On 02/11/2009, at 10:37, Craig Ringer wrote:
> On 2/11/2009 5:21 PM, Raimon Fernandez wrote:
>> Here I'm again ...
>>
>>
>> I'm parsing the DataRow(B), and I'm having problems with NULL values.
>>
>> In the docs I can read they have a -1 value, an no bytes follow
>> them for
>> the value.
>>
>> But I'm getting a 1020 value instead of -1
>
> You're using RealBasic or something, right?
>
> If so, you're probably doing something funky with signed/unsigned
> integer handling and the representation of integers.
>
> -1 is 0xffffffff as a _signed_ 32 bit integer, same in little-endian
> or
> big-endian form. The same hex value as an unsigned integer is
> 4294967295 .
>
> ...
>
> I don't know where you're getting the 1020, but 4294967295 is
> MAXUINT32
> and suggests you're treating the data as an unsigned rather than a
> signed 32 bit integer.
yes, you're right, I had and old legacy code that was processing
incorrectly the received data as string directly than getting the
binary ...
what I see that it was working all the time, except the -1 ...
now it returns -1 ...
thanks !
refards,
raimon