[Python-Dev] is this obselete?: [Patch #100888] Fix UCNs mach ine with >= 32bit longs

[Bill Tutt]

Well, I can't check it in, so assigning it back to me would be fairly
pointless.  Someone should also chime in on the Tools/perfecthash stuff as
well.  If we want to keep it, I'll see if I can come up with an example
using Python's tokens since the data set is way smaller. :)

Bill





 -----Original Message-----
From: 	'Trent Mick' [mailto:trentm@activestate.com] 
Sent:	Tuesday, July 25, 2000 3:55 PM
To:	Bill Tutt
Cc:	Jeremy Hylton; M . -A . Lemburg; Mark Favas; Fredrik Lundh;
python-dev@python.org
Subject:	Re: [Python-Dev] is this obselete?: [Patch #100888] Fix UCNs
mach ine with >= 32bit longs

On Tue, Jul 25, 2000 at 12:08:57PM -0700, Bill Tutt wrote:
> Well, Fredrik's changes aren't checked in yet, so checking it all in
doesn't
> hurt.
> 
> The other relevant question is since Fredrik's code will eventually get
in,
> do we want to yank the code that generates it?
> (Tools/perfecthash) If we keep it, then that part of the patch should go
in.
> If we don't keep it, that's fine too.
> 
> You'd need to test it on Linux64. Win64's integer types are 32bits, and
the
> C code doesn't use long longs.
> The easiest way of testing it of course is to apply the ucnhash.c patch to
> your tree and run test_ucn.py. :)

Okay. I tried it and with the patch test_ucn.py passes fine. I am going to
add a comment to the patch on SourceForge and then assigned it back to....
hmmmm... eenie meenie minie... Bill! Is that alright?


Trent

-- 
Trent Mick
TrentM@ActiveState.com