SectionedSurface Aplitop 1 #106

javierNadal · 2021-04-20T14:45:39Z

Unit test

This scenario is a simple road visualization (IfcSectionedSurface and IfcOpenCrossProfileDef)

Covered topics

tick one/many

Updating changes

Updating

larswik · 2021-05-31T08:26:23Z

SectionedSurface-Aplitop-1/SectionedSurface-Aplitop-1.IFC

+#45=IFCRELNESTS('1qVzEdro5CFf5K20Wsg4rv',#9,$,$,#31,(#33));
+#46=IFCCOMPOSITECURVE((#44),.F.);
+#47=IFCGEOMETRICREPRESENTATIONSUBCONTEXT('Axis','Model',*,*,*,*,#19,$,.GRAPH_VIEW.,$);
+#48=IFCSHAPEREPRESENTATION(#47,'Axis','Curve2D',(#46));


When there is an IfcGradientCurve available, I wonder if the shape representation for an IfcAlignment should refer to this (65) instead if the IfcCompositeCurve?

In a simple sample like this, just a straight line, yes.

But in a more general sample, I still don't have an implementation to build a single IfcCompositeCurve from any 2D alingment and grade line info. So I kept alignment representation as 2D.

I will change it.

It seems as if the IfcProductDefinitionShape on line #49 no longer has an owning IfcProduct. I assume that this should be the IfcAlignment?

I see that IfcAlignmentHorizontal has a shape representation = IfcCompositeCurve.
IfcAlignmentVertical has a shape representation = IfcGradientCurve
I wonder whether IfcAlignment (no shape representation in this file) should have the gradient curve as its representation (Axis,Curve3D)? @SergejMuhic ?

larswik · 2021-05-31T08:27:56Z

SectionedSurface-Aplitop-1/SectionedSurface-Aplitop-1.IFC

+
+#272=IFCPRODUCTDEFINITIONSHAPE($,$,(#273));
+#273=IFCSHAPEREPRESENTATION(#19,'Profiles','',(#274));
+#274=IFCSECTIONEDSURFACE(#65,(#275,#277,#279,#281,#283),(#276,#278,#280,#282,#284),.F.);


It looks to me as if the list of CrossSectionPositions should come as attribute 2 and CrossSections as attribute nr 3. I checked with the KIT checker where the data passes regardless of the ordering between these two?

Yes, you are right. I will swap them.

larswik · 2021-05-31T08:31:14Z

SectionedSurface-Aplitop-1/SectionedSurface-Aplitop-1.IFC

+#274=IFCSECTIONEDSURFACE(#65,(#275,#277,#279,#281,#283),(#276,#278,#280,#282,#284),.F.);
+
+#275=IFCOPENCROSSPROFILEDEF(.CURVE.,$,.T.,(0.031,3.5,3.5,0.056),(-0.667,0.,0.,-0.667),('119','30','0','30','139'));
+#276=IFCPOINTBYDISTANCEEXPRESSION(IFCLENGTHMEASURE(0.),IFCLENGTHMEASURE(3.531),0.,0.,#65);


From looking at the xml data, it would seem that there should be a negative vertical offset needed in this case. Also, i am not sure where the value 3.531 comes from. In the xml at station 0 we have -2.736 and at station 20 we have -3.778.

Value 3.351 is a left displacement of the OpenCrossProfile. First two widths (0.031 & 3.5) are located at left side of alignment.
I will check vertical offsets.

larswik · 2021-05-31T08:34:44Z

SectionedSurface-Aplitop-1/SectionedSurface-Aplitop-1.IFC

+#273=IFCSHAPEREPRESENTATION(#19,'Profiles','',(#274));
+#274=IFCSECTIONEDSURFACE(#65,(#275,#277,#279,#281,#283),(#276,#278,#280,#282,#284),.F.);
+
+#275=IFCOPENCROSSPROFILEDEF(.CURVE.,$,.T.,(0.031,3.5,3.5,0.056),(-0.667,0.,0.,-0.667),('119','30','0','30','139'));


The slope values seem off. Please consult the documentation (x to the left and y up, positive angles from x to y). Typically, the slope value when starting from the left would be relatively close to PI radians.

Also, unique tags would be recommended.

Yes, I'm changing slopes

larswik · 2021-05-31T08:38:00Z

SectionedSurface-Aplitop-1/SectionedSurface-Aplitop-1.IFC

+#277=IFCOPENCROSSPROFILEDEF(.CURVE.,$,.T.,(0.543,3.5,3.5,0.301),(-0.6667,0.,0.,-0.667),('119','30','0','30','119'));
+#278=IFCPOINTBYDISTANCEEXPRESSION(IFCLENGTHMEASURE(30.),IFCLENGTHMEASURE(4.043),0.,0.,#65);
+
+#279=IFCOPENCROSSPROFILEDEF(.CURVE.,$,.T.,(0.663,3.5,3.5,3.154),(-0.667,0.,0.,0.6667),('139','30','0','30','139'));


If it is by design to switch from tag '119' to tag '139', I would recommend adding '139' to the previous cross section or '119' to this. See documentation:

Yes, I will include one more vertex with 139 code in previous/next profile, and 0 width vector.

larswik · 2021-06-07T13:42:09Z

SectionedSurface-Aplitop-1/SectionedSurface-Aplitop-1.IFC

+#45=IFCRELNESTS('1qVzEdro5CFf5K20Wsg4rv',#9,$,$,#31,(#33));
+#46=IFCCOMPOSITECURVE((#44),.F.);
+#47=IFCGEOMETRICREPRESENTATIONSUBCONTEXT('Axis','Model',*,*,*,*,#19,$,.GRAPH_VIEW.,$);
+#48=IFCSHAPEREPRESENTATION(#47,'Axis','Curve2D',(#46));


It seems as if the IfcProductDefinitionShape on line #49 no longer has an owning IfcProduct. I assume that this should be the IfcAlignment?

larswik · 2021-06-07T13:42:39Z

SectionedSurface-Aplitop-1/SectionedSurface-Aplitop-1.IFC

+#46=IFCCOMPOSITECURVE((#44),.F.);
+#47=IFCGEOMETRICREPRESENTATIONSUBCONTEXT('Axis','Model',*,*,*,*,#19,$,.GRAPH_VIEW.,$);
+#48=IFCSHAPEREPRESENTATION(#47,'Axis','Curve2D',(#46));
+#49=IFCPRODUCTDEFINITIONSHAPE($,$,(#48));


This line should be referenced somewhere.

In updated file, #49 is representation of IFCHORIZONTALALINGMENT

larswik · 2021-06-07T13:49:06Z

SectionedSurface-Aplitop-1/SectionedSurface-Aplitop-1.IFC

+#65=IFCGRADIENTCURVE((#63),.F.,#46,$);
+#66=IFCSHAPEREPRESENTATION(#47,'Axis','Curve3D',(#65));
+#67=IFCPRODUCTDEFINITIONSHAPE($,$,(#66));
+#68=IFCFACILITYPART('2VJRykc916eO0F0TUHiUj9',$,$,$,'ROADSEGMENT',#24,#87,$,.COMPLEX.,$,.LONGITUDINAL.);


There's an error here. The 10th (penutimate) attribute should be the predefined type. Suggestion: IFCROADPARTTYPEENUM(.ROADSEGMENT.). The 5th attribute (ObjectType) can thereby be removed (set to $) since it doesn't add anything.

larswik · 2021-06-07T13:50:26Z

SectionedSurface-Aplitop-1/SectionedSurface-Aplitop-1.IFC

@@ -0,0 +1,87 @@
+ISO-10303-21;
+HEADER;
+FILE_DESCRIPTION(('ViewDefinition []'),'2;1');


ViewDefinition can be set to [Ifc4X3NotAssigned] for now.

Fixed IFCROADPARTTYPEENUM(.ROADSEGMENT.) , representation of IFCALIGNMENTHORIZONTAL, and ViewDefinition

larswik · 2021-06-17T12:18:34Z

SectionedSurface-Aplitop-1/SectionedSurface-Aplitop-1.IFC

+#72=IFCPOINTBYDISTANCEEXPRESSION(IFCLENGTHMEASURE(0.),3.531,0.,0.,#65);
+#73=IFCOPENCROSSPROFILEDEF(.CURVE.,'CrossProfile-2',.T.,(0.543,3.5,3.5,0.301),(2.47492598733333,3.141592654,3.141592654,2.47714082675747),('119','30','0','30','119'));
+#75=IFCPOINTBYDISTANCEEXPRESSION(IFCLENGTHMEASURE(30.),4.043,0.,0.,#65);
+#76=IFCOPENCROSSPROFILEDEF(.CURVE.,'CrossProfile-3',.T.,(0.663,3.5,3.5,0.,3.154),(2.47492598733333,3.141592654,3.141592654,3.141592654,3.80836500656817),('119','30','0','30','119','139'));


My interpretation is according to this image:

Is it possible that cross section at 60 should join 119 and 139 (at section 60) instead of 119 and 30, i.e. have the width 0 as last value?
Please also check that my interpretation is correct. Other than that, I think that the file LFTM!

I have added a new IfcOpenCrossProfileDef at station 45. Now I think it looks better.

larswik · 2021-06-17T12:21:59Z

SectionedSurface-Aplitop-1/SectionedSurface-Aplitop-1.IFC

+#45=IFCRELNESTS('1qVzEdro5CFf5K20Wsg4rv',#9,$,$,#31,(#33));
+#46=IFCCOMPOSITECURVE((#44),.F.);
+#47=IFCGEOMETRICREPRESENTATIONSUBCONTEXT('Axis','Model',*,*,*,*,#19,$,.GRAPH_VIEW.,$);
+#48=IFCSHAPEREPRESENTATION(#47,'Axis','Curve2D',(#46));


I see that IfcAlignmentHorizontal has a shape representation = IfcCompositeCurve.
IfcAlignmentVertical has a shape representation = IfcGradientCurve
I wonder whether IfcAlignment (no shape representation in this file) should have the gradient curve as its representation (Axis,Curve3D)? @SergejMuhic ?

jmirtsch · 2021-06-22T11:31:38Z

SectionedSurface-Aplitop-1/SectionedSurface-Aplitop-1.IFC

+#68=IFCFACILITYPART('2VJRykc916eO0F0TUHiUj9',$,$,$,$,#24,#87,$,.COMPLEX.,IFCROADPARTTYPEENUM(.ROADSEGMENT.),.LONGITUDINAL.);
+#69=IFCRELAGGREGATES('0nVIlPiJ1D0Opi4EVcPTpM',#9,$,$,#28,(#68));
+#70=IFCOPENCROSSPROFILEDEF(.CURVE.,'CrossProfile-1',.T.,(0.0310000000000001,3.5,3.5,0.056),(2.49643136367755,3.141592654,3.141592654,2.48087836828573),('119','30','0','30','119'));
+#72=IFCPOINTBYDISTANCEEXPRESSION(IFCLENGTHMEASURE(0.),3.531,0.,0.,#65);


KIT checker doesn't seem to detect this, but shouldn't DistanceAlong be IfcNonNegativeLengthMeasure
and not IfcLengthMeasure ?? This applies too all occurrences.

It should be an IfcNonNegativeLengthMeasure for the IfcPointByDistanceExpression:
TYPE IfcCurveMeasureSelect = SELECT (
IfcParameterValue,
IfcNonNegativeLengthMeasure);
END_TYPE;

One more thing. I discussed with the rest of the team this morning and even if the documentation lacks this information, the conclusion was that it is at least recommended to keep the tags unique within a profile. Would it be possible to e.g. add a "R" and "L" prefix to separate e.g. 30 on the left vs 30 on the right?

…THMEASURE

jmirtsch

It's a great test case and it's close. Please check my comments, it's possible I need to improve my own implementation. I would note that the example isn't a replica of the landxml data, but a simplification.

I think the schema could be improved to allow the section curve to be relocated with start point not at origin, as I'd suggest the intent is to have the middle of the carriage way coincident elevation with the alignment. You've offset laterally, but offsetting vertically then causes problems with nominating vertical profiles (pivot point).

jmirtsch · 2021-06-24T00:22:06Z

SectionedSurface-Aplitop-1/SectionedSurface-Aplitop-1.IFC

+#33=IFCALIGNMENTSEGMENT('0sVreI8Kn6tuw4UYDmeHtA',$,$,$,$,#24,$,#35);
+#34=IFCCARTESIANPOINT((286.47559897,239.79766567));
+#35=IFCALIGNMENTHORIZONTALSEGMENT($,$,#34,1.50346536986339,0.,0.,107.15812752,$,.LINE.);
+#36=IFCDIRECTION((0.999655739909883,0.0262374096515884));


Semantics of Alignment seem to be south to north, where as this direction positions the geometry to near east/west (ie x). Which is correct?

IFCDIRECTION((0.999655739909883,0.0262374096515884));

jmirtsch · 2021-06-24T00:29:39Z

SectionedSurface-Aplitop-1/SectionedSurface-Aplitop-1.IFC

+#72=IFCPOINTBYDISTANCEEXPRESSION(IFCNONNEGATIVELENGTHMEASURE(0.),3.531,0.,0.,#65);
+#73=IFCOPENCROSSPROFILEDEF(.CURVE.,'CrossProfile-2',.T.,(0.543,3.5,3.5,0.301),(2.47492598733333,3.141592654,3.141592654,2.47714082675747),('L119','L30','0','R30','R119'));
+#75=IFCPOINTBYDISTANCEEXPRESSION(IFCNONNEGATIVELENGTHMEASURE(30.),4.043,0.,0.,#65);
+#76=IFCOPENCROSSPROFILEDEF(.CURVE.,'CrossProfile-3',.T.,(0.663,3.5,3.5,3.154),(2.47492598733333,3.141592654,3.141592654,3.80836500656817),('L119','L30','0','R30','R119'));


(0.663,3.5,3.5,3.154)

3.154 for the road side width doesn't seem to match sketch of intent.

Yes, there is a swap between stations 45 and 60

jmirtsch · 2021-06-24T00:58:27Z

SectionedSurface-Aplitop-1/SectionedSurface-Aplitop-1.IFC

+#67=IFCPRODUCTDEFINITIONSHAPE($,$,(#66));
+#68=IFCFACILITYPART('2VJRykc916eO0F0TUHiUj9',$,$,$,$,#24,#87,$,.COMPLEX.,IFCROADPARTTYPEENUM(.ROADSEGMENT.),.LONGITUDINAL.);
+#69=IFCRELAGGREGATES('0nVIlPiJ1D0Opi4EVcPTpM',#9,$,$,#28,(#68));
+#70=IFCOPENCROSSPROFILEDEF(.CURVE.,'CrossProfile-1',.T.,(0.0310000000000001,3.5,3.5,0.056),(2.49643136367755,3.141592654,3.141592654,2.48087836828573),('L119','L30','0','R30','R119'));


Image above might be an incorrect interpretation, but as the last slope ( 2.48087836828573 ) is less than pi, the road side slopes as nominated in ifc would look more like the image below as I understand it. Can you please confirm?

Yes, slope was wrong

javierNadal and others added 11 commits February 12, 2021 09:21

Last alignment changes pending (Meeting Feb 10th)

3f05266

Alignment-Aplitop-1 RC3

3e49056

readme corrections

837118e

More readme corrections

0b4aa07

Merge pull request #1 from bSI-InfraRoom/main

633cb85

Updating changes

Updated with latest schema (12 March 2021)

270d5c1

Wrong file

6ff4431

Merge pull request #2 from bSI-InfraRoom/main

f9fbfcf

Updating

Merge pull request #3 from bSI-InfraRoom/main

286ee6c

Updating

SectionedSurface-Aplitop-1

4f38f6d

Updated png and ifc files

d3d771e

larswik requested changes May 31, 2021

View reviewed changes

Some corrections

8beac91

larswik requested changes Jun 7, 2021

View reviewed changes

Some corrections

7ff84df

Fixed IFCROADPARTTYPEENUM(.ROADSEGMENT.) , representation of IFCALIGNMENTHORIZONTAL, and ViewDefinition

larswik reviewed Jun 17, 2021

View reviewed changes

Added new IfcOpenCrossProfileDef at station 45

b353904

jmirtsch reviewed Jun 22, 2021

View reviewed changes

L&R prefixes in codes. IFCNONNEGATIVELENGTHMEASURE instead of IFCLENG…

bf35bdc

…THMEASURE

jmirtsch reviewed Jun 24, 2021

View reviewed changes

jmirtsch mentioned this pull request Jun 24, 2021

IfcSectionedSurface, IfcSectionedSolidHorizontal and IfcOpenCrossProfileDef Suggested Improvements bSI-InfraRoom/IFC-Specification#92

Closed

Some fixed errors

64fe227

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

SectionedSurface Aplitop 1 #106

SectionedSurface Aplitop 1 #106

javierNadal commented Apr 20, 2021

larswik May 31, 2021

javierNadal Jun 2, 2021

larswik Jun 7, 2021

larswik Jun 17, 2021

larswik May 31, 2021

javierNadal Jun 2, 2021

larswik May 31, 2021

javierNadal Jun 3, 2021

larswik May 31, 2021

javierNadal Jun 3, 2021

larswik May 31, 2021

javierNadal Jun 3, 2021

larswik Jun 7, 2021

larswik Jun 7, 2021

javierNadal Jun 7, 2021

larswik Jun 7, 2021

javierNadal Jun 7, 2021

larswik Jun 7, 2021

javierNadal Jun 7, 2021

larswik Jun 17, 2021

javierNadal Jun 17, 2021 •

edited

Loading

larswik Jun 17, 2021

jmirtsch Jun 22, 2021

larswik Jun 22, 2021

larswik Jun 22, 2021

javierNadal Jun 23, 2021

jmirtsch left a comment

jmirtsch Jun 24, 2021

jmirtsch Jun 24, 2021

javierNadal Jun 29, 2021

jmirtsch Jun 24, 2021

javierNadal Jun 29, 2021

SectionedSurface Aplitop 1 #106

Are you sure you want to change the base?

SectionedSurface Aplitop 1 #106

Conversation

javierNadal commented Apr 20, 2021

Unit test

Covered topics

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

javierNadal Jun 17, 2021 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

jmirtsch left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

javierNadal Jun 17, 2021 •

edited

Loading