I often find myself reusing subprocesses from my main workspace, either for debugging or for creating similar workflows with small changes in a new Workspace. I'm aware that FME has a ton of useful shortcuts, so I was wondering if there's one that lets you automatically select all the necessary(only!) transformers that belong to a previously used process. The idea would be to quickly copy and paste the used chain into a new workspace without having to manually select every necessary transformer or just copy paste all. Is there a shortcut or built-in feature for this, or is manually selecting the chain still the best approach? I feel like this is most likely highly specific to my own taste but I wanted to ask nevertheless since the main process got way too big :D
Thanks in advance! :)
Automatically Mark every 'previous' Transformer for plain copy paste into new Workspace
+1- Contributor
+26- Influencer
I recognize that. Happens to me all the time as well.
What I tend to do is place a bookmark around the Transformer chain I need to re-use. That allows you (on right-click) to select all objects inside the bookmark. Another right-click then copies all of them, including the connections.
Custom transformers are also an option, but I frequently find those to restrictive or have other issues that interfere with my workflow.
I have been thinking about using a dummy workspace with a lot of these re-used chains, and also SQLCreators with complex SQL, to have open as a sort of template to copy and paste from. That does mean having two Flow instances open, so on lighter laptops that might not be a good idea.
If anybody else has any good ideas, I’d be very interested to hear them :-)
+73- Celebrity
The first thought that springs to mind for me is to have these transformers encapsulated in a Custom Trasnformer - Creating a Custom Transformer
+26- Influencer
The first thought that springs to mind for me is to have these transformers encapsulated in a Custom Trasnformer - Creating a Custom Transformer
True, I’ve used/am using them that way as well. My biggest problem with using a Custom Transformer is that almost always the general flow is the same, but details are different (attribute names, buffer distance, that sort of detail). Which then either means changing the custom transformer, which impacts all the previous instances, or adding extra transformers to mitigate the differences. Hence my comment “to restrictive or have other issues that interfere with my workflow”.
That said: it is excellent advice, and I do use them a lot. It’s just that for this particular scenario they’re not always that useful. Maybe if you could ‘explode’ a custom transformer, which would add the contents as ‘normal’ transformers to the workspace...
+26- Influencer
Is it possible to parametrise the changes you need to make each time?
Possibly, especially after the fact. But I usually do not have the time to really make such things generic enough that I can always use them everywhere. Timewise it’s usually a lot faster to simply copy-paste-tweak, and get on with the next job… (I know, that leaves me with a lot of technical debt, but without enough resources I’m already happy if I can get the job done on time...)