Merge pull request #94 from BEL-Public/release-0.7.0

.gitignore

@@ -46,6 +46,8 @@ coverage.xml
 *.cover
 .hypothesis/
 .pytest_cache/
+test.xml
+examples/example_1.mfz
 
 # Translations
 *.mo

CHANGELOG.md

@@ -6,6 +6,14 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+## [0.7.0] - 2022-01-24
+### Added
+- Ability to read bad channels from infoN.xml files.
+- script `mffdiff.py` to compare two MFF files
+
+### Changed
+- cache `Reader._blobs`
+
 ## [0.6.3] - 2021-05-17
 ### Added
 - dependency Deprecated
@@ -55,7 +63,8 @@ In `Reader.get_physical_samples_from_epoch()`:
 - Parse key elements in categories.xml files with `mffpy.xml_files.Categories` class.
 - Incorporate `cached_property` dependency into `mffpy` library.
 
-[Unreleased]: https://github.com/bel-public/mffpy/compare/v0.6.3...HEAD
+[Unreleased]: https://github.com/bel-public/mffpy/compare/v0.7.0...HEAD
+[0.7.0]: https://github.com/bel-public/mffpy/compare/v0.6.3...v0.7.0
 [0.6.3]: https://github.com/bel-public/mffpy/compare/v0.6.2...v0.6.3
 [0.6.2]: https://github.com/bel-public/mffpy/compare/v0.6.1...v0.6.2
 [0.6.1]: https://github.com/bel-public/mffpy/compare/v0.6.0...v0.6.1

bin/mffdiff.py

@@ -0,0 +1,111 @@
+#!/usr/bin/env python
+"""
+mffdiff.py
+
+Compare the content of two MFF files
+"""
+from os.path import join
+from subprocess import check_output
+from mffpy import Reader
+from argparse import ArgumentParser
+
+char_limit = 50
+
+parser = ArgumentParser(description="""
+compare the contents of two MFF files
+""")
+parser.add_argument('leftpath', type=str, help="MFF file to compare")
+parser.add_argument('rightpath', type=str, help="second MFF file to compare")
+args = parser.parse_args()
+
+left_mff = Reader(args.leftpath)
+right_mff = Reader(args.rightpath)
+
+
+def getnested(instance, props, callback=None):
+    """get nested property of instance
+
+    This is a recursive function to access a nested property, e.g. for
+    `instance.a.b.c`, props is `['a', 'b', 'c']`.  `callback` is optionally
+    applied to the final value.
+
+    Parameters
+    ----------
+        instance:
+            instance that has a property `props[0]`
+        props:
+            nested property split into a list
+        callback:
+            is optionally applied to the output value
+    """
+    if props:
+        val = getattr(instance, props[0])
+        return getnested(val, props[1:], callback)
+
+    return callback(instance) if callback else instance
+
+
+def compare_raw(filename):
+    """compare file in MFF using `$ diff`"""
+    leftfile = join(args.leftpath, filename)
+    rightfile = join(args.rightpath, filename)
+    try:
+        check_output(['diff', leftfile, rightfile])
+        status = 'match'
+    except BaseException:
+        status = 'diff'
+
+    print(f">>> {status} @ file '{filename}'")
+
+
+def compare(prop: str, callback=None, info: str = ''):
+    """compare property between left_mff and right_mff
+
+    Parameters
+    ----------
+        prop:
+            string specifying a nested property of `mffpy.Reader` instance
+        callback:
+            post processing of the value of `prop`
+        info:
+            additional note of the nested property
+    """
+    props = prop.split('.')
+    try:
+        left_value = getnested(left_mff, props, callback)
+        right_value = getnested(right_mff, props, callback)
+        left_msg = str(left_value)[:char_limit]
+        right_msg = str(right_value)[:char_limit]
+        if left_value != right_value:
+            msg = f'\t{left_msg} != {right_msg}'
+            status = 'diff'
+        else:
+            msg = ''
+            status = 'match'
+
+    except BaseException as err:
+        status = 'error'
+        msg = str(err)[:char_limit] + '\n'
+        left_value = None
+
+    print(f">>> {status} @ reader_instance.{prop} "
+          f"{': ' + info if info else ''}")
+    if msg:
+        print(msg)
+
+    return left_value if status == 'match' else None
+
+
+simple_props = ['mff_flavor', 'sampling_rates', 'durations', 'startdatetime',
+                'units', 'num_channels', 'categories.categories']
+
+print(f"Comparing {args.leftpath} with {args.rightpath}")
+for prop in simple_props:
+    compare(prop)
+
+epoch_count = compare('epochs.epochs', lambda e: len(e), 'len(..)')
+if epoch_count is not None:
+    for i in range(epoch_count):
+        compare('epochs.epochs', lambda e: e[i].content, f"[{i}].content")
+
+out = compare_raw('signal1.bin')

examples/example_1.mff/info1.xml

@@ -9,6 +9,7 @@
             </EEG>
         </fileDataType>
     </generalInformation>
+    <channels exclusion="badChannels">31 47 65 79</channels>
     <filters>
         <filter>
             <beginTime>0</beginTime>

examples/example_2.json

@@ -12,6 +12,7 @@
             "sensorLayoutName": "HydroCel GSN 256 1.0",
             "montageName": "HydroCel GSN 256 1.0"
         },
+        "channels": [],
         "filters": [],
         "calibrations": {}
     },

makefile

@@ -3,13 +3,9 @@ examples/example_1.mfz: examples/example_1.mff
 	# zip -Z store -r -j ./examples/example_1.mfz ./examples/example_1.mff
 	-python ./bin/mff2mfz.py ./examples/example_1.mff
 
-
-# some tests depend on the existence of a zipped version of
-# 'examples/example_1.mff/'
-test: examples/example_1.mfz
+test:
 	mypy --ignore-missing-imports mffpy
 	pytest --cov
 
 clean:
 	-rm examples/example_1.mfz
-	-rm -rf .cache

mffpy/mffdir.py

@@ -130,8 +130,9 @@ def filename(self, basename: str) -> str:
             if basename in files:
                 return join(self._mffname, basename) + ext
         else:
-            raise ValueError(f"No file with basename {basename} \
-                    in directory {super().__str__()}.")
+            raise FileNotFoundError(
+                f"No file with basename {basename} "
+                f"in directory {super().__str__()}.")
 
     def __contains__(self, filename: str) -> bool:
         return exists(filename)

mffpy/reader.py

@@ -121,7 +121,7 @@ def epochs(self) -> Epochs:
         # Attempt to add category names to the `Epoch` objects in `epochs`
         try:
             categories = self.categories
-        except (ValueError, AssertionError):
+        except (FileNotFoundError, AssertionError):
             print('categories.xml not found or of wrong type. '
                   '`Epoch.name` will default to "epoch" for all epochs.')
             return epochs
@@ -208,7 +208,7 @@ def num_channels(self) -> Dict[str, int]:
             for fn, bin_file in self._blobs.items()
         }
 
-    @property
+    @cached_property
     def _blobs(self) -> Dict[str, bin_files.BinFile]:
         """return dictionary of `BinFile` data readers by signal type"""
         __blobs = {}

mffpy/tests/conftest.py

@@ -0,0 +1,32 @@
+"""
+Copyright 2019 Brain Electrophysiology Laboratory Company LLC
+
+Licensed under the ApacheLicense, Version 2.0(the "License");
+you may not use this module except in compliance with the License.
+You may obtain a copy of the License at:
+
+http: // www.apache.org / licenses / LICENSE - 2.0
+
+Unless required by applicable law or agreed to in writing, software
+distributed under the License is distributed on an
+"AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF
+ANY KIND, either express or implied.
+"""
+
+from glob import glob
+import os.path as op
+from zipfile import ZipFile, ZIP_STORED
+
+import pytest
+
+
+@pytest.fixture(scope='session', autouse=True)
+def ensure_mfz():
+    """Ensure that the mfz file exists."""
+    fname = op.join(
+        op.dirname(__file__), '..', '..', 'examples', 'example_1.mfz')
+    if not op.isfile(fname):
+        with ZipFile(fname, mode='w', compression=ZIP_STORED) as zf:
+            for content_filename in glob(op.join(fname[:-3] + 'mff', '*')):
+                arc_filename = op.basename(content_filename)
+                zf.write(content_filename, arcname=arc_filename)

mffpy/tests/test_dict2xml.py

@@ -12,17 +12,19 @@
 "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF
 ANY KIND, either express or implied.
 """
+from os.path import join
+
 from ..dict2xml import dict2xml, TEXT, ATTR
 
 
-def test_dict2xml():
+def test_dict2xml(tmpdir):
     rootname = 'myroot'
     content = {
         'a': {TEXT: '35', ATTR: {'hello': 'world'}},
         'b': [{TEXT: 'b' + str(i+1)} for i in range(2)]
     }
     elem = dict2xml(content, rootname=rootname)
-    elem.write('test.xml')
+    elem.write(join(str(tmpdir), 'test.xml'))
     root = elem.getroot()
     a = root.find('a')
     bs = root.findall('b')