Stephan Szabo wrote:
> On Tue, 14 Aug 2001, Tom Lane wrote:
>
> > Jan Wieck <[EMAIL PROTECTED]> writes:
> > > And he who needs that kind of long term row identifiers would
> > > be better off with 8-byte sequences anyway - IMNSVHO.
> >
> > What I need is a way to pad the struct declaration
Jan Wieck <[EMAIL PROTECTED]> writes:
> What I'm asking myself all the time is "which platforms do we
> support that doesn't have 8-byte integers?". Could someone
> enlighten me please?
Release a version that doesn't work without 8-byte ints, and I'm sure
we'll find out soon enough
- Original Message -
From: Tom Lane <[EMAIL PROTECTED]>
Sent: Tuesday, August 14, 2001 10:09 AM
> typedef struct FormData_pg_sequence
> {
> NameData sequence_name;
> int64 last_value;
> #ifdef INT64_IS_BUSTED
> int32 pad1;
[snip]
> } FormData_pg_sequence;
>
> This would work, I think,
> typedef struct FormData_pg_sequence
> {
> NameData sequence_name;
> int32 last_value;
> int32 increment_by;
> int32 max_value;
> int32 min_value;
> int32 cache_value;
> int32 log_cnt;
> char is_cycled;
> char is_called;
> } FormData_pg_sequence;
>
> If I just change "int32" to "int64" here, all
On Tue, 14 Aug 2001, Tom Lane wrote:
> Jan Wieck <[EMAIL PROTECTED]> writes:
> > And he who needs that kind of long term row identifiers would
> > be better off with 8-byte sequences anyway - IMNSVHO.
>
> What I need is a way to pad the struct declaration so that it leaves
> 8 bytes per
"Serguei Mokhov" <[EMAIL PROTECTED]> writes:
>> This would work, I think, but my goodness it's an ugly solution.
> Is anything wrong with just having two int32 per value for this case?
Well, we do want it to be int64 on machines where int64 is properly
defined. Or are you suggesting
#ifdef INT
"Joe Conway" <[EMAIL PROTECTED]> writes:
>> What I need is a way to pad the struct declaration so that it leaves
>> 8 bytes per int64 column, no matter what. I thought of
> What if you defined int64 as a union made up of one "long int" member and
> one 8 byte char member, and then always refer t
Tom Lane wrote:
[clip]
>
> This would work, I think, but my goodness it's an ugly solution.
> Has any hacker got a better one?
>
> regards, tom lane
How about:
#ifdef INT64_IS_BUSTED
#define int64aligned(name) int32 name##_; int64 name
#else
#define int64aligned(name) i
- Original Message -
From: Tom Lane <[EMAIL PROTECTED]>
Sent: Tuesday, August 14, 2001 11:28 AM
> "Serguei Mokhov" <[EMAIL PROTECTED]> writes:
> >> This would work, I think, but my goodness it's an ugly solution.
>
> > Is anything wrong with just having two int32 per value for this ca
Jan Wieck <[EMAIL PROTECTED]> writes:
> And he who needs that kind of long term row identifiers would
> be better off with 8-byte sequences anyway - IMNSVHO.
Indeed. I've been looking at switching sequences to be int8, and I see
just one little problem, which is to uphold the promise tha
10 matches
Mail list logo
