import sys

from ij import IJ, ImagePlus
from ij.plugin.filter import Analyzer, MaximumFinder
from ij.process import ImageProcessor

# Fiji Jython interpreter implements Python 2.5 which does not
# provide support for argparse.
error_log = sys.argv[-10]
input_file = sys.argv[-9]
scale_when_converting = sys.argv[-8] == 'yes'
weighted_rgb_conversions = sys.argv[-7] == 'yes'
noise_tolerance = int(sys.argv[-6])
output_type = sys.argv[-5]
exclude_edge_maxima = sys.argv[-4] == 'yes'
light_background = sys.argv[-3]
tmp_output_path = sys.argv[-2]
output_datatype = sys.argv[-1]

# Open the input image file.
input_image_plus = IJ.openImage(input_file)

# Create a copy of the image.
input_image_plus_copy = input_image_plus.duplicate()
image_processor_copy = input_image_plus_copy.getProcessor()
bit_depth = image_processor_copy.getBitDepth()
analyzer = Analyzer(input_image_plus_copy)

# Set the conversion options.
options = []
# The following 2 options are applicable only to RGB images.
if bit_depth == 24:
    if scale_when_converting:
        options.append("scale")
    if weighted_rgb_conversions:
        options.append("weighted")
# Perform conversion - must happen even if no options are set.
IJ.run(input_image_plus_copy, "Conversions...", "%s" % " ".join(options))
if output_type in ['List', 'Count']:
    # W're  generating a tabular file for the output.
    # Set the Find Maxima options.
    options = ['noise=%d' % noise_tolerance]
    if output_type.find('_') > 0:
        output_type_str = 'output=[%s]' % output_type.replace('_', ' ')
    else:
        output_type_str = 'output=%s' % output_type
    options.append(output_type_str)
    if exclude_edge_maxima:
        options.append('exclude')
    if light_background:
        options.append('light')
    # Run the command.
    IJ.run(input_image_plus_copy, "Find Maxima...", "%s" % " ".join(options))
    results_table = analyzer.getResultsTable()
    results_table.saveAs(tmp_output_path)
else:
    # Find the maxima of an image (does not find minima).
    # LIMITATIONS: With output_type=Segmented_Particles
    # (watershed segmentation), some segmentation lines
    # may be improperly placed if local maxima are suppressed
    # by the tolerance.
    mf = MaximumFinder()
    if output_type == 'Single_Points':
        output_type_param = mf.SINGLE_POINTS
    elif output_type == 'Maxima_Within_Tolerance':
        output_type_param = mf.IN_TOLERANCE
    elif output_type == 'Segmented_Particles':
        output_type_param = mf.SEGMENTED
    elif output_type == 'List':
        output_type_param = mf.LIST
    elif output_type == 'Count':
        output_type_param = mf.COUNT
    # Get a new byteProcessor with a normal (uninverted) LUT where
    # the marked points are set to 255 (Background 0). Pixels outside
    # of the roi of the input image_processor_copy are not set. No
    # output image is created for output types POINT_SELECTION, LIST
    # and COUNT.  In these cases findMaxima returns null.
    byte_processor = mf.findMaxima(image_processor_copy,
                                   noise_tolerance,
                                   ImageProcessor.NO_THRESHOLD,
                                   output_type_param,
                                   exclude_edge_maxima,
                                   False)
    # Invert the image or ROI.
    byte_processor.invert()
    if output_type == 'Segmented_Particles' and not light_background:
        # Invert the values in this image's LUT (indexed color model).
        byte_processor.invertLut()
    image_plus = ImagePlus("output", byte_processor)
    IJ.saveAs(image_plus, output_datatype, tmp_output_path)