Manuel Collado wrote: > If you are considering enhancing the automatic scroll policy, I have a > further suggestion: > > The main problem is, IMHO, the independent behavior of the caret versus > the explicit selection. Things would be much simpler if the caret is > forced to be in the scope of the explicit selection, if any. This occurs > almost always, except when the explicit selection contains no text at all. > > In that special case I would suggest replacing the real caret by a > "virtual caret" that points to the beginning of the explicit selection. > This virtual caret could be made visible in the editing pane by a > distinctive mark at the top left corner of the explicit selection, > exactly the same way as the collapsed elements that contains the real > caret. So the caret (either real or virtual) always lies inside the > implicit or explicit selection. > > Once the caret and the explicit selection are made consistent, it should > be easy to implement an automatic scroll policy based on the caret, that > at the same time forces the explicit selection to be visible. > > One more hint: the automatic scroll policy should force the explicit > selection to be fully visible if it is not bigger that the editing > window. Current behavior is to just scroll until the caret is made > visible, so only part of the explicit selection is shown in some cases.
Thank you for the above suggestions. However please note that what you describe is probably not easy to implement in XXE, while the minor enhancement suggested by Philippe Nobili is probaby easy to implement in XXE. -- XMLmind XML Editor Support List [email protected] http://www.xmlmind.com/mailman/listinfo/xmleditor-support
