Re: Signed usage of time_t

> After some investigation I realized what the problem was: in some
> cases the code uses negative values of time_t to (I assume) mean
> "everything's OK".

No, it uses negative time_t to mark unknown data.

> QNX uses uint32_t to represent time_t,

Yeah, I've heard -- and I think that's really stupid.  Actually, they've
changed to an unsigned time_t in version 6 -- previous versions had the
reasonable definition.

> POSIX doesn't specify exactly how time_t should be implemented,

While that's correct, all systems known to man have signed time_t,
except for QNX 6.  As a matter of fact, some Unix interfaces make little
sense with unsigned time_t -- for example time(2), which is defined to
return ((time_t)-1) in case of failure.

> I'm curious, was there any success running Polipo on NetBSD?

Yes, it works fine -- like all reasonable systems, BSD Unix uses signed
time_t.  I've personally tested Polipo on NetBSD 1.6.

> Anyways, I thought I'd bring this to the attention of the Polipo
> developers in case there's an interest in fixing it.

I suggest either compiling with the QNX 5 SDK, or adding the following
to polipo.h:

  typedef signed int polipo_time_t;
  #define time_t polipo_time_t

The proper fix would consist in changing -1 to ((time_t)-1) in places
where it matters, but I think that's overkill.

I'll add a test for signed time_t to Polipo's build process.

-- Juliusz

------------------------------------------------------------------------------
The demand for IT networking professionals continues to grow, and the
demand for specialized networking skills is growing even more rapidly.
Take a complimentary Learning@Cisco Self-Assessment and learn 
about Cisco certifications, training, and career opportunities. 
http://p.sf.net/sfu/cisco-dev2dev