Interesting, your example works, still try mine!
Try to upload an archive with input png and result webp!
Archiv.zip (354.9 KB)
It works as well:
$ gmic test1.webp o foo.webp
[gmic]-0./ Start G'MIC interpreter.
[gmic]-0./ Input file 'test1.webp' at position 0 (1 image 847x1085x1x4).
[gmic]-1./ Output image [0] as webp file 'foo.webp' (1 image 847x1085x1x4).
[gmic]-1./ End G'MIC interpreter.
$ gmic foo.webp
[gmic]-0./ Start G'MIC interpreter.
[gmic]-0./ Input file 'foo.webp' at position 0 (1 image 847x1085x1x4).
[gmic]-1./ Display image [0] = 'foo.webp'.
[0] = 'foo.webp':
size = (847,1085,1,4) [14 Mio of floats].
data = (0,0,0,0,0,0,0,0,0,0,0,0,(...),0,0,0,0,0,0,0,0,0,0,0,0).
min = 0, max = 255, mean = 96.4888, std = 90.9434, coords_min = (0,0,0,0), coords_max = (710,89,0,0).
[gmic]-1./ End G'MIC interpreter.
Hmm, test1.webp is the result of reading rgba.png. Please try that too, reading rgba.png and writing test1.webp!
Karsten, the rgba.png file in your archive is the one I’ve posted above. How could test1.webp (a bird) could be the conversion of rgba.png?
Uuh, I think I should not do such things on Saturday afternoons! Here is the png input
Eisvogel_1.png.zip (1.0 MB)
Looks like it works for me as well:
$ gmic Eisvogel_1.png o foo.webp
[gmic]-0./ Start G'MIC interpreter.
[gmic]-0./ Input file 'Eisvogel_1.png' at position 0 (1 image 847x1085x1x4).
[gmic]-1./ Output image [0] as webp file 'foo.webp' (1 image 847x1085x1x4).
[gmic]-1./ End G'MIC interpreter.
$ gmic foo.webp
[gmic]-0./ Start G'MIC interpreter.
[gmic]-0./ Input file 'foo.webp' at position 0 (1 image 847x1085x1x4).
[gmic]-1./ Display image [0] = 'foo.webp'.
[0] = 'foo.webp':
size = (847,1085,1,4) [14 Mio of floats].
data = (0,0,0,0,0,0,0,0,0,0,0,0,(...),0,0,0,0,0,0,0,0,0,0,0,0).
min = 0, max = 255, mean = 96.4888, std = 90.9434, coords_min = (0,0,0,0), coords_max = (710,89,0,0).
[gmic]-1./ End G'MIC interpreter.
I have the same values mean and std, still here is my screenshot
Makes me a bit crazy, such things.Another X11 thing?
Yes, that’s normal, what you see in just the 3 first channels (R,G,B) of the images.
Use drgba to render it over a checkerboard.
I get the same thing as you, it’s perfectly OK.
1 Like
Puh, thank you. I did not know that webp takes zero transparency into account!
I tried it with “display” (imagemagck) as well as other displays. It was all ok with transparency! webp is really no format to reread in gmic.
If I override an existing command using user.gmic, it works as expected (fixed 
), except for help where it grabs the old output. I am using Version 2.9.8 (pre-release #210409).
Are you talking about the command execution, or the command documentation?
Alter norm and place in user.gmic.
#@cli norm
#@cli : hello
#@cli : $ image.jpg +norm
#@cli : $$
norm :
e hello
>gmic norm
[gmic]-0./ Start G'MIC interpreter.
[gmic]-0./norm/ hello
[gmic]-0./ End G'MIC interpreter.
>gmic h norm
norm:
Compute the pointwise euclidean norm of vector-valued pixels in selected
images.
Example:
[#1] image.jpg +norm
Tutorial: https://gmic.eu/oldtutorial/_norm.shtml
Well no, it works for me:
$ gmic h norm
norm:
hello
Example:
[#1] image.jpg +norm
Tutorial: https://gmic.eu/oldtutorial/_norm.shtml
Maybe this is a Windows-specific issue. Will try on my Windows VM later.
Commit https://github.com/dtschump/gmic/commit/63833ef87843f8dcf104aa0c412fccd7c764f0ba should fix that issue. Please $ gmic update and try again 
All good.
1 Like
Bad output from median_color.
>gmic sp tiger e ${median_color.} q
[gmic]-0./ Start G'MIC interpreter.
[gmic]-1./ Input sample image 'tiger' (1 image 750x500x1x3).
[gmic]-1./ 750,500,1,1,750,500,1,1,750,500,1,1
[gmic]-1./ Quit G'MIC interpreter.
Formatting bug (G'MIC - GREYC's Magic for Image Computing: A Full-Featured Open-Source Framework for Image Processing - Substitution Rules).
1 Like
Fixed, and fixed.
2 Likes
# image size = (1,16777216,1,3) [192 Mio of floats].
# ...
[gmic]-1./ Output image [0] as gmz file '_.gmz', with pixel type 'auto'.
[gmic]-1./ End G'MIC interpreter.
>gmic _.gmz
[gmic]-0./ Start G'MIC interpreter.
[gmic]-0./ Input file '_.gmz' at position 0
[gmic]-0./ *** Error *** Command 'input': File '_.gmz' is not in .gmz format (numbers of images and names do not match).
Works for me 
If it helps, the long image comes from extract_region. I tried again. Now it refuses to do norm after and doesn’t go all the way to saving the gmz.