You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
We should implement a mechanism for automatically wrapping labels for graphol nodes whose label falls within their bounding box by default (class and object property comes to mind).
Motivation
Currently label placement il entirely managed by the user (label position, text wrapping, etc...). This has always been a source of problems when dealing with projects that use very long IRIs for predicates (e.g. for a long time we ignored the newlines that where included in them), and in general leads to bad practices since it mixes content with formatting. Another issue is that the existing behavior could degrade the quality of generated diagrams from the upcoming owl import feature.
From a quick brainstorming session we identified the following possible behavior wrt class nodes:
Make their label bounding box match the class node bounding box
Set the number of label lines equal to what fits said bounding box (using font metrics)
Set label to automatically wrap (preferring patterns like spaces and _)
Use an ellipsis when the label does not fit (probably wise to use the tooltip to display the full label)
IRIs should never allow literal newlines unless escaped (should be already the case), while newlines in annotations should always be preserved.
Example Use Cases
An example of said wrapping:
The text was updated successfully, but these errors were encountered:
Description
We should implement a mechanism for automatically wrapping labels for graphol nodes whose label falls within their bounding box by default (class and object property comes to mind).
Motivation
Currently label placement il entirely managed by the user (label position, text wrapping, etc...). This has always been a source of problems when dealing with projects that use very long IRIs for predicates (e.g. for a long time we ignored the newlines that where included in them), and in general leads to bad practices since it mixes content with formatting. Another issue is that the existing behavior could degrade the quality of generated diagrams from the upcoming owl import feature.
From a quick brainstorming session we identified the following possible behavior wrt class nodes:
IRIs should never allow literal newlines unless escaped (should be already the case), while newlines in annotations should always be preserved.
Example Use Cases
An example of said wrapping:
The text was updated successfully, but these errors were encountered: