And here's another problem: floating displays in mm are output from within a very strange environment called \%ds*fev, which is nowhere defined, which means it has the default parameters, including the line length of 16 cm. This results, for example, in incorrect horizontal alignment of floats when the actual line length differs from the default one.
Does anybody know why this environment is used in the ds@output-float macro? Maybe it should be com- pletely inherited from the current environment? The current line length in ds@output-float is stored in \n[df*old-ll] but is not used anywhere within this macro or any macro called therefrom and is de- stroyed at the end. This seems to be an error (val- ue assinged but never used) because the current in- dent is stored in a similar way but is inherited in the new environment via .in . The problem is demonstrated in the attached example. Use it with -Tps -P-p -PA5. Either it is not an error or nobody is using floats in -mm... -- () ascii ribbon campaign - against html e-mail /\ www.asciiribbon.org - against proprietary attachments
mm-df-bug.roff
Description: Troff document
