[XeTeX] Default values of \pdfpage{width, height} (was: XeTeX and pdfpages: no dice?)
Bruno Voisin
bvoisin at mac.com
Thu Aug 2 22:16:27 CEST 2007
Le 2 août 07 à 20:58, Andreas Matthias a écrit :
> Andreas Matthias wrote:
>
>> There's still one thing that puzzles me: Why doesn't
>> \pdfpage{width,height} have sensible defaults?
>>
>> \documentclass{article}
>> %\pdfpagewidth=\paperwidth
>> %\pdfpageheight=\paperheight
>> \begin{document}
>> aaa
>> \newpage
>> \begingroup
>> \pdfpagewidth=80mm
>> \pdfpageheight=150mm
>> bbb
>> \newpage
>> \endgroup
>> ccc
>> \end{document}
>>
>>
>> I am pretty sure most (or nearly all?) users wouldn't be
>> able to tell what they get from the above example before
>> running it -- with and without the two line uncommented.
>>
>> I really hope that it's not just me for whom the result of this
>> example is highly unexpected.
>
> No opinions? Do all of you agree?
Can't say. Problem is, you need to have a certain degree of
familiarity with the inner working of pdfTeX to expect anything from
the above code: when are \pdfpagewidth and \pdfpageheight taken into
account? is this when the \vbox for a finished page is shipped out?
or when the \vbox for a new page is started? or when the first \hbox
within this \vbox is started?
All this will influence the interaction of \pdfpagewidth,
\pdfpageheight, \newpage, \begingroup, \endgroup and aaa, bbb and ccc
in the above. Personally I've no idea what the above should yield
after typesetting, before trying it.
It may or may not be related, but a few years ago Hartmut Henkel
discussed a bit the pdfTeX initialization procedure in a message to
the OS X TeX list, in relation to the change from pdftex.cfg to
pdftexconfig.tex and the way \mag is applied to the true units
specified in this file:
<http://tug.org/pipermail/macostex-archives/2004-December/012146.html>
Personally I gave up on pdfTeX after that, as I couldn't get the hang
of how the new initialization procedure worked, and I got a bit
heated up by the way the change had broken backwards compatibility.
Bruno Voisin
More information about the XeTeX
mailing list