So I’m pretty new to this ECM thing, so if you see something I’m doing wrong PLEASE point if out.
With that out of the way, I need to know if there is a way to get ECM to create a Job Misc Charge line.
The example I’m working with has one line for a PO, and a second line that needs to be a job misc line. I don’t see a way to set that with the normal flow.
Is that something that will need to be done manually once it gets into epicor?
The second question I have is for the price per code. The vendor uses a per thousand unit price. I’m capturing that in IDC before it get’s passed over. I see that I can add the LINE_CostPer column in my dataset. Is there a way to have the system match using that without customizing the flow? The PO is priced per each, but the invoice is per 1000. While I can calculate it, I would like to keep as close to out of the box as possible.
There is a way to make a MISC Charge manually in the workflow below line items is a drop down. In the AP workflow precreate and validate Invoice there is a Create SubGroup Line Misc Charges Task. The issue I see is how to determine that is misc charge and not a part.
I have not tackled the /1000 pricing and had messed with calculation our price, but always was off from rounding, but at Insights I had decided to avoid the variance thing ECM did and just send in the price and fix in a bpm.
You could have a field for “Misc Charge Amt” (either form IDC or manually entered) which the workflow could then see and fire off the appropriate Intergration datalink to create the AP Misc Line.
As for the Price/Per, I’ve not done that either, and I would like to think that if a value is there, all of the workflow/integration logic would account for it automatically. If it doesn’t then maybe a support ticket to find out what we’re missing, followed by an Idea if we get no solution.
So more updates on the misc charge. I can get the system to create a misc charge on an invoice. However the integration does not return the misc charge in the dataset in either the “Pre-Create invoice” or “Create invoice” integration point. So this means that it will delete that line from your data set at that point. So if you keep adding it every time it removes it, you can push it into epicor. But that won’t work for users.
Does anyone know of something that I could be doing wrong when populating the HeadMisc charges section of the calls? Since it’s creating the invoice in Epicor, I’m guessing that it’s the integration that’s busted. I’ll probably break out the debugging tools and see what I can figure out, but this is kind of unnecessarily painful here.
I sent in this line. It gets deleted on pre-create, and I have to keep adding it but if I do…
when data disappears on me, it’s 99% of the time because I used a field on a datalink’s output, but the datalink didn’t bring back data FOR that field, so it blanks it out/deletes it/empties it. And that could break an entire LINE field group. I’d check if you are using any of the field group fields somewhere they should not be used.
Looking at the “trace” that you can get. It’s the same call, where it leaves with an amount, and comes back without one. There’s too much stuff to get it all in one screenshot, so you’ll have to trust me that it’s the same call.
But you’re saying that if I use anything on that call on the output side, it blanks that out? Is that any field in ANY of the rows coming back? (line header etx.) or just in that field group?
You say it doesn’t return the line - does the line stay in the AP Invoice inside Kinetic? (I didn’t see that above)
Yeah - if a field is listed in the output section of a datalink, and the datalink does not bring back a value for it, then it clears/blanks that field. Yep… dumb, but true. I thought just the opposite of course, so I finally found out why I was having so much trouble.
If I keep adding it so the final “Create invoice” call has that line in there, then it creates the charge. However, it does a “Pre-Create” call to make sure it’s going to work out/check for variances etc. which is an “improvement” in V2 so it doesn’t have to actually create the invoice until everything is kosher. That call removes the lines on me, so by the time it gets around to making it in Epicor, they are removed.
I’ll try and see if there is something I can see about using any of those outputs that would screw it up. I think I’ll try to put in a user pause in there to see if I can “step” through the process and see what step would be removing those lines.
Yes - at least that’s what they say. I’m not at V2 yet, but the intermediate group in AP does cause a few problems.
Hmm. That definitely sounds like the problem area, but now that you are sure of the point of the problem, I wonder if it’s not a problem with the Intergration datalink. Very possible you’ve found a bug that we haven’t yet (or at least no one has posted about. We’re not on V2 yet, and @gpayne’s workflow is very custom)
I might also disable the Pre-create step and see if the line carries to the end. You already know you have a ‘good’ invoice that will get created in the final step, so skip the pre-create and see if that is truly the problem step.
That was it @MikeGross . I removed the outputs on the pre-create and create steps in the workflow and now the lines stay there. I still think it’s a bug on the ECM side. I don’t know why it wouldn’t return those rows, but it seems to work when I removed those 3 outputs.
I don’t really know how to put it to them. The way in which these datasets work is very odd and not like anything I’ve seen in reporting and/or Epicor.
Like, if I submit a dataset through a workflow step/action and I am simply trying to add a value to just one of the fields in the dataset, I still have to specify every other field in the output, or those values disappear… why?
It’s just very weird and cumbersome, but once you get used to it like @MikeGross said it becomes easier to work with. It’s still really tough.
I think the issue is that they don’t really have a mechanism for deleting a row (like a row mod in Epicor). So the only way for the system to know if the row should be deleted is to return all of the rows, and anything that doesn’t match it gets rid of. Obviously this isn’t very efficient, but I get why. In the case with the misc charges though, the values are simply not returned from the integration point. So because of the previously mentioned lack of delete mechanism, simply forgetting to put a dataset into the return ends up deleting things on the ECM side. So a small bug has big consequences.
The method for adding and deleting data to and from a dataset needs to be re-worked to better aline with Epicor standards, however, this was an acquisition and I don’t know how many people are still using this software that use a different ERP system and how much revenue it generates from those ERP systems so changing everything to align with Epicor/Kinetic might not work out.
