On Tue, Mar 04, 2003 at 07:10:56PM +0000, Angus Leeming wrote: > * InsetLatex::latex escapes the contents of the label. Should I create a > write method for LabelInset that does the same with the contents of > cell(0)?
I dont think so, as cell(0).write() produces valid LaTeX per se. > * In math_parser.C, why is the new RefInset(t.cs()) not handled by > createMathInset? Laziness. Of course it should be moved to the factory. > * The string label_ in math_hullinset.[Ch] could now be thrown away. Should > I replace it with a LabelInset*, or could a real inset survive on its own > without such special casing? A real inset would survive, and I don't like the idea of putting a LabelInset* there. This creates a much stronger tie than we currently have with the string "solution". > At the moment, I believe that the MathHullInset should store a LabelInset* > as a straight replacement for string label_. Reasons: the label should be > placed always at the end of the row. Moreover, a row can have only one > label. This is, of course a valid reason. But why do you insist on a LabelInset* and do not use at least the MathAtom wrapper? Andre' -- Those who desire to give up Freedom in order to gain Security, will not have, nor do they deserve, either one. (T. Jefferson)
