Load images into list in a specific order-CodePudding

I have a folder with images that follow this name scheme:

FEATURED_0.png

FEATURED_1.png

FEATURED_2.png

When I want to load them into a list usable by Pillow I do this:

for filename in glob.glob("FEATURED*.png"):
    image = Image.open(filename)
    featured_list.append(image)

This works pretty ok, but has a flaw. It loads the images, regardless of their number. I already tried loading FEATURED_0 regardless of what, moving into a list and then checking if the next Image has a greater number 1, but that failed miserably.

So back to my question. Is there a way to load Images into a list in python, but in a specific order?

CodePudding user response：

If you want to start with 0, load each subsequent number, and stop when you hit a number that's not there, use a while loop:

num = 0
base = "FEATURED_%d.png"
while os.path.isfile(base % num):
    featured_list.append(Image.open(base & num))
    num  = 1

CodePudding user response：

glob doesn't have option to sort filenames. os.list() and os.walk() also can't sort it.

You have to use list = sorted(list) for this

for filename in sorted(glob.glob("FEATURED*.png")):

But sorted() without parameters will put _10 before _2 because it checks char after char and first it compares _ with _ and next 1 with 2 and skip rest.

You may need to use sorted(..., key=your_function) with your_function which extracts number from filename and converts to integer - and sorted will use only this integer to compare names.

def number(filename):
    return int(filename[9:-4])

data = ['FEATURED_0.png', 'FEATURED_1.png', 'FEATURED_2.png', 'FEATURED_10.png']

print(list(sorted(data)))              # `_10` before `_2`
print(list(sorted(data, key=number)))  # `_2` before `_10`

Result

# `_10` before `_2`
['FEATURED_0.png', 'FEATURED_1.png', 'FEATURED_10.png', 'FEATURED_2.png']

# `_2` before `_10`
['FEATURED_0.png', 'FEATURED_1.png', 'FEATURED_2.png', 'FEATURED_10.png']

def number(filename):
    return int(filename[9:-4])

for filename in sorted(glob.glob("FEATURED*.png"), key=number):
    # ... code ...

You may also write it with lambda

sorted( data, key=(lambda x:int(x[9:-4])) )

You can create function which returns something more complex - ie. tuple (extension, number) - and then first you get all .jpg (sorted by number) and next all .png (sorted by number)