add dls runner

islatu/runner.py

@@ -0,0 +1,198 @@
+from islatu import background
+from islatu import corrections
+from islatu import cropping
+from islatu import image
+from islatu import io
+from islatu import refl_data
+from islatu import stitching
+from yaml import load, CLoader
+import datetime
+from os import path
+from ast import literal_eval as make_tuple
+from uncertainties import ufloat
+import numpy as np
+
+function_map = {'gaussian_1d': background.fit_gaussian_1d,
+                'guassian_2d': background.fit_gaussian_2d,
+                'i07': io.i07_dat_parser,
+                'crop': cropping.crop_2d,
+                'crop_peak': cropping.crop_around_peak_2d,
+                }
+
+class Foreperson:
+    def __init__(self):
+        self.instrument = None
+        self.parser = None
+        self.visit_id = None
+        self.year = datetime.datetime.now().year
+        self.directory_path = None
+        self.crop_function = function_map['crop_peak']
+        self.crop_kwargs = None
+        self.bkg_function = function_map['gaussian_1d']
+        self.bkg_kwargs = None
+        self.dcd_normalisation = None
+        self.sample_size = None
+        self.beam_width = None
+        self.n_qvectors = 50
+        self.q_min = None
+        self.q_max = None
+        self.q_step = None
+        self.q_shape = 'linear'
+        self.n_columns = 3
+        self.theta_axis_name = None
+        self.q_axis_name = "qdcd"
+        self.transpose = False
+        self.qz_dimension = 1
+        self.qxy_dimension = 0
+        self.pixel_max = 1e6
+        self.hot_pixel_max = 1e5
+
+    def ready(self, directory):
+        directory_path = directory.format(self.instrument, self.year, self.visit_id)
+        if path.isdir(directory_path):
+            self.directory_path = directory_path
+        else:
+            raise FileNotFoundError("The experiment directory cannot be found.")
+
+def reduce(run_numbers, yaml_file, directory='/dls/{}/data/{}/{}/'):
+    """
+    The runner that parses the yaml file and performs the data reduction.
+
+    run_numbers (:py:attr:`list` of :py:attr:`int`): Reflectometry scans that
+        make up the profile.
+    yaml_file (:py:attr:`str`): File path to instruction set.
+    """
+    the_boss = Foreperson()
+    y_file = open(yaml_file, 'r')
+    recipe = load(y_file, Loader=CLoader)
+    y_file.close()
+    keys = recipe.keys()
+    if 'instrument' in keys:
+        the_boss.instrument = recipe['instrument']
+        the_boss.parser = function_map[recipe['instrument']]
+    else:
+        raise ValueError("No instrument given in {}.".format(yaml_file))
+    if 'visit' in keys:
+        the_boss.visit_id = recipe['visit']['visit_id']
+        if 'year' in recipe['visit'].keys():
+            the_boss.year = recipe['visit']['year']
+    else:
+        raise ValueError("No visit given in {}.".format(yaml_file))
+    if 'crop' in keys:
+        the_boss.crop_function = function_map[recipe['crop']['method']]
+        if 'kwargs' in recipe['crop']:
+            the_boss.crop_kwargs = recipe['crop']['kwargs']
+    if 'background' in keys:
+        the_boss.bkg_function = function_map[recipe['background']['method']]
+        if 'kwargs' in recipe['background']:
+            the_boss.bkg_kwargs = recipe['background']['kwargs']
+    if 'setup' in keys:
+        if 'dcd_normalisation' in recipe['setup'].keys():
+            the_boss.dcd_normalisation = recipe['setup']['dcd_normalisation']
+        if 'sample_size' in recipe['setup'].keys():
+            the_boss.sample_size = make_tuple(recipe['setup']['sample_size'])
+            try:
+                _ = len(the_boss.sample_size)
+                the_boss.sample_size = ufloat(the_boss.sample_size[0], the_boss.sample_size[1])
+            except TypeError:
+                pass
+        else:
+            raise ValueError("No sample_size given in setup of {}.".format(yaml_file))
+        if 'beam_width' in recipe['setup'].keys():
+            the_boss.beam_width = make_tuple(recipe['setup']['beam_width'])
+            try:
+                _ = len(the_boss.beam_width)
+                the_boss.beam_width = ufloat(the_boss.sample_size[0], the_boss.sample_size[1])
+            except TypeError:
+                pass
+        else:
+            raise ValueError("No beam_width given in setup of {}.".format(yaml_file))
+        if 'theta_axis' in recipe['setup'].keys():
+            the_boss.theta_axis_name = recipe['setup']['theta_axis']
+        if 'q_axis' in recipe['setup'].keys():
+            the_boss.q_axis_name = recipe['setup']['theta_axis']
+        if 'transpose' in recipe['setup'].keys():
+            the_boss.transpose = recipe['setup']['transpose']
+            if the_boss.transpose:
+                the_boss.qz_dimension = 0
+                the_boss.qxy_dimension = 1
+        if 'pixel_max' in recipe['setup'].keys():
+            the_boss.pixel_max = recipe['setup']['pixel_max']
+        if 'hot_pixel_max' in recipe['setup'].keys():
+            the_boss.hot_pixel_max = recipe['setup']['hot_pixel_max']
+    else:
+        raise ValueError("No setup given in {}.".format(yaml_file))
+    if 'rebin' in keys:
+        if 'n_qvectors' in recipe['rebin'].keys():
+            the_boss.n_qvectors = recipe['rebin']['n_qvectors']
+        elif 'min' in recipe['rebin'].keys() and 'max' in recipe['rebin'].keys() and 'step' in recipe['rebin'].keys():
+            the_boss.q_min = recipe['rebin']['min']
+            the_boss.q_max = recipe['rebin']['max']
+            the_boss.q_step = recipe['rebin']['step']
+            if 'shape' in recipe['rebin'].keys():
+                the_boss.q_shape = recipe['rebin']['shape']
+        else:
+            raise ValueError("Please define parameters of rebin in {}.".format(yaml_file))
+    if 'output_columns' in keys:
+        the_boss.output_columns = recipe['output_columns']
+
+    the_boss.ready(directory)
+
+    files_to_reduce = []
+    for f in run_numbers:
+        files_to_reduce.append(the_boss.directory_path + f + '.dat')
+
+
+    print("-" * 10)
+    print('File Parsing')
+    print("-" * 10)
+    refl = refl_data.Profile(files_to_reduce, the_boss.parser, the_boss.q_axis_name,
+                             the_boss.theta_axis_name, None, 0, the_boss.pixel_max,
+                             the_boss.hot_pixel_max, the_boss.transpose)
+    print("-" * 10)
+    print('Cropping')
+    print("-" * 10)
+    refl.crop(the_boss.crop_function, the_boss.crop_kwargs)
+
+    print("-" * 10)
+    print('Background Subtraction')
+    print("-" * 10)
+    refl.bkg_sub(the_boss.bkg_function, the_boss.bkg_kwargs)
+
+    print("-" * 10)
+    print('Estimating Resolution Function')
+    print("-" * 10)
+    refl.resolution_function(the_boss.qz_dimension, progress=True)
+
+    print("-" * 10)
+    print('Performing Data Corrections')
+    print("-" * 10)
+    if the_boss.dcd_normalisation is not None:
+        itp = corrections.get_interpolator(the_boss.dcd_normalisation, the_boss.parser)
+        refl.qdcd_normalisation(itp)
+    refl.footprint_correction(the_boss.beam_width, the_boss.sample_size)
+    refl.transmission_normalisation()
+    refl.concatenate()
+    refl.normalise_ter()
+
+    print("-" * 10)
+    print('Rebinning')
+    print("-" * 10)
+    if the_boss.q_min is None:
+        refl.rebin(number_of_q_vectors=the_boss.n_qvectors)
+    else:
+        if the_boss.q_space is 'linear':
+            spacing = np.linspace
+        elif the_boss.q_space is 'log':
+            spacing = np.logspace
+        refl.rebin(new_q=spacing(the_boss.q_min, the_boss.q_max, the_boss.q_step))
+
+    if the_boss.output_columns == 3:
+        data = np.array([refl.q, refl.R, refl.dR]).T
+        np.savetxt(the_boss.directory_path + '/processing/' + run_numbers[0] + '.dat', data, header='q R dR \n{}'.format(recipe))
+    elif the_boss.output_columns == 4:
+        data = np.array([refl.q, refl.R, refl.dR, refl.dq]).T
+        np.savetxt(the_boss.directory_path + '/processing/' + run_numbers[0] + '.dat', data, header='q R dR dq \n{}'.format(recipe))
+    print("-" * 10)
+    print('Reduced Data Stored in Processing Directory')
+    print("-" * 10)