17 Creating and Accessing Data Through URLs

Accessing and Processing Data Through HTTP

Any resources stored inside Oracle XML DB repository can also be retrieved by using the HTTP Server in Oracle XML DB. Oracle Database also includes a servlet that makes table data available through HTTP URLs. The data can be returned as plain text, HTML, or XML.

Any Web-enabled client or application can use the data without SQL programming or any specialized database API. You can retrieve the data by linking to it in a Web page or by requesting it through the HTTP-aware APIs of Java, PL/SQL, or Perl. You can display or process the data through any kind of application, including a regular Web browser or an XML-aware application such as a spreadsheet. The servlet supports generating XML and non-XML content and also transforming the results using XSLT style sheets.

Creating Columns and Storing Data Using UriType

You can create database columns using UriType or its child types, or you can store just the text of each URI or URL and create the object types when needed. When storing a mixture of subtypes in the database, you can define a UriType column that can store various subtypes within the same column.

Because these capabilities use object-oriented programming features such as object types and methods, you can derive your own types that inherit from the Oracle-supplied ones. Deriving new types lets you use specialized techniques for retrieving the data or transforming or filtering it before returning it to the program.

UriFactory Package

When storing just the URI text in the database, you can use the UriFactory package to turn each URI into an object of the appropriate subtype. UriFactory package creates an instance of the appropriate type by checking what kind of URI is represented by a given string. For example, any URI that begins with http:// is considered an HTTP URL. When the UriFactory package is passed such a URI string, it returns an instance of a HttpUriType object.

CREATE table uri_tab (url URIType);
INSERT INTO uri_tab VALUES (httpuritype.createuri('http://www.oracle.com'));  
INSERT INTO uri_tab VALUES
       (dburitype.createuri('/SCOTT/EMP/ROW[ENAME="Jack"]'));
SELECT e.url.getclob() FROM uri_tab e;

Other Sources of Information About URIs and URLs

Before you explore the features in this chapter, you should be familiar with the notation for various kinds of URIs.

What Is a URI?

A URI, or Uniform Resource Identifier, is a generalized kind of URL. Like a URL, it can reference any document, and can reference a specific part of a document. It is more general than a URL because it has a powerful mechanism for specifying the relevant part of the document. A URI consists of two parts:

• URL, that identifies the document using the same notation as a regular URL.

• Fragment, that identifies a fragment within the document. The notation for the fragment depends on the document type. For HTML documents, it has the form #anchor_name. For XML documents, it uses XPath notation.

The fragment appears after the # in the following examples.

Advantages of Using DBUri and XDBUri

The following are advantages of using DBUri and XDBUri:

• Reference style sheets within database-generated Web pages. Oracle-supplied package DBMS_METADATA uses DBUri to reference XSL style sheets. XDBUri can also be used to reference XSLT style sheets stored in Oracle XML DB repository.

• Reference HTML, images and other data stored in the database. The URLs can be used to point to data stored in tables or in the repository hierarchical folders.

• Improved Performance by bypassing the Web server. If you already have a URL in your XML document, then you can replace it with a reference to the database by either:

  • Using a servlet

  • Using a DBUri or XDBUri to bring back the results

  Using DBUri or XDBUri has performance benefits because you interact directly with the database rather than through a Web server.

• Accessing XML Documents in the Database Without SQL. You are not required to know SQL to access an XML document stored in the database. With DBUri you can access an XML document from the database without using SQL.

  Because the files or resources in Oracle XML DB repository are stored in tables, you can access them either through the XDBUri or by using the table metaphor through the DBUri.

  
  

  See Also: PL/SQL Packages and Types Reference, "DBMS_METADATA package"

  
  

Advantages of Using UriType Values

Oracle already provides the PL/SQL package UTL_HTTP and the Java class java.net.URL to fetch URL references. The advantages of defining this new UriType datatype in SQL are:

• Improved Mapping of XML Documents to Columns. Uri-ref support is needed when exploding XML documents into object-relational columns, so that the Uri-ref specified in documents can map to a URL column in the database.

• Unified access to data stored inside and outside the server. Because you can use UriType values to store pointers to HTTP/DB urls, you get a unified access to the data wherever it is stored. This lets you create queries and indexes without having to worry about where the data resides.

  
  

  See Also: "Using UriType Objects"

  
  

UriType Functions

The UriType abstract type supports a variety of functions that can be used over any subtype. Table 17-1 lists the UriType member functions.

getContentType() Function

getContentType() function returns the MIME information for the URL. The HttpUriType de-references the URL and gets the MIME header information. You can use this information to decide whether to retrieve the URL as BLOB or CLOB based on the MIME type. You would treat a Web page with a MIME type of x/jpeg as a BLOB, and one with a MIME type of text/plain or text/html as a CLOB.

declare
  httpuri HttpUriType;
  x clob;  
  y blob;
begin
  httpuri := HttpUriType('http://www.oracle.com/object1');
  if httpuri.getContentType() = 'application-x/bin' then
     y := httpuri.getblob();
  else
     x := httpuri.getclob();
  end if;
end;

getXML() Function

getXML() function returns XMLType information for the result. If the document is not valid XML (or XHTML), then an error is thrown.

Formulating the DBUri

The URL syntax is obtained by specifying XPath-like syntax over a virtual XML visualization of the database. See Figure 17-1, "DBUri: Visual or SQL View, XML View, and Associated XPath":

• The visual model is a hierarchical view of what a current connected user would see in terms of SQL schemas, tables, rows, and columns.

• The XML view contains a root element that maps to the database. The root XML element contains child elements, which are the schemas on which the user has some privileges on any object. The schema elements contain tables and views that the user can see. A child element is ann element that is wholly contained within another, referred to as its parent element. For example <Parent><Child></Child></Parent> illustrates a child element nested within its parent element.

<?xml version="1.0"?>
<oradb SID="ORCL">
 <PUBLIC>
    <ALL_TABLES>
       ..
     </ALL_TABLES>
    <EMP>
      <!-- EMp table -->
    </EMP>
 </PUBLIC>
 <SCOTT>
  <ALL_TABLES>
    ....
  </ALL_TABLES>
  <EMP>
   <ROW>
     <EMPNO>1001</EMPNO>
     <ENAME>John</ENAME>
     <EMP_SALARY>20000</EMP_SALARY>
   </ROW>
   <ROW>
     <EMPNO>2001</EMPNO>
       </ROW>
  </EMP>
  <DEPT>
   <ROW>
     <DEPTNO>200</DEPTNO>
     <DNAME>Sports</DNAME>
   </ROW>
  </DEPT>
</SCOTT>
<JONES>
 <CUSTOMER_OBJ_TAB>
   <ROW>
     <NAME>xxx</NAME>
     <ADDRESS>
        <STATE>CA</STATE>
        <ZIP>94065</ZIP>
     </ADDRESS>
   </ROW>
  </CUSTOMER_OBJ_TAB>
 </JONES>
</oradb>

Figure 17-1 DBUri: Visual or SQL View, XML View, and Associated XPath

Description of the illustration adxdb053.gif

This XML document is constructed at the time you do the query and based on the privileges that you have at that moment.

You can make the following observations from Example 17-4:

• User scott can see the scott database schema and jones database schema. These are schemas on which the user has some table or views that he can read.

• Table emp shows up as EMP with row element tags. This is the default mapping for all tables. The same for dept and the customer_obj_tab table under the jones schema.

• In this release, null elements are absent

• There is also a PUBLIC element under which tables and views are accessible without schema qualification. For example, a SELECT query such as:

  SELECT * FROM emp;

  when queried by user scott, matches the table emp under the scott schema and, if not found, tries to match a public synonym named emp. In the same way, the PUBLIC element contains:

  • All the tables and views visible to users through their database schema

  • All the tables visible through the PUBLIC synonym

Notation for DBUriType Fragments

With the Oracle Database being visualized as an XML tree, you can perform XPath traversals to any part of the virtual document. This translates to any row-column intersection of the database tables or views. By specifying an XPath over the visualization model, you can create references to any piece of data in the database.

DbUri is specified in a simplified XPath format. Currently, Oracle does not support the full XPath or XPointer recommendation for DBURType. The following sections discuss the structure of the DBUri.

You can create DBUri references to any piece of data. You can use the following instances in a column as reference:

• Scalar

• Object

• Collection

• An attribute of an object type within a column. For example:.../ROW[empno=7263]/COL_OBJ/OBJ_ATTR

These are the smallest addressable units. For example, you can use:

/oradb/SCOTT/EMP

or

/oradb/SCOTT/EMP/ROW[empno=7263]

DBUri Syntax Guidelines

There are restrictions on the kind of XPath queries that can be used to specify a reference. In general, the fragment part must:

• Include the user database schema name or specify PUBLIC to resolve the table name without a specific schema.

• Include a table or view name.

• Include the ROW tag for identifying the ROW element.

• Identify the column or object attribute that you wish to extract.

• Include predicates at any level in the path other than the schema and table elements.

• Indicate predicates not on the selection path in the ROW element.

/oradb/scott/purchase_obj_tab/ROW/line_item_list

/oradb/scott/purchase_obj_tab/ROW[pono=100]/line_item_list

• A DBUri must identify exactly a single data value, either an object type or a collection. If the data value is an entire row, then you indicate that by including a ROW node. The DBUri can also point to an entire table. Note that only valid XML can be returned.

/oradb/SCOTT/EMP/ROW[ADDRESS/STATE='CA' OR ADDRESS/STATE='OR']/ADDRESS[CITY='Portland' OR ZIPCODE=94404]/CITY

Identifying the Whole Table

This returns an XML document that retrieves the whole table. The enclosing tag is the name of the table. The row values are enclosed inside a ROW element:

/oradb/schemaname/tablename

/oradb/SCOTT/EMP

<?xml version="1.0"?>
<EMP>
 <ROW>
   <EMPNO>7369</EMPNO>
   <ENAME>Smith</ENAME>
   ... <!-- other columns -->
 </ROW>
 <!-- other rows -->
</EMP>

Identifying a Particular Row of the Table

This identifies a particular ROW element in the table. The result is an XML document that contains the ROW element with its columns as child elements. Use the following syntax:

/oradb/schemaname/tablename/ROW[predicate_expression]

/oradb/SCOTT/EMP/ROW[EMPNO=7369]

<?xml version="1.0"?>
<ROW>
  <EMPNO>7369</EMPNO>
  <ENAME>SMITH</ENAME>
  <JOB>CLERK</JOB>
  <!-- other columns -->
</ROW>

Identifying a Target Column

In this case, a target column or an attribute of a column is identified and retrieved as XML.

Use the following syntax:

/oradb/schemaname/tablename/ROW[predicate_expression]/columnname
/oradb/schemaname/tablename/ROW[predicate_expression]/columnname/attribute1/../attributen

/oradb/SCOTT/EMP/ROW[EMPNO=7369 and DEPTNO=20]/ENAME

<?xml version="1.0"?>
<ENAME>SMITH</ENAME>

/oradb/SCOTT/EMP/ROW[EMPNO=7369]/ADDRESS/STATE

<?xml version="1.0"?>
<STATE>CA</STATE>

Retrieving the Text Value of a Column

In many cases, it can be useful to retrieve only the text values of a column and not the enclosing tags. For example, if XSLT style sheets are stored in a CLOB column, you can retrieve the document text without having any enclosing column name tags. You can use the text() function for this. It specifies that you only want the text value of the node. Use the following syntax:

/oradb/schemaname/tablename/ROW[predicate_expression]/columnname/text()

/oradb/SCOTT/EMP/ROW[EMPNO=7369]/ENAME/text()

How DBUris Differ from Object References

A DBUri can access columns and attributes and is loosely typed Object references can only access row objects. DBUri is a superset of this reference mechanism.

DBUri Applies to a Database and Session

A DBUri is scoped to a database and session. You must already be connected to the database in a particular session context. The schema and permissions needed to access the data are resolved in that context.

Where Can DBUri Be Used?

Uri-ref can be used in a number of scenarios, including those described in the following sections:

In the case of a travel story Web site where you store travel stories in a table, you might create links to related stories. By representing these links in a DBUriType column, you can create intra-database links that let you retrieve related stories through queries.

Applications can use XSLT style sheets to convert XML into other formats. The style sheets are represented as XML documents, stored as CLOBs. The application can use DBUriType objects:

• To access the XSLT style sheets stored in the database for use during parsing.

• To make references, such as import or include, to related XSLT style sheets. You can encode these references within the XSLT style sheet itself.

  
  

  Note: A DBUri is not a general purpose XPointer mechanism to XML data. It is not a replacement for database object references. The syntax and semantics of references differ from those of Uri-refs. It does not enforce or create any new security models or restrictions. Instead, it relies on the underlying security architecture to enforce privileges.

  
  

How to Create an Instance of XDBUriType

XDBUriType is automatically registered with UriFactory so that an XDBUriType instance can be generated by providing the URI to the getURI method.

Currently, XDBUriType is the default UriType generated by the UriFactory.getUri method, when the URI does not have any of the recognized prefixes, such as http://,/DBURI, or /ORADB.

All DBUriType URIs should have a prefix of either /DBURI or /ORADB, case insensitive.

SELECT sys.UriFactory.getUri('/home/scott/doc1.xml') FROM dual;

CREATE TABLE uri_tab (poUrl SYS.UriType, poName VARCHAR2(1000));
  -- We create an abstract type column so any type of URI can be used
 
-- Insert an absolute url into poUrl
-- The factory will create an XDBUriType because there's no prefix.
-- Here, po1.xml is an XML file that is stored in /public/orders/
INSERT INTO uri_tab VALUES
  (UriFactory.getUri('/public/orders/po1.xml'), 'SomePurchaseOrder');
 
-- Get all the purchase orders
SELECT e.poUrl.getClob(), poName FROM uri_tab e;

-- Using PL/SQL, you can access table uri_tab as follows: 
CREATE FUNCTION returnclob()
RETURN CLOB
IS
  a UriType;
BEGIN
  -- Get absolute URL for purchase order named like 'Some%'
   SELECT poUrl INTO a FROM uri_tab WHERE poName LIKE 'Some%';
   RETURN a.getClob();
END;
/

SELECT e.poUrl.getClob() FROM uri_tab e
   WHERE extractValue(e.poUrl.getXML(),'/User') = 'SCOTT';

Storing Pointers to Documents with UriType

As explained earlier, UriType is an abstract type containing a VARCHAR2 attribute that specifies the URI. The object type has functions for traversing the reference and extracting the data.

You can create columns using UriType to store these pointers in the database. Typically, you declare the column using the UriType, and the objects that you store use one or more of the derived types such as HttpUriType.

Table 17-4 lists some useful UriType methods.

CREATE TABLE uri_tab (poUrl SYS.UriType, poName VARCHAR2(200));
       -- We have created abstract type columns; if you know what kind of URIs
       -- you are going to store, you can create the appropriate types.

 -- Insert an absolute URL into SYS.UriType.
 -- The Urifactory creates the correct instance (in this case a HttpUriType)
INSERT INTO uri_tab VALUES 
  (sys.UriFactory.getUri('http://www.oracle.com/cust/po'),'AbsPo');

-- Insert a URL by directly calling the SYS.HttpUriType constructor. 
-- THIS IS *STRONGLY DISCOURAGED*. 
-- Note the absence of the http:// prefix when creating SYS.HttpUriType
-- instance through the default constructor.
INSERT INTO uri_tab VALUES (sys.HttpUriType('proxy.us.oracle.com'),'RelPo');

-- Extract all the purchase orders
SELECT e.poUrl.getClob(), poName FROM uri_tab e;

-- In PL/SQL
CREATE FUNCTION returnclob()
RETURN CLOB
IS
  a SYS.UriType;
BEGIN
  SELECT poUrl INTO a FROM uri_Tab WHERE poName LIKE 'RelPo%';
  RETURN a.getClob();
END;
/

INSERT INTO uri_tab VALUES 
  (UriFactory.getUri(
     '/oradb/SCOTT/PURCHASE_ORDER_TAB/ROW[PONO=1000]'), 
     'ScottPo');

SELECT e.poURL.getclob() FROM uri_tab e;

Registering New UriType Subtypes with the UriFactory Package

The UriFactory package lets you register new UriType subtypes:

• Derive these types using the CREATE TYPE statement in SQL.

• Override the default methods to perform specialized processing when retrieving data, or to transform the XML data before displaying it.

• Pick a new prefix to identify URIs that use this specialized processing.

• Register the prefix using UriFactory.registerURLHandler, so that the UriFactory package can create an instance of your new subtype when it receives a URI starting with the new prefix you defined.

For example, you can invent a new protocol ecom:// and define a subtype of UriType to handle that protocol. Perhaps the subtype implements some special logic for getCLOB, or does some changes to the XML tags or data within getXML. When you register the ecom:// prefix with UriFactory, any calls to UriFactory.getUri generate the new subtype instance for URIs that begin with the ecom:// prefix.

CREATE TABLE url_tab (urlcol varchar2(80));
-- Insert an HTTP URL reference
INSERT INTO url_tab VALUES ('http://www.oracle.com/');
-- Insert a DBUri-ref reference
INSERT INTO url_tab VALUES ('/oradb/SCOTT/EMP/ROW[ENAME="Jack"]');
-- Create a new type to handle a new protocol called ecom://
-- This is just an example template. For it to execute, the implementations
-- of these functions need to be specified.
CREATE TYPE EComUriType UNDER SYS.UriType (
  OVERRIDING MEMBER FUNCTION getClob RETURN CLOB,
  OVERRIDING MEMBER FUNCTION getBlob RETURN BLOB,
  OVERRIDING MEMBER FUNCTION getExternalUrl RETURN VARCHAR2,
  OVERRIDING MEMBER FUNCTION getUrl RETURN VARCHAR2,
  -- Must have this for registering with the URL handler
  STATIC FUNCTION createUri(url IN VARCHAR2) RETURN EcomUriType);
/
-- Register a new handler for the ecom:// prefixes
BEGIN
  -- The handler type name is ECOMUriTYPE; schema is SCOTT
  -- Ignore the prefix case, so that UriFactory creates the same subtype
  -- for URIs beginning with ECOM://, ecom://, eCom://, and so on.
  -- Strip the prefix before calling the createUri function
  -- so that the string 'ecom://' is not stored inside the
  -- ECOMUriTYPE object. (It is added back automatically when
  -- you call ECOMUriTYPE.getURL.)
  urifactory.registerURLHandler (prefix => 'ecom://',
                                 schemaname => 'SCOTT',
                                 typename => 'ECOMURITYPE',
                                 ignoreprefixcase => TRUE,
                                 stripprefix => TRUE);
END;
/
-- Insert this new type of URI into the table
INSERT INTO url_tab VALUES ('ECOM://company1/company2=22/comp');
-- Use the factory to generate an instance of the appropriate
-- subtype for each URI in the table.
SELECT urifactory.getUri(urlcol) FROM url_tab;
-- would now generate
HttpUriType('www.oracle.com'); -- a Http uri type instance
DBUriType('/oradb/SCOTT/EMP/ROW[ENAME="Jack"]', null); -- a DBUriType
EComUriType('company1/company2=22/comp'); -- an EComUriType instance

Rules for Passing Columns or Object Attributes to SYS_DBURIGEN()

The column or attribute passed to the SYS_DBURIGEN() function must obey the following rules:

• Unique mapping: The column or object attribute must be uniquely mappable back to the table or view from which it comes. The only virtual columns allowed are the VALUE and REF operators. The column may come from a TABLE() subquery or an inline view, as long as the inline view does not rename the columns.

• Key columns: Either the rowid or a set of key columns must be specified. The list of key columns is not required to be declared as a unique or primary key, as long as the columns uniquely identify a particular row in the result.

• Same table: All columns referenced in the SYS_DBURIGEN() function must come from the same table or view.

• PUBLIC element: If the table or view pointed by the rowid or key columns does not have a database schema specified, then the PUBLIC keyword is used instead of the schema. When the DBUri is accessed, the table name resolves to the same table, synonym, or view that was visible by that name when the DBUri was created.

• TEXT function: DBUri, by default, retrieves an XML document containing the result. To retrieve only the text value, use the text() keyword as the final argument to the function.

  For example:

  SELECT SYS_DBURIGEN(empno,ename,'text()') FROM scott.emp,
    WHERE empno=7369;

  or you can just generates a URL of the form:

  /SCOTT/EMP/ROW[EMPNO=7369]/ENAME/text()

• Single-column argument: If there is a single-column argument, then the column is used both as the key column to identify the row and as the referenced column.

SELECT SYS_DBURIGEN(empno) FROM emp
  WHERE empno=7369;

/SCOTT/EMP/ROW[EMPNO=7369]/EMPNO,

SYS_DBURIGEN Examples

CREATE TABLE doc_list_tab(docno NUMBER PRIMARY KEY, doc_ref SYS.DBUriType);
-- Insert /SCOTT/EMP/ROW[rowid='xxx']/EMPNO
INSERT INTO doc_list_tab 
  VALUES(1001, 
         (SELECT SYS_DBURIGEN(rowid, empno) FROM emp WHERE empno=100));
-- Insert a Uri-ref to point to the ename column of emp!
INSERT INTO doc_list_tab
  VALUES(1002,
         (SELECT SYS_DBURIGEN(empno, ename) FROM emp WHERE empno=7369));
-- Result of the DBURIGEN looks like /SCOTT/EMP/ROW[EMPNO=7369]/ENAME

When selecting the results of a large column, you might want to retrieve only a portion of the result and create a URL to the column instead. For example, consider the case of a travel story Web site. If all the travel stories are stored in a table, and users search for a set of relevant stories, then you do not want to list each entire story in the result page. Instead, you show the first 100 characters or gist of the story and then return a URL to the full story. This can be done as follows:

CREATE TABLE travel_story (story_name VARCHAR2(100), 
                           story CLOB);
-- Insert Some Value
INSERT INTO travel_story
  VALUES ('Egypt', 'This is the story of my time in Egypt....');

CREATE FUNCTION charfunc(clobval IN CLOB) RETURN VARCHAR2 IS
  res VARCHAR2(20);
  amount NUMBER := 20;
BEGIN
  DBMS_LOB.read(clobval, amount, 1, res);
  RETURN res;
END;
/

CREATE VIEW travel_view AS
  SELECT story_name, charfunc(story) short_story,
         SYS_DBURIGEN(story_name, story, 'text()') story_link
  FROM travel_story;

SELECT * FROM travel_view;
 
STORY_NAME SHORT_STORY          STORY_LINK
---------------------------------------------------
Egypt      This is the story of SYS.DBUriType('/PUBLIC/TRAVEL_STORY/ROW[STORY_NAME='Egypt']/STORY/text()')

You can use SYS_DBURIGEN() in the RETURNING clause of DML statements to retrieve the URL of an object as it is inserted.

CREATE TABLE clob_tab (docid NUMBER, doc CLOB);

CREATE TABLE uri_tab (docs SYS.DBUriType);

DECLARE
  ret SYS.dburitype;
BEGIN
  -- execute the insert and get the url
  EXECUTE IMMEDIATE 
    'INSERT INTO clob_tab VALUES (1,''TEMP CLOB TEST'') 
       RETURNING SYS_DBURIGEN(docid, doc, ''text()'') INTO :1'
    RETURNING INTO ret;
  -- Insert the url into uri_tab
  INSERT INTO uri_tab VALUES (ret);
END;
/

/SCOTT/CLOB_TAB/ROW[DOCID="xxx"]/DOC/text()

DBUri Servlet Mechanism

For the preceding methods, a servlet runs for accessing this information through HTTP. This servlet takes in a path expression following the servlet name as the DBUri reference and produces the document pointed to by the DBUri to the output stream.

The generated document can be a Web page, an XML document, plain text, and so on. You can specify the MIME type so that the browser or other application knows what kind of content to expect:

• By default, the servlet can produce MIME types of text/xml and text/plain. If the URI ends in a text() function, then the text/plain MIME type is used, else an XML document is generated with the MIME type of text/xml.

• You can override the MIME type and set it to binary/x-jpeg or some other value using the contenttype argument to the servlet.

-- Generates a contenttype of text/plain
http://machine.oracle.com:8080/oradb/SCOTT/EMP/ROW[EMPNO=7369]/ENAME/text()
-- Generates a contenttype of text/xml
http://machine.oracle.com:8080/oradb/SCOTT/EMP/ROW[EMPNO=7369/ENAME

Installing DBUri Servlet

DbUriServlet is built into the database, and the installation is handled by the Oracle XML DB configuration file. To customize the installation of the servlet, you must edit it. You can edit the config file, xdbconfig.xml under the Oracle XML DB user, through WebDAV, FTP, from Oracle Enterprise Manager, or in the database. To update the file using FTP or WebDAV, simply download the document, edit it as necessary, and save it back in the database. There are several things that can be customized using the configuration file.

Notice that the servlet is installed at /oradb/* specified in the servlet-pattern tag. The * is necessary to indicate that any path following oradb is to be mapped to the same servlet. The oradb is published as the virtual path. Here, you can change the path that will be used to access the servlet.

DECLARE
  doc XMLType;
  doc2 XMLType;
BEGIN
  doc := dbms_xdb.cfg_get();
  SELECT 
    updateXML(doc,
'/xdbconfig/sysconfig/protocolconfig/httpconfig/webappconfig/servletconfig/
servlet-mappings/servlet-mapping[servlet-name="DBUriServlet"]/servlet-pattern/
text()', 
            '/dburi/*')
    INTO doc2 FROM DUAL;
  DBMS_XDB.cfg_update(doc2);
  COMMIT;
END;
/

DBUri Security

Servlet security is handled by Oracle Database using roles. When users log in to the servlet, they use their database username and password. The servlet will check to make sure the user logging in belongs to one of the roles specified in the configuration file. The roles allowed to access the servlet are specified in the security-role-ref tag. By default, the servlet is available to the special role authenticatedUser. Any user who logs into the servlet with any valid database username and password belongs to this role.

This parameter can be changed to restrict access to any role(s) in the database. To change from the default authenticated-user role to a role that you have created, say servlet-users, run:

DECLARE
  doc XMLType;
  doc2 XMLType;
  doc3 XMLType;
BEGIN
  doc := DBMS_XDB.cfg_get();
  SELECT updateXML(doc,
'/xdbconfig/sysconfig/protocolconfig/httpconfig/webappconfig/servletconfig/
servlet-list/servlet[servlet-name="DBUriServlet"]/security-role-ref/role-name/
text()',
                   'servlet-users')
    INTO doc2 FROM DUAL;
  SELECT updateXML(doc2,
'/xdbconfig/sysconfig/protocolconfig/httpconfig/webappconfig/servletconfig/
servlet-list/servlet[servlet-name="DBUriServlet"]/security-role-ref/role-link/
text()',
                   'servlet-users')
    INTO doc3 FROM DUAL;
  DBMS_XDB.cfg_update(doc3);
  COMMIT;
END;
/

Configuring the UriFactory Package to Handle DBUris

The UriFactory, as explained in "Creating Instances of UriType Objects with the UriFactory Package", takes a URL and generates the appropriate subtypes of the UriType to handle the corresponding protocol. For HTTP URLs, UriFactory creates instances of the HttpUriType. But when you have an HTTP URL that represents a URI path, it is more efficient to store and process it as a DBUriType instance in the database. The DBUriType processing involves fewer layers of communication and potentially fewer character conversions.

After you install OraDBUriServlet, so that any URL such as http://machine-name/servlets/oradb/ gets handled by that servlet, you can configure the UriFactory to use that prefix and create instances of the DBUriType instead of HttpUriType:

begin
 -- register a new handler for the dburi prefix..     
urifactory.registerHandler('http://machine-name/servlets/oradb'
        ,'SYS','DBUriTYPE', true,true);
end;
/

After you execute this block in your session, any UriFactory.getUri() call in that session automatically creates an instance of the DBUriType for those HTTP URLs that have the prefix.