[XeTeX] XeTeX Win32 PDF file output error, second phase

[Jonathan Kew]

On 12 Aug 2006, at 3:25 pm, Maurice Bauhahn wrote:

> Thank you so much for your patience. Happily I've had some degree of
> success!!! A perfectly formed PDF file has been received:-)

Glad to hear it!

>  Admittedly...there
> may be some problem with the Italic font for TTX gives the following
> two messages while compiling the font:
>
> Build\buildttx\out1.pyz/fontTools.misc.textTools:64: FutureWarning:  
> x<<y
> losing bits or changing sign will return a long in Python 2.4 and up
> Build\buildttx\out1.pyz/fcntl:7: DeprecationWarning: the FCNTL  
> module is
> deprecated; please use fcntl

This sounds like an issue within TTX rather than something wrong with  
the actual font. You might like to report this to the TTX developers  
and encourage them to update their code to use more current Python  
modules.

> Subsequently I'm able to generate the PDF file in a two step process
> (but still get the suspect DOS16 errors when I try to run it simply
> as xetex 41Matt.tex...and when run as one step there is no PDF file
> output on the file system despite the message which follows).

I still don't understand why the one-step operation fails for you.  
Are you able to process other test files this way?

The message "Output written on 41Matt.pdf (67 pages)." is wishful  
thinking on xetex's part, I suppose. It doesn't really know what  
output got generated. What it knows is that it wrote 67 pages of xdv  
output (via a pipe) to the xdv-to-pdf driver, and passed 41Matt.pdf  
as the output name to use; so it prints this message in faith that  
the pdf will indeed get generated as requested. But if the driver  
fails, then it's misleading.

> The output (in addition to the lovely PDF file!!!!) when I run:
> xdvipdfmx -vv 41Matt.xdv is:.....

This looks fairly typical now. (To reduce screen clutter, you might  
prefer to run without the -vv option, or even with -q instead, now  
that the font issue is sorted out.)

JK