ChessAnalysisPipeline 0.0.14__py3-none-any.whl → 0.0.16__py3-none-any.whl

CHAP/edd/reader.py

@@ -1,5 +1,894 @@
 #!/usr/bin/env python
 
+# System modules
+import os
+
+# Third party modules
+import numpy as np
+
+# Local modules
+from CHAP.reader import Reader
+
+class EddMapReader(Reader):
+    """Reader for taking an EDD-style .par file and returning a
+    `MapConfig` representing one of the datasets in the
+    file. Independent dimensions are determined automatically, and a
+    specific set of items to use for extra scalar datasets to include
+    are hard-coded in. The raw data is read if detector_names are
+    specified."""
+    def read(self, parfile, dataset_id, detector_names=None):
+        """Return a validated `MapConfig` object representing an EDD
+        dataset.
+
+        :param parfile: Name of the EDD-style .par file containing the
+            dataset.
+        :type parfile: str
+        :param dataset_id: Number of the dataset in the .par file
+            to return as a map.
+        :type dataset_id: int
+        :param detector_names: Detector prefixes for the raw data.
+        :type detector_names: list[str], optional
+        :returns: Map configuration packaged with the appropriate
+            value for 'schema'.
+        :rtype: PipelineData
+        """
+        # Local modules
+        from CHAP.common.models.map import MapConfig
+        from CHAP.pipeline import PipelineData
+        from CHAP.utils.general import list_to_string
+        from CHAP.utils.parfile import ParFile
+        from chess_scanparsers import SMBMCAScanParser as ScanParser
+
+        if detector_names is not None:
+            assert is_str_series(detector_names, raise_error=True)
+
+        parfile = ParFile(parfile)
+        self.logger.debug(f'spec_file: {parfile.spec_file}')
+
+        # Get list of scan numbers for the dataset
+        dataset_ids = np.asarray(parfile.get_values('dataset_id'))
+        dataset_rows_i = np.argwhere(
+            np.where(
+                np.asarray(dataset_ids) == dataset_id, 1, 0)).flatten()
+        scan_nos = [parfile.data[i][parfile.scann_i] for i in dataset_rows_i\
+                    if parfile.data[i][parfile.scann_i] in \
+                        parfile.good_scan_numbers()]
+        self.logger.debug(f'Scan numbers: {list_to_string(scan_nos)}')
+        spec_scans = [dict(spec_file=parfile.spec_file, scan_numbers=scan_nos)]
+
+        # Get scan type for this dataset
+        scan_types = parfile.get_values('scan_type', scan_numbers=scan_nos)
+        if any([st != scan_types[0] for st in scan_types]):
+            msg = 'Only one scan type per dataset is suported.'
+            self.logger.error(msg)
+            raise Exception(msg)
+        scan_type = scan_types[0]
+        self.logger.debug(f'Scan type: {scan_type}')
+
+        # Based on scan type, get independent_dimensions for the map
+        # Start by adding labx, laby, labz, and omega. Any "extra"
+        # dimensions will be sqeezed out of the map later.
+        independent_dimensions = [
+            dict(label='labx', units='mm', data_type='smb_par',
+                 name='labx'),
+            dict(label='laby', units='mm', data_type='smb_par',
+                 name='laby'),
+            dict(label='labz', units='mm', data_type='smb_par',
+                 name='labz'),
+            dict(label='ometotal', units='degrees', data_type='smb_par',
+                 name='ometotal')
+        ]
+        scalar_data = []
+        attrs = {}
+        if scan_type != 0:
+            self.logger.warning(
+                'Assuming all fly axes parameters are identical for all scans')
+            attrs['fly_axis_labels'] = []
+            axes_labels = {1: 'fly_labx', 2: 'fly_laby', 3: 'fly_labz',
+                           4: 'fly_ometotal'}
+            axes_units = {1: 'mm', 2: 'mm', 3: 'mm', 4: 'degrees'}
+            axes_added = []
+            scanparser = ScanParser(parfile.spec_file, scan_nos[0])
+            def add_fly_axis(fly_axis_index):
+                if fly_axis_index in axes_added:
+                    return
+                fly_axis_key = scanparser.pars[f'fly_axis{fly_axis_index}']
+                independent_dimensions.append(dict(
+                    label=axes_labels[fly_axis_key],
+                    data_type='spec_motor',
+                    units=axes_units[fly_axis_key],
+                    name=scanparser.spec_scan_motor_mnes[fly_axis_index]))
+                axes_added.append(fly_axis_index)
+                attrs['fly_axis_labels'].append(axes_labels[fly_axis_key])
+            add_fly_axis(0)
+            if scan_type in (2, 3, 5):
+                add_fly_axis(1)
+            if scan_type == 5:
+                scalar_data.append(dict(
+                    label='bin_axis', units='n/a', data_type='smb_par',
+                    name='bin_axis'))
+                attrs['bin_axis_label'] = axes_labels[
+                    scanparser.pars['bin_axis']].replace('fly_', '')
+
+        # Add in the usual extra scalar data maps for EDD
+        scalar_data.extend([
+            dict(label='SCAN_N', units='n/a', data_type='smb_par',
+                 name='SCAN_N'),
+            dict(label='rsgap_size', units='mm', data_type='smb_par',
+                 name='rsgap_size'),
+            dict(label='x_effective', units='mm', data_type='smb_par',
+                 name='x_effective'),
+            dict(label='z_effective', units='mm', data_type='smb_par',
+                 name='z_effective'),
+        ])
+
+        # Construct initial map config dictionary
+        scanparser = ScanParser(parfile.spec_file, scan_nos[0])
+        map_config_dict = dict(
+            title=f'{scanparser.scan_name}_dataset{dataset_id}',
+            station='id1a3',
+            experiment_type='EDD',
+            sample=dict(name=scanparser.scan_name),
+            spec_scans=[
+                dict(spec_file=parfile.spec_file, scan_numbers=scan_nos)],
+            independent_dimensions=independent_dimensions,
+            scalar_data=scalar_data,
+            presample_intensity=dict(name='a3ic1', data_type='scan_column'),
+            postsample_intensity=dict(name='diode', data_type='scan_column'),
+            dwell_time_actual=dict(name='sec', data_type='scan_column'),
+            attrs=attrs,
+        )
+        map_config = MapConfig(**map_config_dict)
+
+        # Add lab coordinates to the map's scalar_data only if they
+        # are NOT already one of the sqeezed map's
+        # independent_dimensions.
+        lab_dims = [
+            dict(label='labx', units='mm', data_type='smb_par', name='labx'),
+            dict(label='laby', units='mm', data_type='smb_par', name='laby'),
+            dict(label='labz', units='mm', data_type='smb_par', name='labz'),
+            dict(label='ometotal', units='degrees', data_type='smb_par',
+                 name='ometotal')
+        ]
+        for dim in lab_dims:
+            if dim not in independent_dimensions:
+                scalar_data.append(dim)
+
+        # Convert list of scan_numbers to string notation
+        scan_numbers = map_config_dict['spec_scans'][0]['scan_numbers']
+        map_config_dict['spec_scans'][0]['scan_numbers'] = list_to_string(
+            scan_numbers)
+
+        # For now overrule the map type to be always unstructured
+        # Later take out the option of structured entirely from
+        # MapConfig
+        map_config_dict['map_type'] = 'unstructured'
+
+        return map_config_dict
+
+
+class EddMPIMapReader(Reader):
+    """Reader for taking an EDD-style .par file and returning a
+    representing one of the datasets in the file as a NeXus NXentry
+    object. Independent dimensions are determined automatically, and a
+    specific set of items to use for extra scalar datasets to include
+    are hard-coded in."""
+    def read(self, parfile, dataset_id, detector_names):
+        """Return a NeXus NXentry object after validating the
+        `MapConfig` object representing an EDD dataset.
+
+        :param parfile: Name of the EDD-style .par file containing the
+            dataset.
+        :type parfile: str
+        :param dataset_id: Number of the dataset in the .par file
+            to return as a map.
+        :type dataset_id: int
+        :param detector_names: Detector prefixes for the raw data.
+        :type detector_names: list[str]
+        :returns: The EDD map including the raw data packaged with the
+            appropriate value for 'schema'.
+        :rtype: PipelineData
+        """
+        # Third party modules
+        from json import dumps
+        from nexusformat.nexus import (
+            NXcollection,
+            NXdata,
+            NXentry,
+            NXfield,
+            NXsample,
+        )
+
+        # Local modules
+        from CHAP.common.models.map import MapConfig
+        from CHAP.pipeline import PipelineData
+        from CHAP.utils.general import list_to_string
+        from CHAP.utils.parfile import ParFile
+        from chess_scanparsers import SMBMCAScanParser as ScanParser
+
+        assert is_str_series(detector_names, raise_error=True)
+
+        parfile = ParFile(parfile)
+        self.logger.debug(f'spec_file: {parfile.spec_file}')
+
+        # Get list of scan numbers for the dataset
+        dataset_ids = np.asarray(parfile.get_values('dataset_id'))
+        dataset_rows_i = np.argwhere(
+            np.where(
+                np.asarray(dataset_ids) == dataset_id, 1, 0)).flatten()
+        scan_nos = [parfile.data[i][parfile.scann_i] for i in dataset_rows_i\
+                    if parfile.data[i][parfile.scann_i] in \
+                        parfile.good_scan_numbers()]
+        self.logger.debug(f'Scan numbers: {scan_nos}')
+        spec_scans = [dict(spec_file=parfile.spec_file, scan_numbers=scan_nos)]
+
+        # Get scan type for this dataset
+        scan_types = parfile.get_values('scan_type', scan_numbers=scan_nos)
+        if any([st != scan_types[0] for st in scan_types]):
+            msg = 'Only one scan type per dataset is suported.'
+            self.logger.error(msg)
+            raise Exception(msg)
+        scan_type = scan_types[0]
+        self.logger.debug(f'Scan type: {scan_type}')
+
+        # Based on scan type, get independent_dimensions for the map
+        # Start by adding labx, laby, labz, and omega. Any "extra"
+        # dimensions will be sqeezed out of the map later.
+        independent_dimensions = [
+            dict(label='labx', units='mm', data_type='smb_par',
+                 name='labx'),
+            dict(label='laby', units='mm', data_type='smb_par',
+                 name='laby'),
+            dict(label='labz', units='mm', data_type='smb_par',
+                 name='labz'),
+            dict(label='ometotal', units='degrees', data_type='smb_par',
+                 name='ometotal')
+        ]
+        scalar_data = []
+        attrs = {}
+        if scan_type != 0:
+            self.logger.warning(
+                'Assuming all fly axes parameters are identical for all scans')
+            attrs['fly_axis_labels'] = []
+            axes_labels = {1: 'fly_labx', 2: 'fly_laby', 3: 'fly_labz',
+                           4: 'fly_ometotal'}
+            axes_units = {1: 'mm', 2: 'mm', 3: 'mm', 4: 'degrees'}
+            axes_added = []
+            scanparser = ScanParser(parfile.spec_file, scan_nos[0])
+            def add_fly_axis(fly_axis_index):
+                if fly_axis_index in axes_added:
+                    return
+                fly_axis_key = scanparser.pars[f'fly_axis{fly_axis_index}']
+                independent_dimensions.append(dict(
+                    label=axes_labels[fly_axis_key],
+                    data_type='spec_motor',
+                    units=axes_units[fly_axis_key],
+                    name=scanparser.spec_scan_motor_mnes[fly_axis_index]))
+                axes_added.append(fly_axis_index)
+                attrs['fly_axis_labels'].append(axes_labels[fly_axis_key])
+            add_fly_axis(0)
+            if scan_type in (2, 3, 5):
+                add_fly_axis(1)
+            if scan_type == 5:
+                scalar_data.append(dict(
+                    label='bin_axis', units='n/a', data_type='smb_par',
+                    name='bin_axis'))
+                attrs['bin_axis_label'] = axes_labels[
+                    scanparser.pars['bin_axis']].replace('fly_', '')
+
+        # Add in the usual extra scalar data maps for EDD
+        scalar_data.extend([
+            dict(label='SCAN_N', units='n/a', data_type='smb_par',
+                 name='SCAN_N'),
+            dict(label='rsgap_size', units='mm', data_type='smb_par',
+                 name='rsgap_size'),
+            dict(label='x_effective', units='mm', data_type='smb_par',
+                 name='x_effective'),
+            dict(label='z_effective', units='mm', data_type='smb_par',
+                 name='z_effective'),
+        ])
+
+        # Construct initial map config dictionary
+        scanparser = ScanParser(parfile.spec_file, scan_nos[0])
+        map_config_dict = dict(
+            title=f'{scanparser.scan_name}_dataset{dataset_id}',
+            station='id1a3',
+            experiment_type='EDD',
+            sample=dict(name=scanparser.scan_name),
+            spec_scans=[
+                dict(spec_file=parfile.spec_file, scan_numbers=scan_nos)],
+            independent_dimensions=independent_dimensions,
+            scalar_data=scalar_data,
+            presample_intensity=dict(name='a3ic1', data_type='scan_column'),
+            postsample_intensity=dict(name='diode', data_type='scan_column'),
+            dwell_time_actual=dict(name='sec', data_type='scan_column'),
+            attrs=attrs
+        )
+        map_config = MapConfig(**map_config_dict)
+
+        # Squeeze out extraneous independent dimensions (dimensions
+        # along which data were taken at only one unique coordinate
+        # value)
+        while 1 in map_config.shape:
+            remove_dim_index = map_config.shape.index(1)
+            self.logger.debug(
+                'Map dimensions: '
+                + str([dim["label"] for dim in independent_dimensions]))
+            self.logger.debug(f'Map shape: {map_config.shape}')
+            self.logger.debug(
+                'Sqeezing out independent dimension '
+                + independent_dimensions[remove_dim_index]['label'])
+            independent_dimensions.pop(remove_dim_index)
+            map_config = MapConfig(**map_config_dict)
+        self.logger.debug(
+            'Map dimensions: '
+            + str([dim["label"] for dim in independent_dimensions]))
+        self.logger.debug(f'Map shape: {map_config.shape}')
+
+        # Add lab coordinates to the map's scalar_data only if they
+        # are NOT already one of the sqeezed map's
+        # independent_dimensions.
+        lab_dims = [
+            dict(label='labx', units='mm', data_type='smb_par', name='labx'),
+            dict(label='laby', units='mm', data_type='smb_par', name='laby'),
+            dict(label='labz', units='mm', data_type='smb_par', name='labz'),
+            dict(label='ometotal', units='degrees', data_type='smb_par',
+                 name='ometotal')
+        ]
+        for dim in lab_dims:
+            if dim not in independent_dimensions:
+                scalar_data.append(dim)
+
+        # Set up NXentry and add misc. CHESS-specific metadata
+        nxentry = NXentry(name=map_config.title)
+        nxentry.attrs['station'] = map_config.station
+        nxentry.map_config = dumps(map_config.dict())
+        nxentry.spec_scans = NXcollection()
+        for scans in map_config.spec_scans:
+            nxentry.spec_scans[scans.scanparsers[0].scan_name] = \
+                NXfield(value=scans.scan_numbers,
+                        attrs={'spec_file': str(scans.spec_file)})
+
+        # Add sample metadata
+        nxentry[map_config.sample.name] = NXsample(
+            **map_config.sample.dict())
+
+        # Set up default data group
+        nxentry.data = NXdata()
+        independent_dimensions = map_config.independent_dimensions
+        for dim in independent_dimensions:
+            nxentry.data[dim.label] = NXfield(
+                units=dim.units,
+                attrs={'long_name': f'{dim.label} ({dim.units})',
+                       'data_type': dim.data_type,
+                       'local_name': dim.name})
+
+        # Read the raw data and independent dimensions
+        data = [[] for _ in detector_names]
+        dims = [[] for _ in independent_dimensions]
+        for scans in map_config.spec_scans:
+            for scan_number in scans.scan_numbers:
+                scanparser = scans.get_scanparser(scan_number)
+                for i, detector_name in enumerate(detector_names):
+                    if isinstance(detector_name, int):
+                        detector_name = str(detector_name)
+                    ddata = scanparser.get_detector_data(detector_name)
+                    data[i].append(ddata)
+                for i, dim in enumerate(independent_dimensions):
+                    dims[i].append(dim.get_value(
+                        scans, scan_number, scan_step_index=-1, relative=True))
+
+        return map_config_dict
+
+
+class ScanToMapReader(Reader):
+    """Reader for turning a single SPEC scan into a MapConfig."""
+    def read(self, spec_file, scan_number):
+        """Return a dictionary representing a valid map configuration
+        consisting of the single SPEC scan specified.
+
+        :param spec_file: Name of the SPEC file.
+        :type spec_file: str
+        :param scan_number: Number of the SPEC scan.
+        :type scan_number: int
+        :returns: Map configuration dictionary
+        :rtype: dict
+        """
+        # Local modules
+        from chess_scanparsers import SMBMCAScanParser as ScanParser
+
+        scanparser = ScanParser(spec_file, scan_number)
+
+        if scanparser.spec_macro in ('tseries', 'loopscan') or \
+           (scanparser.spec_macro == 'flyscan' and \
+            not len(scanparser.spec_args) ==5):
+            independent_dimensions = [
+                {'label': 'Time',
+                 'units': 'seconds',
+                 'data_type': 'scan_column',
+                 'name': 'Time'}]
+        else:
+            independent_dimensions = [
+                {'label': mne,
+                 'units': 'unknown units',
+                 'data_type': 'spec_motor',
+                 'name': mne}
+                for mne in scanparser.spec_scan_motor_mnes]
+
+        map_config_dict = dict(
+            title=f'{scanparser.scan_name}_{scan_number:03d}',
+            station='id1a3',
+            experiment_type='EDD',
+            sample=dict(name=scanparser.scan_name),
+            spec_scans=[
+                dict(spec_file=spec_file, scan_numbers=[scan_number])],
+            independent_dimensions=independent_dimensions,
+            presample_intensity=dict(name='a3ic1', data_type='scan_column'),
+            postsample_intensity=dict(name='diode', data_type='scan_column'),
+            dwell_time_actual=dict(name='sec', data_type='scan_column')
+        )
+
+        return map_config_dict
+
+
+class SetupNXdataReader(Reader):
+    """Reader for converting the SPEC input .txt file for EDD dataset
+    collection to an approporiate input argument for
+    `CHAP.common.SetupNXdataProcessor`.
+
+    Example of use in a `Pipeline` configuration:
+    ```yaml
+    config:
+      inputdir: /rawdata/samplename
+      outputdir: /reduceddata/samplename
+    pipeline:
+      - edd.SetupNXdataReader:
+          filename: SpecInput.txt
+          dataset_id: 1
+      - common.SetupNXdataProcessor:
+          nxname: samplename_dataset_1
+      - common.NexusWriter:
+          filename: data.nxs
+    ```
+    """
+    def read(self, filename, dataset_id):
+        """Return a dictionary containing the `coords`, `signals`, and
+        `attrs` arguments appropriate for use with
+        `CHAP.common.SetupNXdataProcessor.process` to set up an
+        initial `NXdata` object representing a complete and organized
+        structured EDD dataset.
+
+        :param filename: Name of the input .txt file provided to SPEC
+            for EDD dataset collection.
+        :type filename: str
+        :param dataset_id: Number of the dataset in the .txt file to
+            return
+            `CHAP.common.SetupNXdataProcessor.process`
+            arguments for.
+        :type dataset_id: int
+        :returns: The dataset's coordinate names, values, attributes,
+            and signal names, shapes, and attributes.
+        :rtype: dict
+        """
+        # Columns in input .txt file:
+        # 0: scan number
+        # 1: dataset index
+        # 2: configuration descriptor
+        # 3: labx
+        # 4: laby
+        # 5: labz
+        # 6: omega (reference)
+        # 7: omega (offset)
+        # 8: dwell time
+        # 9: beam width
+        # 10: beam height
+        # 11: detector slit gap width
+        # 12: scan type
+
+        # Following columns used only for scan types 1 and up and
+        # specify flyscan/flymesh parameters.
+        # 13 + 4n: scan direction axis index
+        # 14 + 4n: lower bound
+        # 15 + 4n: upper bound
+        # 16 + 4n: no. points
+        # (For scan types 1, 4: n = 1)
+        # (For scan types 2, 3, 5: n = 1 or 2)
+
+        # For scan type 5 only:
+        # 21: bin axis
+
+        # Parse dataset from the input .txt file.
+        with open(filename) as inf:
+            file_lines = inf.readlines()
+        dataset_lines = []
+        for l in file_lines:
+            vals = l.split()
+            for i, v in enumerate(vals):
+                try:
+                    vals[i] = int(v)
+                except:
+                    try:
+                        vals[i] = float(v)
+                    except:
+                        pass
+            if vals[1] == dataset_id:
+                dataset_lines.append(vals)
+
+        # Start inferring coords and signals lists for EDD experiments
+        self.logger.warning(
+            'Assuming the following parameters are identical across the '
+            + 'entire dataset: scan type, configuration descriptor')
+        scan_type = dataset_lines[0][12]
+        self.logger.debug(f'scan_type = {scan_type}')
+        coords = [
+            dict(name='labx',
+                 values=np.unique([l[3] for l in dataset_lines]),
+                 attrs=dict(
+                     units='mm', local_name='labx', data_type='smb_par')),
+            dict(name='laby',
+                 values=np.unique([l[4] for l in dataset_lines]),
+                 attrs=dict(
+                     units='mm', local_name='laby', data_type='smb_par')),
+            dict(name='labz',
+                 values=np.unique([l[5] for l in dataset_lines]),
+                 attrs=dict(
+                     units='mm', local_name='labz', data_type='smb_par')),
+            dict(name='ometotal',
+                 values=np.unique([l[6] + l[7] for l in dataset_lines]),
+                 attrs=dict(
+                     units='degrees', local_name='ometotal',
+                     data_type='smb_par'))
+        ]
+        signals = [
+            dict(name='presample_intensity', shape=[],
+                 attrs=dict(units='counts', local_name='a3ic1',
+                            data_type='scan_column')),
+            dict(name='postsample_intensity', shape=[],
+                 attrs=dict(units='counts', local_name='diode',
+                            data_type='scan_column')),
+            dict(name='dwell_time_actual', shape=[],
+                 attrs=dict(units='seconds', local_name='sec',
+                            data_type='scan_column')),
+            dict(name='SCAN_N', shape=[],
+                 attrs=dict(units='n/a', local_name='SCAN_N',
+                            data_type='smb_par')),
+            dict(name='rsgap_size', shape=[],
+                 attrs=dict(units='mm', local_name='rsgap_size',
+                            data_type='smb_par')),
+            dict(name='x_effective', shape=[],
+                 attrs=dict(units='mm', local_name='x_effective',
+                            data_type='smb_par')),
+            dict(name='z_effective', shape=[],
+                 attrs=dict(units='mm', local_name='z_effective',
+                            data_type='smb_par')),
+        ]
+        for i in range(23):
+            signals.append(dict(
+                name=str(i), shape=[4096,],
+                attrs=dict(
+                    units='counts', local_name=f'XPS23 element {i}',
+                    eta='unknown')))
+
+        attrs = dict(dataset_id=dataset_id,
+                     config_id=dataset_lines[0][2],
+                     scan_type=scan_type)
+
+        # For potential coordinate axes w/ only one unique value, do
+        # not consider them a coordinate. Make them a signal instead.
+        _coords = []
+        for i, c in enumerate(coords):
+            if len(c['values']) == 1:
+                self.logger.debug(f'Moving {c["name"]} from coords to signals')
+                # signal = coords.pop(i)
+                del c['values']
+                c['shape'] = []
+                signals.append(c)
+            else:
+                _coords.append(c)
+        coords = _coords
+
+        # Append additional coords depending on the scan type of the
+        # dataset. Also find the number of points / scan.
+        if scan_type == 0:
+            scan_npts = 1
+        else:
+            self.logger.warning(
+                'Assuming scan parameters are identical for all scans.')
+            axes_labels = {1: 'scan_labx', 2: 'scan_laby', 3: 'scan_labz',
+                           4: 'scan_ometotal'}
+            axes_units = {1: 'mm', 2: 'mm', 3: 'mm', 4: 'degrees'}
+            coords.append(
+                dict(name=axes_labels[dataset_lines[0][13]],
+                     values=np.round(np.linspace(
+                         dataset_lines[0][14], dataset_lines[0][15],
+                         dataset_lines[0][16]), 3),
+                     attrs=dict(units=axes_units[dataset_lines[0][13]],
+                                relative=True)))
+            scan_npts = len(coords[-1]['values'])
+            if scan_type in (2, 3, 5):
+                coords.append(
+                    dict(name=axes_labels[dataset_lines[0][17]],
+                         values=np.round(np.linspace(
+                             dataset_lines[0][18], dataset_lines[0][19],
+                             dataset_lines[0][20]), 3),
+                         attrs=dict(units=axes_units[dataset_lines[0][17]],
+                                    relative=True)))
+                scan_npts *= len(coords[-1]['values'])
+                if scan_type == 5:
+                    attrs['bin_axis'] = axes_labels[dataset_lines[0][21]]
+
+        # Determine if the datset is structured or unstructured.
+        total_npts = len(dataset_lines) * scan_npts
+        self.logger.debug(f'Total # of points in the dataset: {total_npts}')
+        self.logger.debug(
+            'Determined number of unique coordinate values: '
+            + str({c['name']: len(c['values']) for c in coords}))
+        coords_npts = np.prod([len(c['values']) for c in coords])
+        self.logger.debug(
+            f'If dataset is structured, # of points should be: {coords_npts}')
+        if coords_npts != total_npts:
+            attrs['unstructured_axes'] = []
+            self.logger.warning(
+                'Dataset is unstructured. All coordinates will be treated as '
+                + 'singals, and the dataset will have a single coordinate '
+                + 'instead: data point index.')
+            for c in coords:
+                del c['values']
+                c['shape'] = []
+                signals.append(c)
+                attrs['unstructured_axes'].append(c['name'])
+            coords = [dict(name='dataset_point_index',
+                           values=np.arange(total_npts),
+                           attrs=dict(units='n/a'))]
+        else:
+            signals.append(dict(name='dataset_point_index', shape=[],
+                                attrs=dict(units='n/a')))
+
+        return dict(coords=coords, signals=signals, attrs=attrs)
+
+
+class UpdateNXdataReader(Reader):
+    """Companion to `edd.SetupNXdataReader` and
+    `common.UpdateNXDataProcessor`. Constructs a list of data points
+    to pass as pipeline data to `common.UpdateNXDataProcessor` so that
+    an `NXdata` constructed by `edd.SetupNXdataReader` and
+    `common.SetupNXdataProcessor` can be updated live as individual
+    scans in an EDD dataset are completed.
+
+    Example of use in a `Pipeline` configuration:
+    ```yaml
+    config:
+      inputdir: /rawdata/samplename
+    pipeline:
+      - edd.UpdateNXdataReader:
+          spec_file: spec.log
+          scan_number: 1
+      - common.SetupNXdataProcessor:
+          nxfilename: /reduceddata/samplename/data.nxs
+          nxdata_path: /entry/samplename_dataset_1
+    ```
+    """
+    def read(self, spec_file, scan_number, inputdir='.'):
+        """Return a list of data points containing raw data values for
+        a single EDD spec scan. The returned values can be passed
+        along to `common.UpdateNXdataProcessor` to fill in an existing
+        `NXdata` set up with `common.SetupNXdataProcessor`.
+
+        :param spec_file: Name of the spec file containing the spec
+           scan (a relative or absolute path).
+        :type spec_file: str
+        :param scan_number: Number of the spec scan.
+        :type scan_number: int
+        :param inputdir: Parent directory of `spec_file`, used only if
+            `spec_file` is a relative path. Will be ignored if
+            `spec_file` is an absolute path. Defaults to `'.'`.
+        :type inputdir: str
+        :returs: List of data points appropriate for input to
+            `common.UpdateNXdataProcessor`.
+        :rtype: list[dict[str, object]]
+        """
+        # Local modules
+        from CHAP.utils.parfile import ParFile
+        from chess_scanparsers import SMBMCAScanParser as ScanParser
+
+        if not os.path.isabs(spec_file):
+            spec_file = os.path.join(inputdir, spec_file)
+        scanparser = ScanParser(spec_file, scan_number)
+        self.logger.debug('Parsed scan')
+
+        # A label / counter mne dict for convenience
+        counters = dict(presample_intensity='a3ic0',
+                        postsample_intensity='diode',
+                        dwell_time_actual='sec')
+        # Determine the scan's own coordinate axes based on scan type
+        scan_type = scanparser.pars['scan_type']
+        self.logger.debug(f'scan_type = {scan_type}')
+        if scan_type == 0:
+            scan_axes = []
+        else:
+            axes_labels = {1: 'scan_labx', 2: 'scan_laby', 3: 'scan_labz',
+                           4: 'scan_ometotal'}
+            scan_axes = [axes_labels[scanparser.pars['fly_axis0']]]
+            if scan_type in (2, 3, 5):
+                scan_axes.append(axes_labels[scanparser.pars['fly_axis1']])
+        self.logger.debug(f'Determined scan axes: {scan_axes}')
+
+        # Par file values will be the same for all points in any scan
+        smb_par_values = {}
+        for smb_par in ('labx', 'laby', 'labz', 'ometotal', 'SCAN_N',
+                        'rsgap_size', 'x_effective', 'z_effective'):
+            smb_par_values[smb_par] = scanparser.pars[smb_par]
+
+        # Get offset for the starting index of this scan's points in
+        # the entire dataset.
+        dataset_id = scanparser.pars['dataset_id']
+        parfile = ParFile(scanparser._par_file)
+        good_scans = parfile.good_scan_numbers()
+        n_prior_dataset_scans = sum(
+            [1 if did == dataset_id and scan_n < scan_number else 0 \
+             for did, scan_n in zip(
+                     parfile.get_values(
+                         'dataset_id', scan_numbers=good_scans),
+                     good_scans)])
+        dataset_point_index_offset = n_prior_dataset_scans \
+                                     * scanparser.spec_scan_npts
+        self.logger.debug(
+            f'dataset_point_index_offset = {dataset_point_index_offset}')
+
+        # Get full data point for every point in the scan
+        data_points = []
+        self.logger.info(f'Getting {scanparser.spec_scan_npts} data points')
+        for i in range(scanparser.spec_scan_npts):
+            self.logger.debug(f'Getting data point for scan step index {i}')
+            step = scanparser.get_scan_step(i)
+            data_points.append(dict(
+                dataset_point_index=dataset_point_index_offset + i,
+                **smb_par_values,
+                **{str(_i): scanparser.get_detector_data(_i, i) \
+                   for _i in range(23)},
+                **{c: scanparser.spec_scan_data[counters[c]][i] \
+                   for c in counters},
+                **{a: round(
+                    scanparser.spec_scan_motor_vals_relative[_i][step[_i]], 3)\
+                   for _i, a in enumerate(scan_axes)}))
+
+        return data_points
+
+
+class NXdataSliceReader(Reader):
+    """Reader for returning a sliced verison of an `NXdata` (which
+    represents a full EDD dataset) that contains data from just a
+    single SPEC scan.
+
+    Example of use in a `Pipeline` configuration:
+    ```yaml
+    config:
+      inputdir: /rawdata/samplename
+      outputdir: /reduceddata/samplename
+    pipeline:
+      - edd.NXdataSliceReader:
+          filename: /reduceddata/samplename/data.nxs
+          nxpath: /path/to/nxdata
+          spec_file: spec.log
+          scan_number: 1
+      - common.NexusWriter:
+          filename: scan_1.nxs
+    ```
+    """
+    def read(self, filename, nxpath, spec_file, scan_number, inputdir='.'):
+        """Return a "slice" of an EDD dataset's NXdata that represents
+        just the data from one scan in the dataset.
+
+        :param filename: Name of the NeXus file in which the
+            existing full EDD dataset's NXdata resides.
+        :type filename: str
+        :param nxpath: Path to the existing full EDD dataset's NXdata
+            group in `filename`.
+        :type nxpath: str
+        :param spec_file: Name of the spec file containing whose data
+            will be the only contents of the returned `NXdata`.
+        :type spec_file: str
+        :param scan_number: Number of the spec scan whose data will be
+            the only contents of the returned `NXdata`.
+        :type scan_number: int
+        :param inputdir: Directory containing `filename` and/or
+            `spec_file`, if either one / both of them are not absolute
+            paths. Defaults to `'.'`.
+        :type inputdir: str, optional
+        :returns: An `NXdata` similar to the one at `nxpath` in
+            `filename`, but containing only the data collected by the
+            specified spec scan.
+        :rtype: nexusformat.nexus.NXdata
+        """
+        # Third party modules
+        from nexusformat.nexus import nxload
+
+        # Local modules
+        from CHAP.common import NXdataReader
+        from CHAP.utils.parfile import ParFile
+        from chess_scanparsers import SMBMCAScanParser as ScanParser
+
+        # Parse existing NXdata
+        root = nxload(filename)
+        nxdata = root[nxpath]
+        if nxdata.nxclass != 'NXdata':
+            raise TypeError(
+                f'Object at {nxpath} in {filename} is not an NXdata')
+        self.logger.debug('Loaded existing NXdata')
+
+        # Parse scan
+        if not os.path.isabs(spec_file):
+            spec_file = os.path.join(inputdir, spec_file)
+        scanparser = ScanParser(spec_file, scan_number)
+        self.logger.debug('Parsed scan')
+
+        # Assemble arguments for NXdataReader
+        name = f'{nxdata.nxname}_scan_{scan_number}'
+        axes_names = [a.nxname for a in nxdata.nxaxes]
+        if nxdata.nxsignal is not None:
+            signal_name = nxdata.nxsignal.nxname
+        else:
+            signal_name = list(nxdata.entries.keys())[0]
+        attrs = nxdata.attrs
+        nxfield_params = []
+        if 'dataset_point_index' in nxdata:
+            # Get offset for the starting index of this scan's points in
+            # the entire dataset.
+            dataset_id = scanparser.pars['dataset_id']
+            parfile = ParFile(scanparser._par_file)
+            good_scans = parfile.good_scan_numbers()
+            n_prior_dataset_scans = sum(
+                [1 if did == dataset_id and scan_n < scan_number else 0 \
+                 for did, scan_n in zip(
+                         parfile.get_values(
+                             'dataset_id', scan_numbers=good_scans),
+                         good_scans)])
+            dataset_point_index_offset = n_prior_dataset_scans \
+                                         * scanparser.spec_scan_npts
+            self.logger.debug(
+                f'dataset_point_index_offset = {dataset_point_index_offset}')
+            slice_params = dict(
+                start=dataset_point_index_offset,
+                end=dataset_point_index_offset + scanparser.spec_scan_npts + 1)
+            nxfield_params = [dict(filename=filename,
+                                   nxpath=entry.nxpath,
+                                   slice_params=[slice_params]) \
+                              for entry in nxdata]
+        else:
+            signal_slice_params = []
+            for a in nxdata.nxaxes:
+                if a.nxname.startswith('scan_'):
+                    slice_params = {}
+                else:
+                    value = scanparser.pars[a.nxname]
+                    try:
+                        index = np.where(a.nxdata == value)[0][0]
+                    except:
+                        index = np.argmin(np.abs(a.nxdata - value))
+                        self.logger.warning(
+                            f'Nearest match for coordinate value {a.nxname}: '
+                            f'{a.nxdata[index]} (actual value: {value})')
+                    slice_params = dict(start=index, end=index+1)
+                signal_slice_params.append(slice_params)
+                nxfield_params.append(dict(
+                    filename=filename,
+                    nxpath=os.path.join(nxdata.nxpath, a.nxname),
+                    slice_params=[slice_params]))
+            for name, entry in nxdata.entries.items():
+                if entry in nxdata.nxaxes:
+                    continue
+                nxfield_params.append(dict(
+                    filename=filename, nxpath=entry.nxpath,
+                    slice_params=signal_slice_params))
+
+        # Return the "sliced" NXdata
+        reader = NXdataReader()
+        reader.logger = self.logger
+        return reader.read(name=nxdata.nxname, nxfield_params=nxfield_params,
+                           signal_name=signal_name, axes_names=axes_names,
+                           attrs=attrs)
+
+
 if __name__ == '__main__':
     from CHAP.reader import main
     main()