How can i edit the default layout from " run a workspace" so i can let the user run the workspace from its own without giving him the option to view other repo's and workspaces .
The "html form" method that we had available in server 2015 did help with this a bit.
I am a supporter of "Self Service" so the user can run his workspace on his command and he doesn't have to ask me every time.
i would love to hear your thoughts on this.
If you know your way around Javascript this is rather easy to do with a bit of html and some JS code on a static web page.
Frameworks like Angular make integrating the FME Server REST API relatively simple.
Have a look at the FME Server Playground for lots of examples.
Do you need them to set parameters? If not, just give them a URL to the workspace that will run it.
You could also set up security for users so that each user is only permitted to see specific repositories and workspaces. Security is group-based, so you can create a group of users with similar permissions.
Being more creative, you could set up an email notification - so that a user runs a workspace by simply sending an email to FME Server. Any parameters could be parsed out of the title or content of the email. Or use another notification technique such as Directory Watch (the user just drops a file to be translated into a folder).
Or you could even have the users run a workspace on FME Desktop, that calls the Server workspace with an FMEServerJobSubmitter transformer.
Otherwise, as David says, you'd need to set up a custom web page solution. But I'd be tempted to check out the other options first. There are lots of ways to trigger a workspace without either needing access to the web user interface, or creating a custom interface.
If you know your way around Javascript this is rather easy to do with a bit of html and some JS code on a static web page.
Frameworks like Angular make integrating the FME Server REST API relatively simple.
Have a look at the FME Server Playground for lots of examples.
Thank you for showing me the "Populate Forms Dynamically" docs, this adds some context into solving this!
But what bugs me with this solution, is that i need another system to make this work.
That means, another system to maintain/patch/update/etc.
Do you need them to set parameters? If not, just give them a URL to the workspace that will run it.
You could also set up security for users so that each user is only permitted to see specific repositories and workspaces. Security is group-based, so you can create a group of users with similar permissions.
Being more creative, you could set up an email notification - so that a user runs a workspace by simply sending an email to FME Server. Any parameters could be parsed out of the title or content of the email. Or use another notification technique such as Directory Watch (the user just drops a file to be translated into a folder).
Or you could even have the users run a workspace on FME Desktop, that calls the Server workspace with an FMEServerJobSubmitter transformer.
Otherwise, as David says, you'd need to set up a custom web page solution. But I'd be tempted to check out the other options first. There are lots of ways to trigger a workspace without either needing access to the web user interface, or creating a custom interface.
I am in the situation where it is indeed needed that the user sets a parameter.
Broken into bits it actually runs down to a week number, but this could be any weeknumber that hasn't been submitted and is chronologically after the last one used.
Although i understand the email feature, it would force me to give instant feedback to a user that his "command" has been received and he has to wait for the response when the flow is done.
that is what i like about the webform, the users has control over the moment that he clicks on the "ok" button to submit the job. He then is able to view the progress until the page changes and show it is done. This tends to have, my guess, a positive effect on the users submitting jobs because one can see that the server is running his job.
With email that is not many visual feedback so one would have to build it in the flow.
Thank you for the reply!!
From what i understand from the replies, is that it isn't possible to create a "user oriented run a workspace" front in FME server and that one would need alternative tooling.
Thank you for your answers! greetings from a wonderfull rainy holland!
From what i understand from the replies, is that it isn't possible to create a "user oriented run a workspace" front in FME server and that one would need alternative tooling.
Thank you for your answers! greetings from a wonderfull rainy holland!
That is correct.
Greetings from equally wet Switzerland :-)
From what i understand from the replies, is that it isn't possible to create a "user oriented run a workspace" front in FME server and that one would need alternative tooling.
Thank you for your answers! greetings from a wonderfull rainy holland!
Maybe also worth checking the security options. That is the "proper" way to hide workspaces and repositories, though I can see it might not be the solution in all cases.
Maybe also worth checking the security options. That is the "proper" way to hide workspaces and repositories, though I can see it might not be the solution in all cases.
I can vouch that setting the security options works well for controlling access to repositories. I have several service accounts that can only access their respective repositories, and a couple of other groups, so that different departments can only see and run their workspaces.