On Sat, Dec 29, 2001 at 04:47:52PM +0100, Lars Gullik Bjønnes wrote: > I'd like to have some comments on if this approach seems sound.
The idea looks like seperating the different uses of 'Paragraph * p' into 'container related' (i.e. p points to start of a sequence of linked paragraphs) and 'item related' (i.e. we are only interested in the single paragraph p points to) and you started with the container stuff. This is certainly ok. [I'd probably have started at the other end (i.e. pass 'Paragraph (const) &' whenever a single Paragraph is meant), but that does not really matter.] I cannot tell, however, whether you have found all instances (or if not, left the rest in a consistent state) > (It surely is a _lot_ of work.) I promise to lend you a hand when you start to destroy everything ;-) Andre' -- André Pönitz .............................................. [EMAIL PROTECTED]