When you tried the LineCombiner what parameters did you use? Could you see that any lines were combined, i.e. there were less features coming out the LineCombiner than going in? I think the key thing for splitting into equal segments is to start off with one line.

The IterativeSnipper restarts at each feature, so it needs to be one continuous feature. Aggregating will produce a multi-part feature, which will not help in this instance.

You can use the LineCombiner (what wasn't working?), which will end at the split and have separate features for each fork, but otherwise would be the simplest solution.

If the line combiner truly doesn't work, in may be possible to use the NetworkCostCalculator to add Measures to your original lines as a distance from source (start of your trail) and the snipper in Mode Measure (Value), but it will take some work to make that iterative, and join your the individual features into the 4K segments.

@jdh and @egomm

When I use the LineCombiner I use the default settings, se below, and the output of the translator is 5 features. The original is 10 features, so it has combined half of them.

Below is what the original features look like, but after the LineCombiner the fork looks the same way still. The only difference is that there are only a total of 5 lines in the output. In the fork this means that the highlighted line (OBJECTID:2) had become a much longer feature going north and the line with OBJECTID:3 has become longer going south.

I agree, this must be why the snipping becomes random. Se more details in the new answer I posted.

If you remove OBJECTID 4 before the LineCombiner do you get a single line?

LineCombiner I only get to output at least 5 features from ten instead of 1.

When I try using LineBuilder I get one line output which is great! But the result is kind of crazy:

@egomm If I remove the "loop" OBJECTID:4 then the LineCombiner gives me three lines as an output with the default parameters...

Kullaleden_uppdelning.fmw

In this workspace I have first manually reworked the line so that I have three separate lines.

• One in which the trail ignores the fork in the road/the loop and is one line the whole way
• One in which I have broken up the line that the fork connects to, so that there is online using that path the whole way
• One line that represents the fork/loop

After dividing these lines up with an AttributeFilter, so that I can pick and choose the one I want to test, I put one (of the two complete lines) into the the IterativeSnipper

But the translation fails:

IterativeSnipper_Snipper_3_<Rejected>(TeeFactory): IterativeSnipper_Snipper_3_<Rejected>: Termination Message: 'IterativeSnipper_Snipper_3 output a <Rejected> feature. To continue translation when features are rejected, change 'Workspace Parameters' > Translation > 'Rejected Feature Handling' to 'Continue Translation''

IterativeSnipper_Snipper_3_<Rejected>: Termination Message: 'IterativeSnipper_Snipper_3 output a <Rejected> feature. To continue translation when features are rejected, change 'Workspace Parameters' > Translation > 'Rejected Feature Handling' to 'Continue Translation''

More details:

The below feature caused the translation to be terminated

Storing feature(s) to FME feature store file `G:\\Verksamhet\\Drift och underhåll\\4 Avdelningsstöd\\GIS-ingenjör\\Skötseldatabas\\Projekt\\Kullaleden - certifieringskartor\\Kullaleden_uppdelning_log.ffs'

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Feature Type: `FEATURE_TYPE'

Attribute(encoded: windows-1252): `Beskr_' has value `Kullaleden (SV om Sofiero)'

Attribute(string) : `Id' has value `1'

Attribute(64 bit real) : `Iterative_Snipper_length' has value `85231.1330886951'

Attribute(encoded: UTF-8) : `Iterative_Snipper_maxLength' has value `4000'

Attribute(string) : `Kullale_ID' has value `0'

Attribute(string) : `SHAPE_GEOMETRY' has value `shape_polyline'

Attribute(string) : `fme_feature_type' has value `Kullaleden_bearbetad_utan_loop'

Attribute(string) : `fme_geometry' has value `fme_aggregate'

Attribute(string) : `fme_rejection_code' has value `INVALID_GEOMETRY_TYPE'

Attribute(entangled: string) : `fme_type' has value `fme_line'

entangled to gSHAPE_GEOMETRY]

Attribute(32 bit integer) : `multi_reader_full_id' has value `0'

Attribute(32 bit integer) : `multi_reader_id' has value `0'

Attribute(string) : `multi_reader_keyword' has value `ESRISHAPE_1'

Attribute(string) : `multi_reader_type' has value `ESRISHAPE'

Coordinate System: `_SWEREF-99-13-30_0'

Geometry Type: IFMEMultiCurve

Number of Curves: 3

So the error log indicates that your geometry is still an aggregate. Even though you've separated things off into what you think are single lines, i think it's still worth using the LineCombiner beforehand with Aggregate handling set to Deaggregate

@sofiamvalentin

First make the topology correct.

By using Snapper in end mode and then use LineCombiner (which is a LineJoiner and not a LineBuilder, for the latter combines on the order of the points)

Then a Intersector (if needed use a tolerance).

Use the IterativeSnipper.

Do not aggregate the objects.

Non-single line cannot be split with equi-distance anyway. I think you have only these choices.

• Ignore the 'double-track' (orange and blue), split only remnant two lines (the purple lines) separately.
• Remove one track (the orange one) from the 'double-track', connect remnant three lines to form a single long line (purple - blue - purple), then split it.

Which is your desired one? If the both above don't match your requirement, please explain your requirement specifically.

Yes, that is already done. What you see in the image is one blue line that goes the whole way, and one red line that goes the whole way but with the yellow detour. There is no purple line, that's just where the blue and red ones overlap. I want to create the segmentation along the blue or the red line.

This works! With the Snapper set to End Point Snapping and with a snapping distance of 10, and then the LineJoiner, the topologi problem seems to be solved (I can't understand why there was a problem to begin with, but oh well!)

At the end the translation runs through to give me 22 segments: 20 lines of 3999-4000 m lengths and one segment of 1230.9 m

Thanks to all the suggestions and help!