in-postscript media-type request stripped by pstops?

[blinkety bill]

> Helge Blischke wrote:
> > ...
> > The trashing of stuff between (and including) %%BeginFeature and
> > %%EndFeature comments
> > is a feature that has been introduced by CUPS 1.2. Users still debate if
> > this
> > is to be treated as a bug or as a feature.
>
> Um, CUPS doesn't "trash" feature stuff unless you request N-up or
> scaled output, at which point it assumes that you want to do something
> different with the document than originally requested.

OK, thanks, this is helpful to know, explains something about problems we started getting with the previous server ...

> There were some issues with CUPS adding its feature code after the
> documents' code, overriding what was in the document, but that was
> fixed in 1.2.4...

OK, I checked ramming a test straight to the printer, and compared ramming it just through pstops provided with cupssys-1.2.2 from ubuntu dapper, and you're right -- the beginfeature .. endfeature stuff is preserved.

Here is the command line I used:

cat testdoc_lin.ps | /usr/lib/cups/filter/pstops 0 testuser testtitle 1
testfile > testdoc_lin_pstops.ps

It adds some stuff, but doesn't remove the being/end feature info.

But that doesn't explain why when I print through cups the printer doesn't obey the media request ...

I guess it means somewhere cups is making pstops strip the stuff, perhaps it thinks I want scaled or N-up.

btw, since you are a CUPS developer, I'll mention, it would be very useful for debugging cups if there was a switch or tickbox that an admin can set to make cups do *everything* up to sending the doc to the printer, and have it dump to a file instead. I'm not aware that this can be done ...

I think it's theoretically possible to nab the file from the temp directory, but in practice ... our production server takes several jobs a minute, and thousands of connections a minute are logged to the access_log ... it's very difficult to debug a running CUPS server.

I will fool around some more and see if I can get to the bottom of this.

I am now reluctant to install the alternate cups filter ... there is no information provided on how I should compile it, and the server admin has removed the compiler. Please let me know if anyone is *sure* that either upgrading to 1.2.4 or installing the alternate filter will fix this, or else please provide me with good info on compiling it, or a binary suitable for i386 ubuntu dapper.

I'm happy enough to do this kind of stuff if it will fix it, but at the moment it seems problematic, given the feature stuff actually is in the output I generated, seems the prob is elsewhere. :(

Blinkety Bill

> --
> ______________________________________________________________________
> Michael Sweet, Easy Software Products           mike at easysw dot com
> Internet Printing and Publishing Software        http://www.easysw.com