When highlighting sections of text and labeling them the entire spanned text is captured and stored in the annotation. This may not be a wise use of storage for the annotations. Maybe there should be an option to capture just the first n characters?
This really needs to be addressed now that we are doing section-level annotations that can have very large spans. There is no need to store all this extra text for display purposes. see 1700181 navigate to sections of an annotated text.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
So, here are two thoughts about this bug. One is that the method AnnotationUtil.createAnnotation() already takes a parameter spannedText that overrides the default behavior of using the text captured by the spans. So, for example, if one is creating annotations programmatically with a script (such as when Bill generates section level annotations) then the script can decide just how big the "covered text" should be.
Two is that we should probably consider removing the covered text slot from the annotation class (i.e. "knowtator annotation".knowtator_annotation.text). I'm not sure what the rationale of storing this information is in the first place. It makes things a little easier just because it is possible to make a little more sense of annotations when the their text sources are not immediately available. In the normal course of Knowtator usage, the text sources should always be immediately available.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Logged In: YES
user_id=1052182
Originator: YES
This really needs to be addressed now that we are doing section-level annotations that can have very large spans. There is no need to store all this extra text for display purposes. see 1700181 navigate to sections of an annotated text.
Logged In: YES
user_id=1052182
Originator: YES
So, here are two thoughts about this bug. One is that the method AnnotationUtil.createAnnotation() already takes a parameter spannedText that overrides the default behavior of using the text captured by the spans. So, for example, if one is creating annotations programmatically with a script (such as when Bill generates section level annotations) then the script can decide just how big the "covered text" should be.
Two is that we should probably consider removing the covered text slot from the annotation class (i.e. "knowtator annotation".knowtator_annotation.text). I'm not sure what the rationale of storing this information is in the first place. It makes things a little easier just because it is possible to make a little more sense of annotations when the their text sources are not immediately available. In the normal course of Knowtator usage, the text sources should always be immediately available.