Re: [XBoard-devel] gtk-branch

[h.g. muller]

At 08:38 9-11-2011 +0000, John Cheetham wrote:
> The other problem is that every board square looks identical so we should 
> be able to improve it by introducing some random variations into the 
> positioning when they are loaded onto each square.

The old code did have this, except that it was not random; from where 
tocutthe squares from the bitmap has to be carefully calculated, to make 
sure the exact bitmap provided by the user  is reproduced when thebitmap 
was as wide and high as the board (i.e. N x squareSize). Otherwise a 
Xiangqi board would not properly reproduce.

You were right about the piece menu; one needs to use a new left click 
toselect an item. Thisis rather inconvenient, though. Although I consider 
the piece meu a deprecated feature, and always use sweep selection 
(-pieceMenu false), people that wat to stick to the piece menu would 
probably be outraged they cannot select with the right up-click anymore. 
This destroys whatever convenience the piece menu had, in particular  the 
clever trick of letting a simple static right-click drop a Pawn. Can't we 
arrange it such that upclicks also select menu items?

Note I have tied upa few of the loose ends I signaled the other day, and 
pushed them to hgm.nubati.net (gtk-xt branch). If they dont duplicate 
something John already did, they can be pushed to Savanah. They include:

*) fixing of circle size and position (Atomic-capture animation, 
-showTargetSquares option and seekgraph dots).
*) unstacking on animate dragging (crazyhouse drops and seirawan gatings).
*) connecting various mouse handlers. Detour under-promotion, sweep 
selection in Edit Position, PV walking, and seek-graph hovering now all work.
*) the visibility of the seek graph has been restored by making the 
background color different from the axis & text color.