Document Functions

The _docinfo() function returns information about the current document (message) that is passing through the process flow. Some document information can also be obtained using other functions, such as _iserror(). The _docinfo() function uses the following format:

_docinfo(type)

Example 1

In the following example, the message being received in an arbitrary encoding is to be converted to Base64 encoding:

_base64(_docinfo('encoding'))

Example 2

In a process flow, a File Read Service (com.ibi.agents.XDFileReadAgent) loads a PDF as a bytes file. The Set Document State Service (com.ibi.agents.XDDocAgent) is used to assign a suffix of pdf. The file emit (either in the process flow or as an emit following the process flow) might use the following output name:

/myfiles/file1._docinfo(‘suffix’)

A file named file1.pdf is returned as the output.

The _isroot() function returns true if the root of the current document matches the parameter. It uses the following format:

_isroot(name)

The _isroot() function is often used to drive a routing decision, such as _root('a') or _root('b') to cause the route to be selected for any document with a root of 'a' or 'b'.

The _root() function returns the root element name of the current document. It uses the following format:

_root()

This is equivalent to the _xpath function _xpath(/*/name()). It might be used in a routing decision, such as _cond(_root(),eqc,'a') which selects the route for documents with roots of 'a' or 'A'.

The current document can hold parsed or non-parsed (flat) information. The _isxml function returns true if the current document holds parsed XML. If the document has no contents, then this function returns false. The _isxml function uses the following format:

_isxml()

The current document can hold parsed or non-parsed (flat) information. The _isjson function returns true if the current document holds parsed JavaScript Object Notation (JSON). If the document has no contents, then this function returns false. The _isjson function uses the following format:

_isjson()

The current document can hold parsed or non-parsed (flat) information. The _isflat function returns true if the current document holds non-parsed data. If the document has no contents, then this function returns false. The _isflat function uses the following format:

_isflat()

The current document can have errors posted to it by the system, or can be set into an error state by the actions of an application process. If the document is in error state, default replies are delivered to the configured error addresses, otherwise they are delivered to the configured reply addresses. The _iserror() function returns true if the document is in error state. It uses the following format:

_iserror()

As the document passes through its execution route, a supplied rules-based validation system can be employed. Often the rules relate to eBusiness documents such as SWIFT or EDI. If the rules system has posted a validation error to this document, the _hasruleerr() returns true. It uses the following format:

_hasruleerr()

Certain rules and services can detect schema violations during the life of a document in the system. Whether such a violation has been detected can be tested with this function. A common use of this function is in a process flow test object. This function uses the following format:

_hasschemaerr()

The _iswellformed() function uses the following format:

_iswellformed([format])

The current document can hold parsed or non-parsed information. This function returns true if the current document holds flat information that can be parsed into XML or JSON. If it has not been parsed, then a true response from this function assures that the message can be parsed. If the message already holds parsed information, then this function returns true.

Streamable input is used for handling large documents or documents for which the application desires to split the input into sections under the same transaction. Following the completion of the input stream handling, a final pass is made with a special document containing batch information:

<batch count='n'/>

The end of stream can also be tested with the _iseos() function. A common use of this function is in a process flow test object. This function uses the following format:

_iseos()

Creating a Process Flow to Test for an End of Stream Document

Splitting and streaming or splitting preparsers generate a special control document at the end of a batch to indicate that this batch was completely processed. This End of Stream document is a non-data XML document that is passed into a channel after all data documents. The _iseos() function will return as true if the current document contains an End of Stream message, else it will return as false. The End of Stream document may also contain additional statistics about the splitting process, such as how many iWay documents were generated from the batch.

To create a process flow using iWay Designer to test for an End of Stream document:

1. Create a process flow and add a new Test object.
2. Configure the parameters for this Test object, as shown in the following image.
   
3. Add a new Service object to the process flow that uses the QAAgent.
4. Configure the parameters for this Service object, as shown in the following image.
   
5. Connect the Service object to the True edge of the Test object.

   The new process flow should have a structure that resembles the example in the following image.

   
6. Using the iWay Service Manager Administration Console, create a channel that consists of an XMLSplitpreparser as an inlet, which would return an End of Stream document when the last pass is reached.
7. Add a route to the channel that consists of the process flow, which was created using iWay Designer.
8. Add an outlet to the channel.

   The new channel should have a structure that resembles the example in the following image.

   
9. Use the following input document to test your channel with /a/b as the level string:

   <a>
     <b name="b1">
       <c>value of input element a/b@name=b1/c is 1</c>
       <c>value of input element a/b@name=b1/c is 2</c>
     </b>
     <b name="b2">
       <c>value of input element a/b@name=b2/c is 3</c>
       <c>value of input element a/b@name=b2/c is 4</c>
     </b>
   </a>

   The _iseos() functions returns as true at the third (last) pass and a trace file is created which displays the End of Stream document.

The current document is flattened to a string. If the current document is in XML format, then the decl parameter determines whether an XML declaration will be included in the flattened output. The default value is set to true.

The _flatof() function uses the following format:

_flatof([decl])

Documents can hold attachments. This function has two forms. When used with no parameters, it returns the number of attachments associated with the document. When used with two parameters, it returns the index of the first attachment having the named header equal to the specified value. The _attcnt() function uses the following format:

_attcnt([name,value [,notfound]])

The _atthdr() function returns the value of a specific header in the attachment. The IC (independent case) form tests the attachment header names in a case independent fashion. Number 0 is the first attachment. The _atthdr() function uses the following format:

_atthdr(index, name, default)

_atthdric(index, name, default)

When the _atthdr() function is used in a conditional expression in comparison against a string value, the default value must be specified.

For example, the following conditional expression would not compile, since it requires a default value in the expression:

COND(_atthdr(1,'mysample'),eq,'a')

Instead, the expression must be specified as follows:

COND(_atthdr(1,'mysample','defaultval'),eq,'a')

For example, to return the first header in the attachment, _atthdr(0,'mysample') could be used.

The _attflatof() function returns the content of an attachment as a string. This function uses the following format:

_attflatof(index [,encoding])

Example:

The following sample document contains two attachments:

------=_Part_30_1258108044.1487787594926
Content-Type: application/xml
<?xml version="1.0" encoding="ISO-8859-1" ?><base/>
------=_Part_30_1258108044.1487787594926
Content-Type: application/xml
Content-Transfer-Encoding: 7bit
Content-Disposition: attachment; filename=walter.txt
Content-Length: 32
<part1>attachment data one</part1>
------=_Part_30_1258108044.1487787594926
Content-Type: application/txt
Content-Transfer-Encoding: 7bit
Content-Disposition: attachment; filename=fred.txt
Content-Length: 32
<part2>attachment data two</part2>
------=_Part_30_1258108044.1487787594926--

The function _attflatof(_attbyfname('fred.txt')) will return the following value:

<part2>attachment data two</part2>

The _attbyfname() function accesses the Content-Disposition header of each attachment, and returns the index (base 0) of the first attachment with the requested file name. A value of -1 is returned if the attachment is not found. This function uses the following format:

_attbyfname(filename)

Example:

The following sample document contains two attachments:

------=_Part_30_1258108044.1487787594926
Content-Type: application/xml
<?xml version="1.0" encoding="ISO-8859-1" ?><base/>
------=_Part_30_1258108044.1487787594926
Content-Type: application/xml
Content-Transfer-Encoding: 7bit
Content-Disposition: attachment; filename=walter.txt
Content-Length: 32
<part1>attachment data one</part1>
------=_Part_30_1258108044.1487787594926
Content-Type: application/txt
Content-Transfer-Encoding: 7bit
Content-Disposition: attachment; filename=fred.txt
Content-Length: 32
<part2>attachment data two</part2>
------=_Part_30_1258108044.1487787594926--

The function _attbyfname('fred.txt') will return a value of 1, which indicates that the attachment is found.

The _srcname() function returns the name of the terminate (end node) of the subflow through which the document reached the calling process flow. This allows the subflow to communicate information back to the parent flow.

Imagine a subflow that performs some operation that results in one of three possible situations:

• Success.
• Failure by a security violation.
• Connection error.

The parent flow calls the subflow, and wishes to deal with the result from that flow. Within the subflow, the document that is produced is returned through a terminate or end node, as shown below. The name of that terminate node becomes the edge name followed by the returned document. As this example shows in the image below, there are three activities (called do_<something>) in the outer flow, each of which handles the result returned from the inner subflow.

If the subflow returns multiple documents, each follows the edge of its own name. If more than one of any name is returned, the documents are associated as siblings, and can be split later using the sibling iterator. The source name can be used in a switch statement for more detailed routing.

iWay Service Manager (iSM) manages messages in XML format. This section describes parsed XML functions that belong to the iWay Functional Language (iFL) set. Do not use these functions with documents in other (non-XML) formats, since errors will be generated. Parsed XML is a native format that is managed within iSM.

iWay Service Manager (iSM) manages messages in JavaScript Object Notation (JSON) format. This section describes parsed JSON functions that belong to the iWay Functional Language (iFL) set. Do not use these functions with documents in other (non-JSON) formats, since errors will be generated. Parsed JSON is a native format that is managed within iSM.

To demonstrate the functionality of these parsed JSON function description, consider the following sample JSON message:

{"db":{"vec":[1,2,3],"str":"abc","obj":{"a":1,"b":2}}}