Skip to main content

Hi,

I am trying to take advantage of the Direct URL feature. In my use case, I am including links in messages such as emails, where if the user clicks it, it runs a workspace on FME server with the correct parameters that calls a REST API to create a new project on a particular server. The user gets a summary screen in a web browser that simply reports whether the workspace ran successfully or not, but does not relay any other information. It also asks the user to read the logfile for more info., but there are no logfile links on this page. Here's an example in FME Server 2017.1:

In the above case I added a Terminator so that if the feature comes out of the <Rejected> port of the HTTP caller, the workspace stops and results in an error. In this case I tried to create a project that already existed. Now the workspace does have this exact information in the response body from the server that it interacted with, but we have no way to route this information to the resulting summary page. It would be really useful if this was possible so the user would actually see what the failure is about (the above makes you think you need to re-run it or there is a problem with the workspace). The page doesn't even include a summary of what just ran, so for the end user this is not very useful unless they can infer that information from the URL in their browser. Thanks.

Hi @dbaldacchino,

the direct URL has a query parameter opt_showresult which is set to false by default. If you set this parameter to true in the link that you send out you will see more details about the failure including the Termination Message that you can define in the Terminator transformer in your workspace. Here is an example URL with the parameter set to true:

https://<your_fme_server>/fmejobsubmitter/repository/workspace.fmw?opt_showresult=true&opt;_servicemode=sync

The results page will look something like this:

0684Q00000ArLGnQAN.png


Hi @dbaldacchino,

the direct URL has a query parameter opt_showresult which is set to false by default. If you set this parameter to true in the link that you send out you will see more details about the failure including the Termination Message that you can define in the Terminator transformer in your workspace. Here is an example URL with the parameter set to true:

https://<your_fme_server>/fmejobsubmitter/repository/workspace.fmw?opt_showresult=true&opt;_servicemode=sync

The results page will look something like this:

0684Q00000ArLGnQAN.png

Thanks @GerhardAtSafe, yes I found that yesterday but unfortunately it still doesn't improve things much. What would be much better is if we had control over what to show. For example if I run a workspace that creates a project on some sort of service, I could evaluate the response body and display a summary message with relevant info. Same if it fails for example.

 

 

In my workspace I have terminators with custom messages, but they are not visible if I enable this option. You still have to look at the original log, which isn't linked to from this page (plus the user might not even have rights to view logs). I just think there's some way to improve this with a few additional options/flexibilities. Maybe it could be tied to topics/subscriptions and display data from those if you post to them? Just an idea of course...thanks.
Thanks @GerhardAtSafe, yes I found that yesterday but unfortunately it still doesn't improve things much. What would be much better is if we had control over what to show. For example if I run a workspace that creates a project on some sort of service, I could evaluate the response body and display a summary message with relevant info. Same if it fails for example.

 

 

In my workspace I have terminators with custom messages, but they are not visible if I enable this option. You still have to look at the original log, which isn't linked to from this page (plus the user might not even have rights to view logs). I just think there's some way to improve this with a few additional options/flexibilities. Maybe it could be tied to topics/subscriptions and display data from those if you post to them? Just an idea of course...thanks.
This sounds like a lot of customization. One option I could think of would be a custom front end where you can actually return to the user what you want. Using our FME Server REST API you could submit jobs with the transformation endpoints and prase the response object for what you want to show to the user. You could even have the second workspace only for log file parsing:

 

The first workspace fails -> submit log parser workspace -> return the info you need.

 

 

The big downside is that you would need to do some HTML & JS coding and have a server or S3 bucket to host the page. But the flexibility and the potential to customize would be huge.

 

 

Here are some resources:

 

https://docs.safe.com/fme/html/FME_REST/apidoc/v3/

 

https://playground.fmeserver.com/

 


Reply