Cancel Process Smart Service

I usually recommend, if at all possible, allowing a user to cancel their own task (though i'm a little unclear if this is what you mean here), using a "cancel" button on-form which will route directly to a terminate node in that process.

If you really do need to cancel the process externally, you can access the process ID in at least 2 common ways.  The most common is probably by use of the Query Process Analytics rule called in the context of the process model in question, filtered down by some other identifying info, which can return the process ID as well as all kinds of other metadata depending on what Process Report you use to source the data from.

The second most common way (and easiest, if you set it up in advance), is to have the original process store its own process instance ID to the database shortly after initialization, i.e. in a dedicated column you set up in an existing table & CDT, as in something that specially handles that process.

Here is what it needs to do, lets say user 1 goes in and inputs the data and it creates the process task for it to be reviewed.  But then user 2 goes in and sees it was input wrong so they delete what user 1 put in.  Once is deletes what user 1 put in it needs to cancel that original process task that was created.  What worries me about using the user who created is this, what if the user does three different entries, and it creates three process tasks.  If I try to remove it by the person who submitted it would it remove all three even though it just needs to remove one.

williamm0010 said:

it needs to cancel that original process task

What "original process task" are you referring to here?  By the time User 1 has submitted their information for review, their "task" no longer exists as it's been submitted.  If you're talking about the process belonging to the Reviewer, why wouldn't that review task be contained in the same original parent process thread?  In that case all you'd need to do is terminate that process.

Mike,

Thank you for your help, trying to use the queryProcessAnalytics function to get the process id

Have only been working with Appian for 4 months now and this is the first time we are trying to cancel a process systematically.

I see in the active process the process details.  Under the process nodes tab I do see the process id that I need to get.  Under variables I do see this [id=2568, check=56585, amount=20.00 etc....]  I need to do the process analytics query filtering for the unique id  id=2568 to get the process id for it

Trying to do this in a expression and store it to a variable like shown below, having a issue trying to figure out the correct syntax to do this.

Thank you

did you see my followup questions from earlier?  I'm still super unclear whether you even need to bother with this.  It would be helpful if you could share some screenshots of your process flow indicating what you currently have configured and/or what you hope to accomplish as well.

In any case... You will want to use an expression rule to handle your Query Process Analytics call.  You can open a blank Expression Rule Editor by manually navigating to /suite/design/rule in your environment, and just manually messing with it there.  Eventually you'll want to save it as an expression rule such that you can pass in the unique id (in terms of process data) that you need the process instnace ID for, and as a result it'll pass out the process ID and/or a collection of metadata relating to that process instance.

As an initial step, I'd suggest you create a new / tester Process Report where you can add/edit the various columns as you see fit; then you'd create a constant pointing to that report, and use that constant in your QPA call in your expression rule.

Mike,

Thank you let me explain the process model and what I am seeing.

I log in as user 1, I click on the link to the interface.  I fill out the data and the data is saved into the table.  It then creates a process task and assigns the task to the group that needs to see it.  This process now shows active waiting for approval  Process ID is 268466210

I see the active process which is waiting for someone from the assignee group to look at it.

I sign out and now sign on as user 2, I click on the link to the interface.  I go in and can see what user 1 put in and notice it is not right.  The current process ID I am on that is active is 268466214  I now delete the data that user 1 put in and now need to cancel the active process for the first active process ID 268466210 since the data user 1 put in has been deleted and no longer needs to be approved.

Trying to figure out how user 2 can get the process ID or how the user 2 process can cancel user 1 active process

Let me know if this helps, if you need screen shots still let me know and will get some for you to see.

Thank you Bill

So to confirm, are you saying "User 2" is different from the "Reviewer" who would be comleting the "Waiting for Approval" task initiated as part of the original process flow?

What is User 2 doing to access this information?  When you say "click on the link to the interface", what is the link, what type of link is it, and where / what is the interface?  Are they seeing the item's record entry?  Are they clicking into the "Waiting for Approval" task?  Are they seeing it on some arbitrary report?  Are they starting a Related Action / other arbitrary process model/instance?

There are, of course, various solutions for approaching this, but they all depend on having a good bit of understanding of what the various Appian components are, and do, and what capabilities we have at our disposal; as well as details (hence the above follow-up questions) regarding what the current setup actually looks like.

I don't have anything to add other than, as Mike noted earlier, I always always store the process instance IDs (pp!id) to the CDT along with the request data.  Then when you want to cancel a process, you have the process IDs of the submission, approvals, along with your main data that you can pull to access/cancel any process instances necessary, without additional analytics querying.

So would it be easier then for me to just create a table that holds the unique id for the data and the process id and then write the unique id to the table and the process id so if that data gets delete I can just use the unique id to retrieve the process id for that data to cancel the process?  Thank you

I can't say I've fully digested your use case here yet, but I just store process IDs in the same CDT where the submission or approval data is.  E.g. if the row has all of the unique ID, submission data, approval process ID, you will have it readily available when you are looking at the data to cancel/delete.  Vs having a separate mapping table to manage.  Mapping table would work also, but is a lot more overhead.