XSLT on DecipherMiddleware

DECIPHER: Dynamic XSLTs in BPEL (SOA 12c)

Sat, 12 Mar 2022 07:45:00 +0000

While developing integrations in BPEL 12c, XSLT is most commonly used to transform different XML message structures. We generally come across scenarios wherein based on specific input fields, our transformation logic changes. What we generally do is have all conditions within the same XSLT file. This makes our XSLT look a lot bigger and difficult to maintain. A better way would be to have different XSLTs and move deciding factor out of XSLT. Once we create different XSLTs for specific use cases having the same output format, now the question will come how to configure these XSLTs to be picked at runtime? This blog is curated with answers to such and many more questions.

Let’s take the example of account creation. There is a source application sending account information in an XML document which we need to update in a backend database. Now source system can send different types of address information like home address, business address, or shipping address. We get this information in the addressType input field. In the backend database, for each type of address, there are different fields in the backend database. There could various ways to handle this.

Create one single XSLT to handle all the scenarios using XSLT functions like if, choose-when , etc. One single XSLT file for all the scenarios, BPEL code will be easy to read. But this will lead to a huge XSLT file and will be a bit tricky to maintain. In the future, if we want to add new conditions for only secondary addresses, then regression testing might be required to ensure we don’t hamper the other functionality.

Another way could be to have separate XSLTs for each scenario. As per the example, we have 3 scenarios. The address can be of type home, business, or shipping address.

We can have 3 XSLTs created for each scenario. Now maintainability of Transformation XSLT code will be a lot easier. These can now be configured using an if-else block within the BPEL process to decide which XSLT to pick. The BPEL process will look now similar to below.

Now suppose tomorrow we have a new condition added, then we need to add another block of “elseif” to accommodate this. This will make your BPEL code look bigger.

In SOA 12c we have below two functions that can be used to shorten the BPEL flow and use the XSLT dynamically.

ora:processXSLT
ora:doXSLTransformForDoc

xmlns:ora =“http://schemas.oracle.com/xpath/extension"

Both of the above functions support the location of the XSLT file and input XML. The above functions can be used in conjunction with DVM to fetch the location of XSLT and assign it to the input of the function.
For the same example, let’s consider below DVM file as an example.

We can use the above DVM file to extract the location of XSLT that can be used. Suppose we store the conditional value in a variable called $addType(having one of the values in the home, business, or shipping). We can use DVM lookup to fetch the resulting XSLT file location that can be provided as input to the above-mentioned functions in an Assign activity. In case none of the conditions matches the default.xsl file can be used.

dvm:lookupValue(“dvms/dynamicXSLT.dvm”, “addressType”,$addType, “xsltLocation”, “transformations/default.xsl”)

Let us assume the output value of the above expression is stored in variable $dynamicXSLT and input XML is stored in $input. The expression will be:

ora:processXSLT($dynamicXSLT, $input)
ora:doXSLTransformForDoc($dynamicXSLT, $input)

Both the above functions do a similar task. But there is one major difference. Suppose your XSLT requires extra complex parameters to be passed as input. In such scenarios ora:doXSLTransformForDoc comes real handy. In this function, we can pass key-value pairs of parameters that can be used within the XSLT.

For example, if our XSLT requires a parameter contact(as defined in XSLT) and the value is stored in the $Contact variable in the BPEL process. Now our function will look like.

ora:doXSLTransformForDoc($dynamicXSLT, $input, ‘contact’, $contact)

References: 🔗

https://docs.oracle.com/middleware/12211/soasuite/develop/GUID-B37B91AF-0036-47EC-AB9F-FCBCD2BA55E2.htm#SOASE2117

Please share your valuable feedback 😊😊😊

Tips and Tricks for handling numbers in BPEL or OIC

Sat, 28 Aug 2021 17:52:00 +0000

While integrating with multiple systems and moving data from one system to the other, we generally come across scenarios, wherein we need to change the format of the number or perform operations like floor, ceil, round, divide, multiply, add, etc. on numbers. We will be going through various mathematical functions available in XSLT 1.0, which would be helpful in such scenarios, and also some generic errors which we encounter usually.

Mathematical Operations

Divide
Multiply
Add
Subtract
Square-root
Round
Floor
Ceil
Advanced Number formatting

Common Errors

Mathematical Operations 🔗

1. Divide 🔗

To divide 2 numbers, we can use the div operator to divide two numbers.

Syntax:

$num1 div $num2

Example:

Let us assume we want to divide 120 by 7.

XSL:

<divide><xsl:value-of select="120 div 7"/></divide>

Output

<divide>17.14285714285714285714</divide>

2. Multiply 🔗

To multiply 2 numbers, we can use the * operator.

Syntax:

$num1 * $num2

Example:

Let us assume we want to multiply 120 by 7.

XSL:

<multiply><xsl:value-of select="120 * 7"/></multiply>

Output

<multiply>840</multiply>

3. Add 🔗

To add 2 numbers, we can use the + operator.

Syntax:

$num1 + $num2

Example:

Let us assume we want to add 120 and 7.

XSL:

<add><xsl:value-of select="120 + 7"/></add>

Output

<add>127</add>

4. Subtract 🔗

To subtract one number from another, we can use the - operator.

Syntax:

$num1 - $num2

Example:

Let us assume we want to subtract 7 from 120.

XSL:

<subtract><xsl:value-of select="120 - 7"/></subtract>

Output

<subtract>113</subtract>

5. Square Root 🔗

To find a square root of a number, we can use the oraext:square-root function.

Syntax:

oraext:square-root($number)

Example:

Let us assume we want to find a square root of 121.

XSL:

<square-root><xsl:value-of select="oraext:square-root(121)"/></square-root>

Output

<square-root>11</square-root>

6. Round 🔗

To round a number to the nearest integer we can use the round($number) function available in XSLT.

Syntax:

round($number)

XSL:

<roundInt><xsl:value-of select="round(17.64285714285714285714)"/></roundInt>

Output

<roundInt>18</roundInt>

What if we want to round the number to 2 or 3 decimal places. Since the round function only takes one argument, we cannot pass the number of decimal digits we want to round the number to. To do so we can try the below-mentioned method.

We can multiply the number by 10m, round the number, and then divide the result by 10m. Here, m denotes the number of decimal places.

Pseudo Code:

(round($number * 10 m) )/ 10 m

Example

Let us assume we want to round to two decimal places. Here, the value of m will be 2. 10m will be equal to 100.

XSL:

<roundDec1><xsl:value-of select="round(17.64285714285714285714 * 100 ) div 100"/></roundDec1>

Output

<roundDec1>17.64</roundDec1>

7. Floor 🔗

To return the largest integer less than or equal to the number we can use the floor($number) function available in XSLT.

Syntax:

floor($number)

XSL:

<floorInt><xsl:value-of select="floor(17.64285714285714285714)"/></floorInt>

Output

<floorInt>17</floorInt>

What if we want to perform floor operation on the number and have output with 2 or 3 decimal places. Since the floor function only takes one argument, we cannot pass the number of decimal digits we want to floor the number to. To do so we can try the below-mentioned method.

We can multiply the number by 10m, floor the number, and then divide the result by 10m. Here, m denotes the number of decimal places.

Pseudo Code:

(floor($number * 10 m) )/ 10 m

Example

Let us assume we want to floor the number to three decimal places. Here, the value of m will be 3. 10m will be equal to 1000.

XSL:

<floorDec1><xsl:value-of select="floor(17.64285714285714285714 * 1000 ) div 1000"/></floorDec1>

Output

<floorDec1>17.642</floorDec1>

8. Ceil 🔗

To return the largest integer greater than or equal to the number we can use the ceiling($number) function available in XSLT.

Syntax:

ceiling($number)

XSL:

<ceilInt><xsl:value-of select="ceiling(17.14285714285714285714)"/></ceilInt>

Output

<ceilInt>18</ceilInt>

What if we want to perform ceiling operation on the number and have output with 2 or 3 decimal places. Since the ceiling function only takes one argument, we cannot pass the number of decimal digits we want to ceil the number to. To do so we can try the below-mentioned method.

We can multiply the number by 10m, ceiling the number, and then divide the result by 10m. Here, m denotes the number of decimal places.

Pseudo Code:

(ceiling($number * 10 m) )/ 10 m

Example

Let us assume we want to ceiling the number to three decimal places. Here, the value of m will be 3. 10m will be equal to 1000.

XSL:

<ceilDec1><xsl:value-of select="ceiling(17.14285714285714285714 * 1000 ) div 1000"/></ceilDec1>

Output

<ceilDec1>17.143</ceilDec1>

9. Advanced Number formatting 🔗

We may come across the scenarios especially when dealing with financial or accounting integrations, wherein we want the numbers like 1234.56 , 45 to be formatted like 1,234.56 , $45.00 , etc. One common way, while developing integration which we opt for, is multiple substrings and then multiple concatenations. There is a built-in function format-number, which can come in really handy in such situations.

Syntax:

format-number($number,$format,[$decimal-format])

Note: Here $decimal-format parameter is optional and we have not used it in examples.

Format Character Table

S.No.	Character	Description
1	0	Representation for digit. For 0 after the decimal and before the number, it is present. For example, the number 1 with the format as ‘00.0’ will give 01.0
2	#	Representation for digit. For 0 after the decimal and before the number, it is absent. For example, the number 1 with the format as ‘##.#’ will give 1
3	.	Separator for decimal.
4	,	Group Separator. Suppose we want to separate thousands, we may use like #,###.##, When 1000 is passed as input, the output becomes 1,000.
5	%	Represent number in %
6	;	Separator to define different formats for negative and positive numbers.

Format number performs the round operation by default(which is banker’s rounding) to the number of decimal places defined in format. To understand more on how can we use various formats. Please find below the table consisting of various use cases

S.No.	Use Case	Format	Input Number	Formatted Number
1	Always having two Decimal digits	#.00	123.456	123.46
			123.4	123.40
2	Having a maximum of 2 decimal digits	#.##	123.456	123.46
			123	123
3	Creating a number in Currency Format	$#,###,###.00	123456.789	$123,456.79
4	Display number in percentage	0.13245	##.##%	13.25%
5	Having different formats for positive and negative numbers.	#.000;(#.000)	132.45	132.450
			-132.45	(132.450)

XSL:

<format1><xsl:value-of select="format-number(123.456, '#.00')"/></format1>
<format1-1><xsl:value-of select="format-number(123.4, '#.00')"/></format1-1>
<format2><xsl:value-of select="format-number(123.456, '#.##')"/></format2>
<format2-1><xsl:value-of select="format-number(123, '#.##')"/></format2-1>
<format3><xsl:value-of select="format-number(123456.789, '$#,###,###.00')"/></format3>
<format4><xsl:value-of select="format-number(0.13245, '##.##%')"/></format4>
<format5><xsl:value-of select="format-number(132.45, '#.000;(#.000)')"/></format5>
<format5-1><xsl:value-of select="format-number(-132.45, '#.000;(#.000)')"/></format5-1>

Output

<format1>123.46</format1>
<format1-1>123.40</format1-1>
<format2>123.46</format2>
<format2-1>123</format2-1>
<format3>$123,456.79</format3>
<format4>13.25%</format4>
<format5>132.450</format5>=
<format5-1>(132.450)</format5-1>

Common Errors 🔗

We get XPath errors when the data type doesn’t match the numeric data type.
We get NaN(Not a Number) error when we explicitly try to convert a string or any other data type to a number using the number() function and the string contains some invalid characters apart from what is accepted by the number data type.

XSL to explain NaN errors:

<errors>
 <error><xsl:value-of select="number('123 1')"/></error>
 <error><xsl:value-of select="number('abc')"/></error>
 <error><xsl:value-of select="number('123b')"/></error>
</errors>

Resources: 🔗

XSLT file containing all the XSL statements used in the blog – Link
The output of the above file – Link

Please share your valuable feedback 😊😊😊

Database Xpath-extension Functions in SOA 12c

Wed, 21 Jul 2021 13:18:00 +0000

In SOA, we can connect with databases using a database jca adapter and configuring the data source and connection factory in the WebLogic server. Sometimes, we may want to query the database directly within the BPEL component or XSLT without configuring a database jca adapter. For such use cases, we can take leverage of database functions that can be used within **BPEL Assign Activity **or XSLT transformations.

We have 3 different database functions available, that can be used in different use cases.

oraeext:sequence-next-val()
oraext:lookup-table()
oraext:query-database()

1. oraext:sequence-next-val() 🔗

Syntax:

oraext:sequence-next-val(sequence as string, dataSource as string)

sequence: Specify the sequence available in the database for which the next value needs to be extracted.

dataSource: Specify the Data source JNDI name as configured in WebLogic server or JDBC string in the format jdbc:oracle:thin:username/password@host:port:sid.

Note: Only oracle thin driver is supported if used as JDBC string.

This function is used when we want to fetch the next value of defined sequence in the database. It can be used in cases where we want each request to adhere with a particular sequence id which can be used to identify each unique record where using a db adapter might create overhead and extra transformation. This function is basically an alternative to the below SQL statement.

SELECT SEQUENCE_TEST.NEXTVAL FROM DUAL;

Here, SEQUENCE_TEST is the name of the sequence defined in the database.

Example usage in XSLT :

XSL :

<xsl:template match="/">
 <root>
 <seqJDBCString><xsl:value-of select="**oraext:sequence-next-val('SEQUENCE_TEST','jdbc:oracle:thin:mark1/password@localhost:1521:XE' )** "/></seqJDBCString>
 </root>
</xsl:template>

OUTPUT :

<root>
 <seqJDBCString>**2** </seqJDBCString>
</root>

2. oraext:lookup-table() 🔗

Syntax

oraext:lookup-table(tableName as string, inputColumn as string, inputData as string, outputColumn as string, dataSource as string)

tableName: Name of the table in database, from which value need to be fetched**.**

inputColumn: Column name of the table for which we will supply the data.

inputData: Data that will be provided as input to match a value in inputColumn in the table**.**

outputColumn: Column name of table for which value needs to fetched**.**

dataSource: Specify the Data source JNDI name as configured in WebLogic server or JDBC string in the format jdbc:oracle:thin:username/password@host:port:sid.

**Note: **Only oracle thin driver is supported if used as JDBC string.

Suppose we have a table in our database which contains names of employee corresponding to the empIds. Now, we want to extract the name of employee corresponding to empId which we are getting in input. To do so, one way could be to have a db adapter created and use similar SQL statement as below to retrieve data.

SELECT EMP_NAME FROM EMPLOYEE_TABLE WHERE (EMP_ID=#inputId)

If we want to fetch the value directly in assign activity or xslt, we can use the oraext:lookup-table(). This can be treated as alternative approach to dvm lookup where in dvm files are stored local to composite or in MDS.

EMPLOYEE_TABLE

EMP_ID	EMP_NAME
1002	DecipherMiddleware

INPUT :

<input xmlns="http://test.com/tns">
 <empId>1002</empId>
</input>

XSL :

<xsl:template match="/">
 <root>
 <empName><xsl:value-of select="oraext:lookup-table('EMPLOYEE_TABLE','EMP_ID',/ns0:input/ns0:/empId,'EMP_NAME','jdbc:oracle:thin:mark1/password@localhost:1521:XE' )" /></empName>
 </root>
</xsl:template>

OUTPUT :

<root>
 <empName>DecipherMiddleware</empName>
</root>

3. oraext:query-database() 🔗

oraext:query-database(sqlQuery as string, rowSet as boolean, row as boolean, dataSource as string)

sqlQuery: SQL query that needs to be executed.

rowSet: true(), if we need tag for output rowSet()

row: true(), if we need tag for each row

dataSource: Specify the Data source JNDI name as configured in WebLogic server or JDBC string in the format jdbc:oracle:thin:username/password@host:port:sid.

Note: Only oracle thin driver is supported if used as JDBC string.

This function can be used to execute SQL statements without configuring a JCA adapter in SOA composite. Let us assume we have a category table like below, from where we need to query all columns.

CATEGORY TABLE 🔗

CATEGORYID	CATEGORYNAME	DESCRIPTION
1	Beverages	Soft drinks, coffees, teas, beers, and ales
2	Condiments	Sweet and savory sauces, relishes, spreads, and seasonings
3	Confections	Desserts, candies, and sweet breads
4	Dairy Products	Cheeses
5	Grains/Cereals	Breads, crackers, pasta, and cereal

SQL Query :

SELECT * FROM category;

XSL :

<xsl:template match="/">
 <xsl:variable name="dbData" select="oraext:query-database('SELECT * FROM category', true(),true(),'jdbc:oracle:thin:mark1/password@localhost:1521:XE' )"/>
 <root>
 <categoryData>
 <xsl:copy-of select="$dbData"/>
 </categoryData>
 </root>
</xsl:template>

OUTPUT :

<root>
 <categoryData>
 <ROWSET>
 <ROW num="1">
 <CATEGORYID>1</CATEGORYID>
 <CATEGORYNAME>Beverages</CATEGORYNAME>
 <DESCRIPTION>Soft drinks, coffees, teas, beers, and ales</DESCRIPTION>
 </ROW>
 <ROW num="2">
 <CATEGORYID>2</CATEGORYID>
 <CATEGORYNAME>Condiments</CATEGORYNAME>
 <DESCRIPTION>Sweet and savory sauces, relishes, spreads, and seasonings</DESCRIPTION>
 </ROW>
 <ROW num="3">
 <CATEGORYID>3</CATEGORYID>
 <CATEGORYNAME>Confections</CATEGORYNAME>
 <DESCRIPTION>Desserts, candies, and sweet breads</DESCRIPTION>
 </ROW>
 <ROW num="4">
 <CATEGORYID>4</CATEGORYID>
 <CATEGORYNAME>Dairy Products</CATEGORYNAME>
 <DESCRIPTION>Cheeses</DESCRIPTION>
 </ROW>
 <ROW num="5">
 <CATEGORYID>5</CATEGORYID>
 <CATEGORYNAME>Grains/Cereals</CATEGORYNAME>
 <DESCRIPTION>Breads, crackers, pasta, and cereal</DESCRIPTION>
 </ROW>
 </ROWSET>
 </categoryData>
</root>

Since we have kept rowSet and **row **as true() , we are seeing tags coming in output from function result.

References: 🔗

https://docs.oracle.com/en/middleware/soa-suite/soa/12.2.1.3/develop/xpath-extension-functions.html#GUID-73F8AE17-41AC-474E-8AF3-6027CAEADB55

Please share your valuable feedback in the comments section below 😊

Create Delimited String from XML Nodes and Vice Versa in SOA 12c

Fri, 25 Dec 2020 00:00:00 +0000

A delimited string is a string representation of data separated by a delimiter(e.g. “,”). A simple representation of the delimited string will look like.

Delimited String Examples 🔗

Sample-1-- Delimiter as ","
Value1,Value2,Value3
Sample-2-- Delimiter as "|"
Value1|Value2|Value3

If we want to represent the above set of delimited values in the form of an XML document under some root and parent tag. The representation will look like this.

XML Example 🔗

<root xmlns="http://test.com/sample/xml">
 <Values>
 <value>Value1</value>
 <value>Value2</value>
 <value>Value3</value>
 </Values>
</root>

Sometimes we may need to convert values coming in delimited string to an xml document or vice versa. Since delimited string is similar to csv format. We can create a nxsd schema and then perform translate i.e. native to xml or xml to native within our SOA application. But in the scenarios wherein we just want to convert one delimited string to xml or recurring xml nodes to one delimited string, there we can make use of oracle xpath extension functions:

xmlns:oraext=“http://www.oracle.com/XSL/Transform/java/oracle.tip.pc.services.functions.ExtFunc"

Note: Both functions can be used in XSLT or BPEL assign activity. The above two mentioned examples will be used for explaining these 2 functions.

oraext:create-nodeset-from-delimited-string() 🔗

This function is used to create XML node set from delimited string. Basic syntax of the function is as below.

oraext:create-nodeset-from-delimited-string ($qName, $delimitedString, $delimiter)

$qName: Qualified name of the node for each value to be created along with namespace. For e.g. ‘{http://test.com/sample/xml}value’

$delimitedString: String containing values delimited by a delimiter. Examples as above.

$delimiter: String representation of delimiter. For e.g. ‘,’

Sample XSLT code: 🔗

<xsl:template match="/">
 <ns1:root>
 <ns1:Values>
 <xsl:copy-of select="oraext:create-nodeset-from-delimited-string('{http://test.com/sample/xml}value','Value1,Value2,Value3' , ',' )"/>
 </ns1:Values>
 </ns1:root>
</xsl:template>

Note: Excluding, the XSL header for a better view. Please make sure all namespaces are properly imported in the XSL header. The above code snippet will give output as above example.

oraext:create-delimited-string() 🔗

This function is used to convert nodes to delimited string. Below is the syntax of the function.

oraext:create-delimited-string ($nodeSetValue, $delimiter)

$nodeSetValue: repetitive nodes or combination of nodes for which delimited string needs to be created. For e.g.

XML Sample 🔗

<value>Value1</value>
<value>Value2</value>
<value>Value3</value>

$delimiter: String representation of delimiter. For e.g. ‘,’

Sample code using the above sample

<!--',' as delimiter-->
<xsl:value-of select="oraext:create-delimited-string ($nodeSet/ns1:Values/ns1:value, ',')"/>
<!--'|' as delimiter-->
<xsl:value-of select="oraext:create-delimited-string ($nodeSet/ns1:Values/ns1:value, '|')"/>

xmlns:ns1=“http://test.com/sample/xml"

Here $nodeSet contains the XML data as represented in above example. Above XSL snippets will yield results as shown in delimited string samples.

XML to String and String to XML conversion

Thu, 24 Sep 2020 00:00:00 +0000

Use Cases 🔗

Many a times, we may need to convert xml data and pass it as a string inside an XML tag.
Sometimes from source, we get data which is in string format but actually is xml encoded in string format.

To do so in Oracle BPEL 12c or OSB 12c we can use built in functions.

XML data to String 🔗

Sample input 🔗

<inRoot>
 <t1>Hello</t1>
 <t2>World</t2>
 <t3>One</t3>
</inRoot>

Sample output 🔗

<outRoot> <t1>&lt;inRoot&gt;&lt;t1&gt;Hello&lt;/t1&gt;&lt;t2&gt;World&lt;/t2&gt;&lt;t3&gt;One&lt;/t3&gt;&lt;/inRoot&gt;</t1>
</outRoot>

XSLT 🔗

This can be accomplished by using the function oraext:get-content-as-string(element as node-set). This function is really helpful when we are working on BPEL as we can use this function in ASSIGN activity or in TRANSFORM activity using XSLT.

xmlns:oraext=“http://www.oracle.com/XSL/Transform/java/oracle.tip.pc.services.functions.ExtFunc"

element as node-set: xml data(can be variable, xpath, or a function that returns xml data)

Example: 🔗

<outRoot>
 <t1><xsl:value-of select="oraext:get-content-as-string(/ns0:inRoot)"/></t1>
</outRoot>

XQuery 🔗

Xquery too have a function to convert xml to string i.e. fn-bea:serialize($arg as item()*) as xs:string. This function is really helpful when we are dealing with OSB.

xmlns:fn-bea=“http://www.bea.com/xquery/xquery-functions"

$arg: Variable containing XML data.

Example: 🔗

$input contains xml data that needs to be converted and assigned to one tag in output payload.

<outRoot>
 <t1>{fn-bea:serialize($input)}</t1>
</outRoot>

String to XML 🔗

Sample input 🔗

'&lt;inRoot&gt;&lt;t1&gt;Hello&lt;/t1&gt;&lt;t2&gt;World&lt;/t2&gt;&lt;t3&gt;One&lt;/t3&gt;&lt;/inRoot&gt;'

Sample output 🔗

<inRoot>
 <t1>Hello</t1>
 <t2>World</t2>
 <t3>One</t3>
</inRoot>

As we can see in above example that < and > are replaced with < and >. In xml < and > are invalid characters and escaper characters are used in place of these.

XSLT 🔗

In xslt, we can use built in function oraext:parseXML(stringData).This function is really helpful when we are working on BPEL as we can use this function in ASSIGN activity or in TRANSFORM activity using XSLT.

xmlns:oraext=“http://www.oracle.com/XSL/Transform/java/oracle.tip.pc.services.functions.ExtFunc"

stringData: string value of xml data.

Example: 🔗

<xsl:copy-of select="oraext:parseXML('&lt;inRoot&gt;&lt;t1&gt;Hello&lt;/t1&gt;&lt;t2&gt;World&lt;/t2&gt;&lt;t3&gt;One&lt;/t3&gt;&lt;/inRoot&gt;')"/>

Note: we have used copy-of instead of value-of as we need all the xml nodes to be copied to the target file. If we want to extract the data from xml without xml tags then we can use value-of.

XQuery 🔗

In xquery, we can use fn-bea:inlinedXML($arg as xs:string) as node()*. This function is really helpful when we are dealing with OSB.

xmlns:fn-bea=“http://www.bea.com/xquery/xquery-functions"

$arg: variable containing xml data in string format.

Example: 🔗

fn-bea:inlinedXML('&lt;inRoot&gt;&lt;t1&gt;Hello&lt;/t1&gt;&lt;t2&gt;World&lt;/t2&gt;&lt;t3&gt;One&lt;/t3&gt;&lt;/inRoot&gt;')

Remove Empty tags using XSLT

Sun, 13 Sep 2020 00:00:00 +0000

While performing xslt transform on source XML, we may need not to include empty tags in the output xml payload.

Sample scenarios: 🔗

Target system doesn’t want empty tags to come.
Size of xml payload has become huge due to unwanted empty tags.

Source XML 🔗

<inRoot>
 <t1>Hello</t1>
 <t2></t2>
 <t3>World</t3>
</inRoot>

Output XML 🔗

<outRoot>
 <a1>Hello</a1>
 <a3>World</a3>
</outRoot>

The optimal way to achieve above result is to use conditional if statement in xslt i.e. xsl:if

Sample XSLT 🔗

<xsl:if test="/inRoot/t2!=''">
 <a2>
 <xsl:value-of select="ns0:C1"/>
 </a2>
</xsl:if>

If we have some whitespaces included between tags, the above if-statement may fail and in such scenarios, we may use normalize-space() function in xslt and test condition will become : “normalize-space(/inRoot/t2)!=’’”

Sometimes, we just want to remove empty elements from already transformed xml. Such cases have same input and output schemas. In such scenarios we can take leverage of identity transform using xsl templates and copy methods. See the below sample xslt identity transform to remove empty tags.

Sample XSLT: 🔗

<xsl:template match="node()|@*">
 <xsl:if test="normalize-space(.) !='' or ./@* !=''">
 <xsl:copy>
 <xsl:copy-of select="@*"/>
 <xsl:apply-templates/>
 </xsl:copy>
 </xsl:if>
</xsl:template>

If we are using xslt in JDev, we need to specify input and output schemas and replace xsl:template tag with the above xsl template code.

Format String oraext:format-string

Sat, 05 Sep 2020 00:00:00 +0000

In our integration world, sometimes target system requires the data from source system to be formatted in some specific format which may require data from various input fields along with custom data. To do so we generally use concat() function to concatenate data from various fields and some user specific inputs.

Lets take the below example

Suppose we have a predefined format for the output string:

The abc with employeeID: 1234567 has been successfully registered in the system.

Here, name of the employee($empName=‘abc’) and employee id($empId=‘1234567’) are dynamic values. One way to accomplish this, is to use concat().

concat('The ',$empName,' with employeeID: ',$empId,' has been successfully registered in the system.')

Another way we can achieve this, is using oraext:format-string(string,string,string…)

oraext:format-string ('The {0} with employeeID: {1} has been successfully registered in the system.',$empName, $empId)

In above example, first string defines the output format containing place holders({0},{1}). These place holders will be replaced by the subsequent string values($empName=‘abc’, $empId=‘1234567’). In future, if we need to change the position of these dynamic value in output string, all we need to do is to move the position of these placeholders.

Note: Maximum of 10 arguments are supported for this function.

DateTime formatting using xp20:format-dateTime()

Wed, 02 Sep 2020 00:00:00 +0000

xs:dateTime Format 🔗

In XML, we generally use xs:dateTime(xs:xmlns=“http://www.w3.org/2001/XMLSchema") data type to store date time data. The generic representation of date time, that is supported by xs:dateTime in xml:

YYYY-MM-DDThh:mm:ss[.SSS][Z|(+|-)hh:mm]

Example: 🔗

2019-09-11T10:16:31.943+05:30

Code	Description	Example
YYYY	represents a 4-digit year	2019
MM	represents a 2-digit month	09
DD	represents 2-digit date	11
hh	represents 2-digit hours	10
mm	represents minutes in 2-digits	16
ss	represents seconds in 2-digits	31
SSS	represents milliseconds	943
T	represents time separator	T
Z	represents UTC time zone or 0 time zone offset
(+\|-)	hh:mm represents time zone offset	+5:30

Few more valid examples: 2019-09-11T21:32:52, 2019-09-11T21:32:52+02:00, 2019-09-11T19:32:52Z, 2019-09-11T19:32:52+00:00, 2019-09-11T21:32:52.11179

While working with different systems, we may come across such systems, which require date time in some different formats other than default format as supported by xs:dateTime. To over come such scenarios, function is available in OIC(Oracle Integration Cloud) and Oracle SOA

Format DateTime 🔗

xp20:format-dateTime($dateTime as string, $format as string) xmlns:xp20="http://www.oracle.com/XSL/Transform/java/oracle.tip.pc.services.functions.Xpath20"

$dateTime should contain date time value as per xs:dateTime data type as as string eg: ‘2019-09-11T10:16:31.943+05:30’

$format should contain output format to which we need to convert to. For e.g. if we want output to be ‘11/09/2019’ then value of $format should be [D01]/[M01]/[Y001].

xp20:format-dateTime('2019-09-11T10:16:31.943+05:30', '[D01]/[M01]/[Y001]')

To format current date time, we can use xp20:current-dateTime() in $dateTime

xp20:format-dateTime(xp20:current-dateTime(), '[D01]/[M01]/[Y001]')

Formatting Codes: 🔗

Below are some important codes which can be helpful in generating desired format for target application.

Sample dateTime: 2019-09-03T15:16:31.943+05:30

Code	Description	Example Output
[Y001]	4 digit year	2019
[Y01]	2 digit year	19
[M01]	2 digit month	09
[D01]	2 digit date	03
[H01]	2 digit hour in 24 hour clock	15
[h01]	2 digit hour in 12 hour clock	03
[m01]	2-digit minutes	16
[s01]	2-digit seconds	31
[f001]	3-digit milliseconds	943
[Z]	Time zone offset in the form (+\|-)hhmm	+0530
[z]	Time zone offset in the form GMT(+\|-)hh:mm	GMT+05:30
[Dwo]	Word for ordinal value of day	third
[dwo]	The day of the year, represented in ordinal words	second hundred and forty-sixth
[MNn]	Capitalized month name	September
[P]	am/pm	pm
[PX]	Uppercase am/pm	PM
[FNn]	Capitalized weekday name	Tuesday
[MNn,a-b]	Capitalized month name having atleast “a” charcters and atmost “b” characters [MNn,3-3]:	Sep