Preserving Jobname

[ChrisSigge]

Hello, this is what I want to do:

I insert a folder with PDF Files into the Flow, than I ungroup the Folder to

route the PDFs through an Pitstop Server Preflight, after it I want to regroup all PDFs back into an

Folder an name the folder with the former given name.



There are several options in the Job ungroup configurator to "remember" the jobname,

but i don't see how I can get access to that "remembered" name when put back the PDFs into an Folder with the Job regroup configurator.



Any ideas?

[ArielRauch]

When you ungroup a job folder you have the option to store the name in a PrivateData key (generally named ungroup).

Later in the flow you can use this PrivateData key to rename the files (or job) back to the original name

[ArielRauch]

When you ungroup a job folder you have the option to store the name in a PrivateData key (generally named ungroup).

Later in the flow you can use this PrivateData key to rename the files (or job) back to the original name

[gabrielp]

To elaborate on Ariel's response, Ungroup and Assemble can work in tandem to achieve exactly what you're looking for.



Ungroup configurator:





Assemble configurator:





I find that private data is a life-saver and keeps things from being overly complex. I'm surprised that Switch doesn't have explicit/native private data write or explore abilities or configurators. If you do decide to use PD more, it helps to use a script to write key/val pairs (one of the simplest scripts you can make).

[ChrisSigge]

gabrielp wrote: To elaborate on Ariel's response, Ungroup and Assemble can work in tandem to achieve exactly what you're looking for.



Ungroup configurator:





Assemble configurator:





I find that private data is a life-saver and keeps things from being overly complex. I'm surprised that Switch doesn't have explicit/native private data write or explore abilities or configurators. If you do decide to use PD more, it helps to use a script to write key/val pairs (one of the simplest scripts you can make).




Thank you that worked

[ChrisSigge]

gabrielp wrote: To elaborate on Ariel's response, Ungroup and Assemble can work in tandem to achieve exactly what you're looking for.



Ungroup configurator:





Assemble configurator:





I find that private data is a life-saver and keeps things from being overly complex. I'm surprised that Switch doesn't have explicit/native private data write or explore abilities or configurators. If you do decide to use PD more, it helps to use a script to write key/val pairs (one of the simplest scripts you can make).




But now i have another problem, after the folder is ungrouped I check the pagesize of each pdf,

if the size is not correct that pdf is sent to trash, but than flow doesn't finish anymore.



Is there also an solution for this?

[ArielRauch]

what do you mean by: does not finish anymore?

[ChrisSigge]

ArielRauch wrote: what do you mean by: does not finish anymore?




Hi Ariel,



The remaining files wont get assembled back into the Folder.



That happens if I sort out one of the PDFs that have been in the Folder,

it seems the Assemble configurator is waiting for the missing File.

[gabrielp]

Ah.. that might be because its waiting for more files. I imagine that the ungroup key perhaps saves the digit of the number of files it is expecting so it knows to wait until all X jobs arrive. So you might be able to overwrite that PD key with the new number but that seems tricky.



There is another really nice way to assemble that I use a lot. You don't have to know the number of files but it does involve a timeout which means it adds processing time to your flow. If you can deal with that, it's probably the simplest way to handle it. It looks like this:



1. When the job comes in, assign it some sort of unique ID.

- A good example would be writing the Switch job id (prefix) to private data

- You could also write a timestamp string (with miliseconds) to private data

2. Then you ungroup, disregarding the Ungroup's private data key

3. Then you assemble, set the number of files to 99999 and set the matching scheme to be your unique id (whatever it was).

4. Set the assemble timeout to whatever length of time you'll be sure your processing will be complete, for me, 7 minutes.



Switch will then take all of the files it has with the matching ID and assemble them after 7 minutes. Before the assemble, you maybe delete or add a file -- as long as it has the unique ID, it gets assembled.



Good luck!