Update Operation

For a simple update to a child entity use a Siebel Operation. When a Workflow Process has a Business Object then a Siebel Operation Update step defined for a child Business Component will update all records that are children of the selected record in the Primary Buisiness Component. The update can be filtered to specific records by specifying  a SearchSpec on the Update.

This technique can be expanded through the use of Runtime Events: define an Event on the child BusComp to catch a simple update operation and trigger further processing of the record.

Bear in mind that for this to work then the Object Id process property must specify a valid record in the Primary Business Component. Also, watch out for this gotcha if you’re trying to dynamically update a field.

Sub Process SearchSpec

An alternative way to process more complicated actions is to call a sub-process for multiple rows. In this case a Workflow Process is defined as normal to act on a single record specified through the Object Id. Then to have this workflow run for multiple rows, define a new ‘wrapper’ workflow and add a Sub Process step to call the original process. To this Sub Process step add the Input Argument SearchSpec, supplying a search specification through the Value in the normal way.

When the wrapper workflow is executed, the SearchSpec is applied to the primary Business Component of the sub-process and the child workflow is run once for each row returned.

I spend a lot of time with Workflow: EAI work kind of forces that on you. Any moderately complicated workflow process always ends up needing a stack of workflow process properties. The vanilla processes Siebel provide are good examples of this phenomenon. This is understandable, and equivalent to a scripted function justifiably defining a bunch of local variables. So here’s the thing: when working with processes, why do I always see all the process properties defined to be ‘In/Out’? This is equivalent to a scripted function putting every local variable into the function definition. Crazy talk!

For those who don’t know, the In/Out parameter on a process property defines the purpose of the property. The values are In for input properties, Out for outputs, In/Out (the default) for properties that both accept and return a value, and None for those variables just used internally by the process. Pretty clear cut. Setting this parameter properly makes the workflow easier to understand – particularly helping when calling the workflow as a sub-process. Leaving it as the default In/Out value is lazy config, and I’ve seen too much of it. So there. Someone pass the message on to the developers inside Siebel, please.

P.S. While we’re at it: how about using process property types other than ‘String’ and ‘Hierarchy’ – and that ‘Integration Object’ parameter is pretty useful too…

According to all documentation this should have worked, but instead the workflow was immediately returning a response to the UI the first time it hit the Wait Step. The workflow was continuing on exactly as expected in the background, but it appeared that Siebel was automatically persisting the workflow, making the flow asynchronous.

Took a bit of digging around, but the solution was to change the Mode of the business process. In Interactive Flow or Service Flow mode the Wait step causes the flow to move out of the UI context, but in good old 7.0 Flow compatibility mode it works as expected, blocking the UI until the workflow completes.

Doesn’t fit in very well with Siebel’s mandate of not creating new 7.0 Flows, but what can you do when there are bugs..?

• Runtime Applet.PreInvokeMethod
• Browser Applet_PreInvokeMethod
• Server WebApplet_PreInvokeMethod
• Runtime BusComp.PreInvokeMethod
• Server BusComp_PreInvokeMethod
• Runtime BusComp.InvokeMethod
• Server BusComp_InvokeMethod
• Runtime Applet.InvokeMethod
• Server WebApplet_InvokeMethod
• Browser Applet_InvokeMethod

Essentially, the Runtime Event occurs before the equivalent server script event for the same object, plus the Applet browser script events ‘wrap’ all server side events.All of which makes perfect logical sense, but it’s sometimes handy to see it written down in a list.

Dot notation is what Siebel call your fairly standard ‘parent.child.grandchild’ notation. Imagine a workflow with a hierarchical process property called Employee, which has assigned to it a property set with the properties Login and Password. What dot notation allows you to do is access that Login property directly, by reference to Employee.Login. Simple enough.

This notation works both ways – so you can both retrieve and set properties. To set a property using the Workflow Utilities.Echo method, you need to have the top-level hierarchical process property assigned in the first input variable and then child properties in following variables. As an example: we want to set up the simple Employee Property Set above. We already have the LoginString in a process properties and want the Password ‘geenoos’. Create a Workflow Utilities.Echo step and assign the input variables:

Input Argument	Type	Value	Property Name
Employee	Process Property		Employee
Employee.Login	Process Property		LoginString
Employee.Password	Literal	geenoos

Next – as I’m sure you’ve worked out – if your Employee property set has a child property set, lets say a child PS with a type of Address and properties Street and Postcode, then the Employee’s Street can be accessed with Employee.Address.Street. You can also get at the Property Set’s Value argument, using the notation property.<Value>, which also works at all levels of the hierarchy.

This is all great, useful stuff, but the real kicker comes when you’re working with XML. XML hierarchies in Siebel are just property sets, with attributes as properties and elements as <Values>. By using dot notation in input arguments, output argument and Echo steps, you can construct and edit XML hierarchies on-the-fly, something that previously was impossible without scripting.

The one limitation of dot notation is the 75-character limit to a workflow step argument. This can kick in fairly quickly when you’re dealing with all the ListOfBusinessComponent hierarchies. The workaround is to assign a sub-hierarchy to a secondary process property, do your manipulation, then assign it back in.

All the method does is reflect back the input arguments. Big deal, eh? The thing is, the input arguments can be expressions or business component fields or property set values or functions or anything you can usually reference from a workflow. So you can manipulate process properties without taking any other action, or you can get a current business component field value without doing a query, or you can get at property set child elements using dot notation – all things that I’d struggled to do previously without scripting.

Plenty of good examples on SupportWeb under a search of ‘Workflow AND Echo’. Note that in early versions of Siebel 7 the display name of the method is ‘Return Property Values’, plus there’s a fixable bug in 7.0.

This one is relatively simple but it is worth pointing out. Sometimes its the little things that make you weep for joy as a developer.

In the bad old days of Siebel (i.e. before 7.7), the world was running around clubbing each other over the head, saying “ugh” a lot and wearing animal skins. Then the light shone upon the world and we were given:

WriteRecordNew and WriteRecordUpdated.

These are Runtime Events that I noticed just recently. Many many times in config, you need to do something when a record is written. You place code on the WriteRecord event (so that you can avoid having to worry about the user doing Undo, stepping off a field and recorrecting etc). But then you need to do different things depending on whether the record is new or updated.

Usually what you needed to do was set some global variable when the NewRecord event fired, test that variable in WriteRecord, then clear the variable. It usually worked, but it was not elegant. I have seen implementations where the code on the WriteRecord event runs into the thousands. But we won’t mention those.

Now, the great and glorious Siebel gods have seen fit to now provide us with Runtime Events that distinguish between New and Updated records! Hallelujah brother!

I did this trying to prototype a problem I had for an interface I was working on. When you setup a Workflow Process you must assign it a Business Object, or else a whole heap of functionality within that Workflow Process does not work i.e. you cannot select Business Components etc.

My Workflow Process was triggered by a Runtime Event, and all was fine for the UI contexts where the business component of the Workflow action matched the primary business component of the UI. However, when I was in a different business object, the Row Id passed was that of a different Business Component and so my workflow did not “flow”.

For example, upon creation of a Contact in “My Contacts”, the Row Id passed in was the Row Id of the Contact. However, when you create a Contact in Account->Contacts, then the Row Id passed in is that of the Account record.

Well, there is a simple way around this, ready for it?

Are you sure you’re ready?

Okay, here it is…

In the Runtime Event, you need to set a Profile Attribute to the current Row Id to pass into the Workflow. You can then put as many Runtime Events on as many business objects as you require, as long as it always sets that Profile Attribute for the Row Id.

Too easy. A gentle warm-up from Nathan, there’s more to come…