It appears as though \placefigure doesn't work very well with columnsets if the size of the floats isn't an exact multiple of the underlying grid. More precisely, \placefigure and friends, when used with any of the *tb and tb* options (tblr, tbrl, lrtb, rltb, fxtb) snap the bottom of the figure to the grid rather than the top. Of course, that means that most of the time figures will not be aligned at the top. (See below for an example.) Does anybody know of a fix or workaround for this problem? Reimer Behrends \showframe \showgrid \setuplayout[ backspace=0.75in,topspace=0.75in, width=7.0in,height=9.5in, header=0.0in,footer=0.35in] \setuppapersize[letter][letter] \definecolumnset[two][n=2,distance=12pt] \def\myfig#1{\placefigure[tblr]{none}{ \framed[offset=6pt,width=broad,align=right]{ \switchtobodyfont[10pt] #1 }}} \setupbodyfont[12pt] \starttext \startcolumnset[two] \myfig{\input knuth} \dorecurse{3}{\input knuth}\myfig{\input tufte} \stopcolumnset \stoptext
At 00:50 07/12/2003, you wrote:
It appears as though \placefigure doesn't work very well with columnsets if the size of the floats isn't an exact multiple of the underlying grid. More precisely, \placefigure and friends, when used with any of the *tb and tb* options (tblr, tbrl, lrtb, rltb, fxtb) snap the bottom of the figure to the grid rather than the top. Of course, that means that most of the time figures will not be aligned at the top. (See below for an example.)
Does anybody know of a fix or workaround for this problem?
These things have to do with the fact that captions should snap on the grid; however, you can influence things by setting the grid parameter for a float instance. More on that in details.pdf. Also the next version has a grid key in \externalfigure which can snap scaling to combinations of (height div lineheight) * lines + [0,1] * depth + [0,1] * height which in combination with the float grid options can give you whatever you want. Hans PS. Grid snapping introduced quite some border cases and a lot of effort has gone into providing control over that. An associated new feature is a yet to be documented and for the moment experimental graphic/background synchronization mechanism.
participants (2)
-
Hans Hagen -
Reimer Behrends