[reportlab-users] Is ReportLab suitable for this task?

[Andy Robinson]

> And, did you print this 13,000 page PDF?  If so, did you do 
> it with Acrobat or send it directly to the printer or 
> something else?  My limited understanding of the structure of 
> a PDF is that there is a dictionary at the end of the file 
> with information that is vital to the rendering process and 
> therefore the whole file would have to be read before any 
> rendering could happen.  Can printers do this on their own 
> (Postcript 2.0 capable printers)?
> 
> Don
Most PDF files do need to be read completely. (There is
a specialized "linearized" form, which we don't produce).
However, the fiel format allows extensive reuse and with
repetitive forms-type apps, it's quite feasible to have
a 13000 page document which is just a few megabytes in
size.   

There are two ways in which PDF gets printed.  The traditional
one is that Acrobat Reader converts it back to Postscript
and send this to the printer; or, in many cases the printer
is sent the PDF and converts it to Postscript internally.
This involves writing out about 50kb of Postscript 'headers'
and function calls, and then decompressing and writing out
the page stream of each PDF file, which is basically postscript.
Workgroup printers were beginning to accept PDF directly
around 2000 (the same time as Postscript 3.0).  If your
printer has a network card, read the docs; often you can
FTP a Postscript or PDF file onto it and it will be printed.

There is a more modern approach to rendering PDF 'directly'
and I've heard that Adobe would like to move to this,
but unless you hear to the contrary you should probably
assume your PDFs get converted to Postscript, either by
Acrobat or on the printer.

Hope this helps,

Andy Robinson