We happen to have a little issue with the generation of a mind map when namespaces are concerned.
Let's say wa have a "person" element in a "namespace1" namespace, and a "person" element in a "namespace2" namespace, when I call ref="namespace2:person", in the mindmap it is actually written "namespace1:person", which makes me think maybe CamEditor can sometimes confuse elements with the same name but different namespace.
Has anybody already reported this problem to you? Do you have any idea why that happened and if there is a way to fix it?
Thank you in advance!
PS : I can link you the schemas if needed.
Sounds like a potential bug in the XSLT that generates the mind map.
If you could send us a link to the schema so I can test it - that would be ideal.
Once resolved we can provide the patched XSLT for you to run directly.
Thanks a lot.
Please find attached the schemas (the main one is CDMFR-v1.1.2.xsd).
Hope you can find what's wrong.
OK - I've diagnosed this - it is in fact the import from the XSD schema. Turns out its not "wrong" its just different. CAM is trying to minimize use of namespaces and prefixes. So - in your schema XML prolog - instead of using defaultNamespace - you need to explicitly declare the xmlns:cdm directly. And then of course - all ref="name", type="name" etc need to reference ref="cdm:name" and so on.
I went through and edited all these - and then rebuilt the CAM template. I've attached the results - and the edited schema.
Also - note - when I imported the XSD - I checked off the "Filter non-UTF8" flag - since you have much French text - and you just want to retain that.
Thank you for the answer.
Unfortunately I don't think these changes solved the problem. Indeed, if I check line 347, I can see ref="cdmfr:programDescription" but when I build the CAM model (or look at the one you attached), I still get cdm:programDescription.
Do you get the same behaviour?
OK - I see there is an additional problem. I will have to take a look at the XSLT - see what is tripping up the logic. Thanks for bringing this to our attention. Hopefully it is not a complicated fix... more soon.
Please see attached ZIP of the updated CDM.cam template - I believe I have resolved the schema import issues.
Let us know. We are working on a new v3.2 release with this patch in it - so once that is out - you will be able to process your schema also.
Thanks again for reporting this,
It looks like the namespaces are correct now, but some elements seem to be missing?
When I look at the mind map, the cdm:program part which had 30 children now only has 8, which is a little bit worrying.
Thank you in advance.
OK I will take a look at that and see. That is the help I need with verifying the details here. More soon.
Appears I was not using the original schema, but one I had doctored to make testing run faster - Ooops!
Please see attached for updated ZIP.
Wow that is really great.
I check this afternoon and keep you updated.
Thanks a lot!
I have checked the new model, it looks like it is working just fine. Thanks a lot.
When will the fix be available?
New release CAM Editor v3.2 is already available with this fix in - just download and enjoy!
Thanks again for reporting this and helping us fix it.
Thank you again for the previous help.
After some more deep look into the generated model/mind-map, we still believe there are some worrying points :
1) In the CDMFR-ext schema we define a complexType with <xs:all> (name="prodPeda" line 323), but when I check in the model/mindmap, I cannot find the elements in the "all".
2) When visualizing the mindmap using Freeplane, I can see some elements are written in blue with a "<" before it. I thought it was a way to represent "choices" elements, but I also found some choices elements just written normally so I could'n really figure out what is the reason for the different style.
I am looking forward to any lead you could give us on these two issues.
PS : I posted in this topic because it is using the same schema already linked in this topic, but I can move it if needed.
Please see attached ZIP.
Also - can you give me an example of the xsd:choice not being shown in the MindMap with the < blue marking?
I have looked myself - and so far I only see ones that are now working OK.
Once you confirm this is now OK - we will do a new CAM v.3.2.1 bug fix release with the new fixes in it.
The new model seems to work now for xs:all elements.
Concerning an example of xs:choice not shown in blue, element <xs:element name="domainName"> at line 268 of CDM-ext.xsd is a choice but the choices are not displayed in blue in the mind map.
Thank you in advance
OK - makes sense - let me take a look at what is going on. I know we handle xs:all differently. I saw the blue < choice detail - let me double check to make sure it is matching properly - and if not why.
Thanks again for inspecting all these details.
Do you know when version 3.2.1 with the xs:all fix will be released?
Apologies for delay; I'm finishing up the xs:choice bug currently, and then plan is for 3.2.1 release this coming week.
Did you manage to find why some xs:choice elements are not displayed like the other ones?
Log in to post a comment.