Closed leoleynse closed 2 years ago
Thanks for reporting. Sounds like a reasonable enhancement to have a warning as proposed.
I am not aware of PDOK. Is this ADE publically available?
PDOK (www.pdok.nl) is the Dutch portal for open geo datasets of the government. The BGT-files (topo info) use the extra features of the IMGeo ADE (https://docs.geostandaarden.nl/imgeo/catalogus/imgeo/).
Hello,
I want to ask about my research, which is about converting IFC data to CityGML and CityJSON formats for 3D Cadastre.
I'm having problem converting citygml model to cityjson using citygml tools. The citygml model I got from converting the ifc model on the fzk viewer. The ifc elements I convert to citygml are ifc space and ifc wall. But when I convert it to a cityjson model, the space element doesn't appear in the ninja cityjson viewer. Can you explain why this can happen? does cigyjson not support space elements represented in lod4solid model? because the wall element can still appear (wall element in lod2MultiSurface). here I attach the citygml and cityjson files
Best Regards
Romanza cobacoba_2.zip cobacoba_2.zip
citygml-tools uses CityJSON 1.0 as target, which does neither support rooms nor LoD4 geometries (see the spec here) . Your spaces are not contained in the CityJSON output because they are converted to CityGML LoD4 rooms by the FZKViewer (which is correct). This issue cannot be solved easily because in CityGML 2.0 you can only represent rooms in LoD4.
but when I convert the citygml model to cityjson generated from FME, the model can be rendered completely (including space elements represented in LOD4 MultiSurface) in cityjson web viewer. Is the version of cityjson used by FME different from citygml tools?
the following is proof that the citygml model converted from fme cannot be converted to cityjson in citygml tools, while on fme I can convert citygml to cityjson
Also in first picture, I converted the ifc model into citygml with 3 different software, namely evebim, fme, and fzk. But when I'm going to validate the xml model citygml generated by fme, the citygml tools display says "0 found cityobjects" (still in the image). What does this message mean? even though the xml validation of the citygml model generated by evebim and fzk can be done without any error messages
Well, you map Room
to BuildingPart
. This is a completely different scenario and cannot be compared with the files you initially provided. BuildingParts can also easily be converted with citygml-tools. However, from a semantic point of view, this mapping is incorrect.
Well, citygml-tools says "Found 0 file(s) at..." instead of "0 found cityobjects". This obviously means that the file you provided as input cannot be found at the specified location...
regarding the problem "Found 0 file(s) at.....", how to solve the problem? so that my citygml xml file can be validated
Provide the correct path + filename to your CityGML file. If you just provide a filename without a path, then citygml-tools searches the file in the current working directory by default. If this still doesn't work, try and get rid of the special characters ([
, ]
, #
) in the filename.
it's true that after I renamed the file by deleting the character as you said, I can now validate my citygml file. Thanks very much for your help.
But I want to ask again, about converting citygml to cityjson, I still can't see the feature when loaded into the cityjson ninja web viewer, even though citygml tools can convert it to cityjson ? does cityjson also not support representation of citygml elements if modeled in lod4multisurface ? (my citygml file is lod4multisurface). I still can't understand the cityjson specs regarding LOD representation in cityjson
CityJSON uses an LoD schema defined by TU Delft: https://3d.bk.tudelft.nl/lod/. This schema goes beyond the CityGML LoDs because it has some intermediate levels like LoD 2.1 or 3.2. But it also completely lacks the LoD 4 of CityGML. In general, CityJSON 1.0 lacks the possibility to model interior objects such as rooms.
This has been changed in CityJSON 1.1. And LoD 4 has been dropped with CityGML 3.0. So, conversion should become easier with the latest versions. However, please note that the GML/XML schemas for CityGML have not been published yet. Thus, citygml-tools still works with CityGML 2.0 and CityJSON 1.0.
OK I will re-read resource that related to this.
Next, let me ask again about what you said that mapping Room elements to Building Parts is something that is not true from a semantic point of view. What is the meaning of this statement? because I got the CityGML to Cityjson conversion scheme from : https://community.safe.com/s/question/0D54Q000080hfvVSAQ/problem-with-conversion-of-citygml-to-json
in addition, fme also provides cityjson model results that can be displayed in the ninja viewer
A BuildingPart
in CityGML 2.0 is not meant to model interior structures of a building. It is meant to represent a single building that belongs to a building complex. So, a BuildingPart
must be a building on its own - with walls, roof and a ground surface. That's why a room inside a building should not be represented as BuildingPart
. Note that CityJSON uses the same semantics.
Just because your data validates against the XML/JSON schema or can be visualized with some tool doesn't mean that the semantics are correct. Correct semantics are important if you want to be able to query your data or use it for analyses and simulations that go beyond pure visualization. You wouldn't store a building as Road
feature, would you? Hence, you shouldn't store a room as BuildingPart
.
Maybe this link is also useful: https://en.wiki.quality.sig3d.org/index.php/Modeling_Guide_for_3D_Objects_-_Part_2:_Modeling_of_Buildings_(LoD1,_LoD2,_LoD3)#Building_Part_.28bldg:BuildingPart.29 (note that this document only provides explanations and examples for how to model CityGML buildings. So, it's just informative but not an official part of the CityGML specification)
It would be nice if you got a warning using "non-vanilla" files. Is it possible to implement this option?
I added a warning if a input file contains unsupported non-CityGML content (see here). This will be released with the next major version 2.0 of citygml-tools. Closing this issue.
Hello,
I tried to use citygml-tools to convert the Dutch BGT-files (CityGML-files from PDOK) to CityJSON-files. Unfortunately the output-files didn't have a useful outcome. Reason appeared to be the fact that the PDOK files weren't "vanilla" CityGML files; they contain an ADE.
It would be nice if you got a warning using "non-vanilla" files. Is it possible to implement this option?
Regards,
Leo