On 2-1-2015 17:27, Leyne, Sean wrote:
>
>
If instead I write 32768 bytes in two writes of 16384 I do receive
the correct response. Is the client library buffering the blob writes
somewhere? If so how do I flush?
>>>
>>> Yes, fbclient have 16K buffer per blob. To flush it you sho
02.01.2015 17:46, Vlad Khorsun wrote:
>> AFAIK, client library parse info buffer anyway, so may be it should add
>> unflushed info
>> to the counter?..
>
> For what ?
To satisfy perverts, who used to forget how much info they have just pushed
into the blob.
--
WBR, SD.
> AFAIK, client library parse info buffer anyway, so may be it should add
> unflushed info
> to the counter?..
For what ?
Regards,
Vlad
--
Dive into the World of Parallel Programming! The Go Parallel Website,
spo
02.01.2015 17:30, Vlad Khorsun wrote:
> 02.01.2015 17:01, Mark Rotteveel wrote:
>> On 28-12-2014 14:07, Vlad Khorsun wrote:
>>> 27.12.2014 12:31, Mark Rotteveel wrote:
I have a weird problem. In a test I am writing a blob of 24576 bytes.
This is done in a write (isc_put_segment) of 16384
02.01.2015 17:01, Mark Rotteveel wrote:
> On 28-12-2014 14:07, Vlad Khorsun wrote:
>> 27.12.2014 12:31, Mark Rotteveel wrote:
>>> I have a weird problem. In a test I am writing a blob of 24576 bytes.
>>> This is done in a write (isc_put_segment) of 16384 and 8192 bytes.
>>>
>>> If I subsequently qu
> >> If instead I write 32768 bytes in two writes of 16384 I do receive
> >> the correct response. Is the client library buffering the blob writes
> >> somewhere? If so how do I flush?
> >
> > Yes, fbclient have 16K buffer per blob. To flush it you should
> > close the blob.
>
> Thanks. Ther
On 28-12-2014 14:07, Vlad Khorsun wrote:
> 27.12.2014 12:31, Mark Rotteveel wrote:
>> I have a weird problem. In a test I am writing a blob of 24576 bytes.
>> This is done in a write (isc_put_segment) of 16384 and 8192 bytes.
>>
>> If I subsequently query the blob for the length
>> (isc_info_blob_t
