oracle.xml.parser.v2.SAXParser).
at weblogic.servlet.jsp.PageContextImpl.handlePageException(PageContextImpl.java:417)
Seems that

will not transform XML to HTML via XSL in JSP anymore.
If I run the XSL i get the HTML results I expected in an HTML test page.
So I know the XML and XSL is sound.
XSL and DOM parsing in Java/Oracle was a major pain in the ass in the first place and now this.
Well I've spent hours going through Oralce/Java documentation on XML, DOM and XSL transforms.
All that time spent and still so unstable. I don't think I was ever able to really get what I wanted
which was to transform my XML document to HTML, update via JSP and have the XML document
be updated and ready to view in a summary page. The XML document changes were never really
available unless I closed the app and restarted. So all in all a failure of capabilities compared to my
use of XML in .NET.
So........
Me thinks it will be a good idea to take out the transforms and have my JSP use a script to invoke a
method in a class that extracts it from a message. Just have PL/SQL interface with XML and do the
transform or serialize thing there.
I don't want to give up on XML but I'm tired of trying to jam a square peg into a round hole
with Java and JDeveloper.
Brian

We don't support file references ("cv1_0.dtd"). Only url type references
are supported, of the form:
@ <!DOCTYPE cv1 SYSTEM "http://www.oracle.com/public/cv1_0.dtd>
Note that the url should be accessible to the world without the need to authenticate.
you could put your dtd in the "public" area in your iFS system.
Thanks,
Sirisha
null

Geting xml parsing error on startup

Error Message reads :-
XML Parsing Error: no element found
Location: jar:file:///C:/Program%20Files/Mozilla%20Firefox/omni.ja!/chrome/toolkit/content/global/netError.xhtmlLine Number1, Column 1:
Any ideas, or will i have to format the drive and start from scratch?!?

Do a clean reinstall and delete the Firefox program folder (C:\Program Files\Mozilla Firefox\) before reinstalling a fresh copy of Firefox.
*http://kb.mozillazine.org/Installation_directory
Download a fresh Firefox copy and save the file to the desktop.
*Firefox 15.0.x: http://www.mozilla.org/en-US/firefox/all.html
Uninstall your current Firefox version, if possible, to cleanup the Windows registry and settings in security software.
*Do NOT remove personal data when you uninstall your current Firefox version, because all profile folders will be removed and you will also lose your personal data like bookmarks and passwords from profiles of other Firefox versions.
Remove the Firefox program folder before installing that newly downloaded copy of the Firefox installer.
*It is important to delete the Firefox program folder to remove all the files and make sure that there are no problems with files that were leftover after uninstalling.
*http://kb.mozillazine.org/Uninstalling_Firefox
Your bookmarks and other profile data are stored in the Firefox Profile Folder and won't be affected by an uninstall and (re)install, but make sure that "remove personal data" is NOT selected when you uninstall Firefox.
If you keep having problems then also create a new profile.
*http://kb.mozillazine.org/Profile_folder_-_Firefox
*http://kb.mozillazine.org/Profile_backup
*http://kb.mozillazine.org/Standard_diagnostic_-_Firefox#Clean_reinstall

Bug in Oracle XML Parser?

I have run the same .xsql page using the same .xsl style sheet via URL/Browser and programmatically as referenced in Steve's book. The problem is, the .xsql runs fine via the browser; however, if I call that same file programmatically I get this:
*** Oracle XSQL Programmatic Page Processor 1.0.4.1 (Production)
XSQL-005: XSQL page is not well-formed.
XML parse error at line 5, char 10
End tag does not match start tag 'META'. ***
Now I've validated the .xsl using 3 different XML editors including JDeveloper. All return the file as well-formed. By the way, there are no META tags in the style sheet or the xsql page! Bug?
Regards,
Craig
null

What happening here is the following:
When you run through the browser, the http request invokes the XSQL Servlet, the XSQL page template is found (typically in the cache if it's the 2nd or subsequent time you're asking for the page), and then the result is returned. Everything fine.
When you run through the XSQLRequest class, you are providing a URL to the "raw" XSQL page of "http://foo/bar.xsql". The XSQL Page processor fetches what it's expecting to be the "raw" template XSQL page by retrieving the resource at the given URL. However, since the URL ends with *.xsql, the request for the *.xsql file does NOT return the raw XSQL page template that the XSQLRequest class is expecting, but instead the server processes the template on the server side and returns the result of processing the bar.xsql page to the XSQLRequest class. Since apparently your XSLT stylesheet if producing HTML, what the XSQLRequest class sees for the "raw" XSQL page template is actually the resulting HTML produced by the web server and since this is not wellformed XML, it gives this error.
The solution, either provide the URL to the bar.xsql page as a file URL, or change the extension on the template page to any extension other than *.xsql so that the template won't get processed by the web server before its delivered.

Problem in XML Parsing via oracle procedure...

Hi,
I wrote one oracle procedure for xml parsing.
I have one valid xml file which has "encode UTF-8". The XML file contains some comments also. While we are parsing the xml file at that time it is not parse successfully and also it is not giving any error. After the following line it is skip rest of the codes(lines).
dbms_xmlparser.parseclob(l_parser, l_clob);
At the end of the xml file there are some comments which is like " ".
When I am changing the "encode UTF-8 to ISO-88596-1" & removing the comments which wrote on bottom of the file then its working fine, but the files which we are getting from the system is contains the encode UTF-8 and we don't want to preprocess on that xml files. Even if we will do that via shell script or perl script then it will be overhead to the system and in a single stroke our system will parse more than 5k xml files, so if we will do some preprocess on it, it will take some more time approx 1-2 minutes extra.
So, If someone knows any solution of this problem, then please guide & help me on this.
My xml file structure is as follows:-
<?xml version="1.0" encoding="UTF-8"?>
<mcd xmlns:HTML="http://www.w3.org/TR/REC-xml">
     <child>
          <child1>32.401 V5.5</child1>
          <child2>ZoneGate</child2>
     </child>
     <mc>
          <newid>
               <id>12</id>
          </newid>
          <mindex>
               <date>20111102180000</date>
               <mt>abc1</mt>
               <mt>abc2</mt>
               <mvalue>
                    <r>val_1</r>
                    <r>val_2</r>
               </mvalue>
          </mindex>
     </mc>
</mcd>


<!--
PARAM:EID = 1
PARAM:GId = 3
PARAM:GSId = 0
--!> And my oracle procedure is as follows:-
create or replace procedure loadXMLtotable(dir_name IN varchar2, xmlfile IN varchar2) AS
-- Defining the variables
ecode NUMBER;
emesg VARCHAR2(200);
l_bfile BFILE;
l_clob CLOB;
l_dest_offset INTEGER:=1;
l_src_offset INTEGER:=1;
l_Char_set_id NUMBER := NLS_CHARSET_ID('UTF8');
l_lang_context INTEGER := dbms_lob.default_lang_ctx;
l_warning INTEGER;
l_parser dbms_xmlparser.Parser;
l_doc dbms_xmldom.DOMDocument;
l_nl1 dbms_xmldom.DOMNodeList;
l_nl2 dbms_xmldom.DOMNodeList;
l_n dbms_xmldom.DOMNode;
node1 dbms_xmldom.DOMNode;
colid integer ; -- column id used for identifying which column it belongs.
l_xmltype XMLTYPE;
sub_xmltype XMLTYPE;
num_nodes number;
l_index PLS_INTEGER;
l_subIndex PLS_INTEGER;
starttime Date;
temp_datatime VARCHAR(25);
columnname varchar2(300);
columnvalue varchar2(300);
-- creating a Type which is a type of "Instrument_Details" table RowType, which I created in SVN server
TYPE tab_type IS TABLE OF Instrument_Details%ROWTYPE;
t_tab tab_type := tab_type();
BEGIN
-- Passing the xmlfile and virtual directory name which we gave at the time of directory creation
l_bfile := BFileName('MY_FILES', xmlfile);
dbms_lob.createtemporary(l_clob, cache=>FALSE);
dbms_lob.open(l_bfile, dbms_lob.lob_readonly);
--dbms_lob.loadFromFile(dest_lob => l_clob,
-- src_lob => l_bfile,
-- amount => dbms_lob.getLength(l_bfile));
dbms_lob.loadclobfromfile(l_clob, l_bfile, dbms_lob.getlength(l_bfile),
l_dest_offset, l_src_offset, l_Char_set_id, l_lang_context, l_warning);
dbms_lob.close(l_bfile);
-- make sure implicit date conversions are performed correctly
dbms_session.set_nls('NLS_DATE_FORMAT','''YYYY-MON-DD HH24:MI:SS''');
dbms_output.put_line('Date format set');
-- Create a parser.
l_parser := dbms_xmlparser.newParser;
dbms_output.put_line('output 1');
-- Parse the document and create a new DOM document.
dbms_xmlparser.parseclob(l_parser, l_clob);
*-- Below lines are skipping....*
dbms_output.put_line(' passed parsing');
l_doc := dbms_xmlparser.getDocument(l_parser);
dbms_output.put_line(' passed getdocument');
-- Free resources associated with the CLOB and Parser now they are no longer needed.
dbms_lob.freetemporary(l_clob);
dbms_xmlparser.freeParser(l_parser);
-- Get a list of all the EMP nodes in the document using the XPATH syntax.
l_nl1 := dbms_xslprocessor.selectNodes(dbms_xmldom.makeNode(l_doc),'//mcd/child');
-- Loop through the list and create a new record in a tble collection
FOR cur_sel IN 0 .. dbms_xmldom.getLength(l_nl1) - 1 LOOP
l_n := dbms_xmldom.item(l_nl1, cur_sel);
t_tab.extend;
-- Use XPATH syntax to assign values to he elements of the collection.
dbms_xslprocessor.valueOf(l_n,'child1/text()',t_tab(t_tab.last).country);
-- putting the state and vendorname into the table rowtype
   dbms_xslprocessor.valueOf(l_n,'child2/text()',t_tab(t_tab.last).state);
END LOOP;
-- getting the version and putting into the table rowtype
   l_n := dbms_xslprocessor.selectSingleNode(dbms_xmldom.makeNode(l_doc),'//mcd/mc/newid/id');
dbms_xslprocessor.valueOf(l_n,'id/text()',t_tab(t_tab.last).id);
-- selecting the nodes whose starting tag is "mindex"
l_nl1 := dbms_xslprocessor.selectNodes(dbms_xmldom.makeNode(l_doc),'//mcd/mc/mindex');
-- checking the total number of nodes whose starting through "mi"
num_nodes := dbms_xmldom.getLength(l_nl1);
l_index := 1;
-- For loop to iterate the nodes.
FOR cur_sel IN 0 .. dbms_xmldom.getLength(l_nl1) - 1 LOOP
-- whole current node is selected and storing into the node1 variable
node1 := dbms_xmldom.item(l_nl1, cur_sel);
-- setting the xmltype as AL32UTF8
   l_xmltype := xmltype(l_bfile, nls_charset_id('AL32UTF8'));
-- if selecting parent node containing the mt child node then only proceed else skip that parent node.
    IF (l_xmltype.Existsnode('//mcd/mc/mindex[' || l_index || ']/mt') > 0 and l_xmltype.Existsnode('//mcd/mc/mindex[' || l_index || ']/mvalue/r') > 0) Then
-- fetch the datatime, convert it into to_date format and store it into table rowtype
    temp_datatime := dbms_xslprocessor.valueOf(node1, 'date/text()');
    t_tab(t_tab.last).data_time := to_char(to_date(temp_datatime, 'YYYYMmcDHH24MISS'));
    l_subIndex := 1;
   while (l_xmltype.Existsnode('//mcd/mc/mindex[' || l_index || ']/mt[' || l_subIndex || ']') > 0 and l_xmltype.Existsnode('//mcd/mc/mindex[' || l_index || ']/mvalue/r['|| l_subIndex || ']') > 0 ) LOOP
-- getting mt and corresponging mvalue/r values
   dbms_xslprocessor.valueOf(node1,'mt[' || l_subIndex || ']/text()',columnname);
dbms_xslprocessor.valueOf(node1,'mvalue/r[' || l_subIndex || ']/text()',columnvalue);
l_subIndex := l_subIndex + 1;
-- getting the column to which this mapping belongs.
select columnid into colid from abc_table where columnname=name;
CASE colid
WHEN 1 THEN t_tab(t_tab.last).col1 := columnvalue;
WHEN 2 THEN t_tab(t_tab.last).col2 := columnvalue;
WHEN 3 THEN t_tab(t_tab.last).col3 := columnvalue;
      ELSE dbms_output.put_line('No column mapping for counter ' || columnname) ;
END CASE; -- end of case statement.
END LOOP;
-- Insert data into the real table from the table collection.
FORALL i IN t_tab.first .. t_tab.last
    INSERT INTO test_hem VALUES t_tab(i);
END IF;
l_index := l_index + 1;
COMMIT;
END LOOP;
commit;
EXCEPTION
WHEN OTHERS THEN
ecode := SQLCODE;
emesg := SQLERRM;
dbms_output.put_line(TO_CHAR(ecode) || '-' || emesg);
dbms_lob.freetemporary(l_clob);
dbms_xmlparser.freeParser(l_parser);
dbms_xmldom.freeDocument(l_doc);
END;Thanks in advance for your help...

ANN: XML Parser for Java v2.0.2.6

The v2.0.2.6 of the XML Parser for Java is now available for download. The following features and bug fixes are included:
Changes:
Conformance to the XSLT/XPATH October REC.
New API in XSLStylesheet class:
removeParam(String param)
resetParams()
Bug fixes:
Bug #1111423: OutOfMemory exception, if multiple calls made to document()
Bug #1101028: Unexpected character error in DTD parsing document using Docbook DTD
Bug #1101021: #default not supported in exclude-result-prefixes
Bug #1099830: Extra characters inserted into output using the XML Parser
Bug #1099663: HTML output does not allow only doctype-public to be specified
Bug #1099536: HTML output does not disable escaping for script, style unless lowercase
Bug #1098738: ArrayOutOfBoundsException xsl:if test="not(@a)'"
Bug #1095047: XSLProcessor NPE'S on named templates with non-empty namespaces
Bug #1094971: XSLStylesheet needs methods for removing parameters
Bug #1092351: Using valueof() shuffles order of elements in my source document
Bug #1086663: xsl:sort data-type attribute can now be a namespace-prefixed name
Bug #1086661: xsl:version attribute now required on literal result element
Bug #1064692: Default xml-serialization should use empty-element syntax
Bug #1064689: Current() function doesn't work correctly
This is the sixth production patch release for v2.
Oracle XML Team http://technet.oracle.com
Oracle Technology Network
null

The link has been fixed. You will go to the v2 download page
now. Sorry for the inconvience.
Oracle XML Team
http://technet.oracle.com
Oracle Technology Network
Renilton Oliveira (guest) wrote:
: I didn't find the file for version 2.0.0.0 as well.
: Renilton
: Andrei Filimonov (guest) wrote:
: : I tried to download XML Parser for Java v2 it seems that
only
: v
: : 1.0.1.4 is available. Could you please give an exact URL for
: v2
: : download?
: : Andrei Filimonov
: : Oracle XML Team wrote:
: : : The Oracle XML v2 parser is now available for download
here
: as
: : : an early beta release and is written in Java. It features
: an
: : : improved architecture over the Oracle XML v1 parser and
has
: : : shown better performance on small to large XML documents.
: It
: : : will also be able to format the XML document according to
a
: : : stylesheet, having integrated an XSLT processor.
: : : Version 2 of the XML Parser for Java, besides
incorporating
: an
: : : XSLT processor, has been re-architected from version 1.
This
: : has
: : : resulted in a number of changes to the class names
: especially
: : : those that support Namespaces. See v2changes.txt and
: the .diff
: : : difference files in the sample directory.
: : : Oracle XML Team
: : : http://technet.oracle.com
: : : Oracle Technology Network
null

Problem in IAS Server with generation of PDF with FOP and XML parsing

Dear co.forumers :
We are developing an application that generates a pdf document with FOP Apache utilities, and this PDf is stored in a folder of the server. Firts of all, a DDBB data is retrieved using java ( jdbc) middleware classes ( servlets ), and with these data, the pdf is generated using FOP. In these process, he data retrieved from the DDBB is converted in a XML file , and the with the XSL FOP utility, the pdf is generated.
In the other hand, we have another application that consults data from a java beans, using a JSP, generates a XML with the data of the beans, and with this XML, and with a XSL file generates and html , with the data and fixed text.
Both applications have been deployed as a separated WebApp in the IAS. Both use Xalan as XML parser amb Xerces is used aswell.
So, the problem is , after generating a pdf with the firts application, every html generated via JSP-XML-XSL transformation in the second application is shown completelly bad formated. All fixed text are wrongly composed, and the text that should appear in textFields and ComboBoxes is also bad formated. The XML with the data is well generated, so the problem is in the parsing of these XML and the conversion in html through XSL. We have to re-initiate the server and then everything goes well again. If no pdf is launched, no problem occurs with the other application.
How an a WEB APP pplication can affect to another, installed both in different folders of the IAS ? We tried to change Xalan version and FOP version, but no result.
Is this a Poltergeist ?? and X file , maybe ?
Has anyone had the same problem ?
Pleae, any help will be very wellcomed.

A PDF document may be generated in the JDeveloper OC4J server.
http://www.oracle.com/technology/pub/notes/technote_vohra_fop.html

XML Parser returning XML content

This probably seems like overkill..
Im building a JSP that reads in XML and parses it. I then need to filter for specific values in the XML, and return a chunk of the XML to the output stream.
So far I've got..
      //set-up in and content-type
        FileInputStream file = new FileInputStream (getServletContext().getRealPath("/games_arcade/xml/game_marketing_content.xml"));
        //response.setContentType("text/xml");
        //parse the request
        Document doc = null ;
        try {
               DocumentBuilderFactory docBuilderFact = DocumentBuilderFactory.newInstance();
               DocumentBuilder docBuilder = docBuilderFact.newDocumentBuilder();
               doc = docBuilder.parse(file);
               out.println("Go on ya good thing!!!");
        } catch (ParserConfigurationException pcEx) {
            throw new ServletException("jaxp not configured!", pcEx);
        //get needed data from the xml
        NodeList games = doc.getElementsByTagName("game");
          out.println(output.toString());
          Now Im lost. All the requisites are there, imports etc, Im just stuck with filtering the NodeList.
Or am I completely wrong with this approach?
Any help greatly appreciated.
thanks!
Stephen

Hi,
thanks for the replyand suggestions. Have a bit of an issue with this method, this method (due to technical limitations) is completely JSP, using a standard iPlanet 6.1 classpath. I cant deploy WAR's (and cant use taglibs). Xpath/XSL would be ideal, but how do I write the output as XML?
So are you suggesting
<xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
<xsl:output method="html"/>
    <xsl:template match="/">
        <xsl:for-each select="games/game">
          <xsl:if test="@id = $GAMEID">
               ... output the xml node using value-of's etc...
          </xsl:if>
        </xsl:for-each>
      </xsl:template>
    </xsl:stylesheet>Cheers Dr Clap!
Stephen

XML parsing problems with Oracle XML parser for PL/SQL

I am using the Oracle XML parser for PL/SQL to transform XML into HTML usign XSL. The parser gives me sporadic errors trying to parse various XML documents. For the same XML source, the XMLPARSER will parse without errors one day and the next day will give me errors such as "invalid char in text", "Expected ';'.", or even Java null pointer exceptions.
I have verified that the XML sources are well formed, so I am pretty certain that the problems I am having are bugs in the XMLPARSER.
Is there going to be a XML parser for PL/SQL version 2 that will fix these bugs? If so, when??? If not, is there anything else I could do to fix these problems?
Thanks!

You can use the latest version.

Mapping Error - XML Parser Exception

We have a scenario where an ABAP Client Proxy sends a message to PI. This message is transformed into the mail package format and an email sent via the mail adapter.
The Interface Mapping has 2 steps. The 1st step uses an XSL transformation to format the <Content> as html. This step also sets the <Subject>, <To> and <Content_Type> fields. The 2nd step uses a graphical message mapping to dynamically set the <From> field based on the sender system.
We are experiencing the following intermittent error with this scenario:
<SAP:Category>Application</SAP:Category>
<SAP:Code area="MAPPING">EXCEPTION_DURING_EXECUTE</SAP:Code>
<SAP:P1>com/sap/xi/tf/_EUK_MM_MailFrom_</SAP:P1>
<SAP:P2>com.sap.aii.utilxi.misc.api.BaseRuntimeException</SAP:P2>
<SAP:P3>Fatal Error: com.sap.engine.lib.xml.parser.Parser~</SAP:P3>
This relates to the 2nd mapping step (graphical message mapping) described above.
The message trace provides the following detail
<Trace level="1" type="T">com.sap.aii.utilxi.misc.api.BaseRuntimeException: Fatal Error: com.sap.engine.lib.xml.parser.ParserException: XMLParser: Entity 'lt' undefined(:main:, row:3, col:227) at com.sap.aii.mappingtool.tf3.Transformer.checkParserException
This suggests that the XML produced by the XSL transformation is invalid.
The strange thing is that it is very difficult to replicate the error as it doesn't always occur even with the same payload! If the message is restarted it processes successfully. Testing the interface mapping in the IR, with the same payload, works successfully.
I would welcome any ideas you may have on this.
Thanks
Mark

Hi Mark,
thing i know now why it works at "restart".
When the message comes first you get the character '<' in your xml for example: '&<(0'
and you get the ParserException.
Now the message is stored for queueing in the db. I assume the characters are now stored in the 'right way': '&amp;&lt;(0'
When the message now will be restarted everything works fine.
Regards,
Olaf

XML-Parser for PL/SQL: OutOfMemoryError

When using the XML-Parser for transforming an
XML document into HTML, I sometimes get the
following error:
ORA-29516: Aurora assertion failure: Assertion failure at eox.c:187
Uncaught exception System error: java/lang/OutOfMemoryError
ORA-04030: out of process memory when trying to allocate 25509408 bytes (joxcx
callheap,ioc_allocate ufree)
The operation nevertheless seems to finish correctly. I run Oracle 8.1.6 under WinNT4
(384MB RAM) with a shared_pool_size of about
30MB and a java_pool_size of 40MB.
The original document is stored in a CLOB,
so I use xmlparser.parseClob to get it.
After the parsing is finished, the result is written to a file using xmldom.writeToFile.
Can anybody point me in the right direction?
Many thanks in advance!
Roman ([email protected])
null

May I should have said that the memory error
occurs only when transforming a lot of xml documents at once using a cursor loop.
This is the procedure I use:
CREATE OR REPLACE procedure transformer is
p xmlparser.Parser;
q xmlparser.Parser;
xmldoc xmldom.DOMDocument;
xmldocnode xmldom.DOMNode;
proc xslprocessor.Processor;
ss xslprocessor.Stylesheet;
xsldoc xmldom.DOMDocument;
docfrag xmldom.DOMDocumentFragment;
docfragnode xmldom.DOMNode;
doc0 clob;
doc1 clob;
doc2 clob;
v_buffer long;
dir varchar2(100) := 'e:\grammis\html\parser';
xslfile varchar2(100) := 'detail.xsl';
errfile varchar2(100) := 'fehler.txt';
cursor c1 is select id from tb_infoknoten order by id desc;
begin
q := xmlparser.newParser;
-- parse xsl file
xmlparser.parse(q, dir | | '/' | | xslfile);
-- get document
xsldoc := xmlparser.getDocument(q);
-- make stylesheet
ss := xslprocessor.newStylesheet(xsldoc, dir | | '/' | | xslfile);
-- process xsl
proc := xslprocessor.newProcessor;
xslprocessor.showWarnings(proc, true);
xslprocessor.setErrorLog(proc, dir | | '/' | | errfile);
-- new parser
p := xmlparser.newParser;
-- set some characteristics
xmlparser.setValidationMode(p, FALSE);
xmlparser.setErrorLog(p, dir | | '/' | | errfile);
xmlparser.setPreserveWhiteSpace(p, TRUE);
xmlparser.setBaseDir(p, dir);
for h1 in c1 loop
select inhalt into doc2 from tb_infoknoten where id=h1.id;
DBMS_LOB.CREATETEMPORARY(doc0,TRUE, DBMS_LOB.SESSION);
v_buffer:='<?xml version="1.0" encoding="ISO-8859-1"?>';
dbms_lob.write(doc0,43,1,v_buffer);
DBMS_LOB.OPEN(doc0, DBMS_LOB.LOB_READWRITE);
DBMS_LOB.OPEN(doc2, DBMS_LOB.LOB_READONLY);
DBMS_LOB.COPY(doc0, doc2, DBMS_LOB.GETLENGTH(doc2), 44, 1);
DBMS_LOB.CLOSE(doc0);
DBMS_LOB.CLOSE(doc2);
xmlparser.parseClob(p, doc0);
DBMS_LOB.FREETEMPORARY(doc0);
-- get document
xmldoc := xmlparser.getDocument(p);
docfrag := xslprocessor.processXSL(proc, ss, xmldoc);
docfragnode := xmldom.makeNode(docfrag);
-- writing transformed document
xmldom.writeToFile(docfragnode, dir | | '/' | | h1.id | |'.html');
end loop;
end transformer;
null

Html - XML parser

Similar Messages

Maybe you are looking for