Invoke PLC workflow using cssdk

Hi,

I am trying to invoke PLC workflow using cssdk api. The workflow is getting successfully invoked but is expecting the user to click on Start Input to go ahead. Is there a way I can make the workflow continue without having to click on any link or button like Start Input? If so, could you please share your thoughts as I am new to Interwoven CMS.

Environment:
TS 7.3.2
Linux

Thanks in advance

Find more posts tagged with

Comments

Rick Poulin

The PLC is front-loaded with a bunch of CGI tasks, which means a web browser has to execute them. For the majority of those tasks, there are no server-side equivalents because it simply wasn't designed that way.

If you explain your goal, folks here may be able to suggest a better approach than rewriting the whole PLC as a server-side process.

bhavi1

thanks for your reply rpoulin.

our requirement was to execute PLC through java and cssdk. we have created one custom workflow to achieve this.

nipper

Then you have to rewrite it.

You cannot use CGI tasks, and you have to figure out how to get the proper input without the user interaction

It won't be fast, and it won't be easy. With enough time/money/expertise it can be done.

satyavelaga

Thanks for your reply, Our objective is , when a file is deleted from teamsite,and when we invoke this workflow,This file also should be deleted from runtime ,and all the database references also should be deleted.

We know PLC does this, but it cannot be invoked using CSSDK ,because of the CGI tasks

Can experts help,how to acheive this?

Thanks
Satya

Adam Stoller

If you submit the deleted files through the PLC workflow - in the normal manner as you would modified files, it should do what you want. It still doesn't help you perform the operations automatically without user interaction (CGI Tasks), but it should do the job ...

satyavelaga

Thanks Ghoti for your reply, Since we cannot go with PLC ,because of CGI tasks which requires user interaction, is it possible to achecive the same delete functionality with custom tasks?

-satya

Adam Stoller

Thanks Ghoti for your reply, Since we cannot go with PLC ,because of CGI tasks which requires user interaction, is it possible to achecive the same delete functionality with custom tasks?

-satya

See nipper's response. Amazingly enough - just asking the same question over and over does not often result in different answers.

nipper

See nipper's response. Amazingly enough - just asking the same question over and over does not often result in different answers.

Shockingly even my kids have (finally) learned that lesson.

This can be done. It will be neither cheap, easy or quick. I would likely tell a customer about 4 months with a lot of Java development. Testing would be a nightmare, support after the fact, worse.

Not that it isn't a good idea. I would love it. Many of us would. But the cost of getting there is quite high.

You want to do it ? I hope your project is paid by the hour, because it will be a while. Start by trimming all the fat from a copy of the PLC workflow. If you do not generate HTML, remove it. Remove tagging. Etc.

At that point make certain you have a correct workflow. It had better be perfect. Then identify all the CGI tasks and start rewriting them as Java/External tasks.

By the time you are done, a new PLC will be out and start over.

Rick Poulin

To repeat what nipper and fish already said (which you appear to have missed), is that you can save yourself the nightmare by just training your users to understand that a deletion is just another form of an 'content change' operation, and that this change must be pushed to the runtime to take effect.

Which would you prefer: 5 minutes of user training, or 4 months of semi-supported dev?

ebiz

Can you please post the code snippet used to invoke the PLC workflow. Even if it wasn't able to invoke the CGI task it can help us to understant how to do it.

We have a workflow model "XYZ.ipm" which is being used when anything is submitted from Teamsite UI. But now we also want to use CSSDK API and submit files from backend. I am able to create the DCR , set the extended attributes using the CSSDK API , so the next step would be to use that .ipm and start the deployment where we need the expertise.