Re: Regarding a reversed, multiple sheets per page bug]

[Bartosz Kostrzewa <zoombat runbox com>]

Alex Tsui wrote:
> A workaround that I had in mind was to print a document with the
> desired pages per sheet and in order at first.  From this intermediate
> document, print to PDF again, this time with one page per sheet and in
> reversed order.  This gives the correct order at the end.  Maybe this
> two-step process could be turned into code to deal with the reversed,
> multiple sheets per page case.  Does this seem to be a acceptable fix
> to pursue?

That's an interesting solution. After my patch for printing odd and even
sheets correctly (with multiple pages per sheet) is accepted
(http://bugzilla.gnome.org/show_bug.cgi?id=583429 the "new attempt"), I
wanted to introduce conditions for reverse printing which would
decrement the page counter by pages_per_sheet and then print forward
from the resulting page, decrement again by 2*number_per_sheet and so on...

Your solution is much less error prone and would require only minimal
changes. It would also preserve the correct whitespace at the end of a
collated copy introduced by my patch automatically.

I'm still in favour of simply building a model of the printout and
traversing that by sheet (rather than by page) in ev-print-operation...
it would solve all the problems at once and require much less logic and
jumping around. I haven't received an opinion on that yet though so I
haven't really continued working on it.

-Bartek