Re: [pdf-devel] unit testing & void* & design-doc

[gerel]

On 10/15/07, Mathias Kolehmainen <address@hidden> wrote:
> I have a hard time imagining what you are building without that.  Who
> are the consumers of this library?  What are the inputs and outputs?
> What is the memory management mechanism?

Think for the library as a PDF language Interpreter, and each PDF
document as a program for this interpreter.
There are primitive types (pdf_obj.c, pdf_base.c) upon which more
complex types are defined (pdf_date.c, pdf_rectangle.c, etc) as each
font and image within a PDF document also.
The memory management mechanism I guess, is for managing all those
type objects at runtime (when interpreting the PDF document), and it's
highly related to the xref lib which is like the "interpreter memory
address space" (from the PDF document perspective).
Xref is actually more complex here because the idea in the
implementation is using it as the "interpreter director".

The overall idea is that some application uses this library for
viewing,printing or producing PDF documents.

I hope that helps a bit.

cheers
-gerel