Hehe, i need to read your code before i ramble.
I didnt see this part " r = sample(2)[0];"
In that case this seem like a general bug. You should report it.
As a temporary workaround you could do:
// Sample Data
inline float sample( int row , int col )
{
float3x3 sampleData;
float sampleDataStruc[] =
{0.1f, 0.2f, 0.3f,
0.4f, 0.5f, 0.6f,
0.7f, 0.8f, 0.9f}
;
sampleData.setArray(sampleDataStruc);
float output = sampleData[row][col];
return (output);
};
kernel testTable : ImageComputationKernel<ePixelWise>
{
Image<eRead, eAccessPoint, eEdgeClamped> src;
Image<eWrite> dst;
//The kernel function is run at every pixel to produce the output.
void process() {
float r;
r = sample(2,0);
// Output the color
dst() = r;
}
};
On Fri, Dec 30, 2016 at 4:23 PM, Erwan Leroy <er...@erwanleroy.com> wrote:
> Did you get that to work on your side?
> Even when I build in your suggestion it won't work unless my inline
> function returns a float4.
> I managed to get my array of data to contain float3s, but when I return
> sample() I have to cast it to a float4 otherwise it won't compile.
>
> Erwan
>
> On Dec 30, 2016 9:12 AM, "Mads Lund" <madshl...@gmail.com> wrote:
>
>> It is when you do eComponentWise, but you have set it to ePixelWise, so
>> it does all 4 chans in one go.
>>
>>
>> fre. 30. dec. 2016 kl. 14.06 skrev Erwan Leroy <er...@erwanleroy.com>:
>>
>>> Oh I see.
>>> I thought dst() was a float, and that the blink would run once for each
>>> channel. That's why I was extracting a single float value as r.
>>> Will try again, thanks.
>>>
>>> On Dec 30, 2016 4:02 AM, "Mads Lund" <madshl...@gmail.com> wrote:
>>>
>>> Looks like you are trying to put your float3 into dst() (which is a
>>> float4).
>>> Maybe you should break it up like this:
>>> dst() = float4(r.x,r.y,r.z,1.0f)
>>>
>>> On Fri, Dec 30, 2016 at 3:17 AM, Erwan Leroy <er...@erwanleroy.com>
>>> wrote:
>>>
>>> Hey Guys,
>>> I ran into this problem with blinkscript, I know this mail list is for
>>> python but you guys may have some insights.
>>>
>>> Basically the following script runs just fine (sorry about the stripped
>>> indentation)
>>>
>>> // Sample Data
>>> inline float4 sample( int index )
>>> {
>>>
>>> float4 sampleData[3] = {
>>>
>>> {0.1f, 0.2f, 0.3f, 0.0f},
>>>
>>> {0.4f, 0.5f, 0.6f, 0.0f},
>>>
>>> {0.7f, 0.8f, 0.9f, 0.0f}
>>>
>>> };
>>>
>>> return (float4)(sampleData[index]);
>>>
>>> };
>>>
>>> kernel testTable : ImageComputationKernel<ePixelWise>
>>> {
>>>
>>> Image<eRead, eAccessPoint, eEdgeClamped> src;
>>>
>>> Image<eWrite> dst;
>>>
>>> //The kernel function is run at every pixel to produce the output.
>>>
>>> void process() {
>>>
>>> float r;
>>>
>>> r = sample(2)[0];
>>>
>>>
>>> // Output the color
>>>
>>> dst() = r;
>>>
>>> }
>>>
>>> };
>>>
>>>
>>>
>>>
>>> However my real data sample is float 3 (x,y,z coordinates) but the code
>>> with float3 errors when I try to compile:
>>>
>>>
>>> // Sample Data
>>> inline float3 sample( int index )
>>> {
>>>
>>> float3 sampleData[3] = {
>>>
>>> {0.1f, 0.2f, 0.3f},
>>>
>>> {0.4f, 0.5f, 0.6f},
>>>
>>> {0.7f, 0.8f, 0.9f}
>>>
>>> };
>>>
>>> return (float3)(sampleData[index]);
>>>
>>> };
>>>
>>> kernel testTable : ImageComputationKernel<ePixelWise>
>>> {
>>>
>>> Image<eRead, eAccessPoint, eEdgeClamped> src;
>>>
>>> Image<eWrite> dst;
>>>
>>> //The kernel function is run at every pixel to produce the output.
>>>
>>> void process() {
>>>
>>> float r;
>>>
>>> r = sample(2)[0];
>>>
>>> // Output the color
>>>
>>> dst() = r;
>>>
>>> }
>>>
>>> };
>>>
>>>
>>>
>>> I tried many different syntaxes, but always run into the same problem,
>>> and I can't seem to find the proper documentation for the language, I'm
>>> looking at C++ and OpenCL docs but some stuff is different.
>>>
>>>
>>> Any pointers?
>>>
>>> Thanks
>>>
>>>
>>>
>>> *Erwan* LEROY
>>> www.erwanleroy.com
>>>
>>>
>>>
>>>
>>>
>>> _______________________________________________
>>>
>>>
>>> Nuke-python mailing list
>>>
>>>
>>> Nuke-python@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
>>>
>>>
>>> http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-python
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>> _______________________________________________
>>>
>>>
>>> Nuke-python mailing list
>>>
>>>
>>> Nuke-python@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
>>>
>>>
>>> http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-python
>>>
>>>
>>>
>>>
>>>
>>> _______________________________________________
>>>
>>> Nuke-python mailing list
>>>
>>> Nuke-python@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
>>>
>>> http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-python
>>>
>>> --
>> Best regards. Mads Hagbarth Lund
>>
>> _______________________________________________
>> Nuke-python mailing list
>> Nuke-python@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
>> http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-python
>>
>>
> _______________________________________________
> Nuke-python mailing list
> Nuke-python@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
> http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-python
>
>
_______________________________________________
Nuke-python mailing list
Nuke-python@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-python
