Thread: Re: [Fwd: Re: MaxLongVarChar and new lines in text]

Re: [Fwd: Re: MaxLongVarChar and new lines in text]

From
Åsmund Kveim Lie
Date:
>> Hi,
>>
>> I have sent to the news group (and you) the myLog files, but when I
>> attach them it does not seem like I'm allowed to post them (at least
>> they don’t get through to the news group after six hours). Do you know
>> if I should do it in any other way (e.g. inline them in the mail?).
>>
>> Thank you very much for locking into our problems with new lines. Do
>> you think it’s related to the text array problems?
>>
>
> Please replace the enhanced driver psqlodbc35w.dll by the one at
> http://www.geocities.jp/inocchichichi/psqlodbc/index.html .


Hi,
This one works much better, but it still don't understand the -4
correctly. From the psqlodbc_xx.log it looks like it truncates the value
down to 0 byte:

The 1th item was truncated
The buffer size = 1 and the value is '
'

But it works fine for values > 16383 and the newline is retrieved
correct, so it seems to be a error in the 08.01.0200 version. Is there
anything you fixed that could be patched also in this version of the driver?

Regards,

Åsmund

Attachment

Re: [Fwd: Re: MaxLongVarChar and new lines in text]

From
Hiroshi Inoue
Date:
Åsmund Kveim Lie wrote:
>
>>> Hi,
>>>
>>> I have sent to the news group (and you) the myLog files, but when I
>>> attach them it does not seem like I'm allowed to post them (at least
>>> they don’t get through to the news group after six hours). Do you
>>> know  if I should do it in any other way (e.g. inline them in the
>>> mail?).
>>>
>>> Thank you very much for locking into our problems with new lines. Do
>>> you think it’s related to the text array problems?
>>>
>>
>> Please replace the enhanced driver psqlodbc35w.dll by the one at
>> http://www.geocities.jp/inocchichichi/psqlodbc/index.html .
>
> Hi,
> This one works much better, but it still don't understand the -4
> correctly. From the psqlodbc_xx.log it looks like it truncates the value
> down to 0 byte:

The changed driver replies size 0(means the column size cannot be determined)
for SQLDescribeCol request while it replied (-4) before the change.
Because the parameter is unsigned, the (-4) means the large number 2**32 - 4
in reality.

regards,
Hiroshi Inoue