On Thu, 19 Aug 1999, Allan Rae wrote:
> On Thu, 19 Aug 1999, Juergen Vigna wrote:
>
> > If you select Default I think this is the LaTeX default paper-size
> > and you shouldn't define ANY papersize!!! And I think this is what
> > happens.
I noticed something yesterday that I thought was wierd and upon checking
it today I realized why it was wierd. In the code as it stood before my
patch and still stands now there are lines like:
char real_papersize = buffer->params.papersize;
if (real_papersize == PAPER_DEFAULT)
real_papersize = lyxrc->default_papersize;
This would appear to have been added at the same time as the other
papersize handling code. This means every call to dvips will be of the
form:
dvips ... -t real_papersize output.tex
including when the user has selected papersize PAPER_DEFAULT. Obviously,
custom paper sizes are handled by my new code and aren't affected in the
same manner.
This has its good and bad points. As Juergen pointed out yesterday it
allows a local administrator to force a certain pagesize by default.
However, in the case of IEEEtran.cls when submitting a ps file
electronically dvips needs to be run without a papersize setting so that
only the bounding box info is present in the file. Fortunately, it's very
easy to find and remove this unwanted page size info from a ps file but
it's an extra step.
So what, if anything, should we do about it?
I'd suggest adding another lyxrc variable: \force_default_pagesize bool
with a default setting of "true" to remain compatible with the current
operation. Then we'd only need to change the two or three instances of
the code snippet above to have an extra test in the condition. Thoughts?
Allan. (ARRae)
