I'm trying to use some unicode characters and have encountered a problem I just can't seem to solve. Although the characters display nicely in an .ijx window, the formatting for boxes around them gets messed up--presumably because they're double byte characters. I cannot find a way to solve this, although I've spent hours looking in the documentation and trying various things. Although the context in which I am trying to use them is much more complicated, the matter comes down basically to this problem:
sigma =: 4 u: 963 sigma σ <sigma ┌──┐ │σ│ └──┘ <'s' ┌─┐ │s│ └─┘ The box around the σ spaces incorrectly (by analogy with a box around an ordinary character). The parts of the box not on the σ line appear to believe they need an extra space, even though it doesn't. I have tried converting to U8 characters, among other things, but nothing seems to work. Is there a way to solve this formatting problem? I have tried, for instance, to insert an invisible blank appended to the σ to push the right side of the box over (not a perfect solution, but it would work) but can't even find a way to do that. The best solution would be to find a way to create the σ so that it doesn't make the < think it needs to be wider. Anyone have a way to do that? I am using this, for now, only on the display by employing W =: (1!:2)&2 as in: W 'σ' σ or W sigma σ W <sigma ┌──┐ │σ│ └──┘ Eventually I'll use LaTeX to set type--and I expect to encounter some version of this problem again when mixing TeX and J. Any advice on that? Carl ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
