When uninstalling, GIMP keeps all of your settings. This is very annoying for me, as sometimes, my GIMP has issues. So I either have to reset my PC, or use a complex method. But I understand people who want to keep their settings. So, here's a suggestion: how about an uninstaller that asks you if you want to keep your settings? That way, people can choose if they want their settings kept or not.

If your script is correctly written, in particular its "registration" code, it gets the active image (and the active layer) as parameters:

   

• An important part is the use of the menu named argument. This makes the registration use the new variant. The older variant where the 7th argument is a full path to the menu item (which is still found in some examples) behaves a bit differently.
  
• The "drawable" can be a layer (usual case) but also a layer mask or a channel or even a layer group. Some scripts can work on all types, others may want just a layer. You can use a PF_LAYER instead of a PF_DRAWABLE but the Gimp error message isn't too user-friendly. Better check in your code and react appropriately.
  
• Since the script works on a drawable, the menu item is disabled if there is no active drawable in the image (this can happen after you deleted all layers). You can of course remove the drawable argument if the script doesn't require a specific drawable (runs on all layers, or generates new ones)
  
• Since the script works on an image, the menu item is disabled if there is no image in Gimp.  You can of course remove the image argument if the script doesn't require a specific image (runs on all images, or loads/generates new ones)
  

Hello again, so I have maybe an odd one.

Does anybody know if there are issues with .png's saved with the python pdb.file_png_save()?  In my plugin, any image I "save" using that has issues opening on some viewers, while the same image exported through Gimp's UI using the same settings works perfectly fine.

As a test, I have a simple white image.  If I export it as a .png (16b RGBA) through the UI, works fine everywhere.  But same .xcf saved with this code

Hello all. 

So I have searched and found some info saying this may not be able to be done, but I would really like a way to infer the currently-viewed image in python-fu.

I need this since I have a plugin (think of it more as an extension in other software), that communicates with Gimp from outside.  Mostly is handles loading, saving, and exporting of image files to our pipeline.  Right now I just use:

        currentImage = gimp.image_list()[0]
        currentDrawable = pdb.gimp_image_get_active_layer(currentImage)

which works if there is either one image file open, or want to deal with the last opened.  But would really prefer to be able to have it work on whatever image is "active".  I have read there is nothing in the API that deals with the active, but are there any work-arounds?

Is there something that I am missing?

Thanks in advance,

J.

How do i erase blue pencil lines from a black and white drawing?

Do they stack? For example I am doing a Color Select and it is working perfectly but I do not want it to select the color throughout the entire image. I only want to select the color on a  piece of the image. I tried using Rectangle Select to select the portion of the image that I am interested in and then I tried to use Color Select just in the previously selected area but instead the color was selected throughout the image.

I tried using Select Fuzzy but that still selected too much of the image. There must be a way to have Select Color look for the selected color in just a part of the image.

For example in the image below I want to turn the blocks of green that form a ( and turn them blank. I have gotten SO CLOSE but either a bit too much or a bit too little gets selected.  

   

Here is my best attempt but too much of the blocks remain. 
   
Here too much of the image is removed.
   
It would be nice to have the Select Color tool just look at a small area surrounding the blocks. 
 
Thanks

I have always used an older version of gimp like 2.8 or something like that i just upgraded to 2.10.36. now this is my problem i make clothes for sims and i always used the bump map when doing this in 2.8...I did this by adding a new transparent layer putting the creases in white on that layer and then go to where i want the creases to be on the main colored layer and opening up bump map and clicking on the crease layer so it would bump map it to the colored layer. pretty easy and straight forward now my problem is i go to do that in the new gimp and i get nada nothing ect, i have looked for tutorials that explain how to do this but find none. can anyone help ty in advance

Hello everyone. I need some assistance with getting the Gimp help file to work properly. I tried doing a search for appimage + help file and appimage + user manual and several other variations of this and didn't get any board hits.

I'm using the Gimp AppImage by TasMania17 here:

gimp2-10-36-python2-mm-ubuntu.AppImage

along with Linux Mint 21.3.


Neither the "use the online version" or "use a locally installed copy" of the help system work properly. I would prefer to utilize the "use a locally installed copy" offline option if possible.  I have a .pdf copy of the help file/user manual, which is okay, but it's not integrated into Gimp and is a hassle to pull up each time I need it.

Trying on my own to get things working properly, I installed gimp-help-en 2.10.0-1 from Linux Mint's Software Manager.  I knew it likely would not work, but did manage to get the 57.3 mb  folder of .html files that was installed and saved it in case I could use that in some way.

I also downloaded the gimp-help-2.10.34 .tar file from docs.gimp.org, but really don't know what to do with it.  I've never been too adept with tarball archive files. 

Is it possible to get the integrated help file system to work properly with an appimage?

Hi. I want to make this image transparent: https://imgur.com/a/7z7WtzU

So it looks like this example: https://imgur.com/a/R0I8Z6M

I believe the way to do this is the use color to alpha but it doesn't give the desired result I want.

Anyone know how?

Monash University (the institution I work at as a neuroscientist) decided to no longer provide academic researchers (staff and students) with Adobe Creative cloud (essential things like Illustrator and Adobe DC Pro) because they're "too expensive" for the budget. Staff and students are required to purchase said software out of their own wage. Meanwhile, the Vice Chancellor (on AUD 1.3 mil/year) was given a going away party, which cost  AUD 147,000. 

Alas, I need to re-learn how to do all my skills again on free alternatives (i.e., Inkscape and GIMP).

I need to know how to select a colour channel, add a layer mask to remove the background, and then change the colour of a neural spike trace response. here is what it should look like (see images):