percurnicus / pystamps Goto Github PK

In general I dislike how pystamps method works. I think it should take a list of files, directories or globs as its primary argument. It should loop over the elements of the list, test to see if the element is a directory, if so use it, if its a file, add it to the list of products and then last try and use it as a glob, add any products from that glob to the list of products. I think this argument parser should actually be extracted into a function.

def pystamps(args=None):
    """Run pystamps from python shell or command line with arguments"""
    try:
        if len(args.file) > 1:
            files = args.file
        elif len(args.file) == 1:
            if os.path.isdir(args.file[0]):
                files = glob(os.path.join(str(args.file[0]), '*'))
            elif os.path.isfile(args.file[0]):
                files = glob(str(args.file[0]))
        else:
            files = glob('*')
    except AttributeError:
        if os.path.isdir(args):
            files = glob(os.path.join('%s' % (args), '*'))
        elif args:
            files = glob(args)
        else:
            files = glob('*')

    image_set = ImageSet(files)
    display = MainWindow(image_set.images)
    sys.exit(app.exec_())

A scale of .1 is too small. Try finding a scale that looks good for both and make the scale depend on the screen size. See how scaling works on mer_stamps

See if you can connect using a lambda function like so:

self.next_channel = QtGui.QPushButton("Next")
self.next_channel.clicked.connect(lambda: self.switch_channel(1))

Make a pop-up window from the toolbar to modify quality image flag. Use double line color over a white border to signify the quality flag and that it is selected

See if this works in converting the numpy array to a qimage:
http://flopalm.com/how-to-convert-python-opencv-image-numpy-array-to-pyqt-qpixmap-image/

ImageSet should only take a list of images/filenames. Perform all the globbing/gathering of files in pystamps. This should allow for both command line arguments and python shell usage.

This might fix the lines that appear on the images when scrolling and eliminate the need to update the images when scrolling. Probably will have to disable scroll bars for each image as well during for loop.