Hello
So i have been successful with all other features, but this one seems to me more challenging. The point of interests is 1 and 8 the same on other side of the RAIL . 
Highly obliged with LOGIC / FLOW 
Solved
Complex Geometry MID-POINT Extraction ( 1) & (8) on either sides of RAIL
+9- Supporter
Best answer by natalieatsafe
@vimva679 I just gave the vertex extract a try, and it is indeed an overly complicated workflow, not very useful for what you'd like to do.
I did, however, have some luck playing with the CenterPointExtractor and the BoundsExtractor. Please see the attached template workspace, ExtractBeamGapMidpoints.fmwt, for more details. This workflow produced a midpoint at the two locations that you're looking for, using the sample data you provided above.
Give this a try with your source data and see if you get your expected midpoints! Hopefully it gives you some positive results!
+23- Contributor
Can you provide some sample data? Just the feature you need the point from.
+9- Supporter
caracadrian wrote:
Can you provide some sample data? Just the feature you need the point from.
PM'ed you, also i have updated the image with Point of Interest '1' and '8' mid-point of FRONT facing channel. / tnx for your help and support
+9- Supporter
caracadrian wrote:
Can you provide some sample data? Just the feature you need the point from.
The 2 Point an its coordinates XYZ
+11- Safer
@vimva679 Good question here. I've attached a sketch below to walk through my first idea on this ask. I have not yet tested this out on IFC data as I do not have an appropriate sample available to me at this time.
As long as you have vertices A, B, C, and D, you could extract the midpoint of line AB, and then the midpoint of line CD. You could then construct a line between these two midpoints (the green line in the sketch). Finally, extract the midpoint of this constructed line to acquire Point 2:
Once you have Point 2 as a feature in FME Form, you could use a CoordinateExtractor to acquire the X, Y, and Z of this point of interest. Let me know if this makes sense for your use case!
+9- Supporter
natalieatsafe wrote:
@vimva679 Good question here. I've attached a sketch below to walk through my first idea on this ask. I have not yet tested this out on IFC data as I do not have an appropriate sample available to me at this time.
As long as you have vertices A, B, C, and D, you could extract the midpoint of line AB, and then the midpoint of line CD. You could then construct a line between these two midpoints (the green line in the sketch). Finally, extract the midpoint of this constructed line to acquire Point 2:
Once you have Point 2 as a feature in FME Form, you could use a CoordinateExtractor to acquire the X, Y, and Z of this point of interest. Let me know if this makes sense for your use case!
Hello
Am after mid-point vertex A and C
I could extract vertex of all corners, BUT the challenge to assign / identify which coordinate is of which corner/ surface.
Attached is IFC file for test purpose please.
+11- Safer
- Best Answer
@vimva679 I just gave the vertex extract a try, and it is indeed an overly complicated workflow, not very useful for what you'd like to do.
I did, however, have some luck playing with the CenterPointExtractor and the BoundsExtractor. Please see the attached template workspace, ExtractBeamGapMidpoints.fmwt, for more details. This workflow produced a midpoint at the two locations that you're looking for, using the sample data you provided above.
Give this a try with your source data and see if you get your expected midpoints! Hopefully it gives you some positive results!
+9- Supporter
natalieatsafe wrote:
@vimva679 I just gave the vertex extract a try, and it is indeed an overly complicated workflow, not very useful for what you'd like to do.
I did, however, have some luck playing with the CenterPointExtractor and the BoundsExtractor. Please see the attached template workspace, ExtractBeamGapMidpoints.fmwt, for more details. This workflow produced a midpoint at the two locations that you're looking for, using the sample data you provided above.
Give this a try with your source data and see if you get your expected midpoints! Hopefully it gives you some positive results!
thanks Natalie, i will have go tmrw , but in mean time plz have a look with "Orientation Extractor" transformer. It behaves different with 2022 and 2023 version . I have got 2022.2 FME version and i had to go around extracting Surfaces. Only surface that it gave was 'valid surface orientation' .
Hence what happen was i lost / missing were the geometry of object . So I took a bit of turns to mix and match output to obtain the full geometry without losing any surface of IFC objects. i.e. i merged 'left hand rule' and valid surface orientation' to obtain full geometry.
+9- Supporter
natalieatsafe wrote:
@vimva679 I just gave the vertex extract a try, and it is indeed an overly complicated workflow, not very useful for what you'd like to do.
I did, however, have some luck playing with the CenterPointExtractor and the BoundsExtractor. Please see the attached template workspace, ExtractBeamGapMidpoints.fmwt, for more details. This workflow produced a midpoint at the two locations that you're looking for, using the sample data you provided above.
Give this a try with your source data and see if you get your expected midpoints! Hopefully it gives you some positive results!
great...it works :)
Reply
Rich Text Editor, editor1
Editor toolbars
Press ALT 0 for help
Helpful Members This Week
Enter your E-mail address. We'll send you an e-mail with instructions to reset your password.