For Help about Photoshop SDK!

[rouying76]

I want create a plug-in of Photoshop by the Photoshop's SDK.

This plugin can process the pixel of a opened file, then create a new file, and put the processed pixel into this new file, and save it.

Can I do it with the Photoshop's SDK?

What types plugin can do it?

Only Filter Plug-ins can do it?

Need others Plug-ins?

Where can I find any same sample?

Thanks!

[Chris Cox]

Have you read the documentation that comes with the SDK?

[rouying76]

Read a little.

But I am not good at English.

So read it very very slowly.

In fact, I want to know above questions.

If the SDK can not do that.

I think I will give up learn it.

[Chris Cox]

I'm not even exactly sure what you are asking, probably because something was lost in translation.

You need to read the Photoshop SDK and understand what the different plugin types can do, then you can answer your own question (I hope).

[rouying76]

Thanks！

I will try my best!

[Tom Ruark]

Build the automationfilter (automation plug-in type) and the hidden (filter plug-in type) example. This is a two plug-in example doing work on altering pixels on different layers. The automation plug-in has a dialog that lets you switch layers of the current document and uses a hidden filter plug-in to actually change the pixels on that layer. Photoshop is set up to only allow altering pixels on the front most (active) document and the front most (active) layer of that document. You have to use an automation plug-in to set up the target first. Then you call the filter to process the pixels.

For your case you would have to hold onto the pixels of the first document and then apply them to the second document. Your memory use is going to be very high when operating on large file sets. You might want to duplicate the first image and then run your filter operation on the duplicate.

To answer your question:

You would have to alter the automationfilter to deal with documents instead of layers. You would have to alter the hidden filter to give you back a buffer full of the pixels of the document. The automationfilter would then create a new document and call back to the hidden filter with your pixel buffer and apply that to the new document. Again, I would go with the "duplicate" approach above but you might be leaving out some steps that you want to do and my "duplicate" option will not work for you.

FYI, I had the hidden example doing "give me all the pixels" for the automation plug-in at one point. I thought it would be better to have all the pixel operations inside the filter plug-in type and made the hidden plug-in a command driven filter so that I did not have to copy pixels around (slow). But that example exists in older versions of the SDK. I don't remember which version exactly.