removed unused imports

datacube/api/utils.py

@@ -31,8 +31,6 @@
 from datacube.api.model import Pq25Bands, Ls57Arg25Bands, Satellite, DatasetType, Ls8Arg25Bands, Wofs25Bands, NdviBands, \
     NdwiBands, MndwiBands
 from datacube.api.model import get_bands, EviBands, NbrBands, TciBands
-from scipy.ndimage import map_coordinates
-from eotools.coordinates import convert_coordinates
 from datetime import datetime, date
 
 
@@ -1756,98 +1754,6 @@ def combine_means(mean1, count1, mean2, count2):
     return mean1 * count1 / (count1 + count2) + mean2 * count2 / (count1 + count2)
 
 
-def arbitrary_profile(dataset, xy_points, band=None, cubic=False,
-                      from_map=False):
-    """
-    Get the data associated with an arbitrary set of points that
-    define an arbitrary profile/transect, and the pixel locations
-    associated with the transect.
-
-    :param dataset:
-        An instance of a `DatasetType`.
-
-    :param xy_points:
-        A list of (x, y) co-ordinate paris eg [(x, y), (x, y), (x, y)].
-
-    :param band:
-        The raster band to read from. Default is to read the first band
-        in the list returned by `dataset.bands`.
-
-    :param from_map:
-        A boolean indicating whether or not the input co-ordinates
-        are real world map coordinates. If set to True, then the input
-        xy co-ordinate will be converted to image co-ordinates.
-
-    :return:
-        A 1D NumPy array of lenght determined by the distance between
-        the xy_points; A tuple (y_index, x_index) containing the
-        pixel locations of the transect; and a tuple containing a list
-        of start and end co-ordinates for both x and y fields.
-        The form of the start and end locations is:
-            ([(xstart_1, xend_1),...,(xstart_n-1, xend_n-1)],
-             [(ystart_1, yend_1),...,(ystart_n-1, yend_n-1)]).
-        This form can be directly used in a call to plot() as follows:
-            prf, idx, xy_start_end = arbitrary_profile()
-            plot(xy_start_end[0], xy_start_end[1], 'r-')
-
-    :history:
-        * 2015-07-07 - Base functionality taken from:
-        https://github.com/GeoscienceAustralia/eo-tools/blob/stable/eotools/profiles.py
-        and converted to work directly with the agdc-api.
-    """
-    n_points = len(xy_points)
-    if n_points < 2:
-        msg = "Minimum number of points is 2, received {}".format(n_points)
-        raise ValueError(msg)
-
-    metadata = get_dataset_metadata(dataset)
-    geotransform = metadata.transform
-
-    # Convert to image co-ordinates if needed
-    if from_map:
-        img_xy = convert_coordinates(geotransform, xy_points, to_map=False)
-    else:
-        img_xy = xy_points
-
-    # Read the image band
-    if band is None:
-        band = [bnd for bnd in dataset.bands][0]
-    img = read_dataset_data(dataset, [band])[band]
-
-    # Initialise the arrays to hold the profile
-    profile = numpy.array([], dtype=img.dtype)
-    x_idx = numpy.array([], dtype='int')
-    y_idx = numpy.array([], dtype='int')
-    x_start_end = []
-    y_start_end = []
-
-    # Build up the profile by creating a line segment between each point
-    for i in range(1, n_points):
-        x0, y0 = img_xy[i - 1]
-        x1, y1 = img_xy[i]
-        x_start_end.append((x0, x1))
-        y_start_end.append((y0, y1))
-
-        n_pixels = max(abs(x1 - x0 + 1), abs(y1 - y0 + 1))
-        x = numpy.linspace(x0, x1, n_pixels)
-        y = numpy.linspace(y0, y1, n_pixels)
-        x_idx = numpy.append(x_idx, x)
-        y_idx = numpy.append(y_idx, y)
-
-        # How to interpolate, cubic or nearest neighbour
-        if cubic:
-            transect = map_coordinates(img, (y, x))
-            profile = numpy.append(profile, transect)
-        else:
-            transect = img[y.astype('int'), x.astype('int')]
-            profile = numpy.append(profile, transect)
-
-    x_idx = x_idx.astype('int')
-    y_idx = y_idx.astype('int')
-
-    return (profile, (y_idx, x_idx), (x_start_end, y_start_end))
-
-
 def x_calculate_stack_statistic_count(stack, ndv=NDV, dtype=numpy.int16):
 
     stack = maskify_stack(stack=stack, ndv=ndv)