Re: dumb stacking page breaker

[Joe Neeman]

On Sunday 15 July 2007 08:37, Nicolas Sceaux wrote:
> Joe Neeman <address@hidden> writes:
> > [...]
> > To fit with this, I'd suggest that you move
> > Stacking_page_breaking::compute_page_breaks to something like
> > Page_breaking::pack_systems_on_least_pages and change its interface to
> > match the other system-spacing stuff (ie. take a configuration_index
> > instead of the lines directly and return Spacing_result instead of just
> > the systems per page). Then you can return the uncompressed/compressed
> > line stuff to private within Page_breaking.
>
> Joe,
>
> Here is a new patch accounting for your remarks.

Looks good.

> Should Page_breaking::pack_systems_on_least_pages also compute the
> Page_spacing_result force_, demerits_, etc, slots, even though they are
> not used in the subclass? (to better conform to the protocol designed
> for the other spacing methods).

If it doesn't complicate things too much, I think you should put it in. If it 
gets messy (eg. on the last page), then just put a comment noting the 
difference and we can worry about it if we actually need it in the future. 
The same remark applies to ragged-bottom/ragged-last-bottom support, which 
doesn't seem to be there at the moment.

Apart from that, I'm happy for you to commit it. Although you should probably 
add a regression test, just so that we have the coverage.

Joe