-
Notifications
You must be signed in to change notification settings - Fork 1
Open
Description
When searching for annotations referring to a certain target link, one uses a request in the format:
https:////webannotator/api/annotations?access=read&link=59d08e6a-5cd9-4aed-8aa4-7074c270e635
Which performs a full text search on the value of the XPath node 'annotation/targets/targetInfo/link/text()'.
This creates some complications:
- There is no reliable way to get a list containing just the annotations referring to an exact target URL. Any link that includes the searched link as substring, will also be displayed. e.g. if one annotates the address 'https://someserver.somedomain/products' and also 'https://someserver.somedomain/products/laptops', when requesting the list of annotations available for '/products', one will also get the annotations referring to '/products/laptops'.
Suggestion: Allow differentiation between exact and partial match queries. - The request parameter syntax ...'&link=59d08e6a-5cd9-4aed-8aa4-7074c270e635' actually suggests that an exact match query will be performed.
Suggestion: Keep this syntax for exact match queries and create a new one for partial matches, e.g. '&linkFragment=59d08e6a-5cd9-4aed-8aa4-7074c270e635' - Since it is a partial match query and the searched node 'annotation/targets/targetInfo/link/text()' is not included in the annotationInfoList result, it is not possible to know which kind of match made a certain result appear in the annotationInfoList result without further requesting that specific annotation in full.
Suggestion: Include the searched node 'annotation/targets/targetInfo/link/text()' in the annotationInfoList result of partial match queries.
Metadata
Metadata
Assignees
Labels
No labels