Skip to main content
Solved

"Conditional branching" : run a part of the workflow depending on the value of a user parameter

M
Forum|alt.badge.img

Dear all,

 

I am trying to do the following : I have a yes/no user parameter that determines if additional output files are written or not. At several steps of the workflow, if this parameter is set to 'yes', intermediary files are written (before the final output file).

 

To implement that, before all the intermediary writers, I put a sampler (to go through the filter with a single feature instead of the > 300'000 features of my data) followed by a TestFilter inside which I test the value of the parameter.

 

It works but I don't find it very ''elegant''. And as I am new to FME, I was wondering, what would be the more "FMEic-way" to implement it ?

 

Does a transformer (or something like a junction?) exist which would act as a ''gateway'' and pass the data further downstream ONLY if a certain condition is met ?

 

Thanks in advance for any information

Best answer by virtualcitymatt

mzuer wrote:

but let's say the data you want to write in the file contain a lot of features (>100'000), if you put a Tester between the Writer and the Transformer that precedes the Writer, all the > 100'000 features will be processed by the filter, isn'it ? it is still more efficient than using a sampler before the Tester ? (sorry if the questions sound naive, I am really newbie to FME)

FME (at least when it's processing in BulkMode) can process 100,000+ features very quickly. Attached is an example which creates 1,000,000 features and runs the test and then opens an inspector. The workspace takes only 3 seconds. Without the inspectors its more like 1-2 seconds.

 

View original
Did this help you find an answer to your question?

6 replies

nielsgerrits
VIP
Forum|alt.badge.img+52

I'm not sure how what you implemented works for all features. When I do this I just put a tester in the connection to the FeatureWriter with the condition $(Debug) = Yes. This performs quite well.

virtualcitymatt
Celebrity
Forum|alt.badge.img+34
nielsgerrits wrote:

I'm not sure how what you implemented works for all features. When I do this I just put a tester in the connection to the FeatureWriter with the condition $(Debug) = Yes. This performs quite well.

Same here. The tester shouldn't have much or a measurable impact on performance I wouldn't think

M
Forum|alt.badge.img
  • mzuer
  • Author
  • July 3, 2023
nielsgerrits wrote:

I'm not sure how what you implemented works for all features. When I do this I just put a tester in the connection to the FeatureWriter with the condition $(Debug) = Yes. This performs quite well.

but let's say the data you want to write in the file contain a lot of features (>100'000), if you put a Tester between the Writer and the Transformer that precedes the Writer, all the > 100'000 features will be processed by the filter, isn'it ? it is still more efficient than using a sampler before the Tester ? (sorry if the questions sound naive, I am really newbie to FME)

nielsgerrits
VIP
Forum|alt.badge.img+52
mzuer wrote:

but let's say the data you want to write in the file contain a lot of features (>100'000), if you put a Tester between the Writer and the Transformer that precedes the Writer, all the > 100'000 features will be processed by the filter, isn'it ? it is still more efficient than using a sampler before the Tester ? (sorry if the questions sound naive, I am really newbie to FME)

Sure, but then only the sampled number of features will be written. If that is what you need, no problem.

virtualcitymatt
Celebrity
Forum|alt.badge.img+34
mzuer wrote:

but let's say the data you want to write in the file contain a lot of features (>100'000), if you put a Tester between the Writer and the Transformer that precedes the Writer, all the > 100'000 features will be processed by the filter, isn'it ? it is still more efficient than using a sampler before the Tester ? (sorry if the questions sound naive, I am really newbie to FME)

FME (at least when it's processing in BulkMode) can process 100,000+ features very quickly. Attached is an example which creates 1,000,000 features and runs the test and then opens an inspector. The workspace takes only 3 seconds. Without the inspectors its more like 1-2 seconds.

 

M
Forum|alt.badge.img
  • mzuer
  • Author
  • July 8, 2023

thanks to both of you for your insights !

Reply


Most helpful members this week

Community Stats

30,846
Posts
116,935
Replies
38,669
Members
Terms & ConditionsCookie settings

Cookie policy

We use cookies to enhance and personalize your experience. If you accept you agree to our full cookie policy. Learn more about our cookies.

 
Cookie settings
A product of
Safe Software respectfully acknowledges that we live, learn and work on the traditional and unceded territories of the Kwantlen, Katzie, and Semiahmoo First Nations.