|
|
Re: Exception when accessing xtext index [message #1760858 is a reply to message #1760851] |
Wed, 03 May 2017 16:35   |
I.M. Wiesel Messages: 4 Registered: May 2017 |
Junior Member |
|
|
Thanks for reply, Christian!
My settings is: There are elements of type A which may reference to elements of type B. The instances of A'S and B's are wildy distributed over files in the workspace, but one file may not contain both, A and B, because they are defined in separate grammars.
Now the requirement is that at most one A may reference any B. So a naive approach would be to parse all files with A's for each B. I considered this being not really efficient.
My scope is based on imports of resources, but again, a file with A's imports files with B's, not other way around. So I do not know where B is referenced.
You propose to extend ieobjectdescriptions with the references A->B, is that right?
I did not quite got your other proposal to move the validation to the builder participant. Could you explain this in more detail? And will it be still usable as the "normal", FAST validation?
Thanks!
|
|
|
Re: Exception when accessing xtext index [message #1760859 is a reply to message #1760851] |
Wed, 03 May 2017 16:35   |
I.M. Wiesel Messages: 4 Registered: May 2017 |
Junior Member |
|
|
Thanks for reply, Christian!
My settings is: There are elements of type A which may reference to elements of type B. The instances of A'S and B's are wildy distributed over files in the workspace, but one file may not contain both, A and B, because they are defined in separate grammars.
Now the requirement is that at most one A may reference any B. So a naive approach would be to parse all files with A's for each B. I considered this being not really efficient.
My scope is based on imports of resources, but again, a file with A's imports files with B's, not other way around. So I do not know where B is referenced.
You propose to extend ieobjectdescriptions with the references A->B, is that right?
I did not quite got your other proposal to move the validation to the builder participant. Could you explain this in more detail? And will it be still usable as the "normal", FAST validation?
Thanks!
|
|
|
Re: Exception when accessing xtext index [message #1760860 is a reply to message #1760851] |
Wed, 03 May 2017 16:36   |
I.M. Wiesel Messages: 4 Registered: May 2017 |
Junior Member |
|
|
Thanks for reply, Christian!
My settings is: There are elements of type A which may reference to elements of type B. The instances of A'S and B's are wildy distributed over files in the workspace, but one file may not contain both, A and B, because they are defined in separate grammars.
Now the requirement is that at most one A may reference any B. So a naive approach would be to parse all files with A's for each B. I considered this being not really efficient.
My scope is based on imports of resources, but again, a file with A's imports files with B's, not other way around. So I do not know where B is referenced.
You propose to extend ieobjectdescriptions with the references A->B, is that right?
I did not quite got your other proposal to move the validation to the builder participant. Could you explain this in more detail? And will it be still usable as the "normal", FAST validation?
Thanks!
|
|
|
|
|
|
Powered by
FUDForum. Page generated in 0.02257 seconds