Regarding this: -- TODO: Perhaps the BOS should be a Mark with Forward direction?
It's the way it used to be... However don't think its foolproof either: if you paste some text which does not fit in the window, the point will fall out of the display, but the "snapping" algorithm might not detect it. In general, your solution to do a "point snap" right after a scroll operation is quite appealing, because it minimises the number of layouts. The main downside is that it requires calling-back the UI from the buffer code. I'd like to avoid setting a precedent like this, so that the buffer code remains "pure". E.g. it can be extracted as a library. On top of that, there is the issue that pango layout is in IO, and I don't think it's justified to wrap it in unsafePerformIO: pango "Layout"s are stateful objects. Cheers, JP. On Fri, Aug 14, 2009 at 10:45 PM, Colin McQuillan<c...@community.haskell.org> wrote: > > Fri Aug 14 16:43:17 EDT 2009 m.ni...@gmail.com > * Try to do EOF behavior better - layout should always follow a snapScreenB > Ignore-this: 6d233d961c94e5050e3ec6c26d05cd3d > > M ./src/Yi/Buffer/HighLevel.hs -4 +3 > M ./src/Yi/Core.hs -9 +22 > > > > --~--~---------~--~----~------------~-------~--~----~ Yi development mailing list yi-devel@googlegroups.com http://groups.google.com/group/yi-devel -~----------~----~----~----~------~----~------~--~---
